Damian.Cugley@prg.ox.ac.uk (Damian Cugley) (09/29/90)
From: Hosek, Donald A. <dhosek@frigga.claremont.edu> Message-Id: <8691@jarthur.Claremont.EDU> > There is a draft 256-character character set that came out of the > TeX90 conference at Cork. It is inadequate since it doesn't leave > anywhere near enough empty glyphs for quality typesetting (the > f-ligs of cm are not the full set of possible roman ligatures and > variants, especially for italic typefaces or oldstyle typefaces). Whew, I'm glad someone agrees with me ... I got sent a draft of it & was apalled at how much PostScript has corrupted people's minds (they'll be calling cmcsc10 "cmexpert10" next...) In particular, I thought the inclusion of straight quote marks, visibubble space and other ASCII paraphenalia in *every* text font is foolish and wasteful of character slots - 256 characters is tight enough as it is! Also, trying to include "every" accented character fails 'cos they'll always be leaving *something* out. (What TeX needs is better accenting primitives.) I was tinkering with an idea for a more-than-128-character font encoding which included smallcaps & every symbol that depends on the font - including Old English ones etc. - by having no accented letters. It'd include two of each accent (one styled for UC letters), but would need alterations to TeX-the-program to do hyphenation and accent positioning right :-(. Also, only some characters - letters, punct etc - would have set positions, the rest would vary from font to font, allowing unusual ligatures in unusual fonts. In particular, a "typewriter" or "ASCII" font would include straight quotes, ASCII-style circumflex etc. Does this make sense or am I just weird? /--------------------------------------------------------------------\ | Damian Cugley, Programming Research Group, 11 Keble Rd, Oxford, UK | | Damian.Cugley@prg.oxford.ac.uk =or= ...@oxford.prg in UK | | "It's a good job there are weekdays between weekends to relax in." | \--------------------------------------------------------------------/
spqr@ecs.soton.ac.uk (Sebastian Rahtz) (09/30/90)
In article <DAMIAN.CUGLEY.90Sep29110644@msc3.prg.ox.ac.uk> Damian.Cugley@prg.ox.ac.uk (Damian Cugley) writes:
inclusion of straight quote marks, visibubble space and other ASCII
paraphenalia in *every* text font is foolish and wasteful of character
slots - 256 characters is tight enough as it is! Also, trying to
include "every" accented character fails 'cos they'll always be leaving
*something* out. (What TeX needs is better accenting primitives.)
i am inclined to sympathize, but I think the people who drafted it
were after a pragmatic solution rather than a new international
standard. they *did* throw away some of the silly things in ISO Latin
1, but anyone who decides to tinker with the 32--128 area has got to
be pretty confident. i am not prepared to say that `straight quotes'
are not the norm somewhere in Europe. people want an agreed set of
accented letters *now*, not an ideal approach that might work later.
i dont think the Cork Standard has much of a hope of succeeding. at
best it will replace TeX layout as a standard. and we know how many
people outside TeX use *that*
--
Sebastian Rahtz S.Rahtz@uk.ac.soton.ecs (JANET)
Computer Science S.Rahtz@ecs.soton.ac.uk (Bitnet)
Southampton S09 5NH, UK S.Rahtz@sot-ecs.uucp (uucp)Damian.Cugley@prg.ox.ac.uk (Damian Cugley) (10/01/90)
From: Sebastian Rahtz <spqr@ecs.soton.ac.uk> Message-Id: <SPQR.90Sep30141057@manutius.ecs.soton.ac.uk> > anyone who decides to tinker with the 32--128 area has got to > be pretty confident. But surely we are not talking about a new ASCII extension, we're talking about a new encoding scheme for typesetting fonts; there is no reason for the two to be similar. It's only due to a foolish omission from TeX's internals that we *have* to have, say, the uc letters in 65..90 rather than being able to place them arbitrarily and using `textcodes' (in the style of mathcodes) to map input chars to output chars. Firstly the characters corresponding to <space>, ", |, \, _, {, }, ^, ~, <, > will almost certainly *not* be used directly; these keys will either be active or special in some other way. So there's no need to have them in those particular places in the font. Secondly, apart from {, }, those characters *do not* appear in plain text and so there is no reason to have them in every text font anyway. If people want to use them as math symbols, they belong in a math font; if people want to use them in discusssing the ASCII set they should obtain them from a special ASCII font. Even the proposed Latin8 convention recognizes this (apparently without realizing it) - the neutral apostrophe/singlequote/acute character has been replaced by an apostrophe and the backquote/grave character as been replaced by an inverted comma. This conflicts directly with their assertion that the 95 ASCII "printing characters" are all done as the appropriate ASCII characters. > i am not prepared to say that `straight quotes' are not the norm > somewhere in Europe. Anyone know one way or another?
dhosek@frigga.claremont.edu (Hosek, Donald A.) (10/02/90)
In article <DAMIAN.CUGLEY.90Oct1142230@msc3.prg.ox.ac.uk>, Damian.Cugley@prg.ox.ac.uk (Damian Cugley) writes... >From: Sebastian Rahtz <spqr@ecs.soton.ac.uk> >> anyone who decides to tinker with the 32--128 area has got to >> be pretty confident. >But surely we are not talking about a new ASCII extension, we're talking >about a new encoding scheme for typesetting fonts; there is no reason >for the two to be similar. It's only due to a foolish omission from >TeX's internals that we *have* to have, say, the uc letters in 65..90 >rather than being able to place them arbitrarily and using `textcodes' >(in the style of mathcodes) to map input chars to output chars. >Firstly the characters corresponding to <space>, ", |, \, _, >{, }, ^, ~, <, > will almost certainly *not* be used directly; these >keys will either be active or special in some other way. So there's no >need to have them in those particular places in the font. >Secondly, apart from {, }, those characters *do not* appear in plain >text and so there is no reason to have them in every text font anyway. >If people want to use them as math symbols, they belong in a math font; >if people want to use them in discusssing the ASCII set they should >obtain them from a special ASCII font. Umm, did you know that it's possible to use TeX fonts on a Macintosh like any other font? Blue Sky Research has a PD utility for the conversion from PK to goofy Mac suitcases (or whatever they are). Well, once upon a time when I was "play-testing" Textures, I decided, "I'm typing TeX material, I'll use cm as my text font." Well, I couldn't use cmr since ", |, \, _, {, }, ^, ~, < and > were not in there places, so I decided to use cmtt instead. Guess what I got everytime I hit the space bar! (I don't remember whether I also got random characters with carriage return or not). Not all of us are content to limit our usage of TeX fonts to TeX. There are some sizable markets out there for fonts; if the TeX font coding standard is usable by other applications, maybe TeX people will be able to get other people's fonts in the local coding scheme (exciting prospect, no?). Think about it. The Cork standard still sucks, though. >> i am not prepared to say that `straight quotes' are not the norm >> somewhere in Europe. >Anyone know one way or another? The only language I can think of that uses identical quotes for openings and closings in typeset material is Hebrew. But they have a whole other alphabet. But straight quotes are standard for other applications (e.g., Word for Morons, etc.). Leave 'em in if it'll make it possible for some typeface ecumenism. -dh --- Don Hosek TeX, LaTeX, and Metafont support, consulting dhosek@ymir.claremont.edu installation and production work. dhosek@ymir.bitnet Free Estimates. uunet!jarthur!ymir Phone: 714-625-0147 finger dhosek@ymir.claremont.edu for more info
Damian.Cugley@prg.ox.ac.uk (Damian Cugley) (10/08/90)
From: Hosek, Donald A. <dhosek@frigga.claremont.edu> Message-Id: <8770@jarthur.Claremont.EDU> > Umm, did you know that it's possible to use TeX fonts on a > Macintosh like any other font? Well in that case you'd have to use the standard MacEncoding in order to get Mac's KeyCaps routine to get the quotes, ligatures etc. in the right place? Or can KeyCaps be reconfigured for different encodings? Here a "chars 32-126 are as ASCII" approach is inevitable because almost all the keyboard keys stand for themselves (I'd still cheat and use '` as quotes not quote/backquote myself, except in the tripewriter font). This is an example of the fundamental problems with having a fixed encodoing for characters built into a font - and is one of the places where PostScript wins over TeX hands-down. A given PostScript font can be rearranged in whatever order you like because glyphs are acessed using symbollic names; with TeX you'd have to build into the METAFONT code the necessary admin to produce characters in an ASCII-style ordering. Damian