Combine core into this repositoryv2.18.0

tame-core used to be its own repository, but maintaining it separately does
not provide a whole lot of benefit, and further makes it difficult to see
version dependencies.  I also want to centralize documentation within this
project.

57 files changed, 10127 insertions, 0 deletions

diff --git a/core/.gitignore b/core/.gitignore
new file mode 100644
index 0000000..7a51172
--- /dev/null
+++ b/core/.gitignore
@@ -0,0 +1,7 @@
+# when core is built
+*.xmlo
+*.xmle
+*.js
+*.dep
+*.tmp
+*.html
diff --git a/core/COPYING b/core/COPYING
new file mode 100644
index 0000000..94a9ed0
--- /dev/null
+++ b/core/COPYING
@@ -0,0 +1,674 @@
+                    GNU GENERAL PUBLIC LICENSE
+                       Version 3, 29 June 2007
+
+ Copyright (C) 2007 Free Software Foundation, Inc. <http://fsf.org/>
+ Everyone is permitted to copy and distribute verbatim copies
+ of this license document, but changing it is not allowed.
+
+                            Preamble
+
+  The GNU General Public License is a free, copyleft license for
+software and other kinds of works.
+
+  The licenses for most software and other practical works are designed
+to take away your freedom to share and change the works.  By contrast,
+the GNU General Public License is intended to guarantee your freedom to
+share and change all versions of a program--to make sure it remains free
+software for all its users.  We, the Free Software Foundation, use the
+GNU General Public License for most of our software; it applies also to
+any other work released this way by its authors.  You can apply it to
+your programs, too.
+
+  When we speak of free software, we are referring to freedom, not
+price.  Our General Public Licenses are designed to make sure that you
+have the freedom to distribute copies of free software (and charge for
+them if you wish), that you receive source code or can get it if you
+want it, that you can change the software or use pieces of it in new
+free programs, and that you know you can do these things.
+
+  To protect your rights, we need to prevent others from denying you
+these rights or asking you to surrender the rights.  Therefore, you have
+certain responsibilities if you distribute copies of the software, or if
+you modify it: responsibilities to respect the freedom of others.
+
+  For example, if you distribute copies of such a program, whether
+gratis or for a fee, you must pass on to the recipients the same
+freedoms that you received.  You must make sure that they, too, receive
+or can get the source code.  And you must show them these terms so they
+know their rights.
+
+  Developers that use the GNU GPL protect your rights with two steps:
+(1) assert copyright on the software, and (2) offer you this License
+giving you legal permission to copy, distribute and/or modify it.
+
+  For the developers' and authors' protection, the GPL clearly explains
+that there is no warranty for this free software.  For both users' and
+authors' sake, the GPL requires that modified versions be marked as
+changed, so that their problems will not be attributed erroneously to
+authors of previous versions.
+
+  Some devices are designed to deny users access to install or run
+modified versions of the software inside them, although the manufacturer
+can do so.  This is fundamentally incompatible with the aim of
+protecting users' freedom to change the software.  The systematic
+pattern of such abuse occurs in the area of products for individuals to
+use, which is precisely where it is most unacceptable.  Therefore, we
+have designed this version of the GPL to prohibit the practice for those
+products.  If such problems arise substantially in other domains, we
+stand ready to extend this provision to those domains in future versions
+of the GPL, as needed to protect the freedom of users.
+
+  Finally, every program is threatened constantly by software patents.
+States should not allow patents to restrict development and use of
+software on general-purpose computers, but in those that do, we wish to
+avoid the special danger that patents applied to a free program could
+make it effectively proprietary.  To prevent this, the GPL assures that
+patents cannot be used to render the program non-free.
+
+  The precise terms and conditions for copying, distribution and
+modification follow.
+
+                       TERMS AND CONDITIONS
+
+  0. Definitions.
+
+  "This License" refers to version 3 of the GNU General Public License.
+
+  "Copyright" also means copyright-like laws that apply to other kinds of
+works, such as semiconductor masks.
+
+  "The Program" refers to any copyrightable work licensed under this
+License.  Each licensee is addressed as "you".  "Licensees" and
+"recipients" may be individuals or organizations.
+
+  To "modify" a work means to copy from or adapt all or part of the work
+in a fashion requiring copyright permission, other than the making of an
+exact copy.  The resulting work is called a "modified version" of the
+earlier work or a work "based on" the earlier work.
+
+  A "covered work" means either the unmodified Program or a work based
+on the Program.
+
+  To "propagate" a work means to do anything with it that, without
+permission, would make you directly or secondarily liable for
+infringement under applicable copyright law, except executing it on a
+computer or modifying a private copy.  Propagation includes copying,
+distribution (with or without modification), making available to the
+public, and in some countries other activities as well.
+
+  To "convey" a work means any kind of propagation that enables other
+parties to make or receive copies.  Mere interaction with a user through
+a computer network, with no transfer of a copy, is not conveying.
+
+  An interactive user interface displays "Appropriate Legal Notices"
+to the extent that it includes a convenient and prominently visible
+feature that (1) displays an appropriate copyright notice, and (2)
+tells the user that there is no warranty for the work (except to the
+extent that warranties are provided), that licensees may convey the
+work under this License, and how to view a copy of this License.  If
+the interface presents a list of user commands or options, such as a
+menu, a prominent item in the list meets this criterion.
+
+  1. Source Code.
+
+  The "source code" for a work means the preferred form of the work
+for making modifications to it.  "Object code" means any non-source
+form of a work.
+
+  A "Standard Interface" means an interface that either is an official
+standard defined by a recognized standards body, or, in the case of
+interfaces specified for a particular programming language, one that
+is widely used among developers working in that language.
+
+  The "System Libraries" of an executable work include anything, other
+than the work as a whole, that (a) is included in the normal form of
+packaging a Major Component, but which is not part of that Major
+Component, and (b) serves only to enable use of the work with that
+Major Component, or to implement a Standard Interface for which an
+implementation is available to the public in source code form.  A
+"Major Component", in this context, means a major essential component
+(kernel, window system, and so on) of the specific operating system
+(if any) on which the executable work runs, or a compiler used to
+produce the work, or an object code interpreter used to run it.
+
+  The "Corresponding Source" for a work in object code form means all
+the source code needed to generate, install, and (for an executable
+work) run the object code and to modify the work, including scripts to
+control those activities.  However, it does not include the work's
+System Libraries, or general-purpose tools or generally available free
+programs which are used unmodified in performing those activities but
+which are not part of the work.  For example, Corresponding Source
+includes interface definition files associated with source files for
+the work, and the source code for shared libraries and dynamically
+linked subprograms that the work is specifically designed to require,
+such as by intimate data communication or control flow between those
+subprograms and other parts of the work.
+
+  The Corresponding Source need not include anything that users
+can regenerate automatically from other parts of the Corresponding
+Source.
+
+  The Corresponding Source for a work in source code form is that
+same work.
+
+  2. Basic Permissions.
+
+  All rights granted under this License are granted for the term of
+copyright on the Program, and are irrevocable provided the stated
+conditions are met.  This License explicitly affirms your unlimited
+permission to run the unmodified Program.  The output from running a
+covered work is covered by this License only if the output, given its
+content, constitutes a covered work.  This License acknowledges your
+rights of fair use or other equivalent, as provided by copyright law.
+
+  You may make, run and propagate covered works that you do not
+convey, without conditions so long as your license otherwise remains
+in force.  You may convey covered works to others for the sole purpose
+of having them make modifications exclusively for you, or provide you
+with facilities for running those works, provided that you comply with
+the terms of this License in conveying all material for which you do
+not control copyright.  Those thus making or running the covered works
+for you must do so exclusively on your behalf, under your direction
+and control, on terms that prohibit them from making any copies of
+your copyrighted material outside their relationship with you.
+
+  Conveying under any other circumstances is permitted solely under
+the conditions stated below.  Sublicensing is not allowed; section 10
+makes it unnecessary.
+
+  3. Protecting Users' Legal Rights From Anti-Circumvention Law.
+
+  No covered work shall be deemed part of an effective technological
+measure under any applicable law fulfilling obligations under article
+11 of the WIPO copyright treaty adopted on 20 December 1996, or
+similar laws prohibiting or restricting circumvention of such
+measures.
+
+  When you convey a covered work, you waive any legal power to forbid
+circumvention of technological measures to the extent such circumvention
+is effected by exercising rights under this License with respect to
+the covered work, and you disclaim any intention to limit operation or
+modification of the work as a means of enforcing, against the work's
+users, your or third parties' legal rights to forbid circumvention of
+technological measures.
+
+  4. Conveying Verbatim Copies.
+
+  You may convey verbatim copies of the Program's source code as you
+receive it, in any medium, provided that you conspicuously and
+appropriately publish on each copy an appropriate copyright notice;
+keep intact all notices stating that this License and any
+non-permissive terms added in accord with section 7 apply to the code;
+keep intact all notices of the absence of any warranty; and give all
+recipients a copy of this License along with the Program.
+
+  You may charge any price or no price for each copy that you convey,
+and you may offer support or warranty protection for a fee.
+
+  5. Conveying Modified Source Versions.
+
+  You may convey a work based on the Program, or the modifications to
+produce it from the Program, in the form of source code under the
+terms of section 4, provided that you also meet all of these conditions:
+
+    a) The work must carry prominent notices stating that you modified
+    it, and giving a relevant date.
+
+    b) The work must carry prominent notices stating that it is
+    released under this License and any conditions added under section
+    7.  This requirement modifies the requirement in section 4 to
+    "keep intact all notices".
+
+    c) You must license the entire work, as a whole, under this
+    License to anyone who comes into possession of a copy.  This
+    License will therefore apply, along with any applicable section 7
+    additional terms, to the whole of the work, and all its parts,
+    regardless of how they are packaged.  This License gives no
+    permission to license the work in any other way, but it does not
+    invalidate such permission if you have separately received it.
+
+    d) If the work has interactive user interfaces, each must display
+    Appropriate Legal Notices; however, if the Program has interactive
+    interfaces that do not display Appropriate Legal Notices, your
+    work need not make them do so.
+
+  A compilation of a covered work with other separate and independent
+works, which are not by their nature extensions of the covered work,
+and which are not combined with it such as to form a larger program,
+in or on a volume of a storage or distribution medium, is called an
+"aggregate" if the compilation and its resulting copyright are not
+used to limit the access or legal rights of the compilation's users
+beyond what the individual works permit.  Inclusion of a covered work
+in an aggregate does not cause this License to apply to the other
+parts of the aggregate.
+
+  6. Conveying Non-Source Forms.
+
+  You may convey a covered work in object code form under the terms
+of sections 4 and 5, provided that you also convey the
+machine-readable Corresponding Source under the terms of this License,
+in one of these ways:
+
+    a) Convey the object code in, or embodied in, a physical product
+    (including a physical distribution medium), accompanied by the
+    Corresponding Source fixed on a durable physical medium
+    customarily used for software interchange.
+
+    b) Convey the object code in, or embodied in, a physical product
+    (including a physical distribution medium), accompanied by a
+    written offer, valid for at least three years and valid for as
+    long as you offer spare parts or customer support for that product
+    model, to give anyone who possesses the object code either (1) a
+    copy of the Corresponding Source for all the software in the
+    product that is covered by this License, on a durable physical
+    medium customarily used for software interchange, for a price no
+    more than your reasonable cost of physically performing this
+    conveying of source, or (2) access to copy the
+    Corresponding Source from a network server at no charge.
+
+    c) Convey individual copies of the object code with a copy of the
+    written offer to provide the Corresponding Source.  This
+    alternative is allowed only occasionally and noncommercially, and
+    only if you received the object code with such an offer, in accord
+    with subsection 6b.
+
+    d) Convey the object code by offering access from a designated
+    place (gratis or for a charge), and offer equivalent access to the
+    Corresponding Source in the same way through the same place at no
+    further charge.  You need not require recipients to copy the
+    Corresponding Source along with the object code.  If the place to
+    copy the object code is a network server, the Corresponding Source
+    may be on a different server (operated by you or a third party)
+    that supports equivalent copying facilities, provided you maintain
+    clear directions next to the object code saying where to find the
+    Corresponding Source.  Regardless of what server hosts the
+    Corresponding Source, you remain obligated to ensure that it is
+    available for as long as needed to satisfy these requirements.
+
+    e) Convey the object code using peer-to-peer transmission, provided
+    you inform other peers where the object code and Corresponding
+    Source of the work are being offered to the general public at no
+    charge under subsection 6d.
+
+  A separable portion of the object code, whose source code is excluded
+from the Corresponding Source as a System Library, need not be
+included in conveying the object code work.
+
+  A "User Product" is either (1) a "consumer product", which means any
+tangible personal property which is normally used for personal, family,
+or household purposes, or (2) anything designed or sold for incorporation
+into a dwelling.  In determining whether a product is a consumer product,
+doubtful cases shall be resolved in favor of coverage.  For a particular
+product received by a particular user, "normally used" refers to a
+typical or common use of that class of product, regardless of the status
+of the particular user or of the way in which the particular user
+actually uses, or expects or is expected to use, the product.  A product
+is a consumer product regardless of whether the product has substantial
+commercial, industrial or non-consumer uses, unless such uses represent
+the only significant mode of use of the product.
+
+  "Installation Information" for a User Product means any methods,
+procedures, authorization keys, or other information required to install
+and execute modified versions of a covered work in that User Product from
+a modified version of its Corresponding Source.  The information must
+suffice to ensure that the continued functioning of the modified object
+code is in no case prevented or interfered with solely because
+modification has been made.
+
+  If you convey an object code work under this section in, or with, or
+specifically for use in, a User Product, and the conveying occurs as
+part of a transaction in which the right of possession and use of the
+User Product is transferred to the recipient in perpetuity or for a
+fixed term (regardless of how the transaction is characterized), the
+Corresponding Source conveyed under this section must be accompanied
+by the Installation Information.  But this requirement does not apply
+if neither you nor any third party retains the ability to install
+modified object code on the User Product (for example, the work has
+been installed in ROM).
+
+  The requirement to provide Installation Information does not include a
+requirement to continue to provide support service, warranty, or updates
+for a work that has been modified or installed by the recipient, or for
+the User Product in which it has been modified or installed.  Access to a
+network may be denied when the modification itself materially and
+adversely affects the operation of the network or violates the rules and
+protocols for communication across the network.
+
+  Corresponding Source conveyed, and Installation Information provided,
+in accord with this section must be in a format that is publicly
+documented (and with an implementation available to the public in
+source code form), and must require no special password or key for
+unpacking, reading or copying.
+
+  7. Additional Terms.
+
+  "Additional permissions" are terms that supplement the terms of this
+License by making exceptions from one or more of its conditions.
+Additional permissions that are applicable to the entire Program shall
+be treated as though they were included in this License, to the extent
+that they are valid under applicable law.  If additional permissions
+apply only to part of the Program, that part may be used separately
+under those permissions, but the entire Program remains governed by
+this License without regard to the additional permissions.
+
+  When you convey a copy of a covered work, you may at your option
+remove any additional permissions from that copy, or from any part of
+it.  (Additional permissions may be written to require their own
+removal in certain cases when you modify the work.)  You may place
+additional permissions on material, added by you to a covered work,
+for which you have or can give appropriate copyright permission.
+
+  Notwithstanding any other provision of this License, for material you
+add to a covered work, you may (if authorized by the copyright holders of
+that material) supplement the terms of this License with terms:
+
+    a) Disclaiming warranty or limiting liability differently from the
+    terms of sections 15 and 16 of this License; or
+
+    b) Requiring preservation of specified reasonable legal notices or
+    author attributions in that material or in the Appropriate Legal
+    Notices displayed by works containing it; or
+
+    c) Prohibiting misrepresentation of the origin of that material, or
+    requiring that modified versions of such material be marked in
+    reasonable ways as different from the original version; or
+
+    d) Limiting the use for publicity purposes of names of licensors or
+    authors of the material; or
+
+    e) Declining to grant rights under trademark law for use of some
+    trade names, trademarks, or service marks; or
+
+    f) Requiring indemnification of licensors and authors of that
+    material by anyone who conveys the material (or modified versions of
+    it) with contractual assumptions of liability to the recipient, for
+    any liability that these contractual assumptions directly impose on
+    those licensors and authors.
+
+  All other non-permissive additional terms are considered "further
+restrictions" within the meaning of section 10.  If the Program as you
+received it, or any part of it, contains a notice stating that it is
+governed by this License along with a term that is a further
+restriction, you may remove that term.  If a license document contains
+a further restriction but permits relicensing or conveying under this
+License, you may add to a covered work material governed by the terms
+of that license document, provided that the further restriction does
+not survive such relicensing or conveying.
+
+  If you add terms to a covered work in accord with this section, you
+must place, in the relevant source files, a statement of the
+additional terms that apply to those files, or a notice indicating
+where to find the applicable terms.
+
+  Additional terms, permissive or non-permissive, may be stated in the
+form of a separately written license, or stated as exceptions;
+the above requirements apply either way.
+
+  8. Termination.
+
+  You may not propagate or modify a covered work except as expressly
+provided under this License.  Any attempt otherwise to propagate or
+modify it is void, and will automatically terminate your rights under
+this License (including any patent licenses granted under the third
+paragraph of section 11).
+
+  However, if you cease all violation of this License, then your
+license from a particular copyright holder is reinstated (a)
+provisionally, unless and until the copyright holder explicitly and
+finally terminates your license, and (b) permanently, if the copyright
+holder fails to notify you of the violation by some reasonable means
+prior to 60 days after the cessation.
+
+  Moreover, your license from a particular copyright holder is
+reinstated permanently if the copyright holder notifies you of the
+violation by some reasonable means, this is the first time you have
+received notice of violation of this License (for any work) from that
+copyright holder, and you cure the violation prior to 30 days after
+your receipt of the notice.
+
+  Termination of your rights under this section does not terminate the
+licenses of parties who have received copies or rights from you under
+this License.  If your rights have been terminated and not permanently
+reinstated, you do not qualify to receive new licenses for the same
+material under section 10.
+
+  9. Acceptance Not Required for Having Copies.
+
+  You are not required to accept this License in order to receive or
+run a copy of the Program.  Ancillary propagation of a covered work
+occurring solely as a consequence of using peer-to-peer transmission
+to receive a copy likewise does not require acceptance.  However,
+nothing other than this License grants you permission to propagate or
+modify any covered work.  These actions infringe copyright if you do
+not accept this License.  Therefore, by modifying or propagating a
+covered work, you indicate your acceptance of this License to do so.
+
+  10. Automatic Licensing of Downstream Recipients.
+
+  Each time you convey a covered work, the recipient automatically
+receives a license from the original licensors, to run, modify and
+propagate that work, subject to this License.  You are not responsible
+for enforcing compliance by third parties with this License.
+
+  An "entity transaction" is a transaction transferring control of an
+organization, or substantially all assets of one, or subdividing an
+organization, or merging organizations.  If propagation of a covered
+work results from an entity transaction, each party to that
+transaction who receives a copy of the work also receives whatever
+licenses to the work the party's predecessor in interest had or could
+give under the previous paragraph, plus a right to possession of the
+Corresponding Source of the work from the predecessor in interest, if
+the predecessor has it or can get it with reasonable efforts.
+
+  You may not impose any further restrictions on the exercise of the
+rights granted or affirmed under this License.  For example, you may
+not impose a license fee, royalty, or other charge for exercise of
+rights granted under this License, and you may not initiate litigation
+(including a cross-claim or counterclaim in a lawsuit) alleging that
+any patent claim is infringed by making, using, selling, offering for
+sale, or importing the Program or any portion of it.
+
+  11. Patents.
+
+  A "contributor" is a copyright holder who authorizes use under this
+License of the Program or a work on which the Program is based.  The
+work thus licensed is called the contributor's "contributor version".
+
+  A contributor's "essential patent claims" are all patent claims
+owned or controlled by the contributor, whether already acquired or
+hereafter acquired, that would be infringed by some manner, permitted
+by this License, of making, using, or selling its contributor version,
+but do not include claims that would be infringed only as a
+consequence of further modification of the contributor version.  For
+purposes of this definition, "control" includes the right to grant
+patent sublicenses in a manner consistent with the requirements of
+this License.
+
+  Each contributor grants you a non-exclusive, worldwide, royalty-free
+patent license under the contributor's essential patent claims, to
+make, use, sell, offer for sale, import and otherwise run, modify and
+propagate the contents of its contributor version.
+
+  In the following three paragraphs, a "patent license" is any express
+agreement or commitment, however denominated, not to enforce a patent
+(such as an express permission to practice a patent or covenant not to
+sue for patent infringement).  To "grant" such a patent license to a
+party means to make such an agreement or commitment not to enforce a
+patent against the party.
+
+  If you convey a covered work, knowingly relying on a patent license,
+and the Corresponding Source of the work is not available for anyone
+to copy, free of charge and under the terms of this License, through a
+publicly available network server or other readily accessible means,
+then you must either (1) cause the Corresponding Source to be so
+available, or (2) arrange to deprive yourself of the benefit of the
+patent license for this particular work, or (3) arrange, in a manner
+consistent with the requirements of this License, to extend the patent
+license to downstream recipients.  "Knowingly relying" means you have
+actual knowledge that, but for the patent license, your conveying the
+covered work in a country, or your recipient's use of the covered work
+in a country, would infringe one or more identifiable patents in that
+country that you have reason to believe are valid.
+
+  If, pursuant to or in connection with a single transaction or
+arrangement, you convey, or propagate by procuring conveyance of, a
+covered work, and grant a patent license to some of the parties
+receiving the covered work authorizing them to use, propagate, modify
+or convey a specific copy of the covered work, then the patent license
+you grant is automatically extended to all recipients of the covered
+work and works based on it.
+
+  A patent license is "discriminatory" if it does not include within
+the scope of its coverage, prohibits the exercise of, or is
+conditioned on the non-exercise of one or more of the rights that are
+specifically granted under this License.  You may not convey a covered
+work if you are a party to an arrangement with a third party that is
+in the business of distributing software, under which you make payment
+to the third party based on the extent of your activity of conveying
+the work, and under which the third party grants, to any of the
+parties who would receive the covered work from you, a discriminatory
+patent license (a) in connection with copies of the covered work
+conveyed by you (or copies made from those copies), or (b) primarily
+for and in connection with specific products or compilations that
+contain the covered work, unless you entered into that arrangement,
+or that patent license was granted, prior to 28 March 2007.
+
+  Nothing in this License shall be construed as excluding or limiting
+any implied license or other defenses to infringement that may
+otherwise be available to you under applicable patent law.
+
+  12. No Surrender of Others' Freedom.
+
+  If conditions are imposed on you (whether by court order, agreement or
+otherwise) that contradict the conditions of this License, they do not
+excuse you from the conditions of this License.  If you cannot convey a
+covered work so as to satisfy simultaneously your obligations under this
+License and any other pertinent obligations, then as a consequence you may
+not convey it at all.  For example, if you agree to terms that obligate you
+to collect a royalty for further conveying from those to whom you convey
+the Program, the only way you could satisfy both those terms and this
+License would be to refrain entirely from conveying the Program.
+
+  13. Use with the GNU Affero General Public License.
+
+  Notwithstanding any other provision of this License, you have
+permission to link or combine any covered work with a work licensed
+under version 3 of the GNU Affero General Public License into a single
+combined work, and to convey the resulting work.  The terms of this
+License will continue to apply to the part which is the covered work,
+but the special requirements of the GNU Affero General Public License,
+section 13, concerning interaction through a network will apply to the
+combination as such.
+
+  14. Revised Versions of this License.
+
+  The Free Software Foundation may publish revised and/or new versions of
+the GNU General Public License from time to time.  Such new versions will
+be similar in spirit to the present version, but may differ in detail to
+address new problems or concerns.
+
+  Each version is given a distinguishing version number.  If the
+Program specifies that a certain numbered version of the GNU General
+Public License "or any later version" applies to it, you have the
+option of following the terms and conditions either of that numbered
+version or of any later version published by the Free Software
+Foundation.  If the Program does not specify a version number of the
+GNU General Public License, you may choose any version ever published
+by the Free Software Foundation.
+
+  If the Program specifies that a proxy can decide which future
+versions of the GNU General Public License can be used, that proxy's
+public statement of acceptance of a version permanently authorizes you
+to choose that version for the Program.
+
+  Later license versions may give you additional or different
+permissions.  However, no additional obligations are imposed on any
+author or copyright holder as a result of your choosing to follow a
+later version.
+
+  15. Disclaimer of Warranty.
+
+  THERE IS NO WARRANTY FOR THE PROGRAM, TO THE EXTENT PERMITTED BY
+APPLICABLE LAW.  EXCEPT WHEN OTHERWISE STATED IN WRITING THE COPYRIGHT
+HOLDERS AND/OR OTHER PARTIES PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY
+OF ANY KIND, EITHER EXPRESSED OR IMPLIED, INCLUDING, BUT NOT LIMITED TO,
+THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR
+PURPOSE.  THE ENTIRE RISK AS TO THE QUALITY AND PERFORMANCE OF THE PROGRAM
+IS WITH YOU.  SHOULD THE PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF
+ALL NECESSARY SERVICING, REPAIR OR CORRECTION.
+
+  16. Limitation of Liability.
+
+  IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
+WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MODIFIES AND/OR CONVEYS
+THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES, INCLUDING ANY
+GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING OUT OF THE
+USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED TO LOSS OF
+DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY YOU OR THIRD
+PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER PROGRAMS),
+EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE POSSIBILITY OF
+SUCH DAMAGES.
+
+  17. Interpretation of Sections 15 and 16.
+
+  If the disclaimer of warranty and limitation of liability provided
+above cannot be given local legal effect according to their terms,
+reviewing courts shall apply local law that most closely approximates
+an absolute waiver of all civil liability in connection with the
+Program, unless a warranty or assumption of liability accompanies a
+copy of the Program in return for a fee.
+
+                     END OF TERMS AND CONDITIONS
+
+            How to Apply These Terms to Your New Programs
+
+  If you develop a new program, and you want it to be of the greatest
+possible use to the public, the best way to achieve this is to make it
+free software which everyone can redistribute and change under these terms.
+
+  To do so, attach the following notices to the program.  It is safest
+to attach them to the start of each source file to most effectively
+state the exclusion of warranty; and each file should have at least
+the "copyright" line and a pointer to where the full notice is found.
+
+    <one line to give the program's name and a brief idea of what it does.>
+    Copyright (C) <year>  <name of author>
+
+    This program is free software: you can redistribute it and/or modify
+    it under the terms of the GNU General Public License as published by
+    the Free Software Foundation, either version 3 of the License, or
+    (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+Also add information on how to contact you by electronic and paper mail.
+
+  If the program does terminal interaction, make it output a short
+notice like this when it starts in an interactive mode:
+
+    <program>  Copyright (C) <year>  <name of author>
+    This program comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
+    This is free software, and you are welcome to redistribute it
+    under certain conditions; type `show c' for details.
+
+The hypothetical commands `show w' and `show c' should show the appropriate
+parts of the General Public License.  Of course, your program's commands
+might be different; for a GUI interface, you would use an "about box".
+
+  You should also get your employer (if you work as a programmer) or school,
+if any, to sign a "copyright disclaimer" for the program, if necessary.
+For more information on this, and how to apply and follow the GNU GPL, see
+<http://www.gnu.org/licenses/>.
+
+  The GNU General Public License does not permit incorporating your program
+into proprietary programs.  If your program is a subroutine library, you
+may consider it more useful to permit linking proprietary applications with
+the library.  If this is what you want to do, use the GNU Lesser General
+Public License instead of this License.  But first, please read
+<http://www.gnu.org/philosophy/why-not-lgpl.html>.
diff --git a/core/README.md b/core/README.md
new file mode 100644
index 0000000..9a25315
--- /dev/null
+++ b/core/README.md
@@ -0,0 +1,57 @@
+# TAME Core
+Core library for TAME, providing generic abstractions for common
+operations.
+
+This library has accumulated a bit of cruft, is disorganized, and has
+not been substantially refactored to take advantage of new language
+features.  It is a work in progress.
+
+
+## Features
+- BDD abstraction;
+- Classification match manipulation;
+- Common operations on numbers;
+- Conditional evaluation helpers;
+- Core primitive declarations;
+- Interpolation;
+- Interval mapping;
+- Matrix and vector manipulation;
+- Query matrices as data tables;
+- Value mappings; and
+- Other miscellaneous stuff.
+
+
+## What is TAME?
+TAME is The Adaptive Metalanguage, a programming language and system of tools
+designed to aid in the development, understanding, and maintenance of systems
+performing numerous calculations on a complex graph of dependencies,
+conditions, and a large number of inputs.
+
+This system was developed at R-T Specialty Buffalo to handle the complexity of
+comparative insurance rating systems. It is a domain-specific language (DSL)
+that itself encourages, through the use of templates, the creation of sub-DSLs.
+TAME itself is at heart a calculator—processing only numerical input and
+output—driven by quantifiers as predicates. Calculations and quantifiers are
+written declaratively without concern for order of execution.
+
+The system has powerful dependency resolution and data flow capabilities.
+
+TAME consists of a macro processor (implementing a metalanguage), numerous
+compilers for various targets (JavaScript, HTML documentation and debugging
+environment, LaTeX, and others), linkers, and supporting tools.  The input
+grammar is XML, and the majority of the project (including the macro processor,
+compilers, and linkers) are written in XSLT. There is a reason for that odd
+choice; until an explanation is provided, know that someone is perverted enough
+to implement a full compiler stack in XSLT.
+
+
+## License
+This program is free software: you can redistribute it and/or modify it
+under the terms of the GNU General Public License as published by the Free
+Software Foundation, either version 3 of the License, or (at your option)
+any later version.
+
+This program is distributed in the hope that it will be useful, but WITHOUT ANY
+WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS FOR A
+PARTICULAR PURPOSE.  See the GNU General Public License for more details.
+
diff --git a/core/aggregate.xml b/core/aggregate.xml
new file mode 100644
index 0000000..e6f94e2
--- /dev/null
+++ b/core/aggregate.xml
@@ -0,0 +1,154 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="Aggregating Values">
+
+  <import package="base"          export="true" />
+  <import package="vector/cmatch" export="true" />
+
+
+  Aggregate templates simplify aggregating values through various means.
+  Unless otherwise specified,
+    the default means of aggregation is summation.
+
+
+  <section title="Symbol-Based Aggregation">
+    For large numbers of values,
+      the most convenient way to aggregate is by matching on symbol names.
+    Note that symbols must be available for a match to occur.
+    All imported symbols are immediately available,
+      but \tt{expand-sequence} may need to be used for symbols produced by
+      the same package.
+
+
+    \ref{_aggregate-rate-each_} aggregates values of generators (usually
+      referred to by \tt{rate-each}) through summation.
+    A \tt{rate-each} block is generated to perform the summation.
+
+    Since \tt{rate-each} multiplies its body by \tt{_CMATCH_},
+      zero symbols would normally result in the summation of \tt{_CMATCH_}
+      itself, which is not desirable;
+        this template always includes \ref{ZERO} in the body to defend
+          against this,
+            causing a yield of~$0.00$ if there are no symbol matches.
+
+    <template name="_aggregate-rate-each_"
+              desc="Aggregate generator values by symbol prefix">
+      <param name="@class@"  desc="Iterator class (omit for scalars)" />
+      <param name="@prefix@" desc="Symbol prefix" />
+
+      <param name="@yields@" desc="Scalar yield name (optional)">
+        <text></text>
+      </param>
+
+      <param name="@generates@" desc="Generator name (optional)">
+        <text></text>
+      </param>
+
+
+      <rate-each class="@class@" yields="@yields@"
+                 generates="@generates@" index="k">
+        <c:sum>
+          <!-- prevent summing _CMATCH_ if there are no symbols (see above
+               comments) -->
+          <c:value-of name="ZERO"
+                      label="Guard against zero symbol matches" />
+
+          <inline-template>
+            <for-each>
+              <sym-set name-prefix="@prefix@" type="gen" />
+            </for-each>
+
+            <c:value-of name="@sym_name@" index="k" />
+          </inline-template>
+        </c:sum>
+      </rate-each>
+    </template>
+
+
+    \ref{_aggregate-rate_} is analgous to \ref{_aggregate-rate-each_},
+      handling only scalar~\tt{@yields@}.
+    A \tt{rate} block is generated to aggregate by summation.
+
+    To prevent an empty rate block from being generated if there are no
+      symbol matches,
+        \ref{ZERO} is always included as part of the summation.
+
+    <template name="_aggregate-rate_"
+              desc="Aggregate scalar results by symbol prefix">
+      <param name="@prefix@" desc="Symbol prefix" />
+      <param name="@yields@" desc="Scalar yield name" />
+
+
+      <rate yields="@yields@">
+        <c:sum>
+          <!-- prevent completely empty rate block -->
+          <c:value-of name="ZERO"
+                      label="Guard against zero symbol matches" />
+
+          <inline-template>
+            <for-each>
+              <sym-set name-prefix="@prefix@" type="rate" />
+            </for-each>
+
+            <c:value-of name="@sym_name@" />
+          </inline-template>
+        </c:sum>
+      </rate>
+    </template>
+
+
+    \ref{_aggregate-classify_} aggregates classifications.
+    Keep in mind that classifications act as universal quantifiers by default,
+      meaning zero symbol matches will produce a match and a scalar~$1$;
+        existential quantifiers (\tt{@any@} set to \tt{true}) will \emph{not}
+          match and will produce the scalar~$0$.
+
+    <template name="_aggregate-classify_"
+              desc="Aggregate classification results by symbol prefix">
+      <param name="@prefix@" desc="Symbol prefix" />
+      <param name="@as@"     desc="Classification name" />
+      <param name="@desc@"   desc="Generated classification description" />
+
+      <param name="@yields@" desc="Vector yield name (optional)">
+        <text></text>
+      </param>
+
+      <param name="@any@"
+             desc="Existential classification (default false, universal)">
+        <text></text>
+      </param>
+
+
+      <classify as="@as@" yields="@yields@" desc="@desc@" any="@any@">
+        <inline-template>
+          <for-each>
+            <sym-set name-prefix="@prefix@" type="class" />
+          </for-each>
+
+          <t:match-class name="@sym_name@" />
+        </inline-template>
+      </classify>
+    </template>
+  </section>
+</package>
diff --git a/core/alias.xml b/core/alias.xml
new file mode 100644
index 0000000..7564f1a
--- /dev/null
+++ b/core/alias.xml
@@ -0,0 +1,100 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2017 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         core="true"
+         desc="Aliasing Values">
+
+  These alias templates allow giving a new name to existing values.
+  They are most useful for avoiding too many template-generated variables.
+
+  If an alias name matches the name of the source,
+    then no alias will be generated;
+      this behavior is useful when generating aliases in templates where the
+      caller may or may not choose a name that would otherwise conflict with
+      its internal representation.
+
+  If no description is provided,
+    aliases will inherit the description from the source symbol.
+
+  <template name="_classify-alias_"
+            desc="Alias classification">
+    <param name="@as@"   desc="Classification alias name" />
+    <param name="@from@" desc="Source classification" />
+
+    <param name="@desc@" desc="Classification alias description">
+      <param-sym-value prefix=":class:" name="@from@" value="desc" />
+    </param>
+
+    <param name="@yields@" desc="Classification alias yield">
+      <text></text>
+    </param>
+
+    <param name="@__src_yields@" desc="Source classification yield">
+      <param-class-to-yields name="@from@" />
+    </param>
+
+
+    <unless name="@as@" eq="@from@">
+      <classify as="@as@" desc="@desc@" yields="@yields@">
+        <match on="@__src_yields@" />
+      </classify>
+    </unless>
+  </template>
+
+
+  <template name="_rate-alias_"
+            desc="Alias calculation">
+    <param name="@yields@" desc="Alias name" />
+    <param name="@from@"   desc="Source name" />
+
+
+    <unless name="@from@" eq="@yields@">
+      <rate yields="@yields@">
+        <c:value-of name="@from@" />
+      </rate>
+    </unless>
+  </template>
+
+
+  <template name="_rate-each-alias_"
+            desc="Alias generator">
+    <param name="@generates@" desc="Generator alias name" />
+    <param name="@from@"      desc="Source generator" />
+
+    <param name="@desc@" desc="Generator alias description">
+      <param-sym-value name="@from@" value="desc" />
+    </param>
+
+    <param name="@yields@" desc="Yield alias name">
+      <text>_</text>
+      <param-value name="@generates@" />
+    </param>
+
+
+    <unless name="@generates@" eq="@from@">
+      <rate yields="@yields@">
+        <c:sum of="@from@" desc="@desc@" generates="@generates@" index="k">
+          <c:value-of name="@from@" index="k" />
+        </c:sum>
+      </rate>
+    </unless>
+  </template>
+</package>
diff --git a/core/assert.xml b/core/assert.xml
new file mode 100644
index 0000000..726ee7d
--- /dev/null
+++ b/core/assert.xml
@@ -0,0 +1,110 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2015, 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="Assertions">
+
+  <import package="base" />
+  <import package="vector/cmatch" export="true" />
+
+
+  This package is young;
+    the intent is to provide basic assertions to ensure data
+    integrity.
+
+  \emph{N.B. The behavior of this package changed in version 1.0.0---%
+          \ref{_assert_} now properly fails on non-match,
+            not match.}
+
+  The \ref{_assert_} template generates a~generic assertion using the
+    provided predicates.
+  If the predicates provided to the assertion fail (yields $\bot$),
+    the system immediately terminates.\footnote{
+      Beacuse the system stops processing once a terminating classification
+        yields~$\top$,
+          only one assertion will ever match,
+            even if others would match if execution were to continue.}
+  \tt{@class@} may optionally be used to predicate the asseriton itself---%
+    the assertion will be performed only on respective class matches.
+
+  \ref{_assert_} implements assertions by genearting two classifications---%
+    one to perform the actual assertion,
+      and a terminating classification to ensure that the assertion
+        yields~$\top$.
+
+  The param \ref{assert_ignore} will suppress assertion failures at runtime.
+
+  <param name="assert_ignore" type="boolean" default="0"
+         desc="Ignore assertion failures" />
+
+
+  <template name="_assert_"
+            desc="Terminate on predicate failure">
+    <param name="@values@" desc="Predicates" />
+
+    <param name="@as@" desc="Generated classification name">
+      <text unique="true">-assert-</text>
+    </param>
+
+    <param name="@failure@" desc="Failure description">
+      <text>Assertion</text>
+    </param>
+
+    <param name="@class@" desc="Only perform assertion for respective class
+                                matches (optional)" />
+
+
+    <param name="@neg_as@"
+           desc="Generated name for classification to be negated">
+      <text unique="true">-nassert-</text>
+    </param>
+    <param name="@neg_yields@"
+           desc="Generated yield for classification to be negated">
+      <text unique="true">_nassert</text>
+    </param>
+
+
+    <!-- The actual assertion will be performed by one classification... -->
+    <classify as="@neg_as@" yields="@neg_yields@"
+              desc="{@failure@} (assertion result)">
+      <any>
+        <!-- if class is provided and does not match, then the assertion
+             succeeds automatically -->
+        <if name="@class@">
+          <t:match-class name="@class@" value="FALSE" />
+        </if>
+
+        <all>
+          <param-copy name="@values@" />
+        </all>
+      </any>
+    </classify>
+
+    <!-- ...which is in turn negated for the terminating
+         classification.  The reason for this is that terminating
+         classifications terminate on _match_. -->
+
+    <classify as="@as@" desc="@failure@" terminate="true">
+      <match on="assert_ignore" value="FALSE" />
+      <match on="@neg_yields@" value="FALSE" />
+    </classify>
+  </template>
+</package>
diff --git a/core/base.xml b/core/base.xml
new file mode 100644
index 0000000..099055b
--- /dev/null
+++ b/core/base.xml
@@ -0,0 +1,220 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2015, 2017, 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         core="true"
+         desc="Base features">
+
+  The \pkgself~package exposes common and internal
+  defintions.  Ideally, this package will be included automatically by
+  the compiler to remove repetitive, boilerplate imports.  Importing
+  this package isn't necessary if none of these definitions are
+  needed.
+
+
+
+  <section title="Internal Constants">
+    \ref{_CMATCH_} is a magic constant that contains the result of
+    a~classification match.  This is used implicity by
+    \ref{rate-each}.\footnote{The symbol is \Xi~because it looks like
+    a sideways array.}
+
+    \todo{Remove in favor of a local variable or generated
+    classification; there is no need (anymore) for this to be magic.}
+
+    <const name="_CMATCH_" type="boolean" sym="\Xi"
+           desc="Classification match vector (applicability)">
+      <item value="0"
+            desc="Dummy value; this set is populated upon entering
+                  each rate block" />
+    </const>
+
+
+    The runtime is responsible for populating \ref{__DATE_YEAR__} with
+    a proper value representing the current year.
+
+    \todo{TAME is deterministic with this one exception; remove it and
+    have users use the params from {\tt datetime} instead if they need this
+    datum.}
+
+    <const name="__DATE_YEAR__" magic="true"
+           value="0" type="integer"
+           desc="Current year"
+           sym="\widehat{D^\gamma}" />
+  </section>
+
+
+
+  <section title="Primitive Types">
+    Primitives are defined internally; these definitions simply
+    provide symbols to permit their use.
+
+    <typedef name="integer"
+             desc="Any value in the set of integers"
+             sym="\mathbb{I}">
+      <base-type />
+    </typedef>
+
+
+    <typedef name="float"
+             desc="Any real number (represented as a float)"
+             sym="\mathbb{R}">
+      <base-type />
+    </typedef>
+
+
+    \ref{empty} does not have much use outside of the compiler.
+
+    <typedef name="empty"
+             desc="Empty set"
+             sym="\emptyset">
+      <base-type />
+    </typedef>
+  </section>
+
+
+  <section title="Boolean and Unknown">
+    \ref{boolean} contains the boolean \ref{TRUE} and~\ref{FALSE} values,
+      which map to~$1$ and~$0$ respectively.
+    The \ref{maybe} type is the union of \ref{boolean} and \ref{NOTHING},
+      with a value of~$-1$;\footnote{
+          This is similar in spirit to the Haskell \tt{Maybe} type,
+            or the OCaml \tt{Option} type.
+        }this is commonly used to represent an unknown state or missing
+        value.\footnote{
+          The \ref{nothing}~type is used for the sake of the union;
+            it should not be used directly.}
+
+    <typedef name="maybe" desc="Boolean or unknown value">
+      <union>
+        <typedef name="nothing" desc="Unknown value">
+          <enum type="integer">
+            <item name="NOTHING" value="-1" desc="Unknown or missing value" />
+          </enum>
+        </typedef>
+
+        <typedef name="boolean" desc="Boolean values">
+          <enum type="integer">
+            <item name="TRUE"  value="1" desc="True" />
+            <item name="FALSE" value="0" desc="False" />
+          </enum>
+        </typedef>
+      </union>
+    </typedef>
+
+    The constant \ref{UNKNOWN} is also defined as~$-1$ to serve as an
+      alternative to the term~``nothing''.
+
+    <const name="UNKNOWN" value="-1"
+           desc="Unknown or missing value" />
+  </section>
+
+
+
+  <section title="Convenience">
+    $0$~is a~common value.  Where a value is required (such
+    as a~template argument), \ref{ZERO} may be used.  TAME now
+    supports a~constant-scalar syntax ({\tt #0}; \todo{reference this
+    in documentation}), making this largely unnecessary.
+
+    This is declared as a float to provide compatibility with all
+    types of expressions.
+
+    <const name="ZERO" value="0.00"
+           desc="Zero value" />
+
+
+    In the case where classifications are required, but a~static
+    assumption about the applicability of the subject can be made, we
+    have values that are always~true and always~false.  The use
+    of~\ref{never} may very well be a~code smell, but let us not rush
+    to judgment.\footnote{\ref{never} has been added as an analog
+    to~\ref{always}; its author has never had use for it.  Oh, look,
+    we just used ``never''.}
+
+    <classify as="always"
+              desc="Always true"
+              yields="alwaysTrue"
+              keep="true" />
+
+    <classify as="never"
+              any="true"
+              desc="Never true"
+              yields="neverTrue"
+              keep="true" />
+  </section>
+
+
+
+  <section title="Work-In-Progress">
+    \ref{_todo_} formalizes TODO items and may optionally yield a
+      value~\tt{@value@} for use within calculations.%
+        \footnote{This is different than its previous behavior of always
+                    yielding a scalar~$0$.}
+    All uses of the \ref{_todo_} template will produce a warning composed of
+      its description~\tt{@desc@}.
+
+    <template name="_todo_"
+              desc="Represents work that needs to be done">
+      <param name="@desc@" desc="TODO desc">
+        <text>TODO</text>
+      </param>
+
+      <param name="@value@" desc="Placeholder value" />
+      <param name="@index@" desc="Placeholder value index">
+        <text></text>
+      </param>
+
+      <unless name="@value@">
+        <unless name="@index@" eq="">
+          <error>Using @index@ without @value@</error>
+        </unless>
+      </unless>
+
+
+      <warning>
+        TODO: <param-value name="@desc@" />
+      </warning>
+
+
+      <if name="@value@">
+        <c:value-of name="@value@" index="@index@" />
+      </if>
+    </template>
+
+
+    The \ref{_ignore_} template serves as a~block
+    comment.\footnote{This is useful since XML does not support nested
+    comments, which makes it difficult to comment out code that
+    already has XML comments.}  It may be useful for debugging, but is
+    discouraged for use otherwise.  The \ref{_ignore_/@desc@} param
+    should be used to describe intent.
+
+    <template name="_ignore_"
+              desc="Removes all child nodes (as if commented out)">
+      <param name="@values@" desc="Nodes to comment out" />
+      <param name="@desc@"   desc="Reason for ignore" />
+
+
+      <warning>Ignored block!</warning>
+    </template>
+  </section>
+</package>
+
diff --git a/core/cond.xml b/core/cond.xml
new file mode 100644
index 0000000..1b30199
--- /dev/null
+++ b/core/cond.xml
@@ -0,0 +1,169 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  core="true"
+  desc="Generic conditionals">
+
+  <import package="base" />
+
+
+  <!-- TODO: use reduce once we have support for function passing -->
+  <function name="or" desc="Return value A if non-zero, otherwise B">
+    <param name="or_a" type="float" desc="Value to return if non-zero" />
+    <param name="or_b" type="float" desc="Value to return if A is zero" />
+
+    <c:cases>
+      <!-- return B if A is zero -->
+      <c:case>
+        <c:when name="or_a">
+          <c:eq>
+            <c:const value="0" type="integer" desc="Return B if A is 0" />
+          </c:eq>
+        </c:when>
+
+        <c:value-of name="or_b" />
+      </c:case>
+
+      <!-- return A if non-zero -->
+      <c:otherwise>
+        <c:value-of name="or_a" />
+      </c:otherwise>
+    </c:cases>
+  </function>
+
+
+  <template name="_cond_" desc="Conditional shorthand">
+    <param name="@name@" desc="Value to check" />
+    <param name="@index@" desc="Index of boolean value" />
+    <param name="@value@" desc="Constant value" />
+    <param name="@type@" desc="Value type" />
+
+    <param name="@cond@" desc="Value to compare against (default boolean true)">
+      <text>TRUE</text>
+    </param>
+
+    <param name="@desc@" desc="Value description">
+      <text>Value when </text>
+      <param-value name="@name@" />
+    </param>
+
+    <c:const value="@value@" type="@type@" desc="@desc@">
+      <!-- TODO: non-index option -->
+      <c:when name="@name@" index="@index@">
+        <c:eq>
+          <c:value-of name="@cond@" />
+        </c:eq>
+      </c:when>
+    </c:const>
+  </template>
+
+
+  <template name="_yield-unless_" desc="Yield a value unless another value is set">
+    <param name="@name@" desc="Overriding value" />
+    <param name="@index@" desc="Index" />
+    <param name="@values@" desc="Value to use otherwise" />
+
+    <c:cases>
+      <c:case>
+        <if name="@index@">
+          <c:when name="@name@" index="@index@">
+            <c:gt>
+              <c:const value="0" type="integer" desc="Use override if greater than 0" />
+            </c:gt>
+          </c:when>
+        </if>
+        <unless name="@index@">
+          <c:when name="@name@">
+            <c:gt>
+              <c:const value="0" type="integer" desc="Use override if greater than 0" />
+            </c:gt>
+          </c:when>
+        </unless>
+
+        <if name="@index@">
+          <c:value-of name="@name@" index="@index@" />
+        </if>
+        <unless name="@index@">
+          <c:value-of name="@name@" />
+        </unless>
+      </c:case>
+
+      <c:otherwise>
+        <param-copy name="@values@" />
+      </c:otherwise>
+    </c:cases>
+  </template>
+
+
+  <template name="_default_" desc="Set a default value if a value is not set">
+    <param name="@name@" desc="Param name" />
+    <param name="@index@" desc="Index" />
+    <param name="@default@" desc="Default value to use if empty" />
+
+    <c:cases>
+      <c:case>
+        <c:when name="@name@" index="@index@">
+          <c:eq>
+            <c:const value="0" type="integer" desc="No value" />
+          </c:eq>
+        </c:when>
+
+        <c:const value="@default@" type="integer" desc="Default value" />
+      </c:case>
+
+      <c:otherwise>
+        <c:value-of name="@name@" index="@index@" />
+      </c:otherwise>
+    </c:cases>
+  </template>
+
+
+
+  <template name="_cond-value-each_" desc="Conditional value">
+    <param name="@class@"     desc="Class match" />
+    <param name="@generates@" desc="Variable to generate into" />
+    <param name="@value@"     desc="Value to yield (use either this or const)" />
+    <param name="@const@"     desc="Constant to yield (use either this or value)" />
+    <param name="@count@"     desc="Optional number of times to apply const/value (as a variable)" />
+
+    <param name="@desc@" desc="Optional constant description">
+      <text>Constant value</text>
+    </param>
+
+    <!-- simply returns a constant value for the class match -->
+    <rate-each class="@class@" accumulate="none" generates="@generates@" index="k">
+      <c:product>
+        <if name="@value@">
+          <c:value-of name="@value@" />
+        </if>
+        <unless name="@value@">
+          <c:const value="@const@" type="float" desc="@desc@" />
+        </unless>
+
+        <!-- if this is not provided, then the c:product will be optimized away -->
+        <if name="@count@">
+          <c:value-of name="@count@" index="k" />
+        </if>
+      </c:product>
+    </rate-each>
+  </template>
+</package>
+
diff --git a/core/convention.xml b/core/convention.xml
new file mode 100644
index 0000000..7661888
--- /dev/null
+++ b/core/convention.xml
@@ -0,0 +1,59 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         title="Definition Conventions">
+
+
+  \ref{_naming-convention_} defines a naming convention for a parameter that
+    is enforced at compile-time.
+  Conventions may have prefixes, suffixes, or both.
+  If violated,
+    the compiler will abort in error.
+
+  <template name="_naming-convention_"
+            desc="Error about naming convention violation">
+    <param name="@name@"   desc="Given name" />
+    <param name="@prefix@" desc="Required prefix" />
+    <param name="@suffix@" desc="Required suffix" />
+
+    <if name="@prefix@">
+      <unless name="@name@" prefix="@prefix@">
+        <error>
+          naming convention violation:
+          `<param-value name="@name@" />' must be prefixed with
+          `<param-value name="@prefix@" />'.
+        </error>
+      </unless>
+    </if>
+
+    <if name="@suffix@">
+      <unless name="@name@" suffix="@suffix@">
+        <error>
+          naming convention violation:
+          `<param-value name="@name@" />' must be suffixed with
+          `<param-value name="@suffix@" />'.
+        </error>
+      </unless>
+    </if>
+  </template>
+</package>
diff --git a/core/datetime.xml b/core/datetime.xml
new file mode 100644
index 0000000..76d0bb3
--- /dev/null
+++ b/core/datetime.xml
@@ -0,0 +1,135 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015, 2017 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         xmlns:c="http://www.lovullo.com/calc"
+         core="true"
+         desc="Date and time">
+
+  <import package="base" />
+
+  <import package="assert" />
+  <import package="vector/cmatch" />
+
+
+  Providing the current timestamp externally allows the system to be
+    wholly deterministic.
+  It also captures the state of the data in time,
+    so re-calculating in the future (e.g. using the summary pages)
+    will yield the same result.
+
+  <param name="timestamp_current" type="integer" default="0"
+         desc="Unix timestamp representing the current time of the
+               system (clock time), as it should be perceived" />
+
+  <t:assert failure="Missing Unix timestamp for timestamp_current">
+    <t:match-gt on="timestamp_current" value="ZERO" />
+  </t:assert>
+
+
+  This system does not handle its own arthiemtic to figure out the
+    current date components from the given timestamp.
+
+  <param name="timestamp_year" type="integer" default="0"
+         desc="Current year (clock time), as it should be perceived" />
+
+  <param name="timestamp_month" type="integer" default="0"
+         desc="Current month (clock time), as it should be perceived;
+               1-indexed" />
+
+  <param name="timestamp_day" type="integer" default="0"
+         desc="Current day (clock time), as it should be perceived;
+               1-indexed" />
+
+
+  <!--
+    Determines how many years the given date is relative to the current date
+
+    Note that a positive number will be returned if the given date is
+    in the past, negative if in the future.
+  -->
+  <template name="_age-years_" desc="Age in years of the given date relative to the current">
+    <param name="@generates@" desc="Generator variable, per index" />
+    <param name="@yearset@"   desc="Set of years to iterate on" />
+    <param name="@sym@"       desc="Generator symbol" />
+    <param name="@when@"      desc="Only when this boolean value is true" />
+    <param name="@default@"   desc="Default age when yearset values are empty" />
+
+    <param name="@yields@" desc="Variable to yield into">
+      <text>_</text>
+      <param-value name="@generates@" />
+    </param>
+
+
+    <rate yields="@yields@">
+      <c:sum of="@yearset@" index="k" generates="@generates@" desc="Relative age" sym="@sym@">
+        <c:cases>
+          <if name="@when@">
+            <c:case>
+              <c:when name="@when@" index="k">
+                <c:eq>
+                  <c:value-of name="FALSE" />
+                </c:eq>
+              </c:when>
+
+              <!-- return a default, or 0 if no default is provided -->
+              <if name="@default@">
+                <c:value-of name="@default@" index="k" />
+              </if>
+              <unless name="@default@">
+                <c:const value="0" type="integer" desc="Condition not met, but no default" />
+              </unless>
+            </c:case>
+          </if>
+
+          <!-- if no @when@, then we'll always do this -->
+          <c:otherwise>
+            <c:cases>
+              <c:case>
+                <c:when name="@yearset@" index="k">
+                  <c:gt>
+                    <c:const value="0" type="integer" desc="Only calculate difference if a value is available" />
+                  </c:gt>
+                </c:when>
+
+                <c:sum label="Calculate by subtracting the given year from the current year">
+                  <c:value-of name="timestamp_year" />
+
+                  <c:product>
+                    <c:value-of name="NEGATE" />
+                    <c:value-of name="@yearset@" index="k" />
+                  </c:product>
+                </c:sum>
+              </c:case>
+
+              <!-- we don't have a value; just use the default age we were given -->
+              <if name="@default@">
+                <c:otherwise>
+                  <c:value-of name="@default@" index="k" />
+                </c:otherwise>
+              </if>
+            </c:cases>
+          </c:otherwise>
+        </c:cases>
+      </c:sum>
+    </rate>
+  </template>
+</package>
+
diff --git a/core/dummy.xml b/core/dummy.xml
new file mode 100644
index 0000000..2f1dd47
--- /dev/null
+++ b/core/dummy.xml
@@ -0,0 +1,87 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2016 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         core="true"
+         title="Dummy Values">
+
+  These are dummy values created primarily for testing.
+
+  Vectors are required for iteration;
+    the below generate values useful for doing so in constant,
+    calculation, and classification form.
+
+
+  <const name="VEC1" desc="Vector of length 1">
+    <item value="1" desc="Dummy value" />
+  </const>
+  <const name="VEC2" desc="Vector of length 2">
+    <item value="1" desc="Dummy value" />
+    <item value="2" desc="Dummy value" />
+  </const>
+  <const name="VEC3" desc="Vector of length 3">
+    <item value="1" desc="Dummy value" />
+    <item value="2" desc="Dummy value" />
+    <item value="3" desc="Dummy value" />
+  </const>
+
+
+  <rate yields="_vec1">
+    <c:sum of="VEC1" generates="vec1" index="k"
+           desc="Vector of length 1">
+      <c:const value="1" desc="Dummy" />
+    </c:sum>
+  </rate>
+  <rate yields="_vec2">
+    <c:sum of="VEC2" generates="vec2" index="k"
+           desc="Vector of length 2">
+      <c:const value="1" desc="Dummy" />
+    </c:sum>
+  </rate>
+  <rate yields="_vec3">
+    <c:sum of="VEC3" generates="vec3" index="k"
+           desc="Vector of length 3">
+      <c:const value="1" desc="Dummy" />
+    </c:sum>
+  </rate>
+
+
+  <classify as="length1" desc="Classification of length 1">
+    <match on="vec1">
+      <c:gt>
+        <c:const value="0" desc="Check if defined" />
+      </c:gt>
+    </match>
+  </classify>
+  <classify as="length2" desc="Classification of length 2">
+    <match on="vec2">
+      <c:gt>
+        <c:const value="0" desc="Check if defined" />
+      </c:gt>
+    </match>
+  </classify>
+  <classify as="length3" desc="Classification of length 3">
+    <match on="vec3">
+      <c:gt>
+        <c:const value="0" desc="Check if defined" />
+      </c:gt>
+    </match>
+  </classify>
+</package>
diff --git a/core/extern.xml b/core/extern.xml
new file mode 100644
index 0000000..868d590
--- /dev/null
+++ b/core/extern.xml
@@ -0,0 +1,80 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2017 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         core="true"
+         desc="Extern Definition">
+
+  Externs are symbols that are expected to be defined by package importers
+    before linking.
+  Rather than explicitly declaring extern symbols,
+    these templates provide more human-readable abstractions.
+
+
+  Note that when defining extern classifications with
+    \ref{_classify-extern_},
+      if specifying a yield,
+      the dimensionality is~$1$ unless otherwise provided.
+
+  <template name="_classify-extern_"
+            desc="Define extern for classification">
+    <param name="@as@" desc="Extern classification name" />
+
+    <param name="@yields@" desc="Extern classification yield">
+      <text></text>
+    </param>
+
+    <!-- default assumption is a vector -->
+    <param name="@dim@" desc="Extern classification yield dimensions">
+      <text>1</text>
+    </param>
+
+    <extern name=":class:{@as@}" type="class" dim="0"
+            yields="@yields@" />
+
+    <if name="@yields@">
+      <extern name="@yields@" type="cgen" dim="@dim@"
+              parent=":class:{@as@}" />
+    </if>
+  </template>
+
+
+  <template name="_rate-each-extern_"
+            desc="Define extern for rate-each">
+    <param name="@generates@" desc="Extern generator name" />
+    <param name="@yields@"    desc="Extern yield name" />
+
+    <if name="@generates@">
+      <extern name="@generates@" type="gen" dtype="float" dim="1" />
+    </if>
+
+    <if name="@yields@">
+      <extern name="@yields@" type="rate" dtype="float" dim="0" />
+    </if>
+  </template>
+
+
+  <template name="_rate-extern_"
+            desc="Define extern for rate">
+    <param name="@yields@" desc="Extern yield name" />
+
+    <extern name="@yields@" type="rate" dtype="float" dim="0" />
+  </template>
+</package>
diff --git a/core/insurance.xml b/core/insurance.xml
new file mode 100644
index 0000000..64f93dc
--- /dev/null
+++ b/core/insurance.xml
@@ -0,0 +1,423 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2016, 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         title="Insurance Abstractions">
+
+  <import package="base" />
+  <import package="assert"        export="true" />
+  <import package="convention"    export="true" />
+  <import package="map"           export="true" />
+  <import package="numeric/round" export="true" />
+  <import package="vector/cmatch" export="true" />
+
+
+  These are primitive abstractions for insurance that will be
+  improved upon over time.
+
+  Some notable TODOs:
+
+  \begin{enumerate}
+    \item Enforce naming convention;
+    \item Support scalar results;
+    \item Fail on zero premium unless explicitly stated;
+    \item Fail on negative premium (use a credit template);
+    \item Rounding direction (currently only nearest); and
+    \item Credit and surcharge.
+  \end{enumerate}
+
+  \todo{Template to abstract these {\tt rate-each} generation
+  templates.}
+
+
+  <template name="_premium_"
+            desc="A premium dollar amount">
+    <param name="@values@"    desc="Body" />
+    <param name="@class@"     desc="Predicate" />
+    <param name="@generates@" desc="Generator name" />
+    <param name="@index@"     desc="Generator index" />
+
+    <param name="@no@" desc="Negated predicate">
+      <text></text>
+    </param>
+
+    <param name="@sym@" desc="TeX symbol">
+      <text></text>
+    </param>
+    <param name="@gensym@" desc="Generator TeX symbol">
+      <text></text>
+    </param>
+
+    <!-- required since various companies have various bizzare
+         rounding rules that tend to be a source of bugs; make the
+         developer think about it -->
+    <param name="@round@" desc="Rounding method" />
+
+    <!-- not yet used, but it will at least serve as code
+         documentation for the time being -->
+    <param name="@desc@" desc="Premium description" />
+
+
+    <rate-each class="@class@" no="@no@"
+               generates="@generates@" index="@index@"
+               sym="@sym@" gensym="@gensym@">
+      <!-- TODO: we now have reason for a more concise conditional
+           syntax -->
+      <if name="@round@" eq="dollar">
+        <t:round>
+          <param-copy name="@values@" />
+        </t:round>
+      </if>
+      <unless name="@round@" eq="dollar">
+        <if name="@round@" eq="cent">
+          <c:apply name="round_cents">
+            <c:arg name="round_cents_val">
+              <param-copy name="@values@" />
+            </c:arg>
+          </c:apply>
+        </if>
+
+        <unless name="@round@" eq="cent">
+          <if name="@round@" eq="up">
+            <c:ceil>
+              <param-copy name="@values@" />
+            </c:ceil>
+          </if>
+
+          <unless name="@round@" eq="up">
+            <if name="@round@" eq="down">
+              <c:floor>
+                <param-copy name="@values@" />
+              </c:floor>
+            </if>
+
+            <unless name="@round@" eq="down">
+              <!-- no rounding -->
+              <param-copy name="@values@" />
+            </unless>
+          </unless>
+        </unless>
+      </unless>
+    </rate-each>
+  </template>
+
+
+  \ref{_factor_} defines a calculation that results in a factor
+    which will later be used in a product.
+  There are special considerations for these types of values---%
+    generally, they should not have a value of~$0$ if some sort of calculation
+      condition or lookup is not met,
+        as that would have the effect of wiping out premium.\footnote{
+          Note that every generator index can be non-zero but still sum up
+            to a zero yield,
+              which would trigger this error.}
+  If zero is desired,
+    \tt{@allow-zero@} must be set to \tt{true} to explicitly permit it.
+
+  <template name="_factor_"
+            desc="Factor to multiply against (must be non-zero by default)">
+    <param name="@values@" desc="Body" />
+
+    <param name="@class@" desc="Predicate">
+      <text></text>
+    </param>
+
+    <param name="@generates@" desc="Generator name">
+      <text></text>
+    </param>
+
+    <param name="@index@" desc="Generator index">
+      <text></text>
+    </param>
+
+    <param name="@no@" desc="Negated predicate">
+      <text></text>
+    </param>
+
+    <param name="@sym@" desc="TeX symbol">
+      <text></text>
+    </param>
+    <param name="@gensym@" desc="Generator TeX symbol">
+      <text></text>
+    </param>
+
+    <param name="@yields@" desc="Yield (optional)">
+      <text></text>
+    </param>
+
+    <!-- at least one of generates or yields is required -->
+    <if name="@yields@" eq="">
+      <if name="@generates@" eq="">
+        <error>must provide at least one of @generates or @yields</error>
+      </if>
+    </if>
+
+    <!-- not yet used, but it will at least serve as code
+         documentation for the time being -->
+    <param name="@desc@" desc="Factor description" />
+    <unless name="@desc@">
+      <unless name="@generates@" eq="">
+        <error>
+            a description (@desc@) is required for
+            `<param-value name="@generates@" />'
+        </error>
+      </unless>
+      <if name="@generates@" eq="">
+        <error>
+            a description (@desc@) is required for
+            `<param-value name="@yields@" />'
+        </error>
+      </if>
+    </unless>
+
+    <!-- normally we want factors to default to 1, otherwise they could wipe
+         out premium -->
+    <param name="@allow-zero@" desc="Allow value of zero (default false; see
+                                     also @default@)">
+      <text>false</text>
+    </param>
+
+    <!-- default is _only_ used when a factor is 0, so it makes no sense to
+         set a default to #0 -->
+    <param name="@default@" desc="Default value if 0 (optional)" />
+    <if name="@default@" eq="#0">
+      <error>
+        a value of #0 for @default@ is not meaningful;
+          use @allow-zero@ instead.
+      </error>
+    </if>
+
+    <!-- negative values might be indicitive of a failure to provide a floor
+         on certain factors when summing (for example) -->
+    <param name="@allow-negative@" desc="Allow negative value (default false)">
+      <text>false</text>
+    </param>
+
+    <param name="@_prefix@" desc="Factor type (drives naming convention)">
+      <text>factor</text>
+    </param>
+
+    <unless name="@generates@" eq="">
+      <t:naming-convention name="@generates@" prefix="@_prefix@" />
+    </unless>
+
+    <unless name="@yields@" eq="">
+      <t:naming-convention name="@yields@" prefix="@_prefix@" />
+    </unless>
+
+
+    <!-- factor calculation -->
+    <rate-each class="@class@" no="@no@" yields="@yields@" sym="@sym@"
+               generates="@generates@" index="@index@" gensym="@gensym@">
+      <!-- use a default if provided if the factor expression yields 0 -->
+      <if name="@default@">
+        <c:let>
+          <c:values>
+            <c:value name="factor" type="float"
+                     desc="Factor result before default">
+              <param-copy name="@values@" />
+            </c:value>
+          </c:values>
+
+          <t:map-set name="factor">
+            <t:map from="ZERO" value="@default@" />
+            <t:map-else value="factor" />
+          </t:map-set>
+        </c:let>
+      </if>
+
+      <!-- avoid let generation if no default was provided -->
+      <unless name="@default@">
+        <param-copy name="@values@" />
+      </unless>
+    </rate-each>
+
+
+    <!-- assertion for non-zero -->
+    <unless name="@allow-zero@" eq="true">
+      <!-- assertions are useless if a static default was provided, since we
+           know that zero can then never be yielded-->
+      <unless name="@default@" prefix="#">
+        <unless name="@generates@" eq="">
+          <t:assert failure="{@generates@} must not yield a value of 0 for
+                             any index">
+            <t:match-ne on="@generates@" value="ZERO" />
+          </t:assert>
+        </unless>
+
+        <unless name="@yields@" eq="">
+          <t:assert failure="{@yields@} must not yield a value of 0">
+            <t:match-ne on="@yields@" value="ZERO" />
+          </t:assert>
+        </unless>
+      </unless>
+    </unless>
+
+    <!-- assertion for non-negative -->
+    <unless name="@allow-negative@" eq="true">
+      <unless name="@generates@" eq="">
+        <t:assert failure="{@generates@} must not yield a negative value
+                           for any index">
+          <t:match-gt on="@generates@" value="ZERO" />
+        </t:assert>
+      </unless>
+
+      <unless name="@yields@" eq="">
+        <t:assert failure="{@yields@} must not yield a negative value">
+          <t:match-gt on="@yields@" value="ZERO" />
+        </t:assert>
+      </unless>
+    </unless>
+  </template>
+
+
+  The \ref{_credit_} and \ref{_debit_} templates define factors that are
+    intended to introduce, respectively, credits and surcharges.
+  Each name defined using these templates must be prefixed with ``credit''
+    and ``debit'' respectively.
+  Credits must never exceed a value of~$1$,
+    as they would then be treated as if they were surcharges.
+  Similarly, debits must never have a value less than~$1$.
+  Both must have values greater than~$0$.
+
+  If a value can be either a credit or a debit,
+    it is sometimes called an ``adjustment''.
+  In reality it is just another factor,
+    so just use \ref{_factor_} in such a~case.
+
+  <template name="_credit_"
+            desc="Credit to be used as a factor">
+    <param name="@values@" desc="Credit calculation" />
+
+    <param name="@class@" desc="Predicate">
+      <text></text>
+    </param>
+
+    <param name="@generates@" desc="Generator name">
+      <text></text>
+    </param>
+
+    <param name="@index@" desc="Generator index">
+      <text></text>
+    </param>
+
+    <param name="@no@" desc="Negated predicate">
+      <text></text>
+    </param>
+
+    <param name="@sym@" desc="TeX symbol">
+      <text></text>
+    </param>
+    <param name="@gensym@" desc="Generator TeX symbol">
+      <text></text>
+    </param>
+
+    <param name="@yields@" desc="Yield (optional)">
+      <text></text>
+    </param>
+
+    <param name="@desc@" desc="Factor description">
+      <text></text>
+    </param>
+
+
+    <t:factor _prefix="credit"
+              class="@class@" no="@no@" yields="@yields@" sym="@sym@"
+              generates="@generates@" index="@index@" gensym="@gensym@"
+              desc="@desc@">
+      <param-copy name="@values@" />
+    </t:factor>
+
+    <unless name="@generates@" eq="">
+      <t:assert failure="{@generates@} is a credit but one or more
+                         of its values are greater than 1">
+        <t:match-lte on="@generates@" value="#1" />
+      </t:assert>
+    </unless>
+
+    <unless name="@yields@" eq="">
+      <t:assert failure="{@yields@} is a credit but its value is
+                         greater than 1">
+        <t:match-lte on="@yields@" value="#1" />
+      </t:assert>
+    </unless>
+  </template>
+
+
+  <template name="_debit_"
+            desc="Debit to be used as a factor">
+    <param name="@values@" desc="Credit calculation" />
+
+    <param name="@class@" desc="Predicate">
+      <text></text>
+    </param>
+
+    <param name="@generates@" desc="Generator name">
+      <text></text>
+    </param>
+
+    <param name="@index@" desc="Generator index">
+      <text></text>
+    </param>
+
+    <param name="@no@" desc="Negated predicate">
+      <text></text>
+    </param>
+
+    <param name="@sym@" desc="TeX symbol">
+      <text></text>
+    </param>
+    <param name="@gensym@" desc="Generator TeX symbol">
+      <text></text>
+    </param>
+
+    <param name="@yields@" desc="Yield (optional)">
+      <text></text>
+    </param>
+
+    <param name="@desc@" desc="Factor description">
+      <text></text>
+    </param>
+
+
+    <t:factor _prefix="debit"
+              class="@class@" no="@no@" yields="@yields@" sym="@sym@"
+              generates="@generates@" index="@index@" gensym="@gensym@"
+              desc="@desc@">
+      <param-copy name="@values@" />
+    </t:factor>
+
+    <unless name="@generates@" eq="">
+      <t:assert failure="{@generates@} is a debit but one or more
+                         of its values are less than 1">
+        <t:match-gte on="@generates@" value="#1" />
+      </t:assert>
+    </unless>
+
+    <unless name="@yields@" eq="">
+      <t:assert failure="{@yields@} is a credit but its value is
+                         less than 1">
+        <t:match-gte on="@yields@" value="#1" />
+      </t:assert>
+    </unless>
+  </template>
+</package>
diff --git a/core/map.xml b/core/map.xml
new file mode 100644
index 0000000..1bc0ac2
--- /dev/null
+++ b/core/map.xml
@@ -0,0 +1,188 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+
+<!--
+  Brings c:case statements to life
+
+  The problem with cases is that they are so verbose for simple cases. This
+  makes simple cases simple.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  title="Map Sets">
+
+
+  The problem with \texttt{case} statements is that they are very
+  verbose, which is greatly distracting for smaller cases, and~can
+  take away from the actual intent of the~code.  Map~sets make simple
+  cases simple by allowing concise map definitions.
+
+  If \ref{_map-set_/@name@} is provided, then each map implicitly
+  operates on that reference as its input, unless overridden using
+  \ref{_map_/@name@}.  \ref{_map-set_/@default@} can be used to
+  provide a constant default if no mapping is otherwise available; see
+  also~\ref{_map-else_}.
+
+
+  <template name="_map-set_" desc="Map a set of values">
+    <param name="@name@" desc="Param name" />
+    <param name="@index@" desc="Index" />
+    <param name="@values@" desc="Criteria (map nodes)" />
+    <param name="@label@" desc="Case statement label">
+      <!-- default empty -->
+      <text></text>
+    </param>
+
+    <!-- used in param-meta for _map-else_, so let's make the default
+         clear -->
+    <param name="@default@" desc="Default value">
+      <text></text>
+    </param>
+
+
+    <c:cases label="@label@">
+      <param-copy name="@values@">
+        <param-meta name="map_param"   value="@name@" />
+        <param-meta name="map_index"   value="@index@" />
+        <param-meta name="map_default" value="@default@" />
+      </param-copy>
+
+      <unless name="@default@" eq="">
+        <c:otherwise>
+          <c:const value="@default@" type="integer" desc="No mapping" />
+        </c:otherwise>
+      </unless>
+    </c:cases>
+  </template>
+
+
+  Mappings are defined using \ref{_map_}.  The input defaults to
+  the~reference declared by~\ref{_map-set_/@name@},
+  if~provided.  \ref{_map_/@name@} takes precedence, even if
+  the~former is provided.  Just like \texttt{case}~statements,
+  multiple inputs can be provided by specifying multiple references;
+  this means that each \ref{_map_/@name@} can differ.\footnote{This
+  isn't particularly intuitive or recommended, but it does simplify
+  certain tasks enough to maybe justify this type of use.}
+
+  Just like \texttt{case}~statements themselves, maps are
+  surjective---the codomain implicitly includes $0$~to handle all
+  default cases, unless a default is provided.
+
+
+  <template name="_map_" desc="Map criteria">
+    <param name="@from@" desc="Value to map from" />
+    <param name="@value@" desc="Constant to map to" />
+    <param name="@to@" desc="Named value (use instead of @value@)" />
+
+    <param name="@to-index@" desc="Named value index">
+      <text></text>
+    </param>
+
+    <param name="@type@" desc="Constant value type">
+      <text>float</text>
+    </param>
+
+    <param name="@desc@" desc="Map description">
+      <text>Destination mapping</text>
+    </param>
+
+    <!-- set by parent map-set -->
+    <param name="@name@" desc="Param name">
+      <param-inherit meta="map_param" />
+    </param>
+    <param name="@index@" desc="Index">
+      <param-inherit meta="map_index" />
+    </param>
+
+
+    <c:case>
+      <!-- index provided -->
+      <if name="@index@">
+        <c:when name="@name@" index="@index@">
+          <c:eq>
+            <c:value-of name="@from@" />
+          </c:eq>
+        </c:when>
+      </if>
+      <!-- no index provided -->
+      <unless name="@index@">
+        <c:when name="@name@">
+          <c:eq>
+            <c:value-of name="@from@" />
+          </c:eq>
+        </c:when>
+      </unless>
+
+      <if name="@value@">
+        <c:const value="@value@" desc="@desc@" />
+      </if>
+      <unless name="@value@">
+        <c:value-of name="@to@" index="@to-index@" />
+      </unless>
+    </c:case>
+  </template>
+
+
+  The default condition can be handled in two different ways: via
+  \ref{_map-set_/@default@}, or using \ref{_map-else_}; the latter
+  provides more flexibility and generally reads better.  Both cannot
+  be used together.
+
+
+  <template name="_map-else_" desc="Non-matching map criteria">
+    <param name="@value@" desc="Constant to map to" />
+    <param name="@to@" desc="Named value (use instead of @value@)" />
+
+    <param name="@desc@" desc="Map description">
+      <text>Destination mapping</text>
+    </param>
+
+    <param name="@to-index@" desc="Named value index">
+      <text></text>
+    </param>
+
+    <param name="@_map-default@" desc="_map_-specified default">
+      <param-inherit meta="map_default" />
+    </param>
+
+
+    <!-- provide a more friendly error; otherwise, they'd get an error
+         from having two c:otherwise nodes, which would be confusing -->
+    <unless name="@_map-default@" eq="">
+      <error>
+        A default value was already provided by _map-set_/@default@
+      </error>
+    </unless>
+
+
+    <c:otherwise>
+      <if name="@value@">
+        <c:const value="@value@" desc="@desc@" />
+      </if>
+      <unless name="@value@">
+        <c:value-of name="@to@" index="@to-index@" />
+      </unless>
+    </c:otherwise>
+  </template>
+</package>
+
diff --git a/core/numeric.xml b/core/numeric.xml
new file mode 100644
index 0000000..2a50888
--- /dev/null
+++ b/core/numeric.xml
@@ -0,0 +1,33 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  core="true"
+  desc="Numeric computations">
+
+  <!-- we are a meta-package -->
+  <import package="numeric/boolean" export="true" />
+  <import package="numeric/convert" export="true" />
+  <import package="numeric/minmax"  export="true" />
+  <import package="numeric/round"   export="true" />
+
+  <import package="numeric/common" export="true" />
+</package>
+
diff --git a/core/numeric/boolean.xml b/core/numeric/boolean.xml
new file mode 100644
index 0000000..b2ef700
--- /dev/null
+++ b/core/numeric/boolean.xml
@@ -0,0 +1,39 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  core="true"
+  desc="Numeric computations dealing with boolean algebra">
+
+  <import package="../base" />
+
+
+  <function name="not" desc="Negates a boolean value" sym="\lnot">
+    <param name="not_value" type="boolean" desc="Boolean value to negate" />
+
+    <c:const value="1" type="boolean" desc="Value of 1 if given value is zero">
+      <c:when name="not_value">
+        <c:eq>
+          <c:const value="0" type="boolean" desc="Value to assert against for returning 1" />
+        </c:eq>
+      </c:when>
+    </c:const>
+  </function>
+</package>
diff --git a/core/numeric/common.xml b/core/numeric/common.xml
new file mode 100644
index 0000000..4b8bdcf
--- /dev/null
+++ b/core/numeric/common.xml
@@ -0,0 +1,162 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="General numeric opreations">
+
+  <import package="../base" />
+
+
+
+  <section title="Stepping">
+    Due to the declarative nature of~TAME, recursive operations
+    needing to step are~not all that common.  These operations are
+    useful when converting between 1-indexed and 0-indexed data.
+
+    \ref{_inc_} increments a value by a~single step (default~1).
+
+    <template name="_inc_"
+              desc="Increment value [by 1]">
+      <param name="@values@" desc="Value to increment (nodes)" />
+
+      <param name="@value@" desc="Value to decrement by (default 1)">
+        <text>1</text>
+      </param>
+
+
+      <c:sum>
+        <param-copy name="@values@" />
+
+        <c:const value="@value@"
+                 desc="Increment by 1" />
+      </c:sum>
+    </template>
+
+
+    \ref{_dec_} decrements a value by a~single step (default~1).
+
+    <template name="_dec_"
+              desc="Decrement value [by 1]">
+      <param name="@values@" desc="Value to decrement (nodes)" />
+
+      <param name="@value@" desc="Value to decrement by (default 1)">
+        <text>1</text>
+      </param>
+
+
+      <c:sum>
+        <param-copy name="@values@" />
+
+        <t:negate>
+          <c:const value="@value@"
+                   desc="Decrement by 1" />
+        </t:negate>
+      </c:sum>
+    </template>
+  </section>
+
+
+
+  <section title="Negation">
+    Negation is a common task and it was tedious in older versions of
+    TAME\footnote{Before it was even called TAME, actually.}  For
+    these situations, \ref{NEGATE} is provided to avoid having to use
+    a~\ref{const} expression.  To avoid a {\tt product} expression
+    altogether, use~\ref{_negate_}.
+
+    <const name="NEGATE" value="-1"
+           desc="Negate a value"
+           sym="-" />
+
+
+    <template name="_negate_"
+              desc="Negate an expression">
+      <param name="@values@" desc="Expression to negate" />
+      <param name="@label@"  desc="Application label (default empty)">
+        <text></text>
+      </param>
+
+      <c:product label="@label@">
+        <c:value-of name="NEGATE" />
+        <param-copy name="@values@" />
+      </c:product>
+    </template>
+  </section>
+
+
+
+  <!-- Everything below this line must be moved. -->
+
+
+  <template name="_percent-of_" desc="Take percentage of a value">
+    <param name="@value@" desc="Value to take percentage of" />
+    <param name="@index@" desc="Index" />
+    <param name="@percent@" desc="Percent (as a whole number)" />
+
+    <param name="@desc@" desc="Optional description">
+      <text>Percent to reduce</text>
+    </param>
+
+    <c:product>
+      <c:value-of name="@value@" index="@index@" />
+      <t:ptor value="@percent@" desc="@desc@" />
+    </c:product>
+  </template>
+
+
+
+  <!--
+    Calculates a value based on the given multiplier if the given value falls
+    within the given range
+  -->
+  <function name="rangeadd" desc="Add a value multiplier for a given range">
+    <param name="low"  type="float" desc="Lower bound, inclusive" />
+    <param name="high" type="float" desc="Upper bound, inclusive" />
+    <param name="val"  type="float" desc="Value to compare" />
+    <param name="mult" type="float" desc="Range multiplier" />
+
+    <c:product>
+      <c:when name="val">
+        <c:gte>
+          <c:value-of name="low" />
+        </c:gte>
+      </c:when>
+      <c:when name="val">
+        <c:lte>
+          <c:value-of name="high" />
+        </c:lte>
+      </c:when>
+
+      <!-- atop of the 0.00003 we've already provided = 0.000035 -->
+      <c:value-of name="mult" />
+      <c:sum>
+        <c:value-of name="val" />
+        <!-- subtract -->
+        <c:product>
+          <c:value-of name="NEGATE" />
+          <c:value-of name="low" />
+        </c:product>
+      </c:sum>
+    </c:product>
+  </function>
+</package>
+
diff --git a/core/numeric/convert.xml b/core/numeric/convert.xml
new file mode 100644
index 0000000..4ba5c62
--- /dev/null
+++ b/core/numeric/convert.xml
@@ -0,0 +1,177 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="Converts scalars to another type">
+
+  <import package="../base" />
+
+  <!-- TODO: transitional; remove (used to contain _ptor_ and friends) -->
+  <import package="percent" export="true" />
+
+  <!-- for template applications -->
+  <import package="common" export="true" />
+  <import package="round" export="true" />
+
+
+  <!-- even more trivial, but again, cuts down on code -->
+  <template name="_scalarToAccum_" desc="Simply accumulates a scalar">
+    <param name="@scalar@" desc="Scalar to accumulate" />
+    <param name="@accum@" desc="Accumulator to accumulate into" />
+
+    <!-- this is useless, but required -->
+    <param name="@yields@" desc="Value to yield into, since it's required (useless)">
+      <text>__accum_</text>
+      <param-value name="@accum@" />
+      <text>_</text>
+      <param-value name="@scalar@" />
+    </param>
+
+    <param name="@type@" desc="Accumulation method">
+      <text>all</text>
+    </param>
+
+    <rate yields="@yields@">
+      <accumulate into="@accum@" type="@type@" />
+      <c:value-of name="@scalar@" />
+    </rate>
+  </template>
+
+
+  <!--
+    Map values falling within adjacent intervals
+
+    Interval endpoints are defined using `_endpoint_` nodes.  The
+    first interval is defined as [-∞,@start@), and the last endpoint
+    as (end,∞], where `end` is the value of the final `_endpoint_`.
+    The default maps for these two implicit intervals default to 0,
+    but may be changed with, respectively, @low@ and @high@.
+
+    Permitted children:
+      - _endpoint_* - Endpoint definition.
+  -->
+  <template name="_map-interval_"
+            desc="Map adjacent, closed intervals to scalars">
+    <param name="@values@" desc="Interval definitions" />
+    <param name="@name@"   desc="Reference value" />
+
+    <param name="@start@"
+           desc="Left endpoint of first interval (constant)" />
+
+    <param name="@index@"
+           desc="Reference value index">
+      <text></text>
+    </param>
+
+    <param name="@low@"
+           desc="Default value when less than @start@">
+      <text>#0</text>
+    </param>
+    <param name="@lowindex@"
+           desc="Low value index">
+      <text></text>
+    </param>
+
+    <param name="@high@"
+           desc="Default value when greater than final endpoint value">
+      <text>#0</text>
+    </param>
+    <param name="@highindex@"
+           desc="High value index">
+      <text></text>
+    </param>
+
+
+    <c:let>
+      <!-- c:when does not support shorthand constants (yet); this is
+           a workaround -->
+      <c:values>
+        <c:value name="__intval"
+                 type="float"
+                 desc="Interval reference value">
+          <c:value-of name="@name@" index="@index@" />
+        </c:value>
+      </c:values>
+
+
+      <c:cases>
+        <c:case>
+          <c:when name="__intval">
+            <c:lt>
+              <c:const value="@start@"
+                       type="float"
+                       desc="First interval left endpoint" />
+            </c:lt>
+          </c:when>
+
+          <c:value-of name="@low@" index="@lowindex@" />
+        </c:case>
+
+        <param-copy name="@values@" />
+
+        <c:otherwise>
+          <c:value-of name="@high@" index="@highindex@" />
+        </c:otherwise>
+      </c:cases>
+    </c:let>
+  </template>
+
+
+  <!--
+    Adjacent endpoint definition
+
+    A value that falls within range of the associated interval will be
+    mapped to the value @map@.
+
+    The first endpoint defined is the right endpoint of the interval
+    defined by [start,@value@], where `start` is defined by the parent
+    `_map-interval_`.
+
+    All other endpoints define the interval (prev,@value@].
+  -->
+  <template name="_endpoint_"
+            desc="Define an interval endpoint for _map-interval_">
+    <param name="@value@"
+           desc="Right endpoint of previous interval" />
+    <param name="@map@"
+           desc="Value to which interval is mapped" />
+
+    <param name="@mapindex@"
+           desc="Map value index">
+      <text></text>
+    </param>
+
+
+    <c:case>
+      <c:when name="__intval">
+        <c:lte>
+          <c:const value="@value@"
+                   type="float"
+                   desc="Previous interval right endpoint" />
+        </c:lte>
+      </c:when>
+
+      <c:value-of name="@map@"
+                  index="@mapindex@" />
+    </c:case>
+  </template>
+</package>
diff --git a/core/numeric/minmax.xml b/core/numeric/minmax.xml
new file mode 100644
index 0000000..aa6178f
--- /dev/null
+++ b/core/numeric/minmax.xml
@@ -0,0 +1,344 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="Numeric computations dealing with minimums and maximums">
+
+  <import package="../base" />
+
+
+  <function name="max" desc="Return the greater value">
+    <param name="max1" type="float" desc="First value to compare" />
+    <param name="max2" type="float" desc="Second value to compare" />
+
+    <c:cases>
+      <c:case>
+        <c:when name="max1">
+          <c:gte>
+            <c:value-of name="max2" />
+          </c:gte>
+        </c:when>
+
+        <c:value-of name="max1" />
+      </c:case>
+
+      <c:otherwise>
+        <c:value-of name="max2" />
+      </c:otherwise>
+    </c:cases>
+  </function>
+
+
+  <function name="min" desc="Return the lesser value">
+    <param name="min1" type="float" desc="First value to compare" />
+    <param name="min2" type="float" desc="Second value to compare" />
+
+    <c:cases>
+      <c:case>
+        <c:when name="min1">
+          <c:lte>
+            <c:value-of name="min2" />
+          </c:lte>
+        </c:when>
+
+        <c:value-of name="min1" />
+      </c:case>
+
+      <c:otherwise>
+        <c:value-of name="min2" />
+      </c:otherwise>
+    </c:cases>
+  </function>
+
+
+  <template name="_min-zero_" desc="Does not allow a value below 0 to be yielded">
+    <param name="@values@" desc="Calculation to apply restriction to" />
+    <param name="@label@" desc="Application label">
+      <!-- default empty -->
+      <text></text>
+    </param>
+
+    <c:apply name="max" label="@label@">
+      <c:arg name="max1">
+        <c:const value="0" type="integer" desc="Do not allow a value under 0" />
+      </c:arg>
+
+      <c:arg name="max2">
+        <param-copy name="@values@" />
+      </c:arg>
+    </c:apply>
+  </template>
+
+
+  <!-- trivial, but cuts turns a verbose rate-each and function application
+       into potentially a one-line template application -->
+  <template name="_quickMaxEach_" desc="Quick max application">
+    <param name="@class@" desc="Class to match on" />
+    <param name="@generates@" desc="Value to generate into" />
+    <param name="@a@" desc="Value a" />
+    <param name="@b@" desc="Value b" />
+
+    <param name="@yields@" desc="Yield variable">
+      <text>_</text>
+      <param-value name="@generates@" />
+    </param>
+
+    <rate-each class="@class@" accumulate="none" yields="@yields@" generates="@generates@" index="k">
+      <c:apply name="max">
+        <c:arg name="max1">
+          <c:value-of name="@a@" index="k" />
+        </c:arg>
+        <c:arg name="max2">
+          <c:value-of name="@b@" index="k" />
+        </c:arg>
+      </c:apply>
+    </rate-each>
+  </template>
+
+
+  <template name="_cap_"
+               desc="Cap a value at a given maximum">
+    <param name="@values@" desc="Value expression" />
+    <param name="@name@"   desc="Value to cap at" />
+
+    <param name="@index@" desc="Index of value to cap at">
+      <text></text>
+    </param>
+
+    <param name="@value@"
+              desc="Constant value to cap at (deprecated)" />
+
+    <param name="@desc@" desc="Value description">
+      <text>Maximum value</text>
+    </param>
+
+    <c:apply name="min">
+      <c:arg name="min1">
+        <!-- deprecated -->
+        <if name="@value@">
+          <c:const value="@value@" type="float" desc="@desc@" />
+        </if>
+        <unless name="@value@">
+          <c:value-of name="@name@"
+                      index="@index@"
+                      type="float"
+                      label="@desc@" />
+        </unless>
+      </c:arg>
+
+      <c:arg name="min2">
+        <param-copy name="@values@" />
+      </c:arg>
+    </c:apply>
+  </template>
+
+
+
+  <!-- useful in products -->
+  <template name="_min-value-of_" desc="Return a value or a minimum">
+    <param name="@name@"  desc="Name of value" />
+    <param name="@index@" desc="Value index" />
+    <param name="@label@" desc="Optional label" />
+    <param name="@min@"   desc="Minimum value" />
+
+    <c:apply name="max" label="{@label@}, minimum of 1">
+      <c:arg name="max1">
+        <c:const value="@min@" type="float" desc="Minimum value" />
+      </c:arg>
+
+      <c:arg name="max2">
+        <c:value-of name="@name@" index="@index@" label="@label@" />
+      </c:arg>
+    </c:apply>
+  </template>
+
+
+
+  <template name="_between_" desc="Assert that a value is within the given range, inclusive">
+    <param name="@min@" desc="Minimum value, inclusive" />
+    <param name="@max@" desc="Maximum value, inclusive" />
+    <param name="@desc@" desc="Description" />
+
+    <c:gte>
+      <c:const value="@min@" type="float" desc="{@desc@}; minimum" />
+    </c:gte>
+    <c:lte>
+      <c:const value="@max@" type="float" desc="{@desc@}; maximum" />
+    </c:lte>
+  </template>
+
+
+
+  <!--
+    Performs min and max bumping
+
+    TODO: max is incomplete (should have all the options of min)
+   -->
+  <template name="_bump_" desc="Bump a value if it does not reach a minimum">
+    <!-- if a minimum percentage -->
+    <param name="@percent@"   desc="Percent, as a whole value" />
+    <param name="@value@"     desc="Value to take percentage of, or take whole value of if no percent" />
+    <param name="@vindex@"    desc="Optional index for base" />
+    <param name="@name@"      desc="Provided value" />
+    <param name="@generates@" desc="Variable to generate into" />
+    <param name="@when@"      desc="Conditional bump" />
+    <param name="@class@"     desc="Class to match on" />
+    <param name="@keep@"      desc="Value of keep flag" />
+
+    <!-- alternative to @name@ -->
+    <param name="@const@" desc="Constant value, instead of named" />
+
+    <!-- max values for bumping down -->
+    <param name="@maxpercent@" desc="Maximum percent" />
+
+
+    <rate yields="_{@generates@}" keep="@keep@">
+      <c:sum of="@name@" index="k" generates="@generates@" desc="Bumped value">
+        <c:cases>
+          <!-- if a condition was provided, check it first -->
+          <if name="@when@">
+            <c:case>
+              <c:when name="@when@" index="k">
+                <c:eq>
+                  <c:value-of name="FALSE" />
+                </c:eq>
+              </c:when>
+
+              <!-- just return the value provided -->
+              <c:value-of name="@name@" index="k" />
+            </c:case>
+          </if>
+
+          <!-- if a condition was provided, check it first -->
+          <if name="@class@">
+            <c:case>
+              <c:when name="@class@" index="k">
+                <c:eq>
+                  <c:value-of name="FALSE" />
+                </c:eq>
+              </c:when>
+
+              <!-- just return the value provided -->
+              <c:const value="0" type="float" desc="Zero value" />
+            </c:case>
+          </if>
+
+          <!-- condition was not provided or did not match (we check for the
+               negative of the condition above so that this template will still
+               produce valid output event if @when@ is not provided; the
+               c:cases will be optimized away in that case) -->
+          <c:otherwise>
+            <c:let>
+              <c:values>
+                <c:value name="min" type="float" desc="Minimum">
+                  <unless name="@const@">
+                    <!-- this will produce a percentage of the provided value, unless no
+                         percent was given, in which case the entire value will be used
+                         -->
+                    <c:product>
+                      <!-- if a percent was provided, then take a certain percentage of the value -->
+                      <if name="@percent@">
+                        <!-- given index or default index? -->
+                        <if name="@vindex@">
+                          <c:value-of name="@value@" index="@vindex@" />
+                        </if>
+                        <unless name="@vindex@">
+                          <c:value-of name="@value@" index="k" />
+                        </unless>
+
+                        <c:quotient label="Percent as real number">
+                          <c:const value="@percent@" type="integer" desc="Whole percent" />
+                          <c:const value="100" type="integer" desc="Divisor to convert percent to real number" />
+                        </c:quotient>
+                      </if>
+
+                      <!-- otherwise, use the given value -->
+                      <unless name="@percent@">
+                        <c:value-of name="@name@" index="k" />
+                      </unless>
+                    </c:product>
+                  </unless>
+
+                  <if name="@const@">
+                    <c:const value="@const@" type="float" desc="Constant minimum value" />
+                  </if>
+                </c:value>
+              </c:values>
+
+              <c:let>
+                <c:values>
+                  <c:value name="minbumped" type="float" desc="Bumped to minimum">
+                    <t:maxreduce>
+                      <c:value-of name="min" />
+                      <c:value-of name="@name@" index="k" />
+                    </t:maxreduce>
+                  </c:value>
+
+                  <if name="@maxpercent@">
+                    <c:value name="max" type="float" desc="Maximum">
+                      <c:product>
+                        <!-- given index or default index? -->
+                        <if name="@vindex@">
+                          <c:value-of name="@value@" index="@vindex@" />
+                        </if>
+                        <unless name="@vindex@">
+                          <c:value-of name="@value@" index="k" />
+                        </unless>
+
+                        <c:quotient label="Max percent as real number">
+                          <c:const value="@maxpercent@" type="integer" desc="Whole max percent" />
+                          <c:const value="100" type="integer" desc="Divisor to convert max percent to real number" />
+                        </c:quotient>
+                      </c:product>
+                    </c:value>
+                  </if>
+                </c:values>
+
+                <c:cases>
+                  <!-- if a max percent was provided, then check to ensure it
+                       was not exceeded -->
+                  <if name="@maxpercent@">
+                    <c:case>
+                      <c:when name="@name@" index="k">
+                        <c:gt>
+                          <c:value-of name="max" />
+                        </c:gt>
+                      </c:when>
+
+                      <c:value-of name="max" />
+                    </c:case>
+                  </if>
+
+                  <!-- if no max, just yield the bumped value according to the minimum -->
+                  <c:otherwise>
+                    <c:value-of name="minbumped" />
+                  </c:otherwise>
+                </c:cases>
+              </c:let>
+            </c:let>
+          </c:otherwise>
+        </c:cases>
+      </c:sum>
+    </rate>
+  </template>
+</package>
+
diff --git a/core/numeric/percent.xml b/core/numeric/percent.xml
new file mode 100644
index 0000000..c161422
--- /dev/null
+++ b/core/numeric/percent.xml
@@ -0,0 +1,367 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="Percentage arithmetic">
+
+  <import package="../base" />
+
+
+  Converting between real and~percent is trivial:
+
+  $$
+    \begin{align}
+      r &amp;= p\over100;
+      p &amp;= 100r.
+    \end{align}
+  $$
+
+  ^[Percent difference] (also called ^[relative change]) describes
+  the~change in value as a~percentage (for example: $10$~is $50%$
+  of~$20$, and is~$100%$ of~$5$).  ^[Percent change] recognizes the
+  direction of the change (that is---$10$ is~$20$ reduced by~$50%$, so
+  $p=-50$).
+
+  Since ``change'' sounds like a~verb eclarative contexts, we use
+  the~term ^[percent difference] to mean ^[relative change]; it is
+  defined as:
+
+  $$\delta \over x = {{x_\beta - x_\alpha}\over{x_\alpha}} \times 100 = p.$$
+
+  Notice that this preserves the direction of the change.  So, in the
+  context of the previous example, we can say that $10$
+  \emph{reduces}~$20$ by~$50%$, giving~$p=-50%$.
+
+
+  <template name="_rtop_"
+            desc="Rational number to percentage">
+    <param name="@name@"  desc="Percentage to convert" />
+
+    <param name="@index@" desc="Index">
+      <text></text>
+    </param>
+
+    <param name="@desc@" desc="Optional description">
+      <text>Real number to convert into percentage</text>
+    </param>
+
+    <param name="@difference@"
+           desc="Whether to interpret as a percent difference (default
+                 false)">
+      <text>false</text>
+    </param>
+
+    <param name="@negate@" desc="Whether to negate percentage before
+                                 conversion">
+      <text>false</text>
+    </param>
+
+
+    <c:product>
+      <if name="@negate@" eq="true">
+        <c:const value="-1" desc="Negate result" />
+      </if>
+
+      <c:sum>
+        <c:value-of name="@name@" index="@index@" />
+
+        <if name="@difference@" eq="true">
+          <c:const value="-1"
+                   desc="Percent difference is relative (1 = 100%)" />
+        </if>
+      </c:sum>
+
+      <c:const value="100"
+               desc="Real to percentage" />
+    </c:product>
+  </template>
+
+
+
+  <template name="_ptor_" desc="Percent to rational number">
+    <param name="@name@" desc="Percentage to convert" />
+    <param name="@index@" desc="Index" />
+
+    <param name="@value@" desc="Constant (instead of named value)" />
+    <param name="@desc@" desc="Optional description for constant value">
+      <text>Percent to convert</text>
+    </param>
+
+    <param name="@difference@" desc="Whether to perform a percent
+                                     difference (default false)">
+      <text>false</text>
+    </param>
+
+    <param name="@negate@" desc="Whether to negate percentage before
+                                 conversion">
+      <text>false</text>
+    </param>
+
+    <c:sum>
+      <if name="@difference@" eq="true">
+        <c:const value="1"
+                 desc="Perform percent difference" />
+      </if>
+
+      <c:quotient>
+        <c:product>
+          <if name="@negate@" eq="true">
+            <c:value-of name="NEGATE" />
+          </if>
+
+          <if name="@name@">
+            <c:value-of name="@name@" index="@index@" />
+          </if>
+          <unless name="@name@">
+            <c:const value="@value@" type="float" desc="@desc@" />
+          </unless>
+        </c:product>
+
+        <c:const value="100" type="integer" desc="Convert to rational number" />
+      </c:quotient>
+    </c:sum>
+  </template>
+
+
+
+  <template name="_ptor-each_"
+            desc="Each percent to real number">
+    <param name="@name@"      desc="Percentage to convert" />
+    <param name="@generates@" desc="Value to generate into" />
+
+    <param name="@difference@"
+           desc="Whether to perform a percent difference">
+      <text>false</text>
+    </param>
+
+    <param name="@negate@" desc="Whether to negate percentage before
+                                 conversion">
+      <text>false</text>
+    </param>
+
+
+    <rate yields="_{@generates@}">
+      <c:sum of="@name@"
+             generates="@generates@" index="k"
+             desc="Real representation of {@name@}">
+        <t:ptor difference="@difference@" negate="@negate@"
+                name="@name@" index="k" />
+      </c:sum>
+    </rate>
+  </template>
+
+
+
+  <!--
+    Yields a total percent as an integer and rational number. The
+    integer is useful for displaying a total percent (e.g. for
+    credits/surcharges) and the rational number is useful for using
+    for multiplication in an equation.
+
+    Example:
+
+      <t:total-percent class="foo"
+                       to_int="fooPercentTotal" to_rat="fooTotal"
+                       index="k">
+        <with-param name="@values@">
+          <c:value-of name="fooA" index="k" />
+          <c:value-of name="fooB" index="k" />
+        </with-param>
+      </t:total-percent>
+
+    Let fooA = 5, fooB = -7
+
+    will yield:
+      - fooPercentTotal = -2
+      - fooTotal        = 0.98
+  -->
+  <template name="_total-percent_"
+            desc="Generates a total percentage as an integer and a
+                  rational number">
+    <!-- for use by @values@ -->
+    <param name="@index@" desc="Index variable" />
+
+    <param name="@class@" desc="Classification match" />
+    <param name="@to_int@" desc="Integer total to yield" />
+    <param name="@to_rat@" desc="Rational number to yield" />
+
+    <param name="@values@" desc="Value nodes" />
+
+    <param name="@yields_int@" desc="Yields for integer total">
+      <text>_</text>
+      <param-value name="@to_int@" />
+    </param>
+    <param name="@yields_rat@" desc="Yields for rational total">
+      <text>_</text>
+      <param-value name="@to_rat@" />
+    </param>
+
+
+    <!-- percent total (e.g. 7 for 7%) -->
+    <rate-each class="@class@" yields="@yields_int@"
+               generates="@to_int@" index="@index@">
+      <c:sum>
+        <param-copy name="@values@" />
+      </c:sum>
+    </rate-each>
+
+
+    <!-- as a rational number (to multiply by; e.g. 1.07 for 7%, 0.93 for -7%) -->
+    <rate-each class="@class@" yields="@yields_rat@"
+               generates="@to_rat@" index="k">
+      <c:sum>
+        <c:const value="1" desc="Relative percent" />
+        <t:ptor name="@to_int@" index="k" />
+      </c:sum>
+    </rate-each>
+  </template>
+
+
+
+  <!--
+    This template is useful for, say, when a coverage is partially
+    included, but additional coverage may be purchased. For example,
+    if 10% of coverage X is included, but you can purchase more Y if
+    10% is insufficient, then we can simply reduce coverage Y by 10%
+    of coverage X.
+
+    If the initial value is insufficient, then the value 0 will be
+    produced.
+  -->
+  <template name="_percent-reduction_"
+            desc="Reduces a set of values by a percentage of another set of values">
+    <param name="@index@" desc="Value index" />
+
+    <param name="@reduce@"    desc="Vector to reduce values of" />
+    <param name="@percentof@" desc="Vector to reduce values by a percentage of" />
+    <param name="@percent@"   desc="Percent reduction (as a float)" />
+    <param name="@minval@"    desc="Minimum value to reduce by (optional)" />
+    <param name="@condition@" desc="Boolean conditions (vector for
+                                    each index) under which reduction should be applied" />
+
+    <param name="@desc@" desc="Optional description">
+      <text>Percent to reduce by</text>
+    </param>
+
+    <param name="@label@" desc="Label for max() application">
+      <text>Reduced </text>
+      <param-value name="@reduce@" />
+      <text> by </text>
+      <param-value name="@percent@" />
+      <text>*</text>
+      <param-value name="@percentof@" />
+      <text> (cannot fall below 0)</text>
+    </param>
+
+
+    <c:apply name="max" label="@label@">
+      <!-- if the reduction produces a value less than 0, then simply return 0 -->
+      <c:arg name="max1">
+        <c:const value="0"
+                 desc="0 will be returned if the reduction is too great" />
+      </c:arg>
+
+      <c:arg name="max2">
+        <c:sum>
+          <c:value-of name="@reduce@" index="@index@"
+                      label="Initial value" />
+
+          <c:let>
+            <c:values>
+              <c:value name="reduce" type="float"
+                       desc="Value to reduce by, should condition be met">
+                <c:product>
+                  <c:const value="@percent@" desc="@desc@" />
+                  <c:value-of name="@percentof@" index="@index@" />
+                </c:product>
+              </c:value>
+            </c:values>
+
+            <t:negate>
+              <c:product label="Conditional reduction">
+                <if name="@condition@">
+                  <c:when name="@condition@" index="@index@">
+                    <c:eq>
+                      <c:value-of name="TRUE" />
+                    </c:eq>
+                  </c:when>
+                </if>
+
+                <if name="@minval@">
+                  <c:apply name="max" max1="reduce">
+                    <c:arg name="max2">
+                      <c:const value="@minval@"
+                               desc="Minimum value to reduce by" />
+                    </c:arg>
+                  </c:apply>
+                </if>
+                <unless name="@minval@">
+                  <c:value-of name="reduce" label="Value to reduce by" />
+                </unless>
+              </c:product>
+            </t:negate>
+          </c:let>
+        </c:sum>
+      </c:arg>
+    </c:apply>
+  </template>
+
+
+
+  <template name="_percent-of-set_"
+            desc="Returns the percent of a set of given values">
+    <param name="@class@"     desc="Class match" />
+    <param name="@of@"        desc="Vector of values to take the percent of" />
+    <param name="@percent@"   desc="Percent reduction (as a float)" />
+    <param name="@generates@" desc="Value to generate" />
+
+    <param name="@yields@" desc="Yield">
+      <param-value name="@generates@" />
+      <text>Total</text>
+    </param>
+
+
+    <rate-each class="@class@" yields="@yields@"
+               generates="@generates@" index="k">
+      <c:apply name="round">
+        <c:arg name="roundval">
+          <c:apply name="max">
+            <!-- if the reduction produces a value less than 0, then simply return 0 -->
+            <c:arg name="max1">
+              <c:const value="0"
+                       desc="0 will be returned if the reduction is too great" />
+            </c:arg>
+
+            <c:arg name="max2">
+              <c:product label="Percent value">
+                <c:const value="@percent@" desc="Percent to reduce by" />
+
+                <c:value-of name="@of@" index="k"
+                            label="Value to take percentage of" />
+              </c:product>
+            </c:arg>
+          </c:apply>
+        </c:arg>
+      </c:apply>
+    </rate-each>
+  </template>
+</package>
diff --git a/core/numeric/round.xml b/core/numeric/round.xml
new file mode 100644
index 0000000..319c683
--- /dev/null
+++ b/core/numeric/round.xml
@@ -0,0 +1,264 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="Numeric computations dealing with rounding">
+
+  <import package="../base" />
+
+
+  <!-- template version of round function, supporting arbitrary nodes -->
+  <template name="_round_" desc="Round to the nearest integer">
+    <param name="@values@" desc="Calculation to round (single node)" />
+    <param name="@precision@" desc="Level of precision after decimal point">
+      <text>0</text>
+    </param>
+
+    <c:apply name="round_real" label="Rounded value">
+      <c:arg name="round_real_val">
+        <c:sum>
+          <param-copy name="@values@" />
+        </c:sum>
+      </c:arg>
+
+      <c:arg name="round_real_n">
+        <c:expt>
+          <c:const value="10" type="integer" desc="Decimal base" />
+          <c:const value="@precision@" type="integer" desc="Exponent" />
+        </c:expt>
+      </c:arg>
+    </c:apply>
+  </template>
+
+
+  <!--
+    Let expression exposing one or more of ceiling, floor, and nearest
+    integer of given name
+
+    This is our Swiss Army knife of rounding values.
+
+    The values @high@, @low@, and @nearest@ (can specify all or just one)
+    represent the names of the variables to store, respectively, the
+    ceiling, floor, and nearest integer to the value @name@. Those values
+    are then accessible within the scope of the let expression.
+
+    An @exp@ can be used to provide a base-10 exponent to move the decimal
+    place (negative is right, positive left) before performing rounding.
+
+    The @step@ allows specifying an arbitrary value to round toward. For
+    example, a @step@ of 1500 rounds in 1500 increments. This may be used
+    with @exp@, in which case the step will apply to @name@ after its value
+    has been exponentiated.
+  -->
+  <template name="_let-round_"
+               desc="Produce floor and ceiling values">
+    <param name="@values@"
+              desc="Calculation subject to let expression" />
+
+    <param name="@name@"
+              desc="Value to round" />
+
+    <param name="@index@"
+              desc="Optional value index">
+      <text></text>
+    </param>
+
+    <param name="@high@"
+              desc="Name to yield ceil'd value into" />
+
+    <param name="@low@"
+              desc="Name to yield floor'd value into" />
+
+    <param name="@nearest@"
+              desc="Name to yield nearest integer to value into" />
+
+
+    <param name="@exp@"
+              desc="Offset digits before rounding">
+      <text>#0</text>
+    </param>
+
+    <param name="@step@"
+              desc="Arbitrary step">
+      <text>#1</text>
+    </param>
+
+
+
+    <c:let>
+      <c:values>
+        <c:value name="__div"
+                 type="float"
+                 desc="Exponential/step divisor">
+          <c:product>
+            <c:expt>
+              <c:const value="10" type="integer"
+                       desc="Decimal base" />
+              <c:value-of name="@exp@" />
+            </c:expt>
+
+            <c:value-of name="@step@" />
+          </c:product>
+        </c:value>
+      </c:values>
+
+      <c:let>
+        <c:values>
+          <c:value name="__adjusted"
+                   type="float"
+                   desc="Value adjusted for 10^@exp@ or step @step@">
+            <c:quotient>
+              <c:value-of name="@name@" index="@index@" />
+              <c:value-of name="__div" />
+            </c:quotient>
+          </c:value>
+        </c:values>
+
+        <c:let>
+          <c:values>
+            <if name="@high@">
+              <c:value name="@high@" type="float"
+                       desc="Ceiling of adjusted {@name@}">
+                <c:product>
+                  <c:ceil>
+                    <c:value-of name="__adjusted" />
+                  </c:ceil>
+
+                  <c:value-of name="__div" />
+                </c:product>
+              </c:value>
+            </if>
+
+            <if name="@low@">
+              <c:value name="@low@" type="float"
+                       desc="floor of adjusted {@name@}">
+                <c:product>
+                  <c:floor>
+                    <c:value-of name="__adjusted" />
+                  </c:floor>
+
+                  <c:value-of name="__div" />
+                </c:product>
+              </c:value>
+            </if>
+
+            <if name="@nearest@">
+              <c:value name="@nearest@" type="float"
+                       desc="nearest integer to adjusted {@name@}">
+                <c:product>
+                  <t:round>
+                    <c:value-of name="__adjusted" />
+                  </t:round>
+
+                  <c:value-of name="__div" />
+                </c:product>
+              </c:value>
+            </if>
+          </c:values>
+
+          <!-- body subject to let expression -->
+          <param-copy name="@values@" />
+        </c:let>
+      </c:let>
+    </c:let>
+  </template>
+
+
+  <!-- ceil if > 0.5, otherwise floor; round() in most languages -->
+  <!-- TODO: support left and right delimiters; do \floor \ceil -->
+  <function name="round" desc="Round to the nearest integer">
+    <param name="roundval" type="float" desc="Value to round to nearest integer" />
+
+    <!-- By raising by 0.5 and taking the floor() of the resulting value, we
+         can produce values consistent with common round() implementations. It
+         may be easier to think of this as ceil( val - 0.4 ), but representing
+         subtraction is more verbose than addition, so a floor() implementation
+         is used instead. -->
+    <c:floor>
+      <c:sum>
+        <c:value-of name="roundval" />
+        <c:const value="0.5" type="float" desc="Raises value in a manner that it can be properly rounded by a floor" />
+      </c:sum>
+    </c:floor>
+  </function>
+
+
+  <function name="round_real" desc="Round to the nearest 1/n">
+    <param name="round_real_val" type="float"   desc="Value to round" />
+    <param name="round_real_n"   type="integer" desc="Round to 1/n" />
+
+    <!-- we can achieve this by multiplying by N, rounding, and then dividing
+         by the same N -->
+    <c:quotient>
+      <c:apply name="round">
+        <c:arg name="roundval">
+          <c:product>
+            <c:value-of name="round_real_val" />
+            <c:value-of name="round_real_n" />
+          </c:product>
+        </c:arg>
+      </c:apply>
+
+      <c:value-of name="round_real_n" />
+    </c:quotient>
+  </function>
+
+
+  <!-- this is simply a shorthand for round_real -->
+  <function name="round_cents" desc="Round to the nearest penny">
+    <param name="round_cents_val" type="float" desc="Monetary value" />
+
+    <c:apply name="round_real">
+      <c:arg name="round_real_n">
+        <c:const value="100" type="integer" desc="Round to the nearest 100th" />
+      </c:arg>
+
+      <c:arg name="round_real_val">
+        <c:value-of name="round_cents_val" />
+      </c:arg>
+    </c:apply>
+  </function>
+
+
+  <template name="_ceil-n_" desc="Ceiling on the last n digits">
+    <param name="@values@" desc="Calculation to use (single node)" />
+    <param name="@digits@" desc="Level of precision after decimal point" />
+
+    <c:product>
+      <c:ceil>
+        <c:quotient>
+          <param-copy name="@values@" />
+          <c:expt>
+            <c:const value="10" type="integer" desc="Decimal base" />
+            <c:const value="@digits@" type="integer" desc="Number of digits" />
+          </c:expt>
+        </c:quotient>
+      </c:ceil>
+
+      <c:expt>
+        <c:const value="10" type="integer" desc="Decimal base" />
+        <c:const value="@digits@" type="integer" desc="Number of digits" />
+      </c:expt>
+    </c:product>
+  </template>
+</package>
+
diff --git a/core/state.xml b/core/state.xml
new file mode 100644
index 0000000..a7a5522
--- /dev/null
+++ b/core/state.xml
@@ -0,0 +1,195 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2017, 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="United States">
+
+  <import package="base" />
+  <import package="vector/stub" />
+
+
+  <section title="States">
+    The type \ref{State} contains each of the States in the United~States,
+      with the addition of Washington~DC.
+    The States are 1-indexed and sorted by \emph{abbreviation}\footnote{
+      States are sorted by abbreviation rather than the State name beacuse
+        they are most frequently referenced as such.}.
+
+    <typedef name="State" desc="States">
+      <enum type="integer">
+        <item name="STATE_NONE" value="0"  desc="No State"       />
+        <item name="STATE_AK"   value="1"  desc="Alaska"         />
+        <item name="STATE_AL"   value="2"  desc="Alabama"        />
+        <item name="STATE_AR"   value="3"  desc="Arkansas"       />
+        <item name="STATE_AZ"   value="4"  desc="Arizona"        />
+        <item name="STATE_CA"   value="5"  desc="California"     />
+        <item name="STATE_CO"   value="6"  desc="Colorado"       />
+        <item name="STATE_CT"   value="7"  desc="Connecticut"    />
+        <item name="STATE_DC"   value="8"  desc="Washington DC"  />
+        <item name="STATE_DE"   value="9"  desc="Delaware"       />
+        <item name="STATE_FL"   value="10" desc="Florida"        />
+        <item name="STATE_GA"   value="11" desc="Georgia"        />
+        <item name="STATE_HI"   value="12" desc="Hawaii"         />
+        <item name="STATE_IA"   value="13" desc="Iowa"           />
+        <item name="STATE_ID"   value="14" desc="Idaho"          />
+        <item name="STATE_IL"   value="15" desc="Illinois"       />
+        <item name="STATE_IN"   value="16" desc="Indiana"        />
+        <item name="STATE_KS"   value="17" desc="Kansas"         />
+        <item name="STATE_KY"   value="18" desc="Kentucky"       />
+        <item name="STATE_LA"   value="19" desc="Louisiana"      />
+        <item name="STATE_MA"   value="20" desc="Massachusetts"  />
+        <item name="STATE_MD"   value="21" desc="Maryland"       />
+        <item name="STATE_ME"   value="22" desc="Maine"          />
+        <item name="STATE_MI"   value="23" desc="Michigan"       />
+        <item name="STATE_MN"   value="24" desc="Minnesota"      />
+        <item name="STATE_MO"   value="25" desc="Missouri"       />
+        <item name="STATE_MS"   value="26" desc="Mississippi"    />
+        <item name="STATE_MT"   value="27" desc="Montana"        />
+        <item name="STATE_NC"   value="28" desc="North Carolina" />
+        <item name="STATE_ND"   value="29" desc="North Dakota"   />
+        <item name="STATE_NE"   value="30" desc="Nebraska"       />
+        <item name="STATE_NH"   value="31" desc="New Hampshire"  />
+        <item name="STATE_NJ"   value="32" desc="New Jersey"     />
+        <item name="STATE_NM"   value="33" desc="New Mexico"     />
+        <item name="STATE_NV"   value="34" desc="Nevada"         />
+        <item name="STATE_NY"   value="35" desc="New York"       />
+        <item name="STATE_OH"   value="36" desc="Ohio"           />
+        <item name="STATE_OK"   value="37" desc="Oklahoma"       />
+        <item name="STATE_OR"   value="38" desc="Oregon"         />
+        <item name="STATE_PA"   value="39" desc="Pennsylvania"   />
+        <item name="STATE_RI"   value="40" desc="Rhode Island"   />
+        <item name="STATE_SC"   value="41" desc="South Carolina" />
+        <item name="STATE_SD"   value="42" desc="South Dakota"   />
+        <item name="STATE_TN"   value="43" desc="Tennessee"      />
+        <item name="STATE_TX"   value="44" desc="Texas"          />
+        <item name="STATE_UT"   value="45" desc="Utah"           />
+        <item name="STATE_VA"   value="46" desc="Virginia"       />
+        <item name="STATE_VT"   value="47" desc="Vermont"        />
+        <item name="STATE_WA"   value="48" desc="Washington"     />
+        <item name="STATE_WI"   value="49" desc="Wisconsin"      />
+        <item name="STATE_WV"   value="50" desc="West Virginia"  />
+        <item name="STATE_WY"   value="51" desc="Wyoming"        />
+      </enum>
+    </typedef>
+  </section>
+
+
+
+  <section title="State Operations">
+    For operations that involve taking values of all states where order
+      matters (e.g. for index alignment),
+        querying the symbol table isn't appropriate,
+          as it does not guarantee order.
+    \ref{_for-each-state_} can be used for that purpose;
+      it exposes the following template values to its body:
+
+    \begin{enumerate}
+      \item \tt{@state_const@} contains the State constant;
+      \item \tt{@state_upper@} contains the uppercase two-letter State
+              abbreviation;
+      \item \tt{@state_lower@} contains the lowercase two-letter State
+              abbreviation; and
+      \item \tt{@state_name@} contains the full state name.
+    \end{enumerate}
+
+    This can be used to generate a ^[State vector] by mapping an iteration
+      index to the State constant~\tt{@state_const@}.
+    Note that \ref{STATE_NONE} is not in the list.
+
+    <template name="_for-each-state_"
+              desc="Apply body for each State">
+      <param name="@values@" desc="Body to apply per State" />
+
+
+      <inline-template>
+        <for-each>
+          <set state_const="STATE_AK" state_upper="AK" state_lower="ak" state_name="Alaska"         />
+          <set state_const="STATE_AL" state_upper="AL" state_lower="al" state_name="Alabama"        />
+          <set state_const="STATE_AR" state_upper="AR" state_lower="ar" state_name="Arkansas"       />
+          <set state_const="STATE_AZ" state_upper="AZ" state_lower="az" state_name="Arizona"        />
+          <set state_const="STATE_CA" state_upper="CA" state_lower="ca" state_name="California"     />
+          <set state_const="STATE_CO" state_upper="CO" state_lower="co" state_name="Colorado"       />
+          <set state_const="STATE_CT" state_upper="CT" state_lower="ct" state_name="Connecticut"    />
+          <set state_const="STATE_DC" state_upper="DC" state_lower="dc" state_name="Washington DC"  />
+          <set state_const="STATE_DE" state_upper="DE" state_lower="de" state_name="Delaware"       />
+          <set state_const="STATE_FL" state_upper="FL" state_lower="fl" state_name="Florida"        />
+          <set state_const="STATE_GA" state_upper="GA" state_lower="ga" state_name="Georgia"        />
+          <set state_const="STATE_HI" state_upper="HI" state_lower="hi" state_name="Hawaii"         />
+          <set state_const="STATE_IA" state_upper="IA" state_lower="ia" state_name="Iowa"           />
+          <set state_const="STATE_ID" state_upper="ID" state_lower="id" state_name="Idaho"          />
+          <set state_const="STATE_IL" state_upper="IL" state_lower="il" state_name="Illinois"       />
+          <set state_const="STATE_IN" state_upper="IN" state_lower="in" state_name="Indiana"        />
+          <set state_const="STATE_KS" state_upper="KS" state_lower="ks" state_name="Kansas"         />
+          <set state_const="STATE_KY" state_upper="KY" state_lower="ky" state_name="Kentucky"       />
+          <set state_const="STATE_LA" state_upper="LA" state_lower="la" state_name="Louisiana"      />
+          <set state_const="STATE_MA" state_upper="MA" state_lower="ma" state_name="Massachusetts"  />
+          <set state_const="STATE_MD" state_upper="MD" state_lower="md" state_name="Maryland"       />
+          <set state_const="STATE_ME" state_upper="ME" state_lower="me" state_name="Maine"          />
+          <set state_const="STATE_MI" state_upper="MI" state_lower="mi" state_name="Michigan"       />
+          <set state_const="STATE_MN" state_upper="MN" state_lower="mn" state_name="Minnesota"      />
+          <set state_const="STATE_MO" state_upper="MO" state_lower="mo" state_name="Missouri"       />
+          <set state_const="STATE_MS" state_upper="MS" state_lower="ms" state_name="Mississippi"    />
+          <set state_const="STATE_MT" state_upper="MT" state_lower="mt" state_name="Montana"        />
+          <set state_const="STATE_NC" state_upper="NC" state_lower="nc" state_name="North Carolina" />
+          <set state_const="STATE_ND" state_upper="ND" state_lower="nd" state_name="North Dakota"   />
+          <set state_const="STATE_NE" state_upper="NE" state_lower="ne" state_name="Nebraska"       />
+          <set state_const="STATE_NH" state_upper="NH" state_lower="nh" state_name="New Hampshire"  />
+          <set state_const="STATE_NJ" state_upper="NJ" state_lower="nj" state_name="New Jersey"     />
+          <set state_const="STATE_NM" state_upper="NM" state_lower="nm" state_name="New Mexico"     />
+          <set state_const="STATE_NV" state_upper="NV" state_lower="nv" state_name="Nevada"         />
+          <set state_const="STATE_NY" state_upper="NY" state_lower="ny" state_name="New York"       />
+          <set state_const="STATE_OH" state_upper="OH" state_lower="oh" state_name="Ohio"           />
+          <set state_const="STATE_OK" state_upper="OK" state_lower="ok" state_name="Oklahoma"       />
+          <set state_const="STATE_OR" state_upper="OR" state_lower="or" state_name="Oregon"         />
+          <set state_const="STATE_PA" state_upper="PA" state_lower="pa" state_name="Pennsylvania"   />
+          <set state_const="STATE_RI" state_upper="RI" state_lower="ri" state_name="Rhode Island"   />
+          <set state_const="STATE_SC" state_upper="SC" state_lower="sc" state_name="South Carolina" />
+          <set state_const="STATE_SD" state_upper="SD" state_lower="sd" state_name="South Dakota"   />
+          <set state_const="STATE_TN" state_upper="TN" state_lower="tn" state_name="Tennessee"      />
+          <set state_const="STATE_TX" state_upper="TX" state_lower="tx" state_name="Texas"          />
+          <set state_const="STATE_UT" state_upper="UT" state_lower="ut" state_name="Utah"           />
+          <set state_const="STATE_VA" state_upper="VA" state_lower="va" state_name="Virginia"       />
+          <set state_const="STATE_VT" state_upper="VT" state_lower="vt" state_name="Vermont"        />
+          <set state_const="STATE_WA" state_upper="WA" state_lower="wa" state_name="Washington"     />
+          <set state_const="STATE_WI" state_upper="WI" state_lower="wi" state_name="Wisconsin"      />
+          <set state_const="STATE_WV" state_upper="WV" state_lower="wv" state_name="West Virginia"  />
+          <set state_const="STATE_WY" state_upper="WY" state_lower="wy" state_name="Wyoming"        />
+        </for-each>
+
+        <param-copy name="@values@" />
+      </inline-template>
+    </template>
+
+
+    To ease iteration though ^[State vector]s generated with
+      \ref{_for-each-state_},
+        a 52~vector \ref{NVEC_STATE_ALL} and classification \ref{state-all}
+        are provided.
+
+    <t:n-vector n="52" name="NVEC_STATE_ALL" value="1" />
+
+    <classify as="state-all" yields="stateAll"
+              desc="Every State in a State vector">
+      <match on="NVEC_STATE_ALL" />
+    </classify>
+  </section>
+</package>
diff --git a/core/states.xml b/core/states.xml
new file mode 120000
index 0000000..35a1a26
--- /dev/null
+++ b/core/states.xml
@@ -0,0 +1 @@
+state.xml
\ No newline at end of file
diff --git a/core/symbol.xml b/core/symbol.xml
new file mode 100644
index 0000000..bef7e17
--- /dev/null
+++ b/core/symbol.xml
@@ -0,0 +1,92 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="Symbol Introspection">
+
+  TAME provides powerful symbol table introspection features to assist
+    with metaprogramming.
+  This package provides templates to abstract some of those features and
+    make them easier to use.
+
+  If introspecting on symbols that are defined within the same package,
+    be aware that the symbols may not be available until a future pass;
+      \tt{expand-sequence} may be helpful in that situation.
+
+
+  \ref{_if-symbol_} is a general-purpose template to conditionally expand a
+    body if a symbol matches a given predicate.
+  The predicates currently supported are \tt{type} and \tt{dim}.
+  If no predicate is provided,
+    then the body will be expanded if the symbol exists.\footnote{
+      This is equivalent to \tt{type=""}.}
+
+  <template name="_if-symbol_"
+            desc="Expand body if symbol predicate matches">
+    <param name="@values@" desc="Condition body" />
+    <param name="@name@"   desc="Symbol name" />
+    <param name="@type@"   desc="Symbol type predicate" />
+    <param name="@dim@"    desc="Symbol dimensions predicate" />
+
+    <param name="@_sym_type@" desc="Symbol type lookup">
+      <param-sym-value name="@name@" value="type" ignore-missing="true" />
+    </param>
+    <param name="@_sym_dim@" desc="Symbol dimensions lookup">
+      <param-sym-value name="@name@" value="dim" ignore-missing="true" />
+    </param>
+
+
+    <if name="@type@">
+      <if name="@_sym_type@" eq="@type@">
+        <param-copy name="@values@" />
+      </if>
+    </if>
+    <unless name="@type@">
+      <if name="@dim@">
+        <if name="@_sym_dim@" eq="@dim@">
+          <param-copy name="@values@" />
+        </if>
+      </if>
+
+      <!-- default simply checks to see if the symbol exists -->
+      <unless name="@dim@">
+        <unless name="@_sym_type@" eq="">
+          <param-copy name="@values@" />
+        </unless>
+      </unless>
+    </unless>
+  </template>
+
+
+  \ref{_if-defined_} is the same as \ref{_if-symbol_} with no predicates;
+    it provides more comfortable terminology for a common use case.
+
+  <template name="_if-defined_"
+            desc="Expand body if symbol is defined">
+    <param name="@values@" desc="Condition body" />
+    <param name="@name@"   desc="Symbol name" />
+
+    <t:if-symbol name="@name@">
+      <param-copy name="@values@" />
+    </t:if-symbol>
+  </template>
+</package>
diff --git a/core/tdat.xml b/core/tdat.xml
new file mode 100644
index 0000000..f944420
--- /dev/null
+++ b/core/tdat.xml
@@ -0,0 +1,44 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  core="true"
+  desc="Territory data support (used in conjunction with tdat script)">
+
+  <import package="base" export="true" />
+
+
+  <template name="_terr-code_" desc="Defines a territory code for a given classification">
+    <param name="@class@"     desc="Classification" />
+    <param name="@code@"      desc="Territory code" />
+    <param name="@generates@" desc="Generator" />
+
+    <param name="@yields@" desc="Yield var">
+      <text>_</text>
+      <param-value name="@generates@" />
+    </param>
+
+
+    <rate-each class="@class@" accumulate="none" yields="@yields@" generates="@generates@" index="k">
+      <c:const value="@code@" type="integer" desc="Territory code" />
+    </rate-each>
+  </template>
+</package>
+
diff --git a/core/test/core/aggregate.xml b/core/test/core/aggregate.xml
new file mode 100644
index 0000000..f5e7860
--- /dev/null
+++ b/core/test/core/aggregate.xml
@@ -0,0 +1,264 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         desc="Aggregate Package Specification">
+
+  <import package="../../base" />
+  <import package="../../test/spec" />
+
+  <import package="../../base" />
+  <import package="../../vector/cmatch" />
+  <import package="../../vector/stub" />
+
+  <import package="../../aggregate" />
+
+
+  <rate-each class="nclass3"
+             generates="aggregateGen1" index="k">
+    <c:const value="1" desc="Constant value" />
+  </rate-each>
+
+  <rate-each class="nclass3"
+             generates="aggregateGen2" index="k">
+    <c:value-of name="k" />
+  </rate-each>
+
+
+  <rate yields="aggregateRate1">
+    <c:const value="1" desc="Constant value" />
+  </rate>
+
+  <rate yields="aggregateRate2">
+    <c:const value="3" desc="Constant value" />
+  </rate>
+
+
+  <classify as="agg-class-1"
+            desc="Aggregate test 1">
+    <match on="AGG_1VEC" />
+  </classify>
+
+  <classify as="agg-class-2"
+            desc="Aggregate test 2">
+    <t:match-gt on="AGG_INCVEC" const="0" />
+  </classify>
+
+
+  <t:n-vector n="3" name="AGG_1VEC" value="1" />
+
+  <const name="AGG_INCVEC" desc="Incrementing vector">
+    <item value="0" />
+    <item value="1" />
+    <item value="2" />
+  </const>
+
+
+
+  <t:describe name="aggregate template">
+    <t:describe name="_aggregate-rate-each_">
+      <t:aggregate-rate-each class="nclass3" yields="yieldAggReEmpty"
+                             prefix="doesNotExist"
+                             generates="genAggReEmpty" />
+
+      <t:aggregate-rate-each class="nclass3" yields="yieldAggReNonEmpty"
+                             prefix="aggregateGen"
+                             generates="genAggReNonEmpty" />
+
+
+      <t:describe name="with no symbols">
+        <t:it desc="produces 0">
+          <t:given>
+            <c:sum>
+              <c:value-of name="yieldAggReEmpty" />
+              <c:sum of="genAggReEmpty" />
+            </c:sum>
+          </t:given>
+
+          <t:expect>
+            <t:match-result eq="0" />
+          </t:expect>
+        </t:it>
+      </t:describe>
+
+
+      <t:describe name="with symbols">
+        <t:it desc="sums respective index of each symbol">
+          <t:given>
+            <c:sum of="genAggReNonEmpty" />
+          </t:given>
+
+          <t:expect>
+            <!-- 1 + 2 + 3 -->
+            <t:match-result eq="6" />
+          </t:expect>
+        </t:it>
+
+
+        <t:it desc="yields sum of symbols">
+          <t:given>
+            <c:value-of name="yieldAggReNonEmpty" />
+          </t:given>
+
+          <t:expect>
+            <!-- same as above -->
+            <t:match-result eq="6" />
+          </t:expect>
+        </t:it>
+      </t:describe>
+    </t:describe>
+
+
+    <t:describe name="_aggregate-rate_">
+      <t:aggregate-rate prefix="doesNotExist"  yields="yieldAggRateEmpty" />
+      <t:aggregate-rate prefix="aggregateRate" yields="yieldAggRateNonEmpty" />
+
+      <t:describe name="with no symbols">
+        <t:it desc="yields 0">
+          <t:given>
+            <c:value-of name="yieldAggRateEmpty" />
+          </t:given>
+
+          <t:expect>
+            <t:match-result eq="0" />
+          </t:expect>
+        </t:it>
+      </t:describe>
+
+
+      <t:describe name="with symbols">
+        <t:it desc="yields sum of symbols">
+          <t:given>
+            <c:value-of name="yieldAggRateNonEmpty" />
+          </t:given>
+
+          <t:expect>
+            <t:match-result eq="4" />
+          </t:expect>
+        </t:it>
+      </t:describe>
+    </t:describe>
+
+
+    <t:describe name="_aggregate-classify_">
+      <t:describe name="as a univiersal quantifier">
+        <t:aggregate-classify prefix="does-not-exist" as="class-agg-univ-empty"
+                              desc="Aggregate universal class empty test"
+                              yields="classAggUnivEmpty" />
+        <t:aggregate-classify prefix="agg-class-" as="class-agg-univ-nonempty"
+                              desc="Aggregate class nonempty test"
+                              yields="classAggUnivNonEmpty" />
+
+        <t:describe name="with no symbols">
+          <t:it desc="produces scalar 1">
+            <t:given>
+              <c:value-of name="classAggUnivEmpty" />
+            </t:given>
+
+            <t:expect>
+              <t:match-result eq="1" />
+            </t:expect>
+          </t:it>
+        </t:describe>
+
+
+        <t:describe name="with symbols">
+          <t:it desc="generates matching class">
+            <rate-each class="class-agg-univ-nonempty"
+                       yields="aggUnivNonEmptyCheck"
+                       index="k">
+              <c:const value="1" desc="Truth check" />
+            </rate-each>
+
+            <t:expect>
+              <!-- two non-zero in AGG_INCVEC -->
+              <t:match-eq on="aggUnivNonEmptyCheck" const="2" />
+            </t:expect>
+          </t:it>
+
+
+          <t:it desc="produces vector">
+            <t:given>
+              <c:sum of="classAggUnivNonEmpty" />
+            </t:given>
+
+            <t:expect>
+              <!-- two non-zero in AGG_INCVEC -->
+              <t:match-result eq="2" />
+            </t:expect>
+          </t:it>
+        </t:describe>
+      </t:describe>
+
+
+      <t:describe name="as a existential quantifier">
+        <t:aggregate-classify prefix="does-not-exist" as="class-agg-exist-empty"
+                              desc="Aggregate existersal class empty test"
+                              yields="classAggExistEmpty"
+                              any="true" />
+        <t:aggregate-classify prefix="agg-class-" as="class-agg-exist-nonempty"
+                              desc="Aggregate class nonempty test"
+                              yields="classAggExistNonEmpty"
+                              any="true" />
+
+        <t:describe name="with no symbols">
+          <t:it desc="produces scalar 0">
+            <t:given>
+              <c:value-of name="classAggExistEmpty" />
+            </t:given>
+
+            <t:expect>
+              <t:match-result eq="0" />
+            </t:expect>
+          </t:it>
+        </t:describe>
+
+
+        <t:describe name="with symbols">
+          <t:it desc="generates matching class">
+            <rate-each class="class-agg-exist-nonempty"
+                       yields="aggExistNonEmptyCheck"
+                       index="k">
+              <c:const value="1" desc="Truth check" />
+            </rate-each>
+
+            <t:expect>
+              <!-- all match in AGG_1VEC -->
+              <t:match-eq on="aggExistNonEmptyCheck" const="3" />
+            </t:expect>
+          </t:it>
+
+
+          <t:it desc="produces vector">
+            <t:given>
+              <c:sum of="classAggExistNonEmpty" />
+            </t:given>
+
+            <t:expect>
+              <!-- all match in AGG_1VEC -->
+              <t:match-result eq="3" />
+            </t:expect>
+          </t:it>
+        </t:describe>
+      </t:describe>
+    </t:describe>
+  </t:describe>
+</package>
diff --git a/core/test/core/insurance.xml b/core/test/core/insurance.xml
new file mode 100644
index 0000000..cf052c2
--- /dev/null
+++ b/core/test/core/insurance.xml
@@ -0,0 +1,208 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2016 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  title="Test Insurance Abstractions">
+
+  <import package="../spec" />
+
+  <import package="../../base" />
+  <import package="../../dummy" />
+  <import package="../../vector/cmatch" />
+
+  <!-- SUT -->
+  <import package="../../insurance" />
+
+
+  <t:describe name="_premium_">
+    <t:describe name="@round@">
+      <t:it desc="Performs no round when 'none'">
+        <t:premium class="length1"
+                   generates="premRoundNone" index="k"
+                   round="none">
+          <c:value-of name="#0.5" />
+        </t:premium>
+
+        <t:given name="premRoundNone" />
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+               <c:value-of name="#0.5" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="Rounds to nearest integer when 'dollar' (up)">
+        <t:premium class="length1"
+                   generates="premRoundDollarUp" index="k"
+                   round="dollar">
+          <c:value-of name="#1.5" />
+        </t:premium>
+
+        <t:given name="premRoundDollarUp" />
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+               <c:value-of name="#2" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="Rounds to nearest integer when 'dollar' (down)">
+        <t:premium class="length1"
+                   generates="premRoundDollarDown" index="k"
+                   round="dollar">
+          <c:value-of name="#1.4" />
+        </t:premium>
+
+        <t:given name="premRoundDollarDown" />
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+               <c:value-of name="#1" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="Rounds to nearest cent when 'cent' (up)">
+        <t:premium class="length1"
+                   generates="premRoundCentUp" index="k"
+                   round="cent">
+          <c:value-of name="#1.505" />
+        </t:premium>
+
+        <t:given name="premRoundCentUp" />
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+               <c:value-of name="#1.51" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="Rounds to nearest cent when 'cent' (down)">
+        <t:premium class="length1"
+                   generates="premRoundCentDown" index="k"
+                   round="cent">
+          <c:value-of name="#1.504" />
+        </t:premium>
+
+        <t:given name="premRoundCentDown" />
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+               <c:value-of name="#1.50" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="Performs ceiling when 'up' (low)">
+        <t:premium class="length1"
+                   generates="premRoundCeilLow" index="k"
+                   round="up">
+          <c:value-of name="#1.1" />
+        </t:premium>
+
+        <t:given name="premRoundCeilLow" />
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+               <c:value-of name="#2" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="Performs ceiling when 'up' (high)">
+        <t:premium class="length1"
+                   generates="premRoundCeilHigh" index="k"
+                   round="up">
+          <c:value-of name="#1.7" />
+        </t:premium>
+
+        <t:given name="premRoundCeilHigh" />
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+               <c:value-of name="#2" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="Performs floor when 'down' (low)">
+        <t:premium class="length1"
+                   generates="premRoundFloorLow" index="k"
+                   round="down">
+          <c:value-of name="#1.1" />
+        </t:premium>
+
+        <t:given name="premRoundFloorLow" />
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+               <c:value-of name="#1" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="Performs floor when 'down' (high)">
+        <t:premium class="length1"
+                   generates="premRoundFloorHigh" index="k"
+                   round="down">
+          <c:value-of name="#1.7" />
+        </t:premium>
+
+        <t:given name="premRoundFloorHigh" />
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+               <c:value-of name="#1" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+  </t:describe>
+</package>
diff --git a/core/test/core/numeric/convert.xml b/core/test/core/numeric/convert.xml
new file mode 100644
index 0000000..35998c0
--- /dev/null
+++ b/core/test/core/numeric/convert.xml
@@ -0,0 +1,209 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  desc="Test numeric computations dealing with rounding">
+
+
+  <import package="../../spec" />
+
+  <import package="../../../base" />
+  <import package="../../../numeric/convert" />
+
+
+  <t:describe name="_map-interval_">
+    <t:describe name="with values between intervals">
+      <t:it desc="maps start endpoint to first interval">
+        <t:given>
+          <!-- same value as start -->
+          <t:map-interval name="#5"
+                          start="5">
+            <t:endpoint value="10" map="#1" />
+            <t:endpoint value="20" map="#2" />
+          </t:map-interval>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#1" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="maps first endpoint to first interval">
+        <t:given>
+          <!-- value same as first endpoint -->
+          <t:map-interval name="#10"
+                          start="0">
+            <t:endpoint value="10" map="#1" />
+            <t:endpoint value="20" map="#2" />
+          </t:map-interval>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#1" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="maps value between start and first to first interval">
+        <t:given>
+          <!-- between start and first endpoint -->
+          <t:map-interval name="#5"
+                          start="0">
+            <t:endpoint value="10" map="#1" />
+            <t:endpoint value="20" map="#2" />
+          </t:map-interval>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#1" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="maps value between first and second endpoint to second
+                  interval">
+        <t:given>
+          <!-- between start and first endpoint -->
+          <t:map-interval name="#11"
+                          start="0">
+            <t:endpoint value="10" map="#1" />
+            <t:endpoint value="20" map="#2" />
+          </t:map-interval>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#2" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="maps value after final endpoint to default">
+        <t:given>
+          <!-- between start and first endpoint -->
+          <t:map-interval name="#11"
+                          start="0">
+            <t:endpoint value="10" map="#1" />
+            <t:endpoint value="20" map="#2" />
+          </t:map-interval>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#2" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="in interval [-∞,start) ">
+      <t:it desc="defaults to map value of 0">
+        <t:given>
+          <t:map-interval name="#1"
+                          start="5" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#0" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="maps to low value when provided">
+        <t:given>
+          <t:map-interval name="#1"
+                          start="5"
+                          low="#2" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#2" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="in interval (end,∞] ">
+      <t:it desc="defaults to map value of 0">
+        <t:given>
+          <t:map-interval name="#20"
+                          start="0">
+            <t:endpoint value="10" map="#1" />
+          </t:map-interval>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#0" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="maps to high value when provided">
+        <t:given>
+          <t:map-interval name="#20"
+                          start="0"
+                          high="#25">
+            <t:endpoint value="10" map="#1" />
+          </t:map-interval>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#25" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+  </t:describe>
+</package>
diff --git a/core/test/core/numeric/percent.xml b/core/test/core/numeric/percent.xml
new file mode 100644
index 0000000..6ea987d
--- /dev/null
+++ b/core/test/core/numeric/percent.xml
@@ -0,0 +1,178 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         desc="Test percentage arithemetic">
+
+  <import package="../../spec" />
+
+  <import package="../../../base" />
+  <import package="../../../numeric/percent" />
+
+
+  <t:describe name="_rtop_">
+    <t:describe name="given only a percentage">
+      <t:it desc="converts real into whole percentage">
+        <t:given>
+          <t:rtop name="#0.05" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#5" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="converts integer into whole percentage">
+        <t:given>
+          <t:rtop name="#5" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#500" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="converts negative into negative percentage">
+        <t:given>
+          <t:rtop name="#-0.5" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#-50" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="given a percent difference">
+      These tests are similar to the above, except that they represent
+      the~percent difference between two values.  That is, given the
+      values $5$~and $15$, we arrive at a~$200%$ increase.  So
+      if~$p=200$, then we want $5r=15, r=3=\((15-5)/5\)+1$.
+
+      <t:it desc="converts R into percentage P that will increase N
+                  by P%">
+        <t:given>
+          <t:rtop name="#3" difference="true" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#200" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      Similarly, if we go from~$10$ to~$5$, we have~$p=-100$
+      and~$10r=5$; then~$r=1/2=0.5$.
+
+      <t:it desc="converts negative N into percentage P that will
+                  decrease N by P%">
+        <t:given>
+          <t:rtop name="#0.5" difference="true" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#-50" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="given a value index">
+      <const name="PERCENT_REAL_VECTOR"
+             desc="Test vector">
+        <item value="3" desc="Real number" />
+      </const>
+
+      <t:it desc="converts the scalar value at the given index">
+        <t:given>
+          <t:rtop name="PERCENT_REAL_VECTOR" index="#0" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#300" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="given @negative@">
+      \ref{_rtop_} provides a way to negate the result using a
+      short-hand form for convenience and conciseness.
+
+      <t:it desc="negates resulting percentage">
+        <t:given>
+          <t:rtop name="#0.05" negate="true" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#-5" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="negates resulting difference">
+        <t:given>
+          <t:rtop name="#2" negate="true"
+                  difference="true" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#-100" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+  </t:describe>
+</package>
diff --git a/core/test/core/numeric/round.xml b/core/test/core/numeric/round.xml
new file mode 100644
index 0000000..dc70778
--- /dev/null
+++ b/core/test/core/numeric/round.xml
@@ -0,0 +1,315 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2015, 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  desc="Test numeric computations dealing with rounding">
+
+
+  <import package="../../../test/spec" />
+
+  <import package="../../../base" />
+  <import package="../../../numeric/round" />
+
+
+  <const name="VALUE_LOW" sym="\check{v}"
+         value="5.1" type="float"
+         desc="Low-decimal value" />
+  <const name="VALUE_MID" sym="\bar{v}"
+         value="5.5" type="float"
+         desc="High-decimal value" />
+  <const name="VALUE_HIGH" sym="\hat{v}"
+         value="5.7" type="float"
+         desc="High-decimal value" />
+
+  <const name="VALUE_FLOOR" sym="\lfloor v\rfloor"
+         value="5" type="float"
+         desc="Floor of decimal value" />
+  <const name="VALUE_CEIL" sym="\lceil v\rceil"
+         value="6" type="float"
+         desc="Ceiling of decimal value" />
+
+  <!-- used to test @index; same constant value as above -->
+  <const name="VALUE_VEC" sym="V"
+         type="float"
+         desc="Vector of values">
+    <item value="0" />
+    <item value="5.5" desc="Same as VALUE_MID" />
+  </const>
+  <const name="VALUE_VEC_INDEX" sym="\nu"
+         value="1" type="integer"
+         desc="Index of test value within vector" />
+
+  <const name="VALUE_LARGE" sym="\Omega"
+         value="1234.56" type="float"
+         desc="Value to test rounding offsets" />
+
+  <const name="VALUE_STEP" sym="s"
+         value="43.46" type="integer"
+         desc="Test arbitrary step" />
+
+
+  <!--
+    Eliminate let-round repetition for initial expectations
+  -->
+  <template name="_expect-let-round_"
+            desc="Short-hand rounding expectation">
+    <param name="@name@" desc="Value to round" />
+
+    <!-- choose *one* -->
+    <param name="@high@"    desc="Round high" />
+    <param name="@low@"     desc="Round high" />
+    <param name="@nearest@" desc="Round near" />
+
+    <!-- take whatever we're given and use that as the expected value -->
+    <param name="@__expect@" desc="Expected value">
+      <param-value name="@high@" />
+      <param-value name="@low@" />
+      <param-value name="@nearest@" />
+    </param>
+
+
+    <!-- round based on the provided template param -->
+    <t:given>
+      <if name="@high@">
+        <t:let-round name="@name@" high="rounded_high">
+          <c:value-of name="rounded_high" />
+        </t:let-round>
+      </if>
+
+      <if name="@low@">
+        <t:let-round name="@name@" low="rounded_low">
+          <c:value-of name="rounded_low" />
+        </t:let-round>
+      </if>
+
+      <if name="@nearest@">
+        <t:let-round name="@name@" nearest="rounded_nearest">
+          <c:value-of name="rounded_nearest" />
+        </t:let-round>
+      </if>
+    </t:given>
+
+    <t:expect>
+      <t:match-result>
+        <c:eq>
+          <c:value-of name="@__expect@" />
+        </c:eq>
+      </t:match-result>
+    </t:expect>
+  </template>
+
+
+
+  <t:describe name="_let-round_">
+    <t:describe name="high">
+      <t:it desc="rounds up low decimals">
+        <t:expect-let-round name="VALUE_LOW"
+                            high="VALUE_CEIL" />
+      </t:it>
+      <t:it desc="rounds up high decimals">
+        <t:expect-let-round name="VALUE_HIGH"
+                            high="VALUE_CEIL" />
+      </t:it>
+      <t:it desc="rounds up halves">
+        <t:expect-let-round name="VALUE_MID"
+                            high="VALUE_CEIL" />
+      </t:it>
+      <t:it desc="leaves integers unchanged">
+        <t:expect-let-round name="VALUE_FLOOR"
+                            high="VALUE_FLOOR" />
+        <t:expect-let-round name="VALUE_CEIL"
+                            high="VALUE_CEIL" />
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="low">
+      <t:it desc="rounds down low decimals">
+        <t:expect-let-round name="VALUE_LOW"
+                            low="VALUE_FLOOR" />
+      </t:it>
+      <t:it desc="rounds down high decimals">
+        <t:expect-let-round name="VALUE_HIGH"
+                            low="VALUE_FLOOR" />
+      </t:it>
+      <t:it desc="rounds down halves">
+        <t:expect-let-round name="VALUE_MID"
+                            low="VALUE_FLOOR" />
+      </t:it>
+      <t:it desc="leaves integers unchanged">
+        <t:expect-let-round name="VALUE_FLOOR"
+                            low="VALUE_FLOOR" />
+        <t:expect-let-round name="VALUE_CEIL"
+                            low="VALUE_CEIL" />
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="nearest">
+      <t:it desc="rounds down low decimals">
+        <t:expect-let-round name="VALUE_LOW"
+                            nearest="VALUE_FLOOR" />
+      </t:it>
+      <t:it desc="rounds up high decimals">
+        <t:expect-let-round name="VALUE_HIGH"
+                            nearest="VALUE_CEIL" />
+      </t:it>
+      <t:it desc="rounds up halves">
+        <t:expect-let-round name="VALUE_MID"
+                            nearest="VALUE_CEIL" />
+      </t:it>
+      <t:it desc="leaves integers unchanged">
+        <t:expect-let-round name="VALUE_FLOOR"
+                            nearest="VALUE_FLOOR" />
+        <t:expect-let-round name="VALUE_CEIL"
+                            nearest="VALUE_CEIL" />
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="with multiple names">
+      <t:it desc="produces each value">
+        <t:given>
+          <t:let-round name="VALUE_MID"
+                       low="round_low"
+                       high="round_high"
+                       nearest="round_near">
+            <c:sum>
+              <c:value-of name="round_low" />
+              <c:value-of name="round_high" />
+              <c:value-of name="round_near" />
+            </c:sum>
+          </t:let-round>
+        </t:given>
+
+        <t:expect>
+          <!-- VALUE_FLOOR + VALUE_CEIL + VALUE_CEIL -->
+          <t:match-result eq="17" />
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="with name index">
+      <t:it desc="produces respective rounded value">
+        <t:given>
+          <t:let-round name="VALUE_VEC"
+                       index="VALUE_VEC_INDEX"
+                       low="round_low"
+                       high="round_high"
+                       nearest="round_near">
+            <c:sum>
+              <c:value-of name="round_low" />
+              <c:value-of name="round_high" />
+              <c:value-of name="round_near" />
+            </c:sum>
+          </t:let-round>
+        </t:given>
+
+        <t:expect>
+          <!-- VALUE_FLOOR + VALUE_CEIL + VALUE_CEIL -->
+          <t:match-result eq="17" />
+        </t:expect>
+      </t:it>
+    </t:describe>
+  </t:describe>
+
+
+  <t:describe name="with a non-zero exp N">
+    <t:it desc="rounds before decimal point with positive exp">
+      <t:given>
+        <t:let-round name="VALUE_LARGE"
+                     exp="#2"
+                     low="round_low">
+          <c:value-of name="round_low" />
+        </t:let-round>
+      </t:given>
+
+      <!-- 1234.56 moved over two dec places = 12.3456; round that -->
+      <t:expect>
+        <t:match-result eq="1200" />
+      </t:expect>
+    </t:it>
+
+
+    <t:it desc="rounds after decimal point with negative exp">
+      <t:given>
+        <t:let-round name="VALUE_LARGE"
+                     exp="#-1"
+                     low="round_low">
+          <c:value-of name="round_low" />
+        </t:let-round>
+      </t:given>
+
+      <!-- 1234.56 moved over one dec place = 12345.6; round that -->
+      <t:expect>
+        <t:match-result eq="1234.50" />
+      </t:expect>
+    </t:it>
+  </t:describe>
+
+
+  <t:describe name="with a non-zero step N">
+    <t:it desc="rounds to the nearest arbitrary step">
+      <t:given>
+        <t:let-round name="VALUE_STEP"
+                     step="#15"
+                     low="round_low"
+                     high="round_high">
+          <c:sum>
+            <c:value-of name="round_low" />
+            <c:value-of name="round_high" />
+          </c:sum>
+        </t:let-round>
+      </t:given>
+
+      <t:expect>
+        <!-- 30 + 45 -->
+        <t:match-result eq="75" />
+      </t:expect>
+    </t:it>
+  </t:describe>
+
+
+  <t:describe name="with both step and exp">
+    <t:it desc="steps after exponentiation">
+      <t:given>
+        <t:let-round name="VALUE_STEP"
+                     step="#15"
+                     exp="#-1"
+                     low="round_low"
+                     high="round_high">
+          <c:sum>
+            <c:value-of name="round_low" />
+            <c:value-of name="round_high" />
+          </c:sum>
+        </t:let-round>
+      </t:given>
+
+      <!-- 43.46 * 10^-1 = 434.6 / 15 = 28.973; ceil/floor that, then undo
+           step/exp to yield final result -->
+      <t:expect>
+        <!-- 42 + 43.5 -->
+        <t:match-result eq="85.5" />
+      </t:expect>
+    </t:it>
+  </t:describe>
+</package>
diff --git a/core/test/core/suite.xml b/core/test/core/suite.xml
new file mode 100644
index 0000000..f64aed6
--- /dev/null
+++ b/core/test/core/suite.xml
@@ -0,0 +1,53 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2015, 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  desc="Core test suite"
+  program="true">
+
+  <import package="../../base" />
+  <import package="../spec" />
+
+  <import package="numeric/convert" />
+  <import package="numeric/percent" />
+  <import package="numeric/round" />
+
+  <import package="vector/fold" />
+  <import package="vector/interpolate" />
+  <import package="vector/length" />
+  <import package="vector/stub" />
+
+  <import package="aggregate" />
+  <import package="insurance" />
+  <import package="symbol" />
+  <import package="tplgen" />
+
+  <!-- XXX broken!
+  <import package="ui" />
+  -->
+
+
+  <t:verify-specs result="expectOk" />
+
+  <yield>
+    <c:value-of name="expectOk" />
+  </yield>
+</package>
diff --git a/core/test/core/symbol.xml b/core/test/core/symbol.xml
new file mode 100644
index 0000000..11347b0
--- /dev/null
+++ b/core/test/core/symbol.xml
@@ -0,0 +1,171 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  desc="Vector Folding and Unfolding Specs">
+
+  <import package="../spec" />
+
+  <import package="../../vector/stub" />
+  <import package="../../base" />
+  <import package="../../symbol" />
+
+
+  <t:describe name="_if-symbol_">
+    <t:describe name="given no predicate">
+      <t:it desc="expands body if symbol is defined">
+        <t:given>
+          <c:sum>
+            <t:if-symbol name="TRUE">
+              <c:value-of name="TRUE" />
+            </t:if-symbol>
+            <c:value-of name="FALSE" />
+          </c:sum>
+        </t:given>
+
+        <t:expect>
+          <t:match-result value="TRUE" />
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="does not expand body if symbol is undefined">
+        <t:given>
+          <c:sum>
+            <t:if-symbol name="UNDEFINED">
+              <c:value-of name="TRUE" />
+            </t:if-symbol>
+            <c:value-of name="FALSE" />
+          </c:sum>
+        </t:given>
+
+        <t:expect>
+          <t:match-result value="FALSE" />
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="given type predicate">
+      <t:it desc="expands body if symbol is of given type">
+        <t:given>
+          <c:sum>
+            <t:if-symbol name="TRUE" type="const">
+              <c:value-of name="TRUE" />
+            </t:if-symbol>
+            <c:value-of name="FALSE" />
+          </c:sum>
+        </t:given>
+
+        <t:expect>
+          <t:match-result value="TRUE" />
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="does not expand body if symbol is not of given type">
+        <t:given>
+          <c:sum>
+            <t:if-symbol name="TRUE" type="class">
+              <c:value-of name="TRUE" />
+            </t:if-symbol>
+            <c:value-of name="FALSE" />
+          </c:sum>
+        </t:given>
+
+        <t:expect>
+          <t:match-result value="FALSE" />
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="given dim predicate">
+      <t:it desc="expands body if symbol dimenions matches">
+        <t:given>
+          <c:sum>
+            <t:if-symbol name="NVEC1" dim="1">
+              <c:value-of name="TRUE" />
+            </t:if-symbol>
+            <c:value-of name="FALSE" />
+          </c:sum>
+        </t:given>
+
+        <t:expect>
+          <t:match-result value="TRUE" />
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="does not expand body if symbol does match dimensions">
+        <t:given>
+          <c:sum>
+            <t:if-symbol name="NVEC1" dim="2">
+              <c:value-of name="TRUE" />
+            </t:if-symbol>
+            <c:value-of name="FALSE" />
+          </c:sum>
+        </t:given>
+
+        <t:expect>
+          <t:match-result value="FALSE" />
+        </t:expect>
+      </t:it>
+    </t:describe>
+  </t:describe>
+
+
+  <!-- this duplicates the above _if-symbol_ portion -->
+  <t:describe name="_if-defined_">
+    <t:describe name="given no predicate">
+      <t:it desc="expands body if symbol is defined">
+        <t:given>
+          <c:sum>
+            <t:if-defined name="TRUE">
+              <c:value-of name="TRUE" />
+            </t:if-defined>
+            <c:value-of name="FALSE" />
+          </c:sum>
+        </t:given>
+
+        <t:expect>
+          <t:match-result value="TRUE" />
+        </t:expect>
+      </t:it>
+
+
+      <t:it desc="does not expand body if symbol is undefined">
+        <t:given>
+          <c:sum>
+            <t:if-defined name="UNDEFINED">
+              <c:value-of name="TRUE" />
+            </t:if-defined>
+            <c:value-of name="FALSE" />
+          </c:sum>
+        </t:given>
+
+        <t:expect>
+          <t:match-result value="FALSE" />
+        </t:expect>
+      </t:it>
+    </t:describe>
+  </t:describe>
+</package>
diff --git a/core/test/core/tplgen.xml b/core/test/core/tplgen.xml
new file mode 100644
index 0000000..1eca4aa
--- /dev/null
+++ b/core/test/core/tplgen.xml
@@ -0,0 +1,67 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         desc="Template Generation Package Specification">
+
+  <import package="../../test/spec" />
+
+  <import package="../../base" />
+  <import package="../../tplgen" />
+
+
+  <t:describe name="_for-each-n_">
+    <t:describe name="given a positive step">
+      <t:it desc="produces body N times exposing current N">
+        <t:given>
+          <c:sum>
+            <t:for-each-n start="2" end="6" step="2">
+              <c:const value="@current_n@" desc="Current N" />
+            </t:for-each-n>
+          </c:sum>
+        </t:given>
+
+        <t:expect>
+          <!-- 2 + 4 + 6 -->
+          <t:match-result eq="12" />
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="given a negative step">
+      <t:it desc="produces body N times exposing current N">
+        <t:given>
+          <c:sum>
+            <t:for-each-n start="4" end="1" step="-1">
+              <c:const value="@current_n@" desc="Current N" />
+            </t:for-each-n>
+          </c:sum>
+        </t:given>
+
+        <t:expect>
+          <!-- 4 + 3 + 2 + 1 = 4! (4 factorial) -->
+          <t:match-result eq="10" />
+        </t:expect>
+      </t:it>
+    </t:describe>
+  </t:describe>
+</package>
diff --git a/core/test/core/ui.xml b/core/test/core/ui.xml
new file mode 100644
index 0000000..54d9e99
--- /dev/null
+++ b/core/test/core/ui.xml
@@ -0,0 +1,83 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2016 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  desc="Test UI integration">
+
+  <import package="../spec" />
+
+  <import package="../../base" />
+  <import package="../../ui" />
+
+  <t:describe name="_match-ui-applicable_">
+    The UI package generator will produce classifications of the form
+    {\tt --vis-FIELD}, where {\tt FIELD} has all underscores converted
+    into dashes.  Others aren't supposed to know that, of course.
+
+    <t:it desc="Matches when visibility class is true">
+      <classify as="--vis-vis-true"
+                desc="Field class" />
+
+      <classify as="test-vis-true" desc="Test"
+                yields="testVisTrue">
+        <t:match-ui-applicable on="vis_true" />
+      </classify>
+
+      <t:given name="testVisTrue" />
+      <t:expect>
+        <t:match-result value="TRUE" />
+      </t:expect>
+    </t:it>
+
+
+    <t:it desc="Does not match when visibility class is false">
+      <classify as="--vis-vis-false" any="true"
+                desc="Field class" />
+
+      <classify as="test-vis-false" desc="Test"
+                yields="testVisFalse">
+        <t:match-ui-applicable on="vis_false" />
+      </classify>
+
+      <t:given name="testVisFalse" />
+      <t:expect>
+        <t:match-result value="FALSE" />
+      </t:expect>
+    </t:it>
+
+
+    <t:it desc="Can match against false visibility class">
+      <classify as="--vis-vis-true-false" any="true"
+                desc="Field class" />
+
+      <classify as="test-vis-true-false" desc="Test"
+                yields="testVisTrueFalse">
+        <t:match-ui-applicable on="vis_true_false"
+                               value="FALSE"/>
+      </classify>
+
+      <t:given name="testVisTrueFalse" />
+      <t:expect>
+        <t:match-result value="TRUE" />
+      </t:expect>
+    </t:it>
+  </t:describe>
+</package>
diff --git a/core/test/core/vector/fold.xml b/core/test/core/vector/fold.xml
new file mode 100644
index 0000000..3826476
--- /dev/null
+++ b/core/test/core/vector/fold.xml
@@ -0,0 +1,115 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  desc="Vector Folding and Unfolding Specs">
+
+  <import package="../../spec" />
+
+  <import package="../../../base" />
+  <import package="../../../vector/fold" />
+  <import package="../../../vector/stub" />
+
+
+  <const name="UNFOLD_VEC_MID_EMPTY" type="integer"
+         desc="3-vector with middle unset">
+    <item value="1" desc="Set" />
+    <item value="0" desc="Unset" />
+    <item value="1" desc="Set" />
+  </const>
+
+  <t:vector-to-class name="UNFOLD_VEC_MID_EMPTY" as="unfold-mid-empty" />
+
+  <const name="UNFOLD_VEC_3GROUP" type="integer"
+         desc="3-grouping for test case"
+         values="1,0,2,2,1,2" />
+
+  <!-- given the sequence [0 1 2 3 4 5] -->
+  <const name="UNFOLD_VEC_3GROUP_EXPECTED" type="integer"
+         desc="Expected result"
+         values="1; 0,4; 2,3,5" />
+
+
+  <const name="UNFOLD_3VEC_ONE_EACH" type="integer"
+         desc="3-vector with one in each group">
+  </const>
+
+
+  <t:describe name="_unfold-vector-grouped_">
+    <t:it desc="yields a vector whose length is that of the class">
+      <t:unfold-vector-grouped class="nclass3"
+                               generates="unfoldVecGroupedLenClass"
+                               src="NVEC1"
+                               grouping="NVEC1" />
+
+      <t:given>
+        <c:length-of>
+          <c:value-of name="unfoldVecGroupedLenClass" />
+        </c:length-of>
+      </t:given>
+
+      <t:expect>
+        <t:match-result eq="3" />
+      </t:expect>
+    </t:it>
+
+
+    <t:it desc="uses grouping vector for grouping">
+      <t:unfold-vector-grouped class="nclass3"
+                               generates="unfoldVecGrouped"
+                               src="NVEC6_SEQ"
+                               grouping="UNFOLD_VEC_3GROUP" />
+
+      <!-- TODO: matrix classification matches -->
+    </t:it>
+
+
+    <t:it desc="does not include non-matching groups">
+      <t:unfold-vector-grouped class="unfold-mid-empty"
+                               generates="unfoldVecGroupedNonClassMatch"
+                               src="NVEC3_SEQ"
+                               grouping="NVEC3_SEQ" />
+
+      <t:given>
+        <c:sum>
+          <c:length-of>
+            <c:value-of name="unfoldVecGroupedNonClassMatch" index="#0" />
+          </c:length-of>
+          <c:length-of>
+            <c:value-of name="unfoldVecGroupedNonClassMatch" index="#1" />
+          </c:length-of>
+          <c:length-of>
+            <c:value-of name="unfoldVecGroupedNonClassMatch" index="#2" />
+          </c:length-of>
+
+          <!-- sum a second time to make sure it's 0 -->
+          <c:length-of>
+            <c:value-of name="unfoldVecGroupedNonClassMatch" index="#1" />
+          </c:length-of>
+        </c:sum>
+      </t:given>
+
+      <t:expect>
+        <t:match-result eq="2" />
+      </t:expect>
+    </t:it>
+  </t:describe>
+</package>
diff --git a/core/test/core/vector/interpolate.xml b/core/test/core/vector/interpolate.xml
new file mode 100644
index 0000000..91b5b2e
--- /dev/null
+++ b/core/test/core/vector/interpolate.xml
@@ -0,0 +1,336 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  desc="Interpolation specs">
+
+  <import package="../../../base" />
+  <import package="../../../numeric/common" />
+  <import package="../../../numeric/minmax" />
+  <import package="../../../numeric/round" />
+  <import package="../../../vector/arithmetic" />
+
+  <import package="../../spec" />
+  <import package="../../../vector/table" />
+  <import package="../../../vector/interpolate" />
+
+
+  <const name="INTERP_TABLE_STEP"
+         value="100" type="integer"
+         desc="Test table step" />
+
+  <!-- table containing test data -->
+  <t:create-table name="interp-query-field-test"
+                  desc="Test data for query interpolation">
+    <t:table-column name="key"
+                    index="0"
+                    desc="Lookup value" />
+    <t:table-column name="pred"
+                    index="1"
+                    desc="Random value for predicate testing" />
+    <t:table-column name="value"
+                    index="2"
+                    desc="Value to interpolate" />
+
+    <t:table-rows>
+      <t:table-row>
+        <t:table-value const="100" />
+        <t:table-value const="10" />
+        <t:table-value const="1" />
+      </t:table-row>
+
+      <t:table-row>
+        <t:table-value const="200" />
+        <t:table-value const="20" />
+        <t:table-value const="2" />
+      </t:table-row>
+
+      <!-- two values for predicate testing -->
+      <t:table-row>
+        <t:table-value const="300" />
+        <t:table-value const="30" />
+        <t:table-value const="3" />
+      </t:table-row>
+      <t:table-row>
+        <t:table-value const="300" />
+        <t:table-value const="31" />
+        <t:table-value const="3.5" />
+      </t:table-row>
+
+      <t:table-row>
+        <t:table-value const="400" />
+        <t:table-value const="40" />
+        <t:table-value const="4" />
+      </t:table-row>
+      <t:table-row>
+        <t:table-value const="400" />
+        <t:table-value const="31" />
+        <t:table-value const="4" />
+      </t:table-row>
+    </t:table-rows>
+  </t:create-table>
+
+
+  <const name="INTERP_TABLE_SM_STEP"
+         value="15" type="integer"
+         desc="Small-step test table step" />
+
+  <!-- test another step (we didn't hardcode 100, did we)? -->
+  <t:create-table name="interp-query-field-test-sm"
+                  desc="Test data for query interpolation; sm step">
+    <t:table-column name="key"
+                    index="0"
+                    desc="Lookup value" />
+    <t:table-column name="value"
+                    index="1"
+                    desc="Value to interpolate" />
+
+    <t:table-rows>
+      <t:table-row>
+        <t:table-value const="15" />
+        <t:table-value const="0" />
+      </t:table-row>
+
+      <t:table-row>
+        <t:table-value const="30" />
+        <t:table-value const="10" />
+      </t:table-row>
+    </t:table-rows>
+  </t:create-table>
+
+
+  <!-- by avoiding hardcoding these, we make these tests less fragile
+       (if the test data were to change) -->
+  <section title="Common inferable table values">
+    <rate yields="interpTableMaxKeyValue">
+      <t:maxreduce isvector="true">
+        <t:query-field table="interp-query-field-test"
+                       field="key" />
+      </t:maxreduce>
+    </rate>
+
+    <rate yields="interpTableMaxFieldValue">
+      <t:query-first-field table="interp-query-field-test"
+                           field="value">
+        <t:when field="key">
+          <c:value-of name="interpTableMaxKeyValue" />
+        </t:when>
+      </t:query-first-field>
+    </rate>
+  </section>
+
+
+
+  <t:describe name="_interpolate-query-field_">
+    <t:describe name="exact value query">
+      <!-- a simple expression with no body performs basic interpolation
+           based on a query result...just like that! -->
+      <t:it desc="succeeds without predicates">
+        <t:given>
+          <t:interpolate-query-field table="interp-query-field-test"
+                                     field="value"
+                                     key="key"
+                                     step="INTERP_TABLE_STEP"
+                                     actual="#100" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#1" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <!-- sanity check -->
+      <t:it desc="succeeds with predicates">
+        <t:given>
+          <t:interpolate-query-field table="interp-query-field-test"
+                                     field="value"
+                                     key="key"
+                                     step="INTERP_TABLE_STEP"
+                                     actual="#300">
+            <t:when field="pred">
+              <c:const value="31" type="float"
+                       desc="Test predicate value" />
+            </t:when>
+          </t:interpolate-query-field>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#3.5" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="inexact value query">
+      <t:it desc="performs interpolation on query field">
+        <!-- take that you f@%*#^! PITA algorithms -->
+        <t:given>
+          <t:interpolate-query-field table="interp-query-field-test"
+                                     field="value"
+                                     key="key"
+                                     step="INTERP_TABLE_STEP"
+                                     actual="#150" />
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <!-- this value does not exist in the table; it is
+                   interpolated, just like real magic -->
+              <c:value-of name="#1.5" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+
+
+      <!-- take note that the predicate applies to *both* rows
+           returned by the query -->
+      <t:it desc="performs interpolation considering predicates">
+        <t:given>
+          <t:interpolate-query-field table="interp-query-field-test"
+                                     field="value"
+                                     key="key"
+                                     step="INTERP_TABLE_STEP"
+                                     actual="#350">
+            <t:when field="pred">
+              <c:const value="31" type="float"
+                       desc="Test predicate value" />
+            </t:when>
+          </t:interpolate-query-field>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <!-- note that this interpolates between 3.5 and 4 -->
+              <c:value-of name="#3.75" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <!-- ensure that the step is indeed configurable -->
+    <t:it desc="permits arbitrary step">
+      <t:given>
+        <t:interpolate-query-field table="interp-query-field-test-sm"
+                                   field="value"
+                                   key="key"
+                                   step="INTERP_TABLE_SM_STEP"
+                                   actual="#22.5" />
+      </t:given>
+
+      <t:expect>
+        <t:match-result>
+          <c:eq>
+            <c:value-of name="#5" />
+          </c:eq>
+        </t:match-result>
+      </t:expect>
+    </t:it>
+
+
+    <t:describe name="with actual exceeding maximum">
+      <t:describe name="with no step factor">
+        <t:it desc="yields value of maximum key">
+          <t:given>
+            <t:interpolate-query-field table="interp-query-field-test"
+                                       field="value"
+                                       key="key"
+                                       step="INTERP_TABLE_STEP"
+                                       table_max="interpTableMaxKeyValue"
+                                       actual="#10000" />
+          </t:given>
+
+          <t:expect>
+            <t:match-result>
+              <c:eq>
+                <c:value-of name="interpTableMaxFieldValue" />
+              </c:eq>
+            </t:match-result>
+          </t:expect>
+        </t:it>
+      </t:describe>
+
+
+      <!-- we can infer values past the maximum -->
+      <!-- TODO: these are fragile, because we'd otherwise be
+           duplicating a lot of logic; let's factor some common logic
+           out of _interpolate-query-field_ -->
+      <t:describe name="with step factor">
+        <t:it desc="applies arbitrary factor linearly to difference">
+          <t:given>
+            <!-- note that our step factor differs from the previous
+                 field value step in the table -->
+            <t:interpolate-query-field table="interp-query-field-test"
+                                       field="value"
+                                       key="key"
+                                       step="INTERP_TABLE_STEP"
+                                       step_factor="#2"
+                                       table_max="interpTableMaxKeyValue"
+                                       actual="#600" />
+          </t:given>
+
+          <t:expect>
+            <t:match-result>
+              <c:eq>
+                <c:value-of name="#8" />
+              </c:eq>
+            </t:match-result>
+          </t:expect>
+        </t:it>
+
+
+        <!-- half-step relative to above expectation -->
+        <t:it desc="interpolates linear factor">
+          <t:given>
+            <t:interpolate-query-field table="interp-query-field-test"
+                                       field="value"
+                                       key="key"
+                                       step="INTERP_TABLE_STEP"
+                                       step_factor="#2"
+                                       table_max="interpTableMaxKeyValue"
+                                       actual="#650" />
+          </t:given>
+
+          <t:expect>
+            <t:match-result>
+              <c:eq>
+                <c:value-of name="#9" />
+              </c:eq>
+            </t:match-result>
+          </t:expect>
+        </t:it>
+      </t:describe>
+    </t:describe>
+  </t:describe>
+</package>
diff --git a/core/test/core/vector/length.xml b/core/test/core/vector/length.xml
new file mode 100644
index 0000000..607d0eb
--- /dev/null
+++ b/core/test/core/vector/length.xml
@@ -0,0 +1,143 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2017 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  desc="Interpolation specs">
+
+  <import package="../../spec" />
+
+  <import package="../../../base" />
+  <import package="../../../vector/length" />
+
+
+  <t:describe name="_first-nonempty_">
+    <t:describe name="given no vectors">
+      <t:it desc="yields an empty vector">
+        <t:given>
+          <c:length-of>
+            <t:first-nonempty />
+          </c:length-of>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#0" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="given only one empty vector">
+      <t:it desc="yields an empty vector">
+        <t:given>
+          <c:length-of>
+            <t:first-nonempty>
+              <c:set />
+            </t:first-nonempty>
+          </c:length-of>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#0" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="given only empty vectors">
+      <t:it desc="yields an empty vector">
+        <t:given>
+          <c:length-of>
+            <t:first-nonempty>
+              <c:set />
+              <c:set />
+            </t:first-nonempty>
+          </c:length-of>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#0" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="given first empty vector with non-empty">
+      <t:it desc="yields the second vector">
+        <t:given>
+          <c:car>
+            <t:first-nonempty>
+              <c:set />
+              <c:set>
+                <c:const value="50" desc="Non-empty vector" />
+              </c:set>
+            </t:first-nonempty>
+          </c:car>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#50" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+
+
+    <t:describe name="given two non-empty vectors">
+      <t:it desc="yields the first vector">
+        <t:given>
+          <c:car>
+            <t:first-nonempty>
+              <c:set>
+                <c:const value="60" desc="Non-empty vector" />
+              </c:set>
+              <c:set>
+                <c:const value="70" desc="Non-empty vector" />
+              </c:set>
+            </t:first-nonempty>
+          </c:car>
+        </t:given>
+
+        <t:expect>
+          <t:match-result>
+            <c:eq>
+              <c:value-of name="#60" />
+            </c:eq>
+          </t:match-result>
+        </t:expect>
+      </t:it>
+    </t:describe>
+  </t:describe>
+</package>
diff --git a/core/test/core/vector/stub.xml b/core/test/core/vector/stub.xml
new file mode 100644
index 0000000..0d9c546
--- /dev/null
+++ b/core/test/core/vector/stub.xml
@@ -0,0 +1,49 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         desc="Vector Stub Package Specification">
+
+  <import package="../../../test/spec" />
+
+  <import package="../../../base" />
+  <import package="../../../vector/stub" />
+
+
+  <t:n-vector n="5" name="TEST_N_VEC" value="1" />
+
+
+  <t:describe name="_n-vector_">
+    <t:it desc="generates vector of length N">
+      <t:given>
+        <c:sum of="TEST_N_VEC" />
+      </t:given>
+
+      <t:expect>
+        <t:match-result>
+          <c:eq>
+            <c:value-of name="#5" />
+          </c:eq>
+        </t:match-result>
+      </t:expect>
+    </t:it>
+  </t:describe>
+</package>
diff --git a/core/test/spec.xml b/core/test/spec.xml
new file mode 100644
index 0000000..abb029d
--- /dev/null
+++ b/core/test/spec.xml
@@ -0,0 +1,355 @@
+<?xml version="1.0"?>
+<!--
+  BDD specification framework
+
+    Copyright (C) 2015, 2018 R-T Specialty, LLC.
+
+    This file is part of tame-core.
+
+    tame-core is free software: you can redistribute it and/or modify it
+    under the terms of the GNU General Public License as
+    published by the Free Software Foundation, either version 3 of the
+    License, or (at your option) any later version.
+
+    This program is distributed in the hope that it will be useful,
+    but WITHOUT ANY WARRANTY; without even the implied warranty of
+    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+    GNU General Public License for more details.
+
+    You should have received a copy of the GNU General Public License
+    along with this program.  If not, see <http://www.gnu.org/licenses/>.
+
+  This framework uses specifications, not stories[0].  Its syntax is heavily
+  motivated by popular BDD frameworks such as RSpec (Ruby), Jasmine
+  (JavaScript), and Mocha (JavaScript).
+
+  As an example, consider a specification for a simple stack implementation:
+
+    describe stack
+      describe push
+        it increases the stack size by 1
+        describe to an empty stack
+          it produces a stack of size 1
+      describe pop
+        it decreases the stack size by 1
+        it returns the most recently pushed item
+        describe multiple times
+          it returns elements in reverse order of push
+        describe from an empty stack
+          it produces an error
+
+  These templates expand all definitions flatly into their parent node;
+  descriptions should therefore be in the package root or within nodes that
+  support calculation and classification definitions.
+
+
+  [0]: http://en.wikipedia.org/wiki/Behavior-driven_development
+         #Story_versus_specification
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
+  xsi:schemaLocation="http://www.lovullo.com/rater ../../rater.xsd"
+
+  core="true"
+
+  name="core/test/spec"
+  desc="Behavior-driven development specification system">
+
+  <import package="../base" />
+
+  <import package="../vector/cmatch"
+          export="true" />
+
+
+
+  <template name="_verify-specs_"
+            desc="Declare a template suite">
+    <param name="@result@" desc="Name of boolean result" />
+
+
+    <!--
+      Determines whether all features are conformant to their
+      specifications; this classification is a simple way to determine
+      whether a test suite has passed.
+    -->
+    <classify as="expect-ok"
+              yields="@result@"
+              desc="All features conform to specifications"
+              keep="true">
+      <inline-template>
+        <for-each>
+          <sym-set name-prefix="expect-conform-"
+                   type="class" />
+        </for-each>
+
+        <t:match-class name="{@sym_name@}" />
+      </inline-template>
+    </classify>
+  </template>
+
+
+  <!--
+    Describe a feature specification
+
+    Each specification has a name that describes the feature as concisely as
+    possible. For example, if testing a stack, then `@name@` should simply
+    be "stack".
+
+    Specifications may be arbitrarily nested to group together
+    sub-features. If testing a specific feature of a parent feature, defer
+    as much of the description to the "it" clauses as possible.
+
+    A high-level classification will be generated that will assert on the
+    results of all expectations contained therein.
+
+    Permitted children:
+      - _describe_* - describe sub-features
+      - _it_+       - describe expectations
+  -->
+  <template name="_describe_"
+            desc="Describe a specification">
+    <param name="@name@"   desc="Subject (SUT)" />
+    <param name="@values@" desc="Specification descriptions" />
+
+
+    <param name="@__full_name@"
+           desc="Full subject name, inherited from parents">
+      <param-inherit meta="spec-name" />
+      <param-value name="@name@" />
+      <text> </text>
+    </param>
+
+    <param name="@__prefix@"
+           desc="Generated expectation prefix">
+      <!-- descriptions may be nested; this may or may not exist -->
+      <param-inherit meta="spec-prefix" />
+      <param-value name="@name@" dash="true" identifier="true" />
+      <text>-</text>
+    </param>
+
+    <param name="@__uniq@"
+           desc="Unique id">
+      <text unique="true" />
+    </param>
+
+
+    <expand-sequence>
+      <expand-group>
+        <param-copy name="@values@">
+          <param-meta name="spec-name"
+                      value="@__full_name@" />
+          <param-meta name="spec-prefix"
+                      value="@__prefix@" />
+        </param-copy>
+      </expand-group>
+
+      <!-- joins all generated classifications to provide a higher-level
+           failure if any expectations fail -->
+      <classify as="expect-conform-{@__prefix@}{@__uniq@}"
+                desc="{@__full_name@} meets expectations"
+                keep="true">
+        <inline-template>
+          <for-each>
+            <sym-set name-prefix="expect-that-{@__prefix@}"
+                     type="class" />
+          </for-each>
+
+          <t:match-class name="{@sym_name@}" />
+        </inline-template>
+      </classify>
+    </expand-sequence>
+  </template>
+
+
+
+  <!--
+    Describe a logical group of expectations
+
+    While `_describe_` delimits features (or sub-features), `_it_` delimits
+    logical groups of expectations of those features. Each expectation
+    within an `_it_` group can assert on a common `_given_` value.
+
+    The description should complete the sentence that started at the
+    uppermost `_describe_` ancestor; see the examples at the head of this
+    package.
+
+    Permitted children:
+      - *         - arbitrary definitions for feature test
+      - _given_?  - describe common data for expectations
+      - _expect_+ - describe feature expectation
+  -->
+  <template name="_it_"
+            desc="Describe logical group of expectations">
+    <param name="@desc@"   desc="Description of expectation" />
+    <param name="@values@" desc="Expectations" />
+
+    <param name="@__id@" desc="Unique identifier">
+      <param-inherit meta="spec-prefix" />
+      <param-value name="@desc@" dash="true" identifier="true" />
+    </param>
+
+
+    <param-copy name="@values@">
+      <param-meta name="spec-it"
+                  value="@desc@" />
+      <param-meta name="spec-it-prefix"
+                  value="@__id@" />
+    </param-copy>
+  </template>
+
+
+
+  <!--
+    Describe a value for expectation groups
+
+    The defined value is available to adjacent expectations through use of
+    `_match-result`.
+
+    A `_given_` definition is not required; it exists as a convenient and
+    concise way to represent test data.
+
+    Permitted children:
+      - Any valid calculation
+  -->
+  <template name="_given_"
+            desc="Describe a value for expectation groups">
+    <param name="@values@" desc="Calculation" />
+
+    <param name="@name@" desc="Value to reference (optional)" />
+
+    <param name="@__id@"
+           desc="Unique identifier to avoid symbol conflicts">
+      <text unique="true">given</text>
+    </param>
+
+
+    <!-- name given; no need to generate -->
+    <if name="@name@">
+      <param-meta name="spec-given-id"
+                  value="@name@" />
+    </if>
+
+    <!-- no name given; generate a unique one -->
+    <unless name="@name@">
+      <param-meta name="spec-given-id"
+                  value="@__id@" />
+
+      <rate yields="@__id@">
+        <param-copy name="@values@" />
+      </rate>
+    </unless>
+  </template>
+
+
+
+  <!--
+    Describe a feature expectation
+
+    An expectation tests the adherence of a feature to its specification. It
+    generates a classification and therefore shares identical child nodes.
+
+    Expectations may assert upon any value within scope. The
+    `_given_` template exists to simplify test calculation definitions;
+    asserting on the result of the adjacent `_given_` can be done using
+    `_match-result_`.
+
+    Permitted children:
+      - Any valid classification predicates
+  -->
+  <template name="_expect_"
+            desc="Describe a feature expectation">
+    <param name="@values@"
+           desc="Predicates" />
+
+    <!-- generated by the _describe_ template -->
+    <param name="@__spec_name@"
+           desc="Inherit specification name">
+      <param-inherit meta="spec-name" />
+    </param>
+
+    <!-- generated by the _it_ template -->
+    <param name="@__spec_it@"
+           desc="Inherit specification clause">
+      <param-inherit meta="spec-it" />
+    </param>
+
+    <!-- generate a unique id to avoid class conflicts (be sure to use the
+         ineherited prefix so that they can all be combined into a larger
+         predicate to assert on failing tests) -->
+    <param name="@__id@"
+           desc="Unique identifier">
+      <text>expect-that-</text>
+      <param-inherit meta="spec-it-prefix" />
+      <text unique="true">-</text>
+    </param>
+
+
+    <classify as="@__id@"
+              desc="{@__spec_name@}{@__spec_it@}">
+      <param-copy name="@values@">
+        <param-meta name="spec-expect-id"
+                    value="@__id@" />
+      </param-copy>
+    </classify>
+  </template>
+
+
+  <!--
+    Match against result of an adjacent `_given_` expression
+
+    This has equivalent syntax to the primitive `match`, except that `@on`
+    is determined for you.
+
+    Permitted children:
+      - Anything permitted by the `match` primitive
+  -->
+  <template name="_match-result_"
+            desc="Match against result of an adjacent _given_ expression">
+    <param name="@values@"
+           desc="Calculation as predicate">
+      <text></text>
+    </param>
+
+    <param name="@index@" desc="Constant index">
+      <text></text>
+    </param>
+
+    <param name="@__given_id@"
+           desc="Unique identifier of _given_ expression">
+      <param-inherit meta="spec-given-id" />
+    </param>
+
+    <!-- choose one -->
+    <param name="@value@" desc="Match value" />
+    <param name="@eq@"    desc="Match equality" />
+    <param name="@anyOf@" desc="Match against domain of type" />
+
+
+    <if name="@value@">
+      <match on="@__given_id@" index="@index@" value="@value@" />
+    </if>
+
+    <unless name="@value@">
+      <if name="@eq@">
+        <match on="@__given_id@" index="@index@">
+          <c:eq>
+            <c:value-of name="#{@eq@}" />
+          </c:eq>
+        </match>
+      </if>
+
+      <unless name="@eq@">
+        <if name="@anyOf@">
+          <match on="@__given_id@" index="@index@" anyOf="@anyOf@" />
+        </if>
+        <unless name="@anyOf@">
+          <match on="@__given_id@" index="@index@">
+            <param-copy name="@values@" />
+          </match>
+        </unless>
+      </unless>
+    </unless>
+  </template>
+</package>
diff --git a/core/tplgen.xml b/core/tplgen.xml
new file mode 100644
index 0000000..ce0d699
--- /dev/null
+++ b/core/tplgen.xml
@@ -0,0 +1,77 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="Template Generation">
+
+
+  This package provides various templates for generating templates and
+    iterating using templates.
+  In essense,
+    this package is intended to abstract away certain implementation details
+    and complexities that make groking code more difficult;
+      they should be used when possible to improve code maintenance.
+
+
+  <section title="Iteration">
+    \ref{_for-each-n_} recurisvely produces the body~\tt{@values@} given a
+      numeric range and step.
+    This can also be used to generate sequences at compile-time rather than
+      using functions,
+        provided that the sequence data are static.
+
+    <template name="_for-each-n_"
+              desc="Recursively apply body with counter and step">
+      <param name="@values@" desc="Body" />
+      <param name="@start@"  desc="Counter start" />
+      <param name="@end@"    desc="Counter end" />
+      <param name="@step@"   desc="Counter step" />
+
+
+      <param name="@next_n@" desc="Next iteration value">
+        <param-add name="@start@" value="@step@" />
+      </param>
+
+      <param name="@n_dec@" desc="Current n - 1">
+        <param-add name="@start@" value="-1" />
+      </param>
+
+
+      <!-- inefficient trick to expose @current_n@ to the body -->
+      <inline-template>
+        <for-each>
+          <set current_n="@start@" current_n_dec="@n_dec@" />
+        </for-each>
+
+        <param-copy name="@values@" />
+      </inline-template>
+
+
+      <!-- repeat body for each step, as sibling -->
+      <unless name="@start@" eq="@end@">
+        <t:for-each-n start="@next_n@" end="@end@" step="@step@">
+          <param-copy name="@values@" />
+        </t:for-each-n>
+      </unless>
+    </template>
+  </section>
+</package>
diff --git a/core/ui.xml b/core/ui.xml
new file mode 100644
index 0000000..a2b45d2
--- /dev/null
+++ b/core/ui.xml
@@ -0,0 +1,93 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2016 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         title="UI Integration">
+
+  <import package="base" />
+
+  <!-- contains template dependencies -->
+  <import package="/rater/core/vector/cmatch" export="true" />
+
+
+  This package provides elementary integration with the UI through
+  mystical knowledge of the naming conventions that the compiler uses
+  when generating the UI package.
+
+  You should \emph{never} reference generated values yourself without
+  an abstraction.
+
+
+  <section title="Applicability">
+    An object in the UI is considered to be \dfn{applicable} if the
+    predicate defined by its {\tt @when} attribute matches.  These
+    predicates are generated, and should not be referenced directly;
+    instead, \ref{_match-ui-applicable_} should be used with the
+    question id.
+
+    The term ``applicable'' is used instead of ``visible'' because
+    object can be hidden by other means; a predicate might be true
+    while a field is actually hidden.  Further, a non-matching
+    predicate inhibits other behavior, like running of assertions.
+
+    Note that this template will only be useful with an object
+    (question, display, static, etc) that has a predicate defined;
+    otherwise, compilation will fail.
+
+    <template name="_match-ui-applicable_"
+              desc="Match whether a UI question is visible">
+      <param name="@on@"  desc="Question id" />
+
+      <param name="@value@" desc="Value to match">
+        <text>TRUE</text>
+      </param>
+
+      <param name="@__vis_class@" desc="Generated visibility class name">
+        <text>--vis-</text>
+        <param-value name="@on@" dash="true" />
+      </param>
+
+
+      <t:match-class name="@__vis_class@" value="@value@" />
+    </template>
+
+
+    \ref{_match-ui-set_} determines whether a UI field has a truthful value,
+      which is non-empty and non-zero.
+
+    <template name="_match-ui-set_"
+              desc="Match whether a UI question is set">
+      <param name="@on@"  desc="Question id" />
+
+      <param name="@value@" desc="Whether set">
+        <text>TRUE</text>
+      </param>
+
+      <param name="@__set_class@" desc="Generated set class name">
+        <text>--set-</text>
+        <param-value name="@on@" dash="true" />
+      </param>
+
+
+      <t:match-class name="@__set_class@" value="@value@" />
+    </template>
+  </section>
+</package>
diff --git a/core/vector.xml b/core/vector.xml
new file mode 100644
index 0000000..ab01b0c
--- /dev/null
+++ b/core/vector.xml
@@ -0,0 +1,37 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  core="true"
+  desc="Vector operations">
+
+
+  <!-- we are a meta-package -->
+  <import package="vector/arithmetic" export="true" />
+  <import package="vector/cmatch"     export="true" />
+  <import package="vector/convert"    export="true" />
+  <import package="vector/count"      export="true" />
+  <import package="vector/list"       export="true" />
+  <import package="vector/matrix"     export="true" />
+  <import package="vector/table"      export="true" />
+
+  <import package="vector/common" export="true" />
+</package>
+
diff --git a/core/vector/arithmetic.xml b/core/vector/arithmetic.xml
new file mode 100644
index 0000000..99b84ef
--- /dev/null
+++ b/core/vector/arithmetic.xml
@@ -0,0 +1,174 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="Vector arithmetic">
+
+  <import package="../base" />
+  <import package="../numeric/common" />
+  <import package="../numeric/minmax" />
+  <import package="list" />
+
+
+
+  <section title="Vectors">
+    <!-- for the time being, the vectors must be of the same length, or the first
+         vector must be the longer (otherwise the values will not properly add)
+    -->
+    <template name="_vadd_" desc="Produce a vector resulting from the addition of two vectors">
+      <param name="@a@" desc="First vector to add" />
+      <param name="@b@" desc="Second vector to add" />
+      <param name="@into@" desc="Variable to yield vector into" />
+
+      <param name="@yields@" desc="Value to yield (useless)">
+        <text>__</text>
+        <param-value name="@into@" />
+      </param>
+
+      <param name="@gendesc@" desc="Generator description (for @into@)">
+        <text>Sum of vectors </text>
+        <param-value name="@a@" />
+        <text> and </text>
+        <param-value name="@b@" />
+      </param>
+
+      <param name="@sym@" desc="Generator symbol">
+        <!-- empty by default -->
+        <text></text>
+      </param>
+
+      <rate accumulate="none" yields="@yields@">
+        <c:sum of="@a@" index="k" generates="@into@" desc="@gendesc@" sym="@sym@">
+          <c:value-of name="@a@" index="k" />
+          <c:value-of name="@b@" index="k" />
+        </c:sum>
+      </rate>
+    </template>
+
+
+    <template name="_vsum_" desc="Yield sum of all elements of a vector as a scalar">
+      <param name="@of@"     desc="Vector to sum" />
+      <param name="@yields@" desc="Name of scalar to yield" />
+
+      <rate yields="@yields@">
+        <c:sum of="@of@" />
+      </rate>
+    </template>
+
+
+    \ref{_vproduct_} produces the product of two vectors:
+      $V_k = A_k B_k$.
+
+    <template name="_vproduct_"
+              desc="Vector product">
+      <param name="@vector_a@" desc="First vector" />
+      <param name="@vector_b@" desc="Second vector" />
+
+
+      <c:apply name="_vproduct"
+               vector_a="@vector_a@"
+               vector_b="@vector_b@" />
+    </template>
+
+
+    \ref{_vproduct} is its helper function for recursion.
+
+    <function name="_vproduct"
+              desc="Vector product">
+      <param name="vector_a" type="float" set="vector"
+             desc="First vector" />
+      <param name="vector_b" type="float" set="vector"
+             desc="Second vector" />
+      <param name="k" type="integer"
+             desc="Current index" />
+
+
+      <t:cons-until-empty set="vector_a" index="k" car="value_a">
+        <c:product>
+          <c:value-of name="value_a" />
+
+          <!-- TODO: TAME bug where index variables always compile to args -->
+          <c:value-of name="vector_b">
+            <c:index>
+              <c:value-of name="k" />
+            </c:index>
+          </c:value-of>
+        </c:product>
+      </t:cons-until-empty>
+    </function>
+  </section>
+
+
+
+  <section title="Matrices">
+    \ref{_mproduct_} produces the product of two matrices:
+      $M_{ij} = A_{ij} B_{ij}$.
+
+    <template name="_mproduct_"
+              desc="Matrix product">
+      <param name="@matrix_a@" desc="First matrix" />
+      <param name="@matrix_b@" desc="Second matrix" />
+
+
+      <c:apply name="_mproduct"
+               matrix_a="@matrix_a@"
+               matrix_b="@matrix_b@"
+               k="ZERO" />
+    </template>
+
+
+    \ref{_mproduct} is its helper function for recursion.
+    \ref{_vproduct_} is used to reduce the problem to the product of
+      and array of~vectors.
+
+    <function name="_mproduct"
+              desc="Matrix product">
+
+      <param name="matrix_a" type="float" set="matrix"
+             desc="First matrix" />
+      <param name="matrix_b" type="float" set="matrix"
+             desc="Second matrix" />
+      <param name="k" type="integer"
+             desc="Current index" />
+
+
+      <t:cons-until-empty set="matrix_a" index="k" car="vector_a">
+        <c:let>
+          <c:values>
+            <c:value name="vector_b" type="float" set="vector"
+                     desc="Second vector">
+              <c:value-of name="matrix_b">
+                <c:index>
+                  <c:value-of name="k" />
+                </c:index>
+              </c:value-of>
+            </c:value>
+          </c:values>
+
+          <t:vproduct vector_a="vector_a"
+                      vector_b="vector_b" />
+        </c:let>
+      </t:cons-until-empty>
+    </function>
+  </section>
+</package>
+
diff --git a/core/vector/cmatch.xml b/core/vector/cmatch.xml
new file mode 100644
index 0000000..ff7c961
--- /dev/null
+++ b/core/vector/cmatch.xml
@@ -0,0 +1,254 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015, 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="Vector operations on classification matches">
+
+  <import package="../base" />
+  <import package="arithmetic" export="true" />
+  <import package="minmax"     export="true" />
+
+
+  <!--
+    This wonderful little gem takes a classification match and stores it in a
+    vector, allowing for hassle-free use in calculations without awkward
+    hasany() calls on classification matrices (where a classification yields a
+    matrix, that is). This is useless for classifications that yield vectors,
+    since the result will be the same.
+
+    This template is also useful for combining various classification vectors
+    and matrices, whoose reduction would otherwise be a bit complicated. Of
+    course, the alternative there is also to create a classification that
+    relies on other classifications; use your best judgement.
+
+    This exploits the fact that _CMATCH_'s are always vectors, even if we are
+    matching on a matrix.  As such, the system has already gone through the
+    trouble of reducing the matrix into a vector for us, so all we need to do
+    is store each value into a vector, which can be easily accomplished using a
+    generator.
+
+    This is an excellent example of building a feature atop of the DSL without
+    having to add a new language feature.
+  -->
+  <template name="_cmatch-to-vector_" desc="Vectorizes a classification match">
+    <param name="@class@" desc="Classification match string" />
+    <param name="@generates@" desc="Variable to yield generates (will yield a vector)" />
+    <param name="@keep@" desc="Rate block @keep">
+      <text></text>
+    </param>
+
+    <param name="@yields@" desc="Dummy variable to yield generates (useless, but required)">
+      <text>__</text>
+      <param-value name="@generates@" />
+    </param>
+
+    <param name="@gendesc@" desc="Generator description">
+      <text>Vector containing boolean </text>
+      <param-value name="@class@" />
+      <text> classification matches</text>
+    </param>
+
+    <param name="@sym@" desc="Generator symbol (corresponds to @generates@)">
+      <!-- defaults to nothing -->
+      <text></text>
+    </param>
+
+    <!-- this conversion is as simple as using a generator to yield the value
+         of _CMATCH_ for each index -->
+    <rate class="@class@" accumulate="none" yields="@yields@" always="true" keep="@keep@">
+      <c:sum of="_CMATCH_" index="k" generates="@generates@" desc="@gendesc@" sym="@sym@">
+        <c:value-of name="_CMATCH_" index="k" />
+      </c:sum>
+    </rate>
+  </template>
+
+
+  <template name="_cmatch-to-scalar_" desc="Reduces a classification match into a scalar">
+    <param name="@class@" desc="Classification match string" />
+    <param name="@yields@" desc="Variable to yield into" />
+    <param name="@sym@" desc="Yield symbol (defaults to nothing)">
+      <!-- defaults to nothing -->
+      <text></text>
+    </param>
+
+    <param name="@keep@" desc="Rate block @keep">
+      <text></text>
+    </param>
+
+
+    <rate class="@class@" accumulate="none" yields="@yields@" sym="@sym@" keep="@keep@">
+      <!-- if any single one matches, then we want to yield a 1 -->
+      <c:apply name="maxreduce" maxreduce_set="_CMATCH_" />
+    </rate>
+  </template>
+
+
+  In cases where a classification needs to be based on a result that
+  has been reduced to a scalar, \ref{_cmatch_to-scalar_} does not
+  solve the whole problem: for this, \ref{_classify-scalar_} may be
+  used.  In addition to performing the action of the former (if {\tt
+  @yields} is provided), the resulting classification itself will
+  match on the scalar result.  While this is not strictly
+  necessary---the predicate itself is already scalar---this is
+  important for systems or templates that derive the classification
+  result from the name of the classification.
+
+
+  <template name="_classify-scalar_"
+            desc="Classification with a forced-scalar result">
+    <param name="@values@" desc="Predicates" />
+    <param name="@as@"     desc="Classification name" />
+    <param name="@desc@"   desc="Classification description" />
+
+    <param name="@yields@" desc="Scalar result name">
+      <text>__</text>
+      <param-value snake="true" name="@as@" />
+    </param>
+
+    <param name="@keep@" desc="Whether to force compilation">
+      <text></text>
+    </param>
+
+    <param name="@sym@" desc="Optional yield symbol">
+      <text></text>
+    </param>
+
+
+    <classify as="--{@as@}-pre"
+              yields="__{@yields@}Pre"
+              desc="{@desc@}, pre-scalar">
+      <param-copy name="@values@" />
+    </classify>
+
+    <t:cmatch-to-scalar class="--{@as@}-pre"
+                        yields="__{@yields@}Scalar"
+                        sym="@sym@"
+                        keep="@keep@" />
+
+    <classify as="@as@" yields="@yields@"
+              desc="@desc@"
+              keep="@keep@"
+              sym="@sym@">
+      <match on="__{@yields@}Scalar" />
+    </classify>
+  </template>
+
+
+
+  <!--
+    Counts one for each classification vector match
+  -->
+  <template name="_cmatch-count_" desc="Counts the number of classification matches of a vector">
+    <param name="@class@" desc="Classification match" />
+    <param name="@yields@" desc="Value to yield" />
+
+    <!-- nothing by default -->
+    <param name="@no@" desc="Non-match">
+      <text></text>
+    </param>
+
+
+    <rate class="@class@" no="@no@" yields="@yields@">
+      <c:sum of="_CMATCH_" />
+    </rate>
+  </template>
+
+
+  <!-- because verbose repition is an evil -->
+  <inline-template>
+    <for-each>
+      <set cmp="eq" />
+      <set cmp="ne" />
+      <set cmp="gt" />
+      <set cmp="gte" />
+      <set cmp="lt" />
+      <set cmp="lte" />
+    </for-each>
+
+
+    <template name="_match-{@cmp@}_" desc="Match value {@cmp@}">
+      <param name="@on@" desc="Value to assert" />
+
+      <!-- pick one -->
+      <param name="@const@" desc="Match against constant value" />
+      <param name="@value@" desc="Match against variable" />
+
+      <if name="@const@">
+        <warning>
+          @const@ is deprecated; use @value@ with a #-prefix instead.
+        </warning>
+      </if>
+
+      <match on="@on@">
+        <dyn-node name="c:{@cmp@}">
+          <if name="@const@">
+            <c:const value="@const@" type="float" desc="Comparison" />
+          </if>
+
+          <unless name="@const@">
+            <c:value-of name="@value@" />
+          </unless>
+        </dyn-node>
+      </match>
+    </template>
+  </inline-template>
+
+
+  <template name="_match-class_"
+               desc="Match on a class name (rather than @yields)">
+    <param name="@name@" desc="Classification name" />
+
+    <param name="@__yields@" desc="Classification yield to match on">
+      <param-class-to-yields name="@name@" />
+    </param>
+
+    <param name="@value@" desc="Value to match on">
+      <text></text>
+    </param>
+
+    <match on="@__yields@" value="@value@" />
+  </template>
+
+
+  \ref{_vector-to-class_} converts a vector into a classification.
+
+  <template name="_vector-to-class_"
+            desc="Convert a vector to a classification">
+    <param name="@name@" desc="Source vector" />
+    <param name="@as@"   desc="Classification name" />
+
+    <param name="@yields@" desc="Classification yield">
+      <text></text>
+    </param>
+
+    <param name="@desc@" desc="Classification description">
+      <text>Classification of </text>
+      <param-value name="@name@" />
+    </param>
+
+
+    <classify as="@as@" desc="@desc@" yields="@yields@">
+      <match on="@name@" />
+    </classify>
+  </template>
+</package>
+
diff --git a/core/vector/common.xml b/core/vector/common.xml
new file mode 100644
index 0000000..4cb69e6
--- /dev/null
+++ b/core/vector/common.xml
@@ -0,0 +1,205 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="General vector operations">
+
+  <import package="../base" />
+  <import package="../numeric/common" />
+
+
+  <!-- useful because you can pass a conditional expression as an argument and
+       select the index of the result -->
+  <function name="vival" desc="Select index value from some vector">
+    <param name="vector" type="float" set="vector" desc="Vector" />
+    <param name="index" type="integer" desc="Index" />
+
+    <c:value-of name="vector">
+      <c:index>
+        <c:value-of name="index" />
+      </c:index>
+    </c:value-of>
+  </function>
+
+
+  <function name="mival" desc="Select index value from some matrix">
+    <param name="matrix" type="float" set="matrix" desc="Matrix" />
+    <param name="row" type="integer" desc="Row index" />
+    <param name="col" type="integer" desc="Column index" />
+
+    <c:value-of name="matrix">
+      <c:index>
+        <c:value-of name="row" />
+      </c:index>
+      <c:index>
+        <c:value-of name="col" />
+      </c:index>
+    </c:value-of>
+  </function>
+
+
+  <template name="_map-value_" desc="Map from a value to another using a vector map">
+    <param name="@from@" desc="Name of value to map" />
+    <param name="@index@" desc="Index" />
+    <param name="@using@" desc="Vector to use for mapping" />
+
+    <c:apply name="vival">
+      <c:arg name="vector">
+        <c:value-of name="@using@" />
+      </c:arg>
+
+      <c:arg name="index">
+        <c:value-of name="@from@" index="@index@" />
+      </c:arg>
+    </c:apply>
+  </template>
+
+
+
+  <!--
+    Look up a matrix value using maps for both row and column indexes
+
+    This allows for the same values to be used with different matrices. The maps
+    will be used to map a value to an index within the matrix for either a row
+    or column.
+  -->
+  <function name="mlookup" desc="Matrix value lookup based on two index maps">
+    <param name="matrix" type="float"   set="matrix" desc="Rate matrix indexed by CT and PC" />
+    <param name="rmap"   type="integer" set="vector" desc="Row index map" />
+    <param name="cmap"   type="integer" set="vector" desc="Column index map" />
+
+    <param name="rval" type="integer" desc="Row value (to be fed to map for index lookup)" />
+    <param name="cval" type="integer" desc="Column value (to be fed to map for index lookup)" />
+
+    <c:value-of name="matrix">
+      <!-- row -->
+      <c:index>
+        <c:value-of name="rmap">
+          <c:index>
+            <c:value-of name="rval" />
+          </c:index>
+        </c:value-of>
+      </c:index>
+
+      <!-- column -->
+      <c:index>
+        <c:value-of name="cmap">
+          <c:index>
+            <c:value-of name="cval" />
+          </c:index>
+        </c:value-of>
+      </c:index>
+    </c:value-of>
+  </function>
+
+
+  <function name="first_index" desc="Determine the matching index of a vector; -1 if no match">
+    <param name="vector" type="float"   set="vector" desc="Source vector to search" />
+    <param name="value"  type="integer"              desc="Value to match" />
+    <param name="offset" type="integer"              desc="Vector offset" />
+
+    <c:let>
+      <c:values>
+        <!-- TODO: do not calculate every time -->
+        <c:value name="vlen" type="integer" desc="Vector length">
+          <c:length-of>
+            <c:value-of name="vector" />
+          </c:length-of>
+        </c:value>
+      </c:values>
+
+
+      <c:cases>
+        <c:case>
+          <c:when name="offset">
+            <c:gte>
+              <c:value-of name="vlen" />
+            </c:gte>
+          </c:when>
+
+          <c:const value="-1" type="integer" desc="Not found" />
+        </c:case>
+
+
+        <c:case>
+          <c:when name="vector" index="offset">
+            <c:eq>
+              <c:value-of name="value" />
+            </c:eq>
+          </c:when>
+
+          <!-- found the index -->
+          <c:value-of name="offset" />
+        </c:case>
+
+
+        <c:otherwise>
+          <c:recurse vector="vector" value="value">
+            <c:arg name="offset">
+              <t:inc>
+                <c:value-of name="offset" />
+              </t:inc>
+            </c:arg>
+          </c:recurse>
+        </c:otherwise>
+      </c:cases>
+    </c:let>
+  </function>
+
+
+  The template \ref{_repeat-value_} will generate a vector from the
+    given value for each class match.
+
+  <template name="_repeat-value_"
+            desc="Repeat value for each class match">
+    <param name="@values@"    desc="Calculation producing value (use
+                                    one of this or @value@)" />
+    <param name="@class@"     desc="Classification to consider" />
+    <param name="@value@"     desc="Value to repeat (use one of this
+                                    or @values@)" />
+    <param name="@generates@" desc="Result value (vector)" />
+
+    <!-- if providing @values@ -->
+    <param name="@index@" desc="Generator index">
+      <text>__k</text>
+    </param>
+
+
+    <rate-each class="@class@"
+               generates="@generates@" index="@index@">
+      <if name="@values@">
+        <param-copy name="@values@" />
+      </if>
+      <unless name="@values@">
+        <c:value-of name="@value@" index="@index@" />
+      </unless>
+    </rate-each>
+  </template>
+
+
+  <!-- generates a variable that can be recognized as an empty set (useful for
+       defaults to params that require sets) -->
+  <rate-each class="always" yields="__empty" generates="__emptySet" index="k">
+    <c:const value="0" type="integer" desc="Nothing" />
+  </rate-each>
+</package>
+
diff --git a/core/vector/convert.xml b/core/vector/convert.xml
new file mode 100644
index 0000000..68d67a2
--- /dev/null
+++ b/core/vector/convert.xml
@@ -0,0 +1,26 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  core="true"
+  desc="Convert vectors into other types">
+
+  <import package="../base" />
+</package>
diff --git a/core/vector/count.xml b/core/vector/count.xml
new file mode 100644
index 0000000..447624c
--- /dev/null
+++ b/core/vector/count.xml
@@ -0,0 +1,60 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  core="true"
+  desc="Vector element counting">
+
+  <import package="../base" />
+
+
+  <!-- TODO: deprecate in favor of the new c:length-of -->
+  <function name="length" desc="Counts the number of items in a vector (regardless of value)">
+    <param name="count_set" type="integer" set="vector" desc="Vector to count" />
+
+    <c:sum of="count_set" index="k">
+      <c:const value="1" type="integer" desc="Add 1 for each value in the set" />
+    </c:sum>
+  </function>
+
+
+  <!-- similar to a logical "exists" -->
+  <function name="hasany" desc="Returns 1 or 0 based on whether any values in a boolean vector are set" sym="\exists">
+    <param name="any_set" type="boolean" set="vector" desc="Boolean set to check" />
+
+    <c:ceil>
+      <c:quotient label="Create a fraction > 0 if any value in the given set is 1">
+        <c:sum of="any_set" label="Will be > 0 if any 1s exist in the set" />
+
+        <c:sum>
+          <c:apply name="length">
+            <c:arg name="count_set">
+              <c:value-of name="any_set" />
+            </c:arg>
+          </c:apply>
+
+          <!-- ensure the equation is not undefined if length = 0 -->
+          <c:const value="1" type="integer" desc="Add 1 to ensure equation is always defined" />
+        </c:sum>
+      </c:quotient>
+    </c:ceil>
+  </function>
+</package>
+
diff --git a/core/vector/filter.xml b/core/vector/filter.xml
new file mode 100644
index 0000000..2b8f150
--- /dev/null
+++ b/core/vector/filter.xml
@@ -0,0 +1,540 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="Filtering Vectors and Matrices">
+
+  <import package="../base" />
+  <import package="list" />
+
+
+  <section title="Vector Filtering">
+    <function name="vfilter_lookup"
+              desc="Filter predicate by value and use corresponding index in
+                    source vector as a value">
+      <param name="vector_pred" type="float" set="vector" desc="Vector to filter" />
+      <param name="vector_src" type="float" set="vector" desc="Vector to filter" />
+      <param name="value"  type="float"              desc="Predicate value" />
+      <param name="start_index" type="integer" desc="Starting vector_pred index" />
+
+      <t:cons-until-empty set="vector_pred" only="value" index="start_index">
+        <c:value-of name="vector_src" index="start_index" />
+      </t:cons-until-empty>
+    </function>
+  </section>
+
+
+  <section title="Matrix Filtering">
+    \ref{mfilter} handles complex filtering of matrices.
+    If the requested column~\tt{@col@} is marked as sequential with~\tt{@seq@},
+      a~$O(lg n)$ bisect algorithm will be used;
+        otherwise,
+          it will undergo a~$O(n)$ linear scan.
+
+    <function name="mfilter"
+              desc="Filter matrix rows by column value">
+      <param name="matrix" type="float" set="matrix" desc="Matrix to filter" />
+      <param name="col"    type="integer"            desc="Column index to filter on" />
+      <param name="vals"   type="float"              desc="Column value to filter on" />
+      <param name="seq"    type="boolean"            desc="Is data sequential?" />
+
+      <!-- merge the result of each condition in vals into a single set, which
+           has the effect of supporting multiple conditions on a single column of
+           data (or just one, if they don't want to feel sweet). By performing
+           the lookups separately for each, we preserve the bisect-ability of the
+           condition. -->
+      <t:merge-until-empty set="vals" car="val" glance="TABLE_WHEN_MASK_VALUE">
+        <c:apply name="mrange" matrix="matrix" col="col" val="val" seq="seq">
+          <c:arg name="start">
+            <c:cases>
+              <!-- if we know that the data is sequential, then we may not need to
+                   perform a linear search (if the dataset is large enough and the
+                   column value is relatively distinct) -->
+              <c:case>
+                <c:when name="seq">
+                  <c:eq>
+                    <c:value-of name="TRUE" />
+                  </c:eq>
+                </c:when>
+
+                <c:apply name="bisect" matrix="matrix" col="col" val="val">
+                  <c:arg name="start">
+                    <c:const value="0" desc="Start bisect at beginning" />
+                  </c:arg>
+
+                  <c:arg name="end">
+                    <!-- bisect the length of the matrix -->
+                    <t:dec>
+                      <c:length-of>
+                        <c:value-of name="matrix" />
+                      </c:length-of>
+                    </t:dec>
+                  </c:arg>
+                </c:apply>
+              </c:case>
+
+              <!-- we have no good guess; linear search :x -->
+              <c:otherwise>
+                <c:const value="0" desc="Start at the first element" />
+              </c:otherwise>
+            </c:cases>
+          </c:arg>
+
+          <c:arg name="end">
+            <t:dec>
+              <c:length-of>
+                <c:value-of name="matrix" />
+              </c:length-of>
+            </t:dec>
+          </c:arg>
+        </c:apply>
+      </t:merge-until-empty>
+    </function>
+
+
+    <function name="mrange"
+              desc="Filter matrix rows by column value within a certain
+                    range of indexes (inclusive)">
+      <param name="matrix" type="float" set="matrix" desc="Matrix to filter" />
+      <param name="col" type="integer" desc="Column index to filter on" />
+      <param name="val" type="float" desc="Column value to filter on" />
+      <param name="start" type="integer" desc="Starting index (inclusive)" />
+      <param name="end" type="integer" desc="Ending index (inclusive)" />
+      <param name="seq" type="boolean" desc="Is data sequential?" />
+
+      <c:let>
+        <c:values>
+          <c:value name="curval" type="float" desc="Current value">
+            <c:value-of name="matrix">
+              <c:index>
+                <c:value-of name="start" />
+              </c:index>
+
+              <c:index>
+                <c:value-of name="col" />
+              </c:index>
+            </c:value-of>
+          </c:value>
+        </c:values>
+
+        <!-- nested let needed so that the curval is available to over
+             in the body below -->
+        <c:let>
+          <c:values>
+            <!-- determine if the value we're looking for is over the current value
+                 in a sorted list (meaning that we will not find it) -->
+            <c:value name="over" type="boolean"
+                     desc="Did we pass the potential value in a sorted list?">
+              <c:value-of name="TRUE">
+                <c:when name="seq">
+                  <c:eq>
+                    <c:value-of name="TRUE" />
+                  </c:eq>
+                </c:when>
+
+                <c:when name="curval">
+                  <c:gt>
+                    <c:value-of name="val" />
+                  </c:gt>
+                </c:when>
+              </c:value-of>
+            </c:value>
+          </c:values>
+
+          <c:cases>
+            <!-- if we're done filtering, then return an empty set -->
+            <c:case>
+              <c:when name="start">
+                <c:gt>
+                  <c:value-of name="end" />
+                </c:gt>
+              </c:when>
+
+              <!-- empty set -->
+              <c:set />
+            </c:case>
+
+            <!-- if the data is sequential and the next element is over the
+                 requested value, then we're done -->
+            <c:case>
+              <c:when name="over">
+                <c:eq>
+                  <c:value-of name="TRUE" />
+                </c:eq>
+              </c:when>
+
+              <!-- empty set -->
+              <c:set />
+            </c:case>
+
+
+            <c:otherwise>
+              <c:apply name="_mfilter" matrix="matrix" col="col" val="val"
+                       start="start" end="end" seq="seq">
+                <c:arg name="cur">
+                  <c:value-of name="matrix">
+                    <!-- current row -->
+                    <c:index>
+                      <c:value-of name="start" />
+                    </c:index>
+
+                    <!-- requested column -->
+                    <c:index>
+                      <c:value-of name="col" />
+                    </c:index>
+                  </c:value-of>
+                </c:arg>
+              </c:apply>
+            </c:otherwise>
+          </c:cases>
+        </c:let>
+      </c:let>
+    </function>
+
+
+    <function name="_mfilter" desc="mfilter helper">
+      <param name="matrix" type="float" set="matrix" desc="Matrix to filter" />
+      <param name="col" type="integer"               desc="Column index to filter on" />
+      <param name="val" type="float"                 desc="Column value to filter on" />
+      <param name="start" type="integer"             desc="Starting index (aka current index)" />
+      <param name="end" type="integer"               desc="Ending index" />
+      <param name="seq" type="integer"               desc="Is data sequential?" />
+
+      <param name="cur" type="float"                 desc="Current value" />
+
+      <c:cases>
+        <c:case>
+          <c:when name="cur">
+            <c:eq>
+              <c:value-of name="val" />
+            </c:eq>
+          </c:when>
+
+          <c:cons>
+            <c:value-of name="matrix">
+              <c:index>
+                <c:value-of name="start" />
+              </c:index>
+            </c:value-of>
+
+            <c:apply name="mrange" matrix="matrix" col="col" val="val"
+                     end="end" seq="seq">
+              <c:arg name="start">
+                <c:sum>
+                  <c:value-of name="start" />
+                  <c:const value="1" desc="Check next element" />
+                </c:sum>
+              </c:arg>
+            </c:apply>
+          </c:cons>
+        </c:case>
+
+        <c:otherwise>
+          <c:apply name="mrange" matrix="matrix" col="col" val="val"
+                   end="end" seq="seq">
+            <c:arg name="start">
+              <c:sum>
+                <c:value-of name="start" />
+                <c:const value="1" desc="Check next element" />
+              </c:sum>
+            </c:arg>
+          </c:apply>
+        </c:otherwise>
+      </c:cases>
+    </function>
+
+
+    <section title="Bisecting">
+      Perform an~$O(lg n)$ bisect on a data set.
+
+      This is intended to limit recursion on very large data sets (and
+        consequently will increase performance).
+      This will bisect up until a certain point (the gap),
+        unless it finds the value in question.
+      After finding the value,
+        it will perform an~$O(n)$ linear backward search to find the first
+          occurrence of the value.
+      If the value is not found,
+        it will halt at the gap and return the first index of the gap,
+          which we will consider its "best guess",
+            at which point a linear search can be performed by the caller to
+              determine if the value does in fact exist at all.
+
+      (The reason this operates so oddly is because of its caller;
+        we could rid the gap and make this a general-purpose function if need be.
+      Technically,
+        the gap is useless and saves $lg g$ steps,
+          which may be very small.)
+
+
+      <const name="TABLE_WHEN_MASK_VALUE" type="float" value="-0.0000001"
+             desc="Used to mask when conditions" />
+      <const name="MFILTER_BISECT_GAP_MAX" type="integer" value="10"
+             desc="Quit bisect if size is less than or equal to this value" />
+
+
+      <function name="bisect"
+                desc="Bisect a matrix toward the requested column value">
+        <param name="matrix" type="float" set="matrix" desc="Matrix to bisect" />
+        <param name="col" type="integer" desc="Column index to filter on" />
+        <param name="val" type="float" desc="Column value to filter on" />
+        <param name="start" type="integer" desc="Start index" />
+        <param name="end" type="integer" desc="Start end" />
+
+        <c:let>
+          <c:values>
+            <!-- the gap represents the number of indexes between the current start
+                 and end indexes -->
+            <c:value name="gap" type="integer" desc="Gap between current start and end">
+              <c:sum>
+                <c:value-of name="end" />
+
+                <t:negate>
+                  <c:value-of name="start" />
+                </t:negate>
+              </c:sum>
+            </c:value>
+          </c:values>
+
+          <!--
+            At this point, we need to determine if we should continue the bisect or
+            halt. The purpose of the gap is based on the understanding that (with
+            our use cases) we will arrive, almost always, at one of two scenarios:
+            we found the value, but it's part of a larger set of the same values, or
+            the value we are looking for may not even exist at all.
+
+            The gap just limits recursion (but just barely) at smaller levels, since
+            bisect will take lg(n) steps). Increase the gap limit to decrease the
+            number of steps, or decrease it to 1 if you want a complete bisection.
+          -->
+          <c:cases>
+            <!-- give up if we've reached our gap limit -->
+            <c:case>
+              <c:when name="gap">
+                <c:lte>
+                  <c:value-of name="MFILTER_BISECT_GAP_MAX" />
+                </c:lte>
+              </c:when>
+
+              <!-- we tried our best; return our current position -->
+              <c:value-of name="start" />
+            </c:case>
+
+            <!-- we have not yet reached our gap limit; keep going -->
+            <c:otherwise>
+              <c:let>
+                <c:values>
+                  <c:value name="mid_index" type="integer" desc="Middle index">
+                    <!-- to determine the new mid index, add half of the gap to the
+                         current index -->
+                    <c:sum>
+                      <c:value-of name="start" />
+                      <c:ceil>
+                        <c:quotient>
+                          <c:value-of name="gap" />
+                          <c:const value="2" desc="Bisect" />
+                        </c:quotient>
+                      </c:ceil>
+                    </c:sum>
+                  </c:value>
+                </c:values>
+
+                <c:let>
+                  <c:values>
+                    <c:value name="mid" type="float" desc="Middle value">
+                      <c:value-of name="matrix">
+                        <!-- row -->
+                        <c:index>
+                          <c:value-of name="mid_index" />
+                        </c:index>
+
+                        <!-- column -->
+                        <c:index>
+                          <c:value-of name="col" />
+                        </c:index>
+                      </c:value-of>
+                    </c:value>
+                  </c:values>
+
+                  <c:cases>
+                    <!-- if the middle value is lower than our value, then take the upper half -->
+                    <c:case>
+                      <c:when name="mid">
+                        <c:lt>
+                          <c:value-of name="val" />
+                        </c:lt>
+                      </c:when>
+
+                      <c:recurse start="mid_index" />
+                    </c:case>
+
+                    <!-- similarily, take the lower half if we over-shot -->
+                    <c:case>
+                      <c:when name="mid">
+                        <c:gt>
+                          <c:value-of name="val" />
+                        </c:gt>
+                      </c:when>
+
+                      <c:recurse end="mid_index" />
+                    </c:case>
+
+                    <!-- if we have an exact match, that doesn't necessarily mean that we
+                         have every value; we may have intersected a set of them -->
+                    <c:otherwise>
+                      <!-- this will return an exact index: the first index
+                           containing the element we've been looking for (it is a
+                           linear backwards search) -->
+                      <c:apply name="foremost" matrix="matrix" col="col" i="mid_index" />
+                    </c:otherwise>
+                  </c:cases>
+                </c:let>
+              </c:let>
+            </c:otherwise>
+          </c:cases>
+        </c:let>
+      </function>
+
+
+      <function name="foremost"
+                desc="Search backwards for the first occurrance in a sorted list">
+        <param name="matrix" type="float" set="matrix" desc="Matrix to bisect" />
+        <param name="col" type="integer"               desc="Column index to search on" />
+        <param name="i" type="integer"                 desc="Current index" />
+
+        <c:let>
+          <c:values>
+            <!-- we calculate this rather than accept it via an argument so that
+                 this function may be called directly in a more convenient manner
+                 -->
+            <c:value name="val" type="float" desc="Current value">
+              <c:value-of name="matrix">
+                <!-- row -->
+                <c:index>
+                  <c:value-of name="i" />
+                </c:index>
+
+                <!-- column -->
+                <c:index>
+                  <c:value-of name="col" />
+                </c:index>
+              </c:value-of>
+            </c:value>
+
+            <c:value name="prev" type="float" desc="Previous value">
+              <c:value-of name="matrix">
+                <!-- row -->
+                <c:index>
+                  <t:dec>
+                    <c:value-of name="i" />
+                  </t:dec>
+                </c:index>
+
+                <!-- column -->
+                <c:index>
+                  <c:value-of name="col" />
+                </c:index>
+              </c:value-of>
+            </c:value>
+          </c:values>
+
+          <c:cases>
+            <!-- if we have no more indexes to check, then we're done -->
+            <c:case>
+              <c:when name="i">
+                <c:eq>
+                  <c:const value="0"
+                           desc="Did we check the final (first) index?" />
+                </c:eq>
+              </c:when>
+
+              <!-- well, then, we're done -->
+              <c:value-of name="i" />
+            </c:case>
+
+            <!-- if the previous column value is the same value, then continue checking -->
+            <c:case>
+              <c:when name="prev">
+                <c:eq>
+                  <c:value-of name="val" />
+                </c:eq>
+              </c:when>
+
+              <c:recurse>
+                <c:arg name="i">
+                  <t:dec>
+                    <c:value-of name="i" />
+                  </t:dec>
+                </c:arg>
+              </c:recurse>
+            </c:case>
+
+            <!-- otherwise, we've found the foremost index -->
+            <c:otherwise>
+              <c:value-of name="i" />
+            </c:otherwise>
+          </c:cases>
+        </c:let>
+      </function>
+
+
+      <template name="_mask-unless_"
+                desc="Mask a value unless the condition is truthful">
+        <param name="@values@" desc="Body" />
+        <param name="@name@"   desc="Scalar to check" />
+        <param name="@index@"  desc="Optional index" />
+        <param name="@desc@"   desc="Optional description" />
+
+        <c:cases>
+          <!-- if masked -->
+          <c:case>
+            <!-- no index provided -->
+            <unless name="@index@">
+              <c:when name="@name@">
+                <c:eq>
+                  <c:value-of name="FALSE" />
+                </c:eq>
+              </c:when>
+            </unless>
+
+            <!-- index provided -->
+            <if name="@index@">
+              <c:when name="@name@" index="@index@">
+                <c:eq>
+                  <c:value-of name="FALSE" />
+                </c:eq>
+              </c:when>
+            </if>
+
+            <!-- TODO: configurable mask via meta and/or param -->
+            <c:value-of name="TABLE_WHEN_MASK_VALUE" />
+          </c:case>
+
+          <!-- if not masked -->
+          <c:otherwise>
+            <param-copy name="@values@" />
+          </c:otherwise>
+        </c:cases>
+      </template>
+    </section>
+  </section>
+</package>
diff --git a/core/vector/fold.xml b/core/vector/fold.xml
new file mode 100644
index 0000000..3582198
--- /dev/null
+++ b/core/vector/fold.xml
@@ -0,0 +1,131 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="Folding and Unfolding of Vectors">
+
+  <import package="../base" />
+  <import package="filter" export="true" />
+
+
+  The term ``fold'' is also referred to as ``reduce''---%
+    they are synonymous.
+  Unless otherwise specified,
+    folding occurs left-to-right.
+
+  \emph{Unfolding} is the opposite of a reduction---%
+    it \emph{generates} values from existing values.
+
+
+  <section title="Matrix Folding">
+    \ref{_fold-matrix_} folds a matrix into a vector by summing each of
+      the~rows.
+
+    <template name="_fold-matrix_"
+              desc="Folds matrix into a vector by summing rows">
+      <param name="@name@"      desc="Matrix to fold" />
+      <param name="@class@"     desc="Iteration vector of desired length" />
+      <param name="@generates@" desc="Generator name (to yield)" />
+
+      <param name="@yields@" desc="Yield name">
+        <text></text>
+      </param>
+
+
+      <rate-each class="@class@" yields="@yields@"
+                 generates="@generates@" index="k">
+        <c:let>
+          <c:values>
+            <c:value name="row" type="float" set="vector"
+                     desc="Matrix row">
+              <c:value-of name="@name@" index="k" />
+            </c:value>
+          </c:values>
+
+          <c:sum of="row" />
+        </c:let>
+      </rate-each>
+    </template>
+  </section>
+
+
+  <section title="Matrix Generation">
+    \ref{_unfold-vector-grouped_} generates a matrix from a vector---%
+      that is, it generates vectors within a vector---%
+      by grouping values.
+
+    The \tt{@class@} is used both as a predicate and as a determination of
+      the resulting vector's length
+        (the~number of rows in the resulting matrix).
+    If non-matching,
+      no columns will be produced for that respective row.
+
+    \tt{@src@} is the vector to be unfolded,
+      containing the raw values to be grouped.
+
+    \tt{@grouping@} \should be the same length as~\ref{@src@} and determines
+      the group~(row) in which the respective value should appear.
+
+    \ref{@generates@} names the resulting matrix and~\ref{@desc@} provides
+      its description.
+
+    <template name="_unfold-vector-grouped_"
+              desc="Unfold vector into a matrix by grouping">
+      <param name="@class@"     desc="Iteration vector of desired length" />
+      <param name="@src@"       desc="Source vector" />
+      <param name="@grouping@"  desc="Grouping vector" />
+      <param name="@generates@" desc="Generator name (to yield)" />
+
+      <param name="@desc@" desc="Generator description">
+        <text>Unfolded vector </text>
+        <param-value name="@src@" />
+        <text> grouped by </text>
+        <param-value name="@grouping@" />
+      </param>
+
+      <param name="@lengthv@" desc="Length vector (of desired length)">
+        <param-class-to-yields name="@class@" />
+      </param>
+
+
+      <rate yields="_{@generates@}">
+        <c:sum of="@lengthv@" dim="matrix"
+               generates="@generates@" index="k"
+               desc="@desc@">
+          <c:cases>
+            <c:case label="Unfold on class vector match">
+              <c:when name="@lengthv@" index="k" />
+              <c:apply name="vfilter_lookup"
+                       vector_pred="@grouping@" vector_src="@src@"
+                       value="k"
+                       start_index="#0" />
+            </c:case>
+
+            <c:otherwise label="Ignore on class vector non-match">
+              <c:set />
+            </c:otherwise>
+          </c:cases>
+        </c:sum>
+      </rate>
+    </template>
+  </section>
+</package>
diff --git a/core/vector/interpolate.xml b/core/vector/interpolate.xml
new file mode 100644
index 0000000..d12d6c3
--- /dev/null
+++ b/core/vector/interpolate.xml
@@ -0,0 +1,402 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="Interpolation">
+
+  <import package="../base" />
+  <import package="../numeric/common" />
+
+
+  <template name="_interpolate_" desc="Interpolate the first two values in a vector">
+    <param name="@values@" desc="Set" />
+    <param name="@low@" desc="Lower value" />
+    <param name="@high@" desc="Upper value" />
+    <param name="@actual@" desc="Actual value" />
+
+
+    <c:let>
+      <c:values>
+        <!-- allows us to reference the set values -->
+        <c:value name="orig_set" type="float" set="vector" desc="The given set">
+          <param-copy name="@values@" />
+        </c:value>
+      </c:values>
+
+      <c:let>
+        <c:values>
+          <!-- ensure that the set is ordered such that the lower value is the first index -->
+          <c:value name="set" type="float" set="vector" desc="Ordered set">
+            <c:let>
+              <c:values>
+                <c:value name="a" type="float" desc="First set value">
+                  <c:car>
+                    <c:value-of name="orig_set" />
+                  </c:car>
+                </c:value>
+
+                <c:value name="b" type="float" desc="Second set value">
+                  <c:value-of name="orig_set">
+                    <c:index>
+                      <c:const value="1" type="integer" desc="Second index" />
+                    </c:index>
+                  </c:value-of>
+                </c:value>
+              </c:values>
+
+              <c:cases>
+                <!-- when a > b, reorder -->
+                <c:case>
+                  <c:when name="a">
+                    <c:gt>
+                      <c:value-of name="b" />
+                    </c:gt>
+                  </c:when>
+
+                  <c:set label="Re-ordered set such that the lower value is first in the vector">
+                    <c:value-of name="b" />
+                    <c:value-of name="a" />
+                  </c:set>
+                </c:case>
+
+                <!-- already ordered -->
+                <c:otherwise>
+                  <c:value-of name="orig_set" />
+                </c:otherwise>
+              </c:cases>
+            </c:let>
+          </c:value>
+
+          <!-- determine the skip to use for the vecstep call -->
+          <c:value name="skip" type="float" desc="First value to be used as skip">
+            <c:value-of name="@low@" />
+          </c:value>
+
+          <!-- determine the step to use for the vecstep call -->
+          <c:value name="step" type="float" desc="Use difference between the first two as the step">
+            <c:sum label="Step between the low and high values">
+              <c:value-of name="@high@" />
+
+              <t:negate>
+                <c:value-of name="@low@" />
+              </t:negate>
+            </c:sum>
+          </c:value>
+        </c:values>
+
+        <c:cases>
+          <!-- check to see if interpolation is even necessary; in particular,
+               this will prevent a step of 0 to vecstep, which would eventually
+               result in a division by 0 -->
+          <c:case>
+            <c:when name="step">
+              <c:eq>
+                <c:const value="0" type="integer" desc="No step indicates identical values" />
+              </c:eq>
+            </c:when>
+
+            <!-- just return the first value; it's exact and no interpolation is necessary -->
+            <c:value-of name="set">
+              <c:index>
+                <c:const value="0" type="integer" desc="First index" />
+              </c:index>
+            </c:value-of>
+          </c:case>
+
+          <!-- values are inexact; interpolation is required -->
+          <c:otherwise>
+            <!-- give the values computed above, we can re-use vecstep on the first two values on the vector -->
+            <c:apply name="vecstep" set="set" skip="skip" step="step" value="@actual@" />
+          </c:otherwise>
+        </c:cases>
+      </c:let>
+    </c:let>
+  </template>
+
+
+  <!--
+    Perform interpolation on the results of a table query
+
+    Not only is interpolation itself obnoxious, but determining the values
+    to look up from a table in order to get the data *to* interpolate also
+    results in a great deal of boilerplate code. This makes you not want to
+    kill yourself. At least not because of this interpolation query
+  -->
+  <template name="_interpolate-query-field_"
+               desc="Interpolate table data">
+    <param name="@table@"  desc="Table to query" />
+    <param name="@field@"  desc="Table field to query" />
+    <param name="@key@"    desc="Predicate subject column" />
+    <param name="@step@"   desc="Key step" />
+    <param name="@values@" desc="Query predicates" />
+
+    <param name="@actual@"
+              desc="Actual value" />
+
+    <param name="@index@" desc="Actual value index">
+      <text></text>
+    </param>
+
+    <param name="@table_max@"
+              desc="Maximum value in table (should be a multiple of @step)" />
+
+    <!-- TODO: accept a function to calculate factor -->
+    <param name="@step_factor@"
+              desc="Factor to use per step after maximum to infer value" />
+    <param name="@step_factor_index@"
+              desc="Index of step factor value">
+      <text></text>
+    </param>
+
+
+    <c:let>
+      <c:values>
+        <!-- if the requested limit exceeds the maximum value we support in
+             the table, then look up the maximum -->
+        <c:value name="lookup" type="float"
+                 desc="Value to retrieve and interpolate">
+          <!-- maximum provided -->
+          <if name="@table_max@">
+            <t:cap name="@table_max@">
+              <c:value-of name="@actual@" />
+            </t:cap>
+          </if>
+
+          <!-- no maximum -->
+          <unless name="@table_max@">
+            <c:value-of name="@actual@" index="@index@" />
+          </unless>
+        </c:value>
+      </c:values>
+
+
+      <c:sum>
+        <!-- lookup -->
+        <t:let-round name="lookup"
+                     step="@step@"
+                     high="high"
+                     low="low">
+          <!-- query and interpolate -->
+          <t:interpolate low="low"
+                         high="high"
+                         actual="lookup">
+            <t:query-field table="@table@"
+                           field="@field@">
+              <!-- query for upper and lower values for interpolation -->
+              <t:when field="@key@">
+                <c:value-of name="low" />
+                <c:value-of name="high" />
+              </t:when>
+
+              <param-copy name="@values@" />
+            </t:query-field>
+          </t:interpolate>
+        </t:let-round>
+
+        <!-- step factor -->
+        <if name="@step_factor@">
+          <c:let>
+            <c:values>
+              <!-- additional value (of key) that needs to be infered -->
+              <c:value name="add" type="float"
+                       desc="Additional value">
+                <c:sum>
+                  <c:value-of name="@actual@" index="@index@" />
+                  <t:negate>
+                    <c:value-of name="lookup" />
+                  </t:negate>
+                </c:sum>
+              </c:value>
+            </c:values>
+
+
+            <c:product>
+              <c:value-of name="@step_factor@"
+                          index="@step_factor_index@" />
+              <c:quotient>
+                <c:value-of name="add" />
+                <c:value-of name="@step@" />
+              </c:quotient>
+            </c:product>
+          </c:let>
+        </if>
+      </c:sum>
+    </c:let>
+  </template>
+
+
+  <!--
+    Calculates a floating-point representation of an arbitrary position within a
+    vector of values, where the value may exist between two elements within the
+    vector.
+
+    For example, given the following vector with the value of 12.5:
+
+      [ 5   ]
+      [ 10  ]
+             <- value is somewhere in here
+      [ 15  ]
+      [ ... ]
+      [ 100 ]
+
+    (Since the vector is 0-indexed, and we know that the index of 10 and 15 are
+    1 and 2 respectively, a value halfway between them would have an index of
+    exactly 1.5; so that's the value we're looking for.)
+
+    To calculate its position, given a 0-indexed vector, we need to know (a) the
+    step between each value and (b) the offset of the first step (the skip). In
+    the above case, index 0 represents 5, so let's assume that the skip provided
+    to us is 5. In this case, we have:
+
+      12.5 - 5 = 7.5
+
+    Which is the value without the skip. We can now simply divide it by the
+    step:
+
+      7.5 / 5 = 1.5, Q.E.D.
+  -->
+  <function name="vecpos" desc="Calculate the position of a value within a vector of a given step">
+    <param name="step"  type="float" desc="Step between each of the values" />
+    <param name="skip"  type="float" desc="Amount skipped before first element of vector" />
+    <param name="value" type="float" desc="Arbitrary value" />
+
+    <c:quotient>
+      <c:sum>
+        <c:value-of name="value" />
+        <t:negate>
+          <c:value-of name="skip" />
+        </t:negate>
+      </c:sum>
+
+      <c:value-of name="step" />
+    </c:quotient>
+  </function>
+
+
+  <!--
+    Calculates any arbitrary value given a vector of values and the step between
+    those values
+
+    This function is best demonstrated with an example. Consider the following
+    vector: [ 0, 5, 10, ..., 100 ]T. From this, we can see that the step is 5
+    and the skip is 0. (If the vector would have started at 5, then the skip
+    would have been 5.) The value is any arbitrary value.
+
+    This performs an index calculation and then calls vecstepi for further
+    processing; see vecstepi and vecpos for further information. Once the
+    relative position within the vector is calculated, we no longer need the
+    step and skip amounts, since they were only needed to determine the value's
+    position.
+  -->
+  <function name="vecstep" desc="Calculate a value that falls between a vector of values at a given step">
+    <param name="set"   type="float" set="vector" desc="Vector of values" />
+    <param name="step"  type="float"              desc="Step between each of the values" />
+    <param name="skip"  type="float"              desc="Amount skipped before first element of vector" />
+    <param name="value" type="float"              desc="Arbitrary value" />
+
+    <!-- call the function that will do the actual work -->
+    <c:apply name="vecstepi" set="set">
+      <!-- this is why the function call is necessary; we do not support
+           variable assignments, so the only way to have the same effect is to
+           invoke another function with the assignment as an argument (this
+           value is used frequently) -->
+      <c:arg name="pos">
+        <c:apply name="vecpos" step="step" skip="skip" value="value" />
+      </c:arg>
+    </c:apply>
+  </function>
+
+
+  <!--
+    Calculates any arbitrary value given a vector of values and an arbitrary
+    index
+
+    (Continuing from vecstep): if we were given the same vector [ 0, 5, 10, ...,
+    100 ]T and the calculated index 2 (assuming that the vectors are 0-indexed),
+    that would yield the same upper and lower value (5 and 5). The result would
+    then be ( 5 + ( 5-5 * 2-2 ) ) = 5. That's not very impressive, so let's
+    consider a more complicated case.
+
+    Consider that the given position is 2.5. By taking the floor and ceiling of
+    this position, we arrive at indexes 2 and 3, which yields the values 10 and
+    15 respectively. Our goal is to determine what the value (v) at position 2.5
+    should be:
+      - Start with the lower value; we'll be adding atop of this.
+      - Consider the remaining values (upper - lower): 15 - 10 = 5.
+        This means that we'll be adding some value between 0 and 5 such that the
+        value is 50% of the way (remember, we had 2.5) between the two. So,
+        we're looking for the value 2.5.
+      - To get this value, we can multiply the difference in the upper and lower
+        bounds by the decimal portion of our position (because 5 * 0.50 = 2.5).
+        To get the decimal portion, we can simply subtract the floor of the
+        position from the position itself: 2.5 - 2 = 0.5.
+      - So, we have: 5 + ( 15-10 * 2.5-2 ) = 5 + ( 5 * 0.5 ) = 5 + 2.5 = 7.5.
+  -->
+  <function name="vecstepi" desc="Calculate a value that falls between a vector of values at an arbitrary index">
+    <param name="set"  type="float" set="vector" desc="Vector of values" />
+    <param name="pos"  type="float"              desc="Position of value within the vector (may be between values)" />
+
+    <c:sum>
+      <!-- add the closest lower value... -->
+      <c:value-of name="set">
+        <c:index>
+          <c:floor label="Lower index">
+            <c:value-of name="pos" />
+          </c:floor>
+        </c:index>
+      </c:value-of>
+
+      <c:product label="Partial value to apply atop of the lower base value">
+        <c:sum label="Subtract the upper and lower values to get the increment per step">
+          <c:value-of name="set" label="Upper value">
+            <c:index>
+              <c:ceil>
+                <c:value-of name="pos" />
+              </c:ceil>
+            </c:index>
+          </c:value-of>
+
+          <t:negate>
+            <c:value-of name="set" label="Lower value">
+              <c:index>
+                <c:floor>
+                  <c:value-of name="pos" />
+                </c:floor>
+              </c:index>
+            </c:value-of>
+          </t:negate>
+        </c:sum>
+
+        <!-- determine the % between the two values -->
+        <c:sum label="Dropping the whole number from the position (yielding only the decimal) gives us the % difference">
+          <c:value-of name="pos" />
+          <t:negate>
+            <c:floor>
+              <c:value-of name="pos" />
+            </c:floor>
+          </t:negate>
+        </c:sum>
+      </c:product>
+    </c:sum>
+  </function>
+</package>
+
diff --git a/core/vector/length.xml b/core/vector/length.xml
new file mode 100644
index 0000000..68637c9
--- /dev/null
+++ b/core/vector/length.xml
@@ -0,0 +1,137 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2017 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="Operations Based on Vector Length">
+
+  <import package="../base" />
+  <import package="../numeric/common" export="true" />
+
+
+  See the respective test specification for examples.
+
+
+  <section title="Non-empty Vectors">
+    \ref{_first-nonempty_} will yield the result of the first toplevel
+      expression that is not an empty vector.
+    This template yields an empty vector if no non-empty vectors are found.
+
+    <template name="_first-nonempty_"
+              desc="Return the first non-empty vector">
+      <param name="@values@" desc="List of vectors" />
+
+
+      <c:let>
+        <c:values>
+          <!-- avoid having to copy @values@ multiple times -->
+          <c:value name="_list" type="float"
+                   desc="Result of body expression">
+            <c:set>
+              <param-copy name="@values@" />
+            </c:set>
+          </c:value>
+        </c:values>
+
+
+        <c:apply name="_first_nonempty">
+          <c:arg name="list">
+            <c:value-of name="_list" />
+          </c:arg>
+
+          <c:arg name="index">
+            <c:const value="0" desc="First element" />
+          </c:arg>
+
+          <c:arg name="length">
+            <c:length-of>
+              <c:value-of name="_list" />
+            </c:length-of>
+          </c:arg>
+        </c:apply>
+      </c:let>
+    </template>
+
+
+    Its helper function is \ref{_first_nonempty},
+      which recurses through each vector element until a non-empty vector is
+        encountered.
+
+    <function name="_first_nonempty"
+              desc="Return the first non-empty vector">
+      <param name="list" type="float" set="vector"
+             desc="List of vectors to process" />
+
+      <param name="index" type="integer"
+             desc="Current index (for recursion), decrementing" />
+      <param name="length" type="integer"
+             desc="Length of list" />
+
+
+      <c:let>
+        <c:values>
+          <c:value name="cur_len" type="integer"
+                   desc="Length of current vector">
+            <c:length-of>
+              <c:value-of name="list" index="index" />
+            </c:length-of>
+          </c:value>
+        </c:values>
+
+
+        <c:cases>
+          <!-- if none were found (end of list), return empty vector -->
+          <c:case>
+            <c:when name="index">
+              <c:gte>
+                <c:value-of name="length" />
+              </c:gte>
+            </c:when>
+
+            <c:set />
+          </c:case>
+
+          <!-- non-empty vector, return it -->
+          <c:case>
+            <c:when name="cur_len">
+              <c:gt>
+                <c:const value="0" desc="Vector length" />
+              </c:gt>
+            </c:when>
+
+            <c:value-of name="list" index="index" />
+          </c:case>
+
+          <!-- vector was empty, keep going (decrement index) -->
+          <c:otherwise>
+            <c:recurse>
+              <c:arg name="index">
+                <t:inc>
+                  <c:value-of name="index" />
+                </t:inc>
+              </c:arg>
+            </c:recurse>
+          </c:otherwise>
+        </c:cases>
+      </c:let>
+    </function>
+  </section>
+</package>
diff --git a/core/vector/list.xml b/core/vector/list.xml
new file mode 100644
index 0000000..eafbbdc
--- /dev/null
+++ b/core/vector/list.xml
@@ -0,0 +1,255 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015, 2017 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="Treating vectors as lists">
+
+  <import package="../base" />
+
+  <import package="../numeric/common" export="true" />
+  <import package="../when"           export="true" />
+
+
+
+  <!--
+    This abstraction relieves the developer of one of the most common
+    lisp-style recursions: recursing over a set's cdr until empty.
+
+    This template recurses and, as such, should only be used within functions.
+  -->
+  <template name="_cons-until-empty_" desc="Generate empty base case for functions that recurse on cdrs of sets">
+    <param name="@values@" desc="Body" />
+
+    <param name="@set@" desc="Set to operate on" />
+    <param name="@car@" desc="Variable in which to store car of the set">
+      <text>__car</text>
+    </param>
+    <param name="@cdr@" desc="Variable in which to store the cdr of the set">
+      <text>__cdr</text>
+    </param>
+    <param name="@index@" desc="Index variable" />
+
+    <param name="@base@" desc="Base set to return (otherwise an empty set)" />
+    <param name="@glance@" desc="Glance at (but do nothing with) this value; recurse without action" />
+    <param name="@only@" desc="Process only this value; otherwise recurse without action" />
+
+    <!-- intended for use by merge-until-empty to reduce duplicate code; not to
+         be set via direct template applications -->
+    <param name="@merge@" desc="Perform merge instead of cons; system-use only" />
+
+
+    <c:let>
+      <c:values>
+        <!-- TODO: it'd be nice if the DSL made values unique for us, unless
+             they're used, to prevent potential conflicts with template callers
+             -->
+        <c:value name="__valn" type="integer" desc="Number of values">
+          <c:length-of>
+            <c:value-of name="@set@" />
+          </c:length-of>
+        </c:value>
+      </c:values>
+
+      <c:cases>
+        <c:case>
+          <c:when name="__valn">
+            <c:eq>
+              <c:const value="0" type="integer" desc="When there are no more elements in the set" />
+            </c:eq>
+          </c:when>
+
+          <!-- if a base set was provided, return that; otherwise, return an empty set -->
+          <if name="@base@">
+            <c:value-of name="@base@" />
+          </if>
+          <unless name="@base@">
+            <!-- return an empty set -->
+            <c:set />
+          </unless>
+        </c:case>
+
+        <c:otherwise>
+          <c:let>
+            <c:values>
+              <c:value name="@car@" type="float" desc="Car of set">
+                <c:car>
+                  <c:value-of name="@set@" />
+                </c:car>
+              </c:value>
+
+              <c:value name="@cdr@" type="float" desc="Cdr of set">
+                <c:cdr>
+                  <c:value-of name="@set@" />
+                </c:cdr>
+              </c:value>
+            </c:values>
+
+            <!-- this case statement will be optimized away if we have no
+                 @glance@ or @only@ value -->
+            <c:cases>
+              <!-- if we have a glancing value, then immediately recurse
+                   without processing if we have a match -->
+              <if name="@glance@">
+                <c:case>
+                  <t:when-eq name="@car@" value="@glance@" />
+
+                  <c:recurse>
+                    <c:arg name="@set@">
+                      <c:value-of name="@cdr@" />
+                    </c:arg>
+
+                    <if name="@index@">
+                      <c:arg name="@index@">
+                        <t:inc>
+                          <c:value-of name="@index@" />
+                        </t:inc>
+                      </c:arg>
+                    </if>
+                  </c:recurse>
+                </c:case>
+              </if>
+
+              <!-- if we should only recurse when a value matches, ignore
+                   non-match -->
+              <if name="@only@">
+                <c:case>
+                  <t:when-ne name="@car@" value="@only@" />
+
+                  <c:recurse>
+                    <c:arg name="@set@">
+                      <c:value-of name="@cdr@" />
+                    </c:arg>
+
+                    <if name="@index@">
+                      <c:arg name="@index@">
+                        <t:inc>
+                          <c:value-of name="@index@" />
+                        </t:inc>
+                      </c:arg>
+                    </if>
+                  </c:recurse>
+                </c:case>
+              </if>
+
+              <!-- otherwise, process as normal -->
+              <c:otherwise>
+                <unless name="@merge@">
+                  <!-- here's our recursive operation: cons the result of processing
+                       this car with the result of recursively processing the cdr
+                       (note that c:recurse will recurse on the function that applied
+                       this template, _not_ on the template itself)-->
+                  <c:cons>
+                    <param-copy name="@values@" />
+
+                    <c:recurse>
+                      <c:arg name="@set@">
+                        <c:value-of name="@cdr@" />
+                      </c:arg>
+
+                      <if name="@index@">
+                        <c:arg name="@index@">
+                          <t:inc>
+                            <c:value-of name="@index@" />
+                          </t:inc>
+                        </c:arg>
+                      </if>
+                    </c:recurse>
+                  </c:cons>
+                </unless>
+
+                <if name="@merge@">
+                  <!-- the order is different from the cons above to maintain
+                       consistency in the returned set -->
+                  <c:apply name="vmerge">
+                    <c:arg name="vector">
+                      <c:recurse>
+                        <c:arg name="@set@">
+                          <c:value-of name="@cdr@" />
+                        </c:arg>
+
+                        <if name="@index@">
+                          <c:arg name="@index@">
+                            <t:inc>
+                              <c:value-of name="@index@" />
+                            </t:inc>
+                          </c:arg>
+                        </if>
+                      </c:recurse>
+                    </c:arg>
+
+                    <c:arg name="onto">
+                      <param-copy name="@values@" />
+                    </c:arg>
+                  </c:apply>
+                </if>
+              </c:otherwise>
+            </c:cases>
+          </c:let>
+        </c:otherwise>
+      </c:cases>
+    </c:let>
+  </template>
+
+
+
+  <!--
+    Like cons-until-empty, except that it merges each application instead of cons-ing it.
+
+    This template recurses and, as such, should only be used within functions.
+  -->
+  <template name="_merge-until-empty_" desc="Generate empty base case for functions that recursively merge on cdrs of sets">
+    <param name="@values@" desc="Body" />
+
+    <param name="@set@" desc="Set to operate on" />
+    <param name="@car@" desc="Variable in which to store car of the set" />
+    <param name="@cdr@" desc="Variable in which to store the cdr of the set">
+      <text>__cdr</text>
+    </param>
+
+    <param name="@glance@" desc="Glance at (but do nothing with) this value; recurse without action" />
+    <param name="@only@"   desc="Process only this value; otherwise recurse without action" />
+
+    <!-- to reduce duplicate template code, we simply set a merge flag on cons-until-empty -->
+    <t:cons-until-empty set="@set@" car="@car@" cdr="@cdr@"
+                        glance="@glance@" only="@only@" merge="true">
+      <param-copy name="@values@" />
+    </t:cons-until-empty>
+  </template>
+
+
+
+  <!--
+    Merges two vectors
+  -->
+  <function name="vmerge" desc="Merge two vectors (does not remove duplicates)">
+    <param name="vector" type="float" set="vector" desc="Vector to merge" />
+    <param name="onto"   type="float" set="vector" desc="Vector to append to" />
+
+    <!-- the template handles much of this for us: just keep cons'ing the car
+         until we have nothing more to cons, then that gets cons'd onto the
+         base -->
+    <t:cons-until-empty set="vector" car="car" base="onto">
+      <c:value-of name="car" />
+    </t:cons-until-empty>
+  </function>
+</package>
+
diff --git a/core/vector/matrix.xml b/core/vector/matrix.xml
new file mode 100644
index 0000000..894dd89
--- /dev/null
+++ b/core/vector/matrix.xml
@@ -0,0 +1,108 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="Matrix (vector of vectors) arithmetic">
+
+  <import package="../base" />
+  <import package="list" />
+
+
+  <!--
+    Extracts a vector of rates (a value for each location) associated with the
+    requested class and line codes
+
+    This seems more complicated/confusing than it really is, so it helps to
+    look at an example. Consider that we have the following data passed to this
+    template (assuming that this is the data that will exist at runtime):
+
+      R = [ [ "A", "B" ], [ "C" ], [ "D", "E" ] ]
+      C = [ [  1 ,  0  ], [  0  ], [  0,   1  ] ]
+      L = "gl"
+
+    Given matrix R (@rate_matrix@) and the classification matches C, for any
+    index that matches the given line L, a dot product will be performed. Let
+    us assume that indexes 0 and 2 match GL; we would then have the following
+    dot product:
+
+      R = [ [ "A", "B" ], ..., [ "D", "E" ] ]
+      C = [ [  1 ,  0  ], ..., [  0,   1  ] ]
+        = [   "A",                    "E"   ]
+
+    And so the resulting vector would be [ "A", "E" ]. To put this into an
+    example within the context of dwelling, let C above be the classification
+    set for all vacant buildings. Then, [ "A", "E" ] would be the rate
+    associated with the vacant building class code for locations 0 and 2
+    respectively.
+
+    If multiple matches are found per location, then the two rates will be
+    summed (as that is how dot products work), which is obviously not good.
+    However, this should not be a problem, because you should not have
+    duplicate class codes per location. Fix your data if that happens.
+
+    (Note that the dot product is actually performed for each index and then
+    multiplied by the classification match __CMATCH__, so while the effect
+    above is true, the result is achieved in a slightly different manner.)
+  -->
+  <!-- TODO: more generic name and documentation; we are in the core, after all -->
+  <template name="_rate-vector_" desc="Matches against a matrix of classes and a matrix of rates and returns a vector representing the rates per location">
+    <param name="@line@" desc="Line code classification string" />
+    <param name="@class_set@" desc="Matrix containing line classification matches" />
+    <param name="@into@" desc="Variable to which the vector will be assigned" />
+
+    <!-- this would be the sum of all the rates, which really is useless, but the XML requires it -->
+    <param name="@yields@" desc="Variable to yield to (useless, but required)">
+      <text>__</text>
+      <param-value name="@into@" />
+    </param>
+
+    <param name="@rate_matrix@" desc="Matrix containing the rates for the given line">
+      <text>rate_</text>
+      <param-value name="@line@" />
+    </param>
+
+    <rate-each class="@line@" accumulate="none" yields="@yields@" generates="@into@" index="k">
+      <!-- take the dot product of the two vectors (each part of a larger matrix)
+           to get the rate for the associated class code -->
+      <c:product dot="true" label="Dot product between the class and rate vectors for each location will yield the respective rate per location">
+        <c:value-of name="@class_set@" index="k" />
+        <c:value-of name="@rate_matrix@" index="k" />
+      </c:product>
+    </rate-each>
+  </template>
+
+
+  <function name="mcol" desc="Retrieve a matrix column as a vector">
+    <param name="matrix" type="float"   set="matrix" desc="Matrix" />
+    <param name="col"    type="integer"              desc="Column index (0-indexed)" />
+
+    <!-- generate a vector containing only the requested column from each row -->
+    <t:cons-until-empty set="matrix" car="row">
+      <c:value-of name="row">
+        <c:index>
+          <c:value-of name="col" />
+        </c:index>
+      </c:value-of>
+    </t:cons-until-empty>
+  </function>
+</package>
+
diff --git a/core/vector/minmax.xml b/core/vector/minmax.xml
new file mode 100644
index 0000000..5fa1239
--- /dev/null
+++ b/core/vector/minmax.xml
@@ -0,0 +1,236 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  title="Maximum and Minimum Elements">
+
+  <import package="../base" />
+  <import package="../numeric/common" />
+  <import package="../numeric/minmax" />
+
+
+  <section title="Vector Reduction">
+    Core currently only offers a~maximum reduction on
+    a~vector.  \todo{Add a~minimum reduction.}  \ref{_maxreduce_}
+    provides a convenient template-based abstraction.
+
+    <template name="_maxreduce_"
+              desc="Reduce a set to its maximum">
+      <param name="@values@"   desc="Values to reduce" />
+      <param name="@isvector@" desc="Set to 'true' if the nodes should
+                                     not be wrapped in c:set" />
+
+      <param name="@label@" desc="Application label">
+        <!-- default empty -->
+        <text></text>
+      </param>
+
+
+      <c:apply name="maxreduce" label="@label@">
+        <c:arg name="maxreduce_set">
+          <!-- if we were not provided with a vector (default), create
+               one out of the given nodes -->
+          <unless name="@isvector@" eq="true">
+            <c:set>
+              <param-copy name="@values@" />
+            </c:set>
+          </unless>
+
+          <!-- if they told us that they have provided a vector, then
+               do not create one -->
+          <if name="@isvector@" eq="true">
+            <param-copy name="@values@" />
+          </if>
+        </c:arg>
+      </c:apply>
+    </template>
+
+
+    \ref{maxreduce} is the~function encapsulated by the template and
+    may be applied directly.
+
+    \todo{This function was originally written before TAME supported
+    let~expressions and other convenience templates.  It has since
+    been refactored slightly, but can be made to be more concise.}
+
+    <function name="maxreduce" desc="Reduce a set to its maximum">
+      <param name="maxreduce_set" type="float" set="vector"
+             desc="Set to find max of" />
+
+      <c:let>
+        <c:values>
+          <c:value name="n" type="integer"
+                   desc="Length of set to reduce">
+            <c:length-of>
+              <c:value-of name="maxreduce_set" />
+            </c:length-of>
+          </c:value>
+        </c:values>
+
+
+        <c:cases>
+          <!-- if we have no values to reduce, then simply return 0 -->
+          <c:case>
+            <c:when name="n">
+              <c:eq>
+                <c:const value="0"
+                         desc="When there are no elements" />
+              </c:eq>
+            </c:when>
+
+            <c:const value="0" type="integer" desc="No value" />
+          </c:case>
+
+          <!-- we have values; perform reduction -->
+          <c:otherwise>
+            <c:apply name="_maxreduce">
+              <c:arg name="_maxreduce_set">
+                <c:value-of name="maxreduce_set" />
+              </c:arg>
+
+              <c:arg name="_maxreduce_i">
+                <t:dec>
+                  <c:length-of>
+                    <c:value-of name="maxreduce_set" />
+                  </c:length-of>
+                </t:dec>
+              </c:arg>
+            </c:apply>
+          </c:otherwise>
+        </c:cases>
+      </c:let>
+    </function>
+
+
+    \ref{_maxreduce} is a~helper function that recursively determines
+    the maximum value of the given set.  It should not be invoked
+    directly; use \ref{maxreduce} or~\ref{_maxreduce_} instead.
+
+    For example, given the vector~\vector{2,4,3,1}, here's the
+    recursive application:
+
+    \begingroup
+      \def\max{\textrm{max}}
+
+      \begin{equation}
+                \max( 1, \max( 3, max( 4, 2 ) ) )
+      \\ \yield \max( 1, \max( 3, 4 ) )
+      \\ \yield \max( 1, 4 )
+      \\ \yield 4
+      \end{equation}
+    \endgroup
+
+    \todo{If we eventually support first-class functions, then this
+    can simply be a~generic reduce function that accepts max/min/etc.}
+
+    \todo{This function was written before local variables generated
+    their own unique symbols; strip the prefixes.}
+
+    <function name="_maxreduce"
+              desc="Recursively reduce a set to its maximum (called by maxreduce)">
+      <param name="_maxreduce_set" type="float" set="vector"
+             desc="Set to find max of" />
+      <param name="_maxreduce_i"   type="integer"
+             desc="Index" />
+
+      <c:cases>
+        <!-- base case: if we're on the last index, do not recurse -->
+        <c:case>
+          <c:when name="_maxreduce_i">
+            <c:eq>
+              <c:const value="0"
+                       desc="Return when only one element remains" />
+            </c:eq>
+          </c:when>
+
+          <!-- return the first value -->
+          <c:value-of name="_maxreduce_set">
+            <c:index>
+              <c:const value="0"
+                       desc="First item in set" />
+            </c:index>
+          </c:value-of>
+        </c:case>
+
+        <!-- we have more elements in the set; recursively determine
+             the maximum value -->
+        <c:otherwise>
+          <c:apply name="max">
+            <!-- the first element to compare is our index -->
+            <c:arg name="max1">
+              <c:value-of name="_maxreduce_set">
+                <c:index>
+                  <c:value-of name="_maxreduce_i" />
+                </c:index>
+              </c:value-of>
+            </c:arg>
+
+            <!-- and we'll compare to the recursive application of the
+                 same set on the previous index -->
+            <c:arg name="max2">
+              <c:apply name="_maxreduce">
+                <c:arg name="_maxreduce_set">
+                  <c:value-of name="_maxreduce_set" />
+                </c:arg>
+
+                <c:arg name="_maxreduce_i">
+                  <c:sum>
+                    <c:value-of name="_maxreduce_i" />
+                    <c:const value="-1" type="integer" desc="Decrement index by 1" />
+                  </c:sum>
+                </c:arg>
+              </c:apply>
+            </c:arg>
+          </c:apply>
+        </c:otherwise>
+      </c:cases>
+    </function>
+  </section>
+
+
+
+  <!-- TODO: Everything below this line must be moved.  This
+       technically operates on a matrix as well. -->
+
+  <!-- simplifies retrieving the max of a set of values -->
+  <template name="_maxOfEach_" desc="Take the max of the given set of values">
+    <param name="@class@" desc="Class to match on" />
+    <param name="@values@" desc="Individual values without set" />
+    <param name="@generates@" desc="Value to generate into" />
+    <param name="@index@" desc="Index to use for rate-each" />
+
+    <param name="@yields@" desc="Yield variable">
+      <text>_</text>
+      <param-value name="@generates@" />
+    </param>
+
+    <rate-each class="@class@" accumulate="none" yields="@yields@" generates="@generates@" index="@index@">
+      <c:apply name="maxreduce">
+        <c:arg name="maxreduce_set">
+          <c:set>
+            <param-copy name="@values@" />
+          </c:set>
+        </c:arg>
+      </c:apply>
+    </rate-each>
+  </template>
+</package>
diff --git a/core/vector/stub.xml b/core/vector/stub.xml
new file mode 100644
index 0000000..718a14c
--- /dev/null
+++ b/core/vector/stub.xml
@@ -0,0 +1,95 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="Stub Vectors and Classifications">
+
+  <import package="../base" />
+  <import package="../tplgen" export="true" />
+
+
+  Stub vectors and classifications of various lengths---%
+    as well as the means of producing arbitrary lengths---%
+    are provided for convenience and testing.
+
+
+  <section title="Stub Vectors">
+    Ten n-vectors with each element containing the value~$0$ are
+      provided.
+
+    <t:for-each-n start="1" end="10" step="1">
+      <t:n-vector n="@current_n@" name="NVEC{@current_n@}"
+                  value="0" />
+    </t:for-each-n>
+
+
+    <section title="Sequence Vectors">
+      Sequence n-vectors are of length~$n$ and have a sequence of values
+        from~$0$ to~$n-1$.
+
+      <t:for-each-n start="1" end="10" step="1">
+        <t:n-vector n="@current_n@" name="NVEC{@current_n@}_SEQ" />
+      </t:for-each-n>
+    </section>
+  </section>
+
+
+
+  <section title="Stub Classifications">
+    Ten n-classes generated from the respective n-vectors are provided.
+
+    <t:for-each-n start="1" end="10" step="1">
+      <classify as="nclass{@current_n@}" yields="nClass{@current_n@}"
+                desc="Classification {@current_n@}-vector">
+        <match on="NVEC{@current_n@}" value="ZERO" />
+      </classify>
+    </t:for-each-n>
+  </section>
+
+
+
+  <section title="Vector Generation">
+    \ref{_n-vector_} generates vector constants of length~$n$ where each
+      element is assigned the provided value (or~$0$ if unspecified).
+
+    <template name="_n-vector_"
+              desc="Generate vector of length n">
+      <param name="@n@"     desc="Number of vector elements" />
+      <param name="@name@"  desc="Resulting vector constant name" />
+
+      <param name="@value@"
+             desc="Value of elements (optional; default n-1)" />
+
+
+      <const name="@name@" desc="{@n@}-vector" set="vector">
+        <t:for-each-n start="1" end="@n@" step="1">
+          <if name="@value@">
+            <item value="@value@" desc="Vector value" />
+          </if>
+          <unless name="@value@">
+            <item value="@current_n_dec@" desc="Vector value" />
+          </unless>
+        </t:for-each-n>
+      </const>
+    </template>
+  </section>
+</package>
diff --git a/core/vector/table.xml b/core/vector/table.xml
new file mode 100644
index 0000000..38dc0f7
--- /dev/null
+++ b/core/vector/table.xml
@@ -0,0 +1,494 @@
+<?xml version="1.0" encoding="ISO-8859-1"?>
+<!--
+  Copyright (C) 2015, 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+  xmlns:c="http://www.lovullo.com/calc"
+  xmlns:t="http://www.lovullo.com/rater/apply-template"
+  core="true"
+  desc="Functions for performing table lookups">
+
+  <import package="../base" />
+
+  <!-- since templates are inlined, we need to make these symbols available to
+       avoid terrible confusion -->
+  <import package="../numeric/common" export="true"/>
+  <import package="common" export="true" />
+  <import package="filter" export="true" />
+  <import package="matrix" export="true" />
+
+  <!--
+    Create a constant table
+
+    This definition must appear within a `constants' block.
+
+    Permitted children:
+      - _table_column_+ - Column definitions
+      - _table_rows_    - Begin table data definition
+  -->
+  <template name="_create-table_"
+               desc="Create an arbitrary table for querying">
+    <param name="@name@"   desc="Table name" />
+    <param name="@values@" desc="Table definition" />
+
+    <param name="@desc@" desc="Table description">
+      <text></text>
+    </param>
+
+    <param name="@__tid@"
+              desc="Internal table identifier">
+      <param-value name="@name@" upper="true" snake="true" />
+    </param>
+
+
+    <param-copy name="@values@">
+      <param-meta name="create-table-id" value="@__tid@" />
+      <param-meta name="create-table-name" value="@name@" />
+      <param-meta name="create-table-desc" value="@desc@" />
+    </param-copy>
+  </template>
+
+
+  <!--
+    Declare a table column name
+
+    A column definition assigns a field name to a column index. If the
+    column always contains ordered (sequenced) data, then @seq@ should
+    be set to "true"; this allows a more efficient query strategy to
+    be used.
+
+    If a table is especially large, the first column should be treated
+    as the index and always be sequenced.
+  -->
+  <template name="_table-column_"
+               desc="Declare name for table column">
+    <param name="@name@"  desc="Column name" />
+    <param name="@index@" desc="Column index (0-indexed)" />
+
+    <param name="@desc@" desc="Column description">
+      <param-value name="@name@" />
+    </param>
+
+    <!-- use carefully; leave alone unless data is definately sorted,
+         or query results may be incorrect -->
+    <param name="@seq@" desc="Column is sorted (sequential)">
+      <text></text>
+    </param>
+
+    <param name="@__tid@"
+              desc="Internal table identifier">
+      <param-inherit meta="create-table-id" />
+    </param>
+
+    <!-- FIXME: this doesn't contain @__tid@ because of a bug in the
+         preprocessor; this is fixed in the new DSL -->
+    <param name="@__constname@"
+              desc="Name of internal constant used for column lookup">
+      <text>RATE_TABLE_</text>
+      <param-value name="@name@" upper="true" snake="true" />
+    </param>
+
+
+    <!-- column index identifier -->
+    <const name="{@__tid@}_{@__constname@}"
+              value="@index@" type="integer"
+              desc="@desc@" />
+
+    <!-- column sequential flag to permit query optimizations -->
+    <if name="@seq@" eq="true">
+      <const name="{@__tid@}_{@__constname@}_IS_SEQ"
+                value="1" type="integer"
+                desc="{@name@} is sequenced" />
+    </if>
+    <unless name="@seq@" eq="true">
+      <const name="{@__tid@}_{@__constname@}_IS_SEQ"
+                value="0" type="integer"
+                desc="{@name@} is unordered" />
+    </unless>
+  </template>
+
+
+  <!--
+    Begin table data definition (rows)
+
+    Each _table-row_ child inserts a constant row into the table. Note
+    that all data must be constant.
+
+    Use only one of @data@ or children to define rows.
+
+    Permitted children:
+      - _table-row_* - Table row definitions
+  -->
+  <template name="_table-rows_"
+            desc="Begin table data definition">
+    <param name="@values@" desc="Row definitions" />
+    <param name="@data@"   desc="GNU Octave/MATLAB-style data definition" />
+
+    <param name="@__tid@"
+           desc="Table identifier">
+      <param-inherit meta="create-table-id" />
+    </param>
+
+    <param name="@__tname@"
+           desc="Table name as provided by caller">
+      <param-inherit meta="create-table-name" />
+    </param>
+
+    <param name="@__desc@"
+           desc="Table description">
+      <param-inherit meta="create-table-desc" />
+    </param>
+
+
+    <if name="@data@">
+      <const name="{@__tid@}_RATE_TABLE"
+             type="float"
+             desc="{@__tname@} table; {@__desc@}"
+             values="@data@" />
+    </if>
+    <unless name="@data@">
+      <const name="{@__tid@}_RATE_TABLE"
+             type="float"
+             desc="{@__tname@} table; {@__desc@}">
+        <param-copy name="@values@" />
+      </const>
+    </unless>
+  </template>
+
+
+  <!--
+    Define a constant table row
+
+    Rows will be inserted into the table in the order in which they
+    appear. Note that, if the column is marked as sequenced, it is
+    important that the order is itself sequenced.
+
+    All values must be constant.
+
+    Permitted children:
+      _table-value_+ - Row column value
+  -->
+  <template name="_table-row_"
+               desc="Define a constant table row (ordered)">
+    <param name="@values@" desc="Column values" />
+
+
+    <set desc="Row">
+      <param-copy name="@values@" />
+    </set>
+  </template>
+
+
+  <!--
+    Set a column value for the parent row
+
+    Column value order should match the defined column order. All
+    values must be constants.
+  -->
+  <template name="_table-value_"
+               desc="Table column value for row (ordered)">
+    <param name="@const@" desc="Constant column value" />
+
+
+    <item value="@const@" desc="Column value" />
+  </template>
+
+
+  <template name="_query-first-field_" desc="Return the requested field from the first row returned by a query">
+    <param name="@table@" desc="Table (matrix) to query" />
+    <param name="@values@" desc="Query parameters" />
+
+    <!-- use either field or name[+index] -->
+    <param name="@field@" desc="Column to select (variable/constant); do not use with @name@" />
+    <param name="@name@" desc="Name of field to query (variable/constant); overrides @field@" />
+    <param name="@index@" desc="Optional index for field name lookup (using @name@)" />
+
+
+    <c:car label="First row of query result">
+      <t:query-field table="@table@" field="@field@" name="@name@" index="@index@">
+        <param-copy name="@values@" />
+      </t:query-field>
+    </c:car>
+  </template>
+
+
+  <!--
+    Query-style syntax for matrix searches
+
+    Overhead of this algorithm is minimal. Let us consider that a direct access
+    to an element in a vector is O(1). Therefore, given a set of p predicates
+    in a direct matrix access, the time would be O(p).
+
+    In the case of this query method, we perform first a bisect and then a
+    linear search backward to the first element matching the given predicate.
+    Therefore, per field, the best case average is O(lg n), where n is the
+    number of rows; this assumes that no backward linear search is required.
+    Should such a search be required, the worst case per field is O(lg n + m),
+    where m is the length of the largest subset.
+
+    Once again considering p predicates, each predicate P_k where k > 0
+    (0-indexed) will operate on a subset of the first predicate. As already
+    discussed, the worst-case scenerio for the length of the subset is m. Since
+    further queries on this subset are rarely likely to use the bisect
+    algorithm, we're looking at O(m) time per field. Therefore, the total query
+    time for the best-case scenerio is still O(lg n) if m is sufficiently small
+    and O(lg n + pm) if m is sufficiently large.
+
+    An important case to note is when m approaches (or is equal to) n; in such
+    a case, the algorithm degrades to a worst-case linear search of O(pn) and
+    best case of O(n) if early predicates are sufficiently effective at
+    reducing the subsets to further predicates: That is, the bisect algorithm
+    either cannot be used or is ineffective. For large m, this may cause a
+    stack overflow due to the recursive nature of the algorithm. Is is
+    therefore important to order the first column of the table such that it is
+    both sorted and produces small m.  Additionally, it is ideal for the first
+    predicate to query the first field to quickly reduce the size of the set
+    for the next predicate.
+  -->
+  <template name="_query-field_" desc="Return the requested field from rows returned by a query">
+    <param name="@table@" desc="Table (matrix) to query" />
+    <param name="@values@" desc="Query parameters" />
+
+    <!-- use one or the other -->
+    <param name="@field@" desc="Column to select (variable/constant); do not use with @name@" />
+
+    <param name="@index@" desc="Optional index for field name lookup (using @name@)" />
+    <!-- by default, if @field@ is provided instead of @name@, the field
+         constant will be generated (same concept as the 'when' template) -->
+    <param name="@name@" desc="Name of field to query (variable/constant); overrides @field@">
+      <!-- convert @table@ to uppercase and snake case (replace - with _) -->
+      <param-value name="@table@" upper="true" snake="true" />
+      <text>_RATE_TABLE_</text>
+      <param-value name="@field@" upper="true" snake="true" />
+    </param>
+
+
+    <c:apply name="mcol" label="Query result">
+      <!-- the matrix (vector of rows) returned by the query -->
+      <c:arg name="matrix">
+        <t:query-row table="@table@">
+          <param-copy name="@values@" />
+        </t:query-row>
+      </c:arg>
+
+      <!-- the field (column) to retrieve; 0-based index -->
+      <c:arg name="col">
+        <!-- no index lookup needed -->
+        <unless name="@index@">
+          <c:value-of name="@name@" />
+        </unless>
+
+        <!-- index lookup required -->
+        <if name="@index@">
+          <c:value-of name="@name@" index="@index@" />
+        </if>
+      </c:arg>
+    </c:apply>
+  </template>
+
+
+  <template name="_query-row_" desc="Query a table (matrix) for a row (vector) of values">
+    <param name="@table@" desc="Table (matrix)" />
+    <param name="@values@" desc="Query parameters" />
+
+    <!-- this defaults to a table name constant as generated from the csv2xml
+         script; either this or @table@ should be used -->
+    <param name="@matrix@" desc="Matrix to look up from">
+      <!-- convert @table@ to uppercase and snake case (replace - with _) -->
+      <param-value name="@table@" upper="true" snake="true" />
+      <text>_RATE_TABLE</text>
+    </param>
+
+    <c:apply name="_mquery">
+      <c:arg name="matrix">
+        <c:value-of name="@matrix@" />
+      </c:arg>
+
+      <c:arg name="criteria">
+        <c:set>
+          <param-copy name="@values@">
+            <param-meta name="table_basename" value="@matrix@" />
+          </param-copy>
+        </c:set>
+      </c:arg>
+
+      <c:arg name="i">
+        <!-- begin with the last predicate (due to the way we'll recurse, it
+             will be applied *last* -->
+        <t:dec>
+          <c:length-of>
+            <c:set>
+              <param-copy name="@values@">
+                <param-meta name="table_basename" value="@matrix@" />
+              </param-copy>
+            </c:set>
+          </c:length-of>
+        </t:dec>
+      </c:arg>
+    </c:apply>
+  </template>
+
+
+  <template name="_when_" desc="Create field predicate for query definition">
+    <param name="@id@" desc="Field index" />
+    <param name="@values@" desc="Field value (provide only one node)" />
+    <param name="@sequential@" desc="Is data sequential?" />
+
+    <!-- @name@ may be provided directly, or @field@ may be used when the
+         basename is available (set by a query template), giving the illusion of
+         querying the table columns by name directly (magic!); pure syntatic
+         sugary goodness -->
+    <param name="@field@" desc="Field name (to be used with base)" />
+    <param name="@name@" desc="Field name (as a variable/constant)">
+      <param-inherit meta="table_basename" />
+      <text>_</text>
+      <param-value name="@field@" upper="true" snake="true" />
+    </param>
+
+    <param name="@seqvar@" desc="Var/constant containing whether field is sequential">
+      <param-inherit meta="table_basename" />
+      <text>_</text>
+      <param-value name="@field@" upper="true" snake="true" />
+      <text>_IS_SEQ</text>
+    </param>
+
+    <c:set label="Conditional for {@field@}">
+      <!-- the first element will represent the column (field) index -->
+      <unless name="@name@">
+        <c:const value="@id@" type="integer" desc="Field index" />
+      </unless>
+      <if name="@name@">
+        <c:value-of name="@name@" />
+      </if>
+
+      <!-- the second element will represent the expected value(s) -->
+      <c:set>
+        <param-copy name="@values@" />
+      </c:set>
+
+      <!-- the final element will represent whether or not this field is sequential -->
+      <if name="@sequential@">
+        <c:const value="@sequential@" type="boolean" desc="Whether data is sequential" />
+      </if>
+      <unless name="@sequential@">
+        <!-- if a field name was given, we can get the sequential information
+             that was already generated for us -->
+        <if name="@field@">
+          <c:value-of name="@seqvar@" />
+        </if>
+        <!-- otherwise, default to non-sequential -->
+        <unless name="@field@">
+          <c:value-of name="FALSE" />
+        </unless>
+      </unless>
+    </c:set>
+  </template>
+
+
+
+  <!--
+    These functions make the magic happen
+
+    They are hideous. Look away.
+  -->
+
+
+  <!-- this function is intended to be called by the _query_ template, not directly -->
+  <function name="_mquery" desc="Query for vectors using a set of column criteria">
+    <param name="matrix" type="float" set="matrix" desc="Matrix to query" />
+    <param name="criteria" type="float" set="matrix" desc="Query criteria" />
+
+    <param name="i" type="integer" desc="Current criteria index" />
+
+    <c:cases>
+      <c:case>
+        <c:when name="i">
+          <c:eq>
+            <!-- it's important that we allow index 0, since that is a valid
+                 predicate -->
+            <c:const value="-1" type="integer" desc="We're done." />
+          </c:eq>
+        </c:when>
+
+        <!-- we're done; stick with the result -->
+        <c:value-of name="matrix" />
+      </c:case>
+
+      <c:otherwise>
+        <c:apply name="mfilter">
+          <!-- matrix to search -->
+          <c:arg name="matrix">
+            <!-- >> recursion happens here << -->
+            <c:apply name="_mquery">
+              <c:arg name="matrix">
+                <c:value-of name="matrix" />
+              </c:arg>
+
+              <c:arg name="criteria">
+                <c:value-of name="criteria" />
+              </c:arg>
+
+              <c:arg name="i">
+                <t:dec>
+                  <c:value-of name="i" />
+                </t:dec>
+              </c:arg>
+            </c:apply>
+          </c:arg>
+
+          <!-- field (column) -->
+          <c:arg name="col">
+            <c:value-of name="criteria">
+              <c:index>
+                <c:value-of name="i" />
+              </c:index>
+              <c:index>
+                <c:const value="0" type="integer" desc="Field id" />
+              </c:index>
+            </c:value-of>
+          </c:arg>
+
+          <!-- value(s) to search for -->
+          <c:arg name="vals">
+            <c:value-of name="criteria">
+              <c:index>
+                <c:value-of name="i" />
+              </c:index>
+              <c:index>
+                <c:const value="1" type="integer" desc="Field value" />
+              </c:index>
+            </c:value-of>
+          </c:arg>
+
+          <!-- if it's sequential, we can cut down on the search substantially -->
+          <c:arg name="seq">
+            <c:value-of name="criteria">
+              <c:index>
+                <c:value-of name="i" />
+              </c:index>
+              <c:index>
+                <c:const value="2" type="integer" desc="Sequential flag" />
+              </c:index>
+            </c:value-of>
+          </c:arg>
+        </c:apply>
+      </c:otherwise>
+    </c:cases>
+  </function>
+</package>
+
diff --git a/core/when.xml b/core/when.xml
new file mode 100644
index 0000000..a5a647e
--- /dev/null
+++ b/core/when.xml
@@ -0,0 +1,76 @@
+<?xml version="1.0"?>
+<!--
+  Copyright (C) 2018 R-T Specialty, LLC.
+
+  This file is part of tame-core.
+
+  tame-core is free software: you can redistribute it and/or modify it
+  under the terms of the GNU General Public License as
+  published by the Free Software Foundation, either version 3 of the
+  License, or (at your option) any later version.
+
+  This program is distributed in the hope that it will be useful,
+  but WITHOUT ANY WARRANTY; without even the implied warranty of
+  MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+  GNU General Public License for more details.
+
+  You should have received a copy of the GNU General Public License
+  along with this program.  If not, see <http://www.gnu.org/licenses/>.
+-->
+<package xmlns="http://www.lovullo.com/rater"
+         xmlns:c="http://www.lovullo.com/calc"
+         xmlns:t="http://www.lovullo.com/rater/apply-template"
+         core="true"
+         desc="When Matching">
+
+  <import package="base" />
+
+
+  This package deals with matching in calculations (such as case statements
+    or values);
+    for classifications, see \tt{vector/cmatch}.
+
+
+  <inline-template>
+    <for-each>
+      <set cmp="eq" />
+      <set cmp="ne" />
+      <set cmp="gt" />
+      <set cmp="gte" />
+      <set cmp="lt" />
+      <set cmp="lte" />
+    </for-each>
+
+
+    <template name="_when-{@cmp@}_" desc="Value predicate {@cmp@}">
+      <param name="@name@" desc="Name to assert against" />
+
+      <param name="@index@" desc="Name index (optional)">
+        <text></text>
+      </param>
+
+      <!-- former exists only for consistency with cmatch templates and will
+           produce an error -->
+      <param name="@const@" desc="Match against constant value" />
+      <param name="@value@" desc="Match against variable" />
+
+      <param name="@value_index@" desc="Value index (optional)">
+        <text></text>
+      </param>
+
+
+      <if name="@const@">
+        <error>
+          @const@ is deprecated; use @value@ with a #-prefix instead.
+        </error>
+      </if>
+
+
+      <c:when name="@name@" index="@index@">
+        <dyn-node name="c:{@cmp@}">
+          <c:value-of name="@value@" index="@value_index@" />
+        </dyn-node>
+      </c:when>
+    </template>
+  </inline-template>
+</package>