[omega] Draft of paper for EuroTeX 2005

[Gábor BELLA]

On Sun, 2005-03-06 at 15:48 +0100, Antonio Regidor García wrote:
> Hi,
> 
> I have still to read again and think calmly about the paper but I have
> a question: the character inside a sign will be allways a Unicode
> character, or more complex structures will be allowed, like kanji
> defined recursively from their radicals? (Maybe in this case instead of
> <character, properties, glyph> we will need something like
> <pseudo_character_glyph, properties> or <radical_tree, properties,
> glyph>).

Hello Antonio,

I don't know if Yannis has already answered your question but yes, the
Unicode character will always be included in a sign (unless in the rare
cases when it is a "characterless" sign, i.e., a glyph that does not
have a corresponding character). This however shouldn't stop us from
including structures like the radical tree along with it (à la CDL, for
example: http://www.wenlin.com/cdl/ ). Indeed we have thought about this
possibility which seems very interesting: having more complex structures
such as graphs as property values would be useful in cases like this
one. The exact way such a thing should be implemented remains to be seen
though. (The actual byte-level format for representing signs is still an
open question.)

Regards,
Gábor