% /usr/spool/ftp/pub/doc/tex-for-arabic.txt, Fri Oct 4 17:33:19 1991 % Edit by Nelson H. F. Beebe ; PS:TEX-FOR-ARABIC.TXT.5, 4-Jun-90 18:01:19, ; Edit by Nelson H. F. Beebe TeX for Arabic [04-Oct-1991] This file collects all the information I have been able to find on TeX for Arabic. For a recent survey of the Arabic encoding problem, and the status of Arabic word processing, see @Article{Tayli:arabic-wp, author = "Murat Tayli and Abdulla I. Al-Salamah", title = "Building Bilingual Microcomputer Systems", journal = CACM, volume = "33", number = "5", pages = "495--504", year = "1990", } @Article{Becker:arabic-wp, author = "Joseph D. Becker", title = "{Arabic Word Processing}", journal = CACM, year = "1987", volume = "30", number = "7", pages = "600--610", month = jul, } Some early work relating to standardization of keyboards and encoding, and even writing, is @Article{Kubba:arabic, author = "G.A. Kubba", title = "{The Impact of Computers on Arabic Writing, Character Processing, and Teaching}", journal = "Information Processing", year = "1980", volume = "80", pages = "961--965", } Pierre Mackay, is a past President of the TeX Users Group, as well as being a Professor and scholar of Arabic and Turkish at the University of Washington in Seattle, WA. For some years, he has been involved in the design of Arabic scripts with Metafont, and together with Don Knuth, developed TeX-XeT referred to in the text. Here are some related literature references: @Article{Mackay:byte-arabic-typesetting, author = "Pierre Mackay", title = "{Typesetting Problem Scripts}", journal = "Byte", year = "1986", volume = "11", number = "2", pages = "201--218", month = feb, } @Article{MacKay:TB4-2-76, author = "Pierre MacKay", title = "{{\TeX\ for Arabic script}}", journal = TUGboat, year = "1983", volume = "4", number = "2", pages = "76", month = Sep, } @Article{Knuth:TB8-1-14, author = "Donald Knuth and Pierre MacKay", title = "{{Mixing right-to-left texts with left-to-right texts}}", journal = TUGboat, year = "1987", volume = "8", number = "1", pages = "14", month = Apr, } @string{EPODD = "Electronic Publishing---Origination Dissemination and Design"} @Article{EPODD:Becker:triroff, author = "Zeev Becker and Daniel Berry", title = "{{\tt triroff}}, an adaptation of the device-independent {{\tt troff}} for formatting tri-directional text", journal = EPODD, year = "1989", volume = "2", number = "3", pages = "119--142", month = oct, } ------------------------------------------------------------------------ From TeXhax Digest V91 #40 (Sunday, September 15, 1991): Date: Thu, 29 Aug 1991 15:33:53 CDT From: "George D. Greenwade" Subject: ArabTeX files available at FILESERV Keywords: ArabTeX, FILESERV Yesterday, I forwarded along a post to INFO-TeX about ArabTeX. I have had a few private inquiries about getting these files (some who had no ftp access, others who had trouble getting ftp access). I have installed the basic files for access on FILESERV, and they are available for anonymous ftp retrieval from Niord.SHSU.edu [192.92.115.8]. I am attaching the main ArabTeX description file which has retrieval instructions, as well as the file lists for the related METAFONT sources required and the style files. Regards, George %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% George D. Greenwade, Ph.D. Bitnet: BED_GDG@SHSU Department of Economics and Business Analysis THEnet: SHSU::BED_GDG College of Business Administration Voice: (409) 294-1266 P. O. Box 2118 FAX: (409) 294-3612 Sam Houston State University Internet: bed_gdg@SHSU.edu Huntsville, TX 77341 bed_gdg%SHSU.decnet@relay.the.net %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%% ArabTeX ------- The ArabTeX package includes the preliminary distribution of files for Prof. Klaus Lagally's ArabTeX, a LaTeX extension for high-quality arabic writing. They are designed to work in concert with two other packages available on FILESERV, ArabTeX_STYLE (the style files used by ArabTeX), and ArabTeX_MF (the METAFONT sources to produce the NASH10 fonts used by ArabTeX). For sites with ftp access, the pre-built NASH10 fonts (pk and tfm files) are available from Niord.SHSU.edu [192.92.115.8] in the directory [.ARABTEX.NASH10]. The home ftp site for this package is ifi.informatik.uni-stuttgart.de [129.69.211.1]. You may retrieve the entire package of 5 files by including the command: SENDME ArabTeX in the body of a mail message to to FILESERV@SHSU.BITNET (FILESERV@SHSU.edu). A complete distribution of this version of ArabTeX requires all of the files in the ArabTeX and ArabTeX_STYLE packages, as well as either the ArabTeX_MF files or the prebuilt fonts. If, for some reason, you should only need one of these files, say ArabTeX.TEX, use the command: SENDME ArabTeX.TEX in your MAIL message to FILESERV. To retrieve ArabTeX_MF and ArabTeX_STYLE, include the commands: SENDME ArabTeX_MF SENDME ArabTeX_STYLE Files in this package: (1 Block = 512 bytes) File Blocks Save file as: ARABTEX.MAN 9 ARABTEX.MAN ARABTEX.READ_ME 4 READ.ME ARABTEX.SULTAN_PS 125 SULTAN.PS ARABTEX.SULTAN_TEX 8 SULTAN.TEX ARABTEX.TEX 15 ARABTEX.TEX Approximate total blocks in full ArabTeX package = 161 =============================================================================== Files in ArabTeX_MF (retrieve with SENDME ArabTeX_MF): Files in this package: (1 Block = 512 bytes) File Blocks Save file as: ARABTEX_MF.NASH10_MF 1 NASH10.MF ARABTEX_MF.NASH12_MF 1 NASH12.MF ARABTEX_MF.NASH14_MF 1 NASH14.MF ARABTEX_MF.NASH17_MF 1 NASH17.MF ARABTEX_MF.NASH20_MF 1 NASH20.MF ARABTEX_MF.NASH25_MF 1 NASH25.MF ARABTEX_MF.NASH72_MF 1 NASH72.MF ARABTEX_MF.NASHBASE_MF 17 NASHBASE.MF ARABTEX_MF.NASHCHAR_MF 31 NASHCHAR.MF ARABTEX_MF.NASHCODE_MF 3 NASHCODE.MF ARABTEX_MF.NASHDIA_MF 4 NASHDIA.MF ARABTEX_MF.NASHDIG_MF 4 NASHDIG.MF ARABTEX_MF.NASHFONT_MF 2 NASHFONT.MF ARABTEX_MF.NASHLIG_MF 9 NASHLIG.MF ARABTEX_MF.NASHSPEC_MF 9 NASHSPEC.MF ARABTEX_MF.NASH_MF 3 NASH.MF ARABTEX_MF.READ_ME 3 READ.ME Approximate total blocks in full ArabTeX_MF package = 92 =============================================================================== Files in ArabTeX_STYLE (retrieve with SENDME ArabTeX_STYLE): Files in this package: (1 Block = 512 bytes) File Blocks Save file as: ARABTEX_STYLE.AFONTS_STY 3 AFONTS.STY ARABTEX_STYLE.APARSE_STY 23 APARSE.STY ARABTEX_STYLE.ARABTEX_STY 16 ARABTEX.STY ARABTEX_STYLE.ASCAN_STY 6 ASCAN.STY ARABTEX_STYLE.ATRANS_STY 7 ATRANS.STY ARABTEX_STYLE.AWRITE_STY 19 AWRITE.STY ARABTEX_STYLE.READ_ME 2 READ.ME Approximate total blocks in full ArabTeX_STYLE package = 76 ------------------------------------------------------------------------ From TeXhax Digest V90 #47: Date: Thu, 3 May 90 17:12:11 -0700 From: mackay@cs.washington.edu (Pierre MacKay) Subject: Arabic TeX Keywords: TeX, Arabic Basically, Arabic TeX is still nowhere. The new TeX3.0 is a really promising foundation, but it is not a bidirectional system. The TeX-XeT change file needs to be upgraded so that it will match TeX3.0, and I am about half way through that. Fonts are still the chief problem, but Dominic Wujastyk is coordinating work on a basic Arabic font. Editing is yet another matter. Ideally, GNU Emacs should be internationalized for this, but it will take real effort. The GNU Emacs lisp can handle any text in which the display buffer is the same as the one being edited, but it does appear to have that restriction. For context sensitive languages, I think it is going to be necessary to separate the two, and have graphic image tokens in the display buffer that are constantly related to the edit buffer, but not the same. Other than that, GNU lisp looks as if it can handle any of the rest of the job quite effectively. It might be possible even to put bidirectionality into the display, but that is going to be rather a stinker. I am just starting to lay out the possibilities for full context evaluation using the extended ligature constructs of TeX3.0 and Virtual Font capabilities. Up till now, I could see little hope for anything other than a preprocessing stage to feed TeX, but I believe that the new ligature system can actually do the trick in a single stage. The bidirectional feature, however, must still be grafted on to TeX (it is essentially a mechanical task, since Donald Knuth said when he announced the plans for TeX3.0 that he was sure they would not materially affect the application of the XeT patch). So that is the slightly discouraging news. Other editing systems than GNU Emacs exist for Arabic, but they are essentially "word processors." In the long run, I would not want to be restricted to the capacities of a "word-processor" for Arabic, any more than for English. Some may be able to put up with WORD or FULLWRITE to compose TeX input, but I am not one of them. Pierre MacKay ------- ------------------------------------------------------------------------ [The End] ------------------------------------------------------------------------