Return-Path: Received: from video.uic.vsu.ru ([62.76.169.38] verified) by vsu.ru (CommuniGate Pro SMTP 3.5.9) with ESMTP id 851328 for CyrTeX-en@vsu.ru; Tue, 21 May 2002 11:37:27 +0400 Sender: vvv@video.uic.vsu.ru To: (Cyrillic TeX Users Group) Subject: Re: Russian/Polish/German...without switching References: From: Vladimir Volovich Date: Tue, 21 May 2002 11:39:00 +0400 In-Reply-To: (Laurent Siebenmann's message of "Tue, 21 May 2002 02:49:46 +0100 (WEST)") Message-ID: Lines: 65 User-Agent: Gnus/5.090006 (Oort Gnus v0.06) Emacs/21.1 (sparc-sun-solaris2.8) MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii "LS" == Laurent Siebenmann writes: >> but you can add definitions like >> \def\setcyr{\fontencoding{T2A}\selectfont} >> \def\setlat{\fontencoding{T1}\selectfont} LS> That leaves us switching between two environments rather than the LS> expected three. LS> Having done this sort of programming, I would guess that all LS> switching could be eliminated via a more complex UTF8 macro LS> package. i had some ideas on improving the utf-8 input encoding package which will provide "transparent" font encoding switching on demand. i think i know a fairly good method for this, but i didn't have time to implement it yet. But - if we think about this more, it will become obvious that a multilingual document must have language switching macros at least for good hyphenation. E.g., the text should look like some text in default language (e.g. english) \switchtorussian some text in russian (using utf-8 input encoding) \switchtopolish some text in polish (using utf-8 input encoding) \switchtogerman some text in german (using utf-8 input encoding) \switchtoenglish ... where \switchto commands can also look in babel style, e.g. \begin{otherlanguage}{russian} some text in russian (using utf-8 input encoding) \end{otherlanguage} The necessity of switching LANGUAGE in the markup of the document (not input encoding which may be UTF-8 and contain ALL letters of all needed languages, and not font encoding which may be implemented in a smarter utf-8 package) is necessary because obviously there is no other way TeX can grok the language: only the author of text knows that. And without these language-switching markup commands, hyphenation will be wrong. Note, that even Omega still needs these markup language-switching commands, by the same reasons! So, we do agree that markup commands like \switchtorussian, \switchtogerman, etc are necessary - for good hyphenation. But then, what prevents us to include automatic font encoding switching inside them? And we can see that it is not that bad to have not so smart utf-8.def which requires font encoding switching commands, -- these commands can be put inside language switching commands which ARE REQUIRED for quality typesetting -- be it TeX or Omega. Best, v.