home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!olivea!mintaka.lcs.mit.edu!ai-lab!wheat-chex!glenn
- From: glenn@wheat-chex.ai.mit.edu (Glenn A. Adams)
- Newsgroups: comp.fonts
- Subject: Re: Any UNICODE Fonts ?
- Message-ID: <1hie8bINNjb7@life.ai.mit.edu>
- Date: 26 Dec 92 20:10:51 GMT
- References: <9212220823.AA38608@chaos.intercon.com> <1992Dec23.104810.26913@cs.ruu.nl> <1hap2qINNk33@life.ai.mit.edu>
- Organization: MIT Artificial Intelligence Laboratory
- Lines: 28
- NNTP-Posting-Host: wheat-chex.ai.mit.edu
-
-
- In article <1hap2qINNk33@life.ai.mit.edu> bkph@wheat-chex.ai.mit.edu (Berthold K.P. Horn) writes:
- >(Ghee, maybe TeX will be the first typesetting language that can actually
- >use a more or less full UNICODE font? But then again maybe not, since
- >no TeXie would want to, since it would presumably would contain no
- >ligatures).
-
- If a "UNICODE font" is a font that has only the nominal forms of the
- Unicode character set, then this would be true. However, as has been
- pointed out already, it is a grave misuse of the term Unicode, since
- the glyph space required for displaying Unicode data may well be larger
- than 2^24 or more, having perhaps many 10^6 ligatures in it. For example,
- displaying a medium complexity Arabic script like Ruq`ah might require
- 3-5 million ligatures if one were to precompose all ligatures in the
- font. [Practically speaking, the display of Ruq`ah and Nastaliq would
- use glyph busiting - a process of decomposing glyphs into smaller graphical
- components. This technology is already used by DynaLab in Taiwan to
- achieve an 8-10 times reduction in memory requirements for large Han
- fonts.]
-
- Glenn Adams
- Cambridge, Mass.
-
- P.S. Unicode 1.1 will have ff, fi, and ffi ligatures due to the ISO10646
- merger. However, I would recommend against using them except for backward
- compatibility with simple text rendering technology.
-
-
-