TODO

==================
  Instructions For Helpers
==================

Ok, things are a lot more miscellaneous now, and require a lot more
care and attention.

First off, 

  YOU MUST READ "README-tags"

in its entirety.  The rest of this won't make sense until you do.
If that document seems confusing, feel free to bug
rlpowell/camgusmis or zort, but it may mean you're not cut out to
help at this stage.

 ------

IF YOU CHANGE, ADD, OR REMOVE ANY ACTUAL CONTENT, I WILL EAT YOUR
FAMILY.  WITH STEAK SAUCE.  MMMMMM.

If you make a change that is not *solely* XML tags, that's a content
change, and I will hunt you down and bad things will occur.

 ------

==================
PDF Thoughts
==================

Try incorporating the example section of
http://linuxlibertine.sourceforge.net/Libertine-XeTex-EN.pdf , and
also
http://www.math.washington.edu/tex-archive/macros/xetex/latex/polyglossia/polyglossia.pdf
; see jbovlaste's pdf export for a working version maybe

==================
  Uncategorized Issues By Chapter
==================
2. S11: "gismu:" as a line by itself looks pretty lame.
2. Example 2.9 could do with some re-formatting (compare 5.3 in http://dag.github.com/cll/2/5/ )
4. Example 4.4 is formatted differently from other nearby and similar examples
4. Examples 4.24, 4.26-4.38, 4.52-4.70 are missing content
	- The content in question gets deleted by xml/docbook2html_preprocess.xsl
4. Examples 4.46-4.51 are broken, and 4.51 is not present in the red book
4. Broken HTML character codes on the last line of page 79 of the red book
	- This is from the <quote xml:lang="el"/> tag not being handled properly in the docbook->html conversion
5. Table of language acronyms in section 14 could use reformatting to save space
7. Section cmavo lists are missing series information
	- This seems to be due to the <series> tag not being handled by the xml->docbook conversion
7. Example 7.39 is hella ugly.  Like, seriously
7. Example 7.62, line 3, differs from the red book
8. Example 8.45, line 3, is missing quotes around a word
8. Extraneous spaces in the () parts after Example 8.49
9. Example 9.40 has the last two lines on one line
9. BAI tables at end of chapter are poorly formatted compared to the red book
10. pg. 223 of red book: duplicate notes on section content due to the new form for x-refs
10. Example 10.84 does not fit in with the rest of the examples
10. Example 10.147's fourth line should be at the end of the third
11. pg. 259 of the red book, definitions list is super ugly
12. pg. 288, extra space after a (
13. pg. 385, attitudinal scale cmavo table is missing a *lot*
13. Various examples have the last two lines on one line
13. Section 7: table has values flowing into extra columns instead of rows
13. Vocatives table puts the large paragraphs explaining the vocatives onto the existing table rows
14. Logical group letters are merely capitals; they need to stand out much better
14. Example 14.106 is missing a row
15. Diagram on pg. 375 could use some work
16. After Example 16.93: plurals with spaces between the valsi and the "s"
17. Letters table is ridiculously tall
18. Examples are missing numeric rows
18. Section 6 cmavo list: mo'o is in the wrong row
18. Section 8 cmavo list is missing every PA+PA row
18. Section 22, the Examples have a leading 'li' that the red book does not have
18. Section 23's selma'o table has xrefs that are not showing up
18. Section 24 is ridiculously tall
19. Examples 4.2 and 4.6 have their rows out of order
19. Example 19.33 is missing a row
19. Examples 19.52-19.54, 19.67, 19.71-19.74, 19.77, 19.81 are missing quotes
19. Examples 19.61, 19.62 are missing emphasis
20. All the examples are using different markup
20. xrefs here can be a little verbose

 ------

==================
I Don't Even
==================
24-22:44 < rlpowell> *Neat*.
24-22:45 < rlpowell> The HTML and the LaTeX number the examples differently.
24-22:45 < rlpowell> *THAT'S* gonna need to change.
[change that *AFTER* handling all the proofread stuff, or it's going to suck)

- ---------

16-12:05 <    rlpowell> Random thought that never occured to me before so I'm writing it down for myself:
                        in the CLL there's the whole "underline the sumti and double-underline the
                        selbri" thing, which real typesetting software gets very cranky about.
16-12:05 <    rlpowell> It suddenly occurs to me that everything I'm doing is in UTF-8; I can just *draw
                        my own fucking lines*.
16-12:05 <    rlpowell> With the actual UTF-8 line drawing characters.

Or, draw boxes!  Regular for selbri and dashed for sumti.

==================
  Non-Visual Issues
==================

Maybe add an interlinear-gloss-manual? for manual grouping that
isn't chapter 2 style; otherwise it's hard to see how to make
something like this work:

    le    se nelci      cu   cafne
    The   liked-thing   is-frequent.

 ------
 
The <definition> tags within tables have their <valsi> component in another column of the table.  We need to decide whether is this is
satisfactory, or if something else should be done with them.

 ------
 
<lojbanization> tables are shit

 ------
 
All the imported/manual indexterm entries for single words or
selma'o need to be checked; they should be covered adequately by
nearby <cmavo-list> or <valsi> entries, both of which create index
entries.  If you don't know how to search for only the relevant
entries in your particular text editor, come on #jbopre and ask.

 ------

Clean up the "FIXME: TAG SPOT" bits.  Those were put there as part
of the semi-automated index insertion project.  If there's an
<indexterm> next to them, great; kill it.  If there isn't, maybe
look around nearby and make sure that obvious <valsi> tags and
similar are in use.

 ------

Similarily, entries like <indexterm type="lojban-word-imported">
(NOTE the type) probably need to be destroyed; there should be a bit
of Lojban near them that can be wrapped in <valsi>, to exactly the
same effect as the <indexterm> entry.  If you can't find such a
correspondance, add a FIXME comment.

 ------

Hunt down all FIXME:s and <!-- ... --> with todo type comments.
That's all going to be advanced shit, though.

==================
  Visual Issues
==================

After we're done with all that, we need to actually make it look
nice.  That's going to be quite a bit of work, and mostly on the
XSLT end of things, but here are some things you can do to maybe
help (you're certainly free to hack the XSLT if you know how,
though).

Make sure the <interlinear-gloss> bits look right in the output (which
is why the columns are so spaced out).

==================
  Data
==================
- It seems that there is a problem with the cmavo lists at the
  beginning of sections having missing entries; in particular, ones
  wwith +/* in the selma'o, maybe?; they need to be manually checked
- lojban words, lojban phrases, terms of art ("abstraction"),
  others?... should each have their own index
- list the members of each selma'o in chapter 20
- <phrase role="IPA">∞</phrase> is *not* IPA
- In the mediaobjects in chapter 1, can we do better than <alt>The
  picture for chapter 20</alt> in terms of non-numerical cross
  referencing?
- there's a bunch of problems in 18.xml ; a lot of math stuff has
  gotten truncated in the final form.  they basically all need to be
  checked; having a CLL to hand would help
- Fix all FIXMEs; in particular the output of:
  grep 'FIXME:' [0-9]*.xml | grep -v 'TAG SPOT'

18-17:44 < tsani> rlpowell: I think that having the glossary split into [cmavo], [brivla], and
          [cmene] sections might be worth while
18-17:44 < tsani> rlpowell: where the cmavo definitions say in which section the cmavo is
          discussed

==================
  Display
==================
- links to examples should *say* "example N"; easy to test from the
  index
- links to sections should say the number of section and chapter
- examples should be 5.2.1 rather than 5.6 for the first example in
  section 2 of chapter 5
- print: Add a header or footer to the left or the right page that
  has the number of the chapter you're in.

- for chapter 13, needs code changes to handle things like this:

      <cmavo-entry>
        <cmavo>fu'i</cmavo>
        <gismu>[frili]</gismu>
        <attitudinal-scale point="sai">with help</attitudinal-scale>
        <attitudinal-scale point="sai">easily</attitudinal-scale>
        <attitudinal-scale point="cu'i">without help</attitudinal-scale>
        <attitudinal-scale point="nai">with opposition</attitudinal-scale>
        <attitudinal-scale point="nai">with difficulty</attitudinal-scale>
      </cmavo-entry>

  the way the red book does

==================
  Both/Other
==================
- make sure backwards-compatible anchors exist
  - Ideally, make sure they are autogenerated as part of the HTML
    production.
- Similarily, generate more-readable anchors based on current
  section number and such

==================
Content Issues That Need Fixing
==================

DON'T TOUCH THESE.  The timing on these should be handled very carefully

2. Description of formatting for interlinear-gloss needs to be fixed to match the formatting changes
2. section 2.4 should be dropped eventually.