This makes sense to me. The source files I am working from are
electronic dictionaries (not print dictionaries), so the <entry>
approach is appropriate. It makes a lot of sense to handle different
dictionaries differently. I will focus on the English-Vietnamese
dictionary first, and as the filters take shape I can decide whether to
convert them to <entryFree> or not.
Incidentally, I had some local friends over yesterday, and I showed
them what BibleCS can do with dictionaries (I had created modules for
the two Vietnamese dictionaries using ThML and used them for my own
purposes until the announcement about TEI support), and they were
excited to have a copy. Dictionary functionality, after the ability to
search the Bible, is probably the most important function of Bible
software, so I am excited about how things are developing.
Daniel
Chris Little wrote:
Daniel Owens wrote:
Like I said, for what it's worth... Perhaps Lexique will give you some
food for thought. It's a nice program for producing a nicely typeset
dictionary quickly and with minimal expertise. I might create Perl
scripts to go between TEI and their format if I ever have the need.
I should have started with a question. When tagging dictionaries, then,
is the onus on the module creator to add punctuation and other means of
distinguishing elements? That would certainly be simpler for the
front-end developers, but if that is the case it would be helpful to
know what elements are styled in what way so that module creators don't
double-up on that sort of thing. Or if I use the <hi> element and a
stylesheet does otherwise, will it matter? Should I just "style-away"
using TEI?
As DM says, <entry>-based dictionaries and <entryFree>-based
dictionaries are quite different and we will probably handle them
separately. A third option that exists, especially important for
<entryFree> dictionaries that come from printed sources, is entries that
make extensive use of the rend attribute (which is the case in my 1913
Webster's).
With <entryFree>-based dictionaries that make extensive use of rend, we
should generate very little decoration other than what is specifically
identified in the text. With other <entryFree>-based dictionaries, we
will want to generate decoration like italicized <pron>, bolded <orth>,
or small-caps <re> or <xr> links.
<entry>-based dictionaries are completely different, and we'll probably
generate decoration, punctuation, paragraphing & linebreaks, etc.
So, in terms of how much decoration & punctuation you should add, it
should depend on which kind of entry you use and how much decoration you
want to do.
TEI support at this point is very preliminary. Just as soon as I'm
finished encoding the 1913 Webster's, I'll take a look at it, the
Viet-Anh dictionary, the NAS lexicons, and possibly some of the old
Perseus/Germanic Lexicon Project dictionaries and try to develop some
nice filters for RTF, HTMLHREF, & WEBIF. I can't say what it will
necessarily look like, but I probably won't use any color or mid-line
textsize changes.
In any case, we have a while before 1.5.12 will be ready, so we can do
some testing before we settle on the generated rendering.
--Chris
_______________________________________________
sword-devel mailing list: sword-devel@crosswire.org
http://www.crosswire.org/mailman/listinfo/sword-devel
Instructions to unsubscribe/change your settings at above page
--
PMBX license 1502
|
