Re: [wEr\ Ar\ ju: fr6m] ?

> X-WebMail-UserID:  alrivera@alumni.southern.edu
> X-EXP32-SerialNo: 00002783
> Date:         Sun, 11 Nov 2001 19:31:01 -0500
> From: Muke Tever <alrivera@...>
>
> >===== Original Message From Constructed Languages List
> <CONLANG@...> =====
> >Should you really use ligatures in the IPA? I thought, given one
> >glyph was supposed to equal one sound, you shouldn't. Isn't that
> >the whole philosophy behind only using script g and not text g?
> You do, however, want good ligatures for connecting "modifier
> letters" such as the rhotic hook and "combining diacritics" such as
> the voiceless ring below to letters. At least, if you want them
> properly placed. Good typography demandeth it. ;p
Exactly --- try adding a diacritic below /y/, and see it ends up.

It would also be nice just to have characters for implosive and
retroflex hooks and then let the display system pick the right
alternative letter shapes based on information in the font. As it is,
you have to know where to find the combined letters yourself --- or if
you're lucky, the Unicode tables will tell you.

There's also the little remark on the IPA chart that diacritics should
go above characters with descenders. I was just skimming the OpenType
spec earlier tonight, and that's something you'd be able to specify as
a context-dependent glyph replacement. But if anything out there would
actually use the info, I don't know. PageMaker, perhaps.

(It can't be fixed at the Unicode level, the extra diacritics simply
aren't there).

Lars Mathiesen (U of Copenhagen CS Dep) <thorinn@...> (Humour NOT marked)

Reply