README-tags

      ============
      Introduction
      ============

The docbook for the CLL uses a lot of our own custom/homebrew xml tags.

No really.  A *lot*.

These are all auto-converted to basic docbook.  Our customization of
docbook itself is very minor; just the usual CSS and
docbook2html_config.xsl stuff.

The basic point of all the custom tags is to make it easy to enter
our own special-case stuff, and also to maintain semantic
distinctions that we might want to mark up later, whether we want to
mark them up specially right now or not.  The actual markup is done
with the "role" attribute and CSS, basically.

The following is a list of the custom tags and when to use them.
Mostly it's just in the form of examples; ask rlpowell/camgusmis if
you want more detail.

IMPORTANT: These tags (both name and structure) are not sacred; if
you can see a better way to do things, please let rlpowell/camgusmis
know.

      Simple Phrase Markup
      ====================

These are used to markup inline phrases, mostly in Lojban.

  oldjbophrase
  ------------

FIXME: Remove when not needed.

This exists only as an artifact of the conversion process.  Feel
free to help.  These tags should be turned into one of the other
entries in this section.

  jbophrase
  ---------

Example: "a man biting a dog at a specified place and time. But
Lojbanic events may be much more <quote>spread out</quote> than
that: <jbophrase>mi vasxu</jbophrase> (I breathe) is something
which is true during the whole of my life"

This is used for any sequence of Lojban words that is simply present
inline as normal text.  For now at least, these are entered in a
special Lojban phrase glossary; we'll see how useful that turns out
to be.  (<--- FIXME: rephrase when decided)

  jbophrase validity
  ------------------

A <jbophrase> that contains invalid Lojban should be <jbophrase
valid="false">, and one that contains strange or surprising Lojban
should be <jbophrase valid="iffy">, to give us the option of marking
them up with special colours or other indicators later.


  valsi
  -----

Example: "It is also possible to put the tense somewhere else in the
bridi by adding <valsi>ku</valsi> after it."

This is used for a single Lojban word when it is referenced in the
text (that is, when discussing the word, not when using its actual
meaning).  This results in a link from the Lojban glossary.

NB: This means that anything wrapped in valsi must be findable in
jbovlaste.

A special case here is compound cmavo.  Where the compound has its
own meaning that isn't necessarily composable from the parts, use a
single tag, like so:

  <valse>.uinai</valsi>

Where it is decomposable, use two adjacent, like so:

  <valse>je</valsi><valsi>nai</valsi>

  valsi validity
  --------------

<valsi valid="false">ro'irre'o</valsi> should be used where an
example morphologically illegal word is given, as in that example.
  
  cmevla
  ------

Like "valsi" (including validity) but with no expectation that it
can be found in a dictionary.

  diphthong
  ---------

Example: "(for example, <diphthong>io</diphthong> is pronounced
<quote>yo</quote>)"

Used to markup diphthong in discussion of morphology.

  letteral
  --------

Example: "they all end in the letter <letteral>o</letteral>, which
is otherwise a rare letter in Lojban gismu."

Used to markup individual letters when their use in Lojban is being
referred to by the text.
  
  rafsi
  -----

Example: "<para>In making a lujvo that contains <rafsi>jax-</rafsi>
for a selbri that contains <valsi>jai</valsi>,"

Used to markup a rafsi when referred to as such.
  
  morphology
  ----------

Example: "As a result, <morphology>bf</morphology> is forbidden, and
so is <morphology>sd</morphology>"

Used to markup any morphological examples not otherwise covered.
  
  inlinemath
  ----------

Example: <inlinemath>(1000 * 6) - (500 * 0) + (100 * 0) - (10 * 15) - 3 = 5847</inlinemath>

Used to wrap math that appears inline; just shorthand for docbook's
<inlineequation><mathphrase>
  
  math
  ----
  
Example: <math>(1000 * L) - (500 * A) + (100 * H) - (10 * R) - V</math>

Used to wrap math that appears as its own paragrah; just shorthand
for docbook's <informalequation><mathphrase>

  grammar-template
  ----------------

Example:

  The syntax of jeks is:</para>
  <grammar-template>
    [na] [se] JA [nai]
  </grammar-template>
  <para>parallel to eks and giheks.</para>

Used to show (usually very simplified versions of) the Lojban formal
grammatical productions.

  definition
  ----------

Example:

  <definition>
    <valsi>bridi</valsi> <content>x1 is a predicate relationship with relation x2 (abstraction) among arguments (sequence/set) x3</content>
  </definition>

Used to denote a free-standing definition for a Lojban word (as opposed to the
cmavo-list sections).

      The cmavo Lists
      ===============

At the top of most sections, and sometimes embedded throughout, is a list of
cmavo, with (at least) the cmavo, a selmaho, and a description.

Basic Example:

    <cmavo-list>
      <cmavo-entry>
        <cmavo>bo</cmavo>
        <selmaho>BO</selmaho>
        <description>closest scope grouping</description>
      </cmavo-entry>
    </cmavo-list>

Example that covers the attitudinal version:

    <cmavo-entry>
      <cmavo>.u'i</cmavo>
      <attitudinal-scale point="sai">amusement</attitudinal-scale>
      <attitudinal-scale point="nai">weariness</attitudinal-scale>
    </cmavo-entry>
    
Example that covers the sumtcita version:

    <cmavo-entry>
      <cmavo>bai</cmavo>
      <gismu>bapli</gismu>
      <modal-place>compelled by</modal-place>
      <modal-place se="se">compelling</modal-place>
    </cmavo-entry>

THere's also a <cmavo-list-head> which works just like a normal
table/list head, and only actually occurs once so far.

      Examples
      ========

A truly staggering percentage of the CLL consists of what it calls
"exmaples" (they usually aren't examples of anything really, but
that's neither here nor there).  These come in a few set patterns,
which we've codified.  Each of these comes with a bunch of sub-tags,
as well, that are only valid inside examples.

Note that the <jbo> tags here can also take valid="false" and
valid="iffy", like <jbophrase>.

  interlinear-gloss-example
  -------------------------

(This is a technical linguistics term for word-by-word
pseudo-translations; see
http://www.eva.mpg.de/lingua/resources/glossing-rules.php )

Example:

    <example role="interlinear-gloss-example" xml:id="example-do-mamta-mi">
      <title>
        <anchor xml:id="c5e1d1"/>
      </title>
      <interlinear-gloss>
        <jbo>do mamta mi</jbo>
        <gloss>You are-a-mother-of me</gloss>
        <natlang>You are my mother</natlang>
      </interlinear-gloss>
    </example>

Inside the <interlinear-gloss>, <jbo> marks raw Lojban, <gloss> marks natural
language text (normally English) that matches the Lojban word for word, and
<natlang> marks natural language text (normally English) that more
loosely/colloquially translates the Lojban.

  pronunciation-example
  ---------------------

Example:

    <example role="pronunciation-example" xml:id="example-random-id-k2B4">
      <title>
        <anchor xml:id="c3e3d1"/>
      </title>
      <pronunciation>
        <jbo>.i.ai.i.ai.o</jbo>
        <ipa><phrase role="IPA">[ʔi ʔaj ʔi ʔaj ʔo]</phrase></ipa>
        <natlang>Ee! Eye! Ee! Eye! Oh!</natlang>
      </pronunciation>
    </example>

Similar in concept, but the <ipa> section matches the <jbo> section
in terms of sounds rather than words.

  lojbanization-example
  ---------------------

Example:

    <example role="lojbanization-example" xml:id="example-random-id-DQju">
      <title>
        <indexterm type="example-imported"><primary>cobra</primary><secondary>example</secondary></indexterm>
        <anchor xml:id="c4e7d6"/>
      </title>
      <lojbanization>
        <natlang>cobra</natlang>
        <jbo>kobra <comment>Lojbanize</comment></jbo>
        <jbo>sinc,r,kobra <comment>prefix rafsi</comment></jbo>
      </lojbanization>
    </example>

Used for demonstration of conversion of a natural language word or
name into a Lojban word.  Note the <comment> tag that can be
associated with particular lines.

  lujvo-example
  -------------

Example:

    <example xml:id="example-random-id-qjbP" role="lujvo-example">
      <title>
        <indexterm type="example-imported"><primary>supper</primary><secondary>example</secondary></indexterm>
        <anchor xml:id="c4e6d5"/>
      </title>
      <lujvo-making>
        <jbo>vancysanmi</jbo>
        <veljvo>vanci sanmi</veljvo>
        <gloss><quote>evening meal</quote></gloss>
        <natlang>or <quote>supper</quote></natlang>
      </lujvo-making>
    </example>

Used to show conversion between a lujvo and the words used to make
it, and its meaning; note the new <veljvo> internal tag.

  compound-cmavo-example
  ----------------------

Example:

    <example xml:id="example-random-id-qIYK" role="compound-cmavo-example">
      <title>
        <anchor xml:id="c4e2d2"/>
      </title>
      <compound-cmavo>
        <jbo>punaijecanai</jbo>
        <jbo>pu nai je ca nai</jbo>
      </compound-cmavo>
    </example>

Used to show the breakup of compound cmavo.

      Special Cases
      =============

<phrase role="logical-vowel">A</phrase> denotes one of the A, E, O
or U vowels associated with Lojbanic logic operations.

<lujvo-making>
  <jbo>zbasai</jbo>
  <rafsi>zba + sai</rafsi>
  <score><inlinemath>(1000 * 6) - (500 * 0) + (100 * 0) - (10 * 15) - 3 = 5847</inlinemath></score>
</lujvo-making>