Return-Path: Received: from [129.175.52.4] (HELO matups.math.u-psud.fr) by vsu.ru (CommuniGate Pro SMTP 3.5.9) with ESMTP id 924144; Sun, 16 Jun 2002 07:25:36 +0400 Received: from stats.math.u-psud.fr (beryl.math.u-psud.fr [129.175.54.194]) by matups.math.u-psud.fr (8.11.6/jtpda-5.3.3) with ESMTP id g5G3PZY12178 ; Sun, 16 Jun 2002 05:25:35 +0200 (MEST) Received: (from sieben@localhost) by stats.math.u-psud.fr (8.11.6+Sun/8.11.6) id g5G4OqD09901; Sun, 16 Jun 2002 05:24:52 +0100 (WEST) Date: Sun, 16 Jun 2002 05:24:52 +0100 (WEST) From: Laurent Siebenmann Message-Id: <200206160424.g5G4OqD09901@beryl.math.u-psud.fr> To: CyrTeX-en@vsu.ru, vvv@vsu.ru Subject: Re: Russian/Polish/German...without switching Hi Vladimir and others > first, i wonder, how you can make a distinction (say color) in your > editor for German A and Polish A for text files?? text files do not > have any additional structure besides characters. Knuth once threw a curveball in a TeX meeting by arguing that font glyphs should include arbitrary default coloring; that ran counter too all current work in TeX -- which imposed color by external color-switch commands and (usually) a color stack. Maybe I too am swimming against the current. But recently, most operating systems have begun to invent new notions of font for the coming era of 16 and 32 bit fonts. I am daring to imagine that some of the screen fonts will be able have default color for the characters. It is in such a world, that my notion of the "stacked 16 bit multilingual" screenfont lives most comfortably. There would be no language switches in multilingual text. Rather the parts of the big font for Russian, French and English respectively would be disjoint and color coded. If keyboards are to stay roughly as they are today, one might switch language with one of the dozen or more function keys always available (mostly unused it seems). But such a switch would still leave one in the same 16 or 32 bit font, and every typed character would hopefully appear onscreen as a visually recognisable glyph-with-color. That is a simplistic scheme that any typist instantly understands and every literate person will know how to read. There is a conceptual difference between the insertion of switching commands and the use of "switched" type. Something like the difference between a function and its derivative. It makes a difference both to the typist/author and to the TeX programmer-typographer. At any rate, for the day I venture into the 16 bit font world, as a multilingual manuscript creator I hope I will have something as clean and simple to offer the typist, be that me or a secretary. I have still to argue that the lot of the programmer-typographer for such "stacked multilingual typing" can be a happy one. Cheers Laurent S. PS. Thanks for the hints on russification of emacs; I still have to get up to speed in the emacs world. Emacs finally has proved stable for me on Mac and Wintel. Since it is the one (??) text editor working stably on Mac PC and linux it commands attention -- in spite of a user interface with many distressingly antedeluvian aspects. PS. Leif's report of how Nisus writer can realize the visual aspects of the above typing using just the Macintosh font technology of the 1980's and the color controls of the 1990's is interesting and accurate (I believe). There is a free version of Nisus Writer available at http://www.nisus.com. Further, it would be possible to convert such Nisus typing to a straight 16 bit text file (with a single font), modulo programming a reasonably simple Nisus specific converter.