Julian Bradfield writes:
I have some ideas, but they have to wait for me to finish the
However, you mention OOo and Firefox - does anything (apart from
Emacs!) do better? It would be nice to have an example of material
that Firefox deals with badly, but something else deals with well.
Well, TeX does pretty damn well, as you'd expect. Omega might be even
better but Omega is still kinda painful to work with. I don't have an
example text, offhand, but IIRC Firefox and OOo both occasionally
forget that *all* spaces at line breaks should be deleted when
wrapping Japanese, and only TeX gets the thinspaces around foreign
words in Japanese correct IMO (but I don't know what real Japanese
speakers think about that).
I can't imagine why one would want to use Emacs to do typesetting
Real typesetting, of course not. I wrote "improve capabilities", not
"attain world domination"! ;-)
>For what it's worth, I would have no trouble reading Japanese
>such as 将来 (future) or 将軍 (shogun == generalissimo) if written
>with the reference glyph (in the context of other simplified Han).
And are you doing that by "error correction", i.e. seeing a wrong
shape, but matching it to a nearby word that you know, or do you read
as one might read Gothic (Fraktur) script, recognizing that it's the
right letters, but in a style one doesn't know well?
The latter. Two out of three right components in the right place
ain't bad, after all (especially given that Unicode standardization is
dependent on government work! ;-) The standard simplifications of
Simplified Chinese are mostly similar to the deformations that result
in (handwritten) abbreviated kanji, or Japanese kana. 4-stroke Claw
-> 3-stroke Evening is plausible, however in Chinese I'm not conscious
of other characters such as 受 which don't get that treatment (had to
look it up).
XEmacs-Beta mailing list