W3C home > Mailing lists > Public > www-math@w3.org > November 2002

[MathML compeditor] John Forkosh: announcing mimetex

From: Michael Kohlhase <Michael_Kohlhase@asuka.mt.cs.cmu.edu>
Date: Fri, 15 Nov 2002 09:36:35 -0500
Message-Id: <200211151436.JAA06111@tux.w3.org>
To: www-math@w3.org

Dear all, 

I just saw this posting on comp.text.tex, and it seemed relevant to this


   Dr. Michael Kohlhase,               Office: Newell Simon Hall 4623
   Adjunct Associate Professor            5000 Forbes Avenue, 
   School of Computer Science             Pittsburgh, Pa 15213-3891, USA.
   Carnegie Mellon University          tel/fax: +1 412 268 5749/6298           
   http://www.cs.cmu.edu/~kohlhase     e-mail: <kohlhase+@cs.cmu.edu>          

------- Forwarded Message

Forwarded: Fri, 15 Nov 2002 09:34:25 -0500
Forwarded: "CCaps Developers <+dist+~sutner/dl/ccaps-dev.dl@andrew.cmu.edu> "
Path: cantaloupe.srv.cs.cmu.edu!newsfeed.srv.cs.cmu.edu!nntp.ece.cmu.edu!usenet01.sei.cmu.edu!news.cis.ohio-state.edu!news.ems.psu.edu!news.cse.psu.edu!uwm.edu!vixen.cso.uiuc.edu!news.uchicago.edu!newsswitch.lcs.mit.edu!bloom-beacon.mit.edu!panix!forkosh
From: John Forkosh <john@invalid.com>
Newsgroups: comp.text.tex
Subject: announcing mimetex
Date: Fri, 15 Nov 2002 14:05:50 +0000 (UTC)
Organization: PANIX Public Access Internet and UNIX, NYC
Lines: 60
Message-ID: <ar2uvu$n33$1@reader1.panix.com>
NNTP-Posting-Host: panix3.panix.com
X-Trace: reader1.panix.com 1037369150 23651 (15 Nov 2002 14:05:50 GMT)
X-Complaints-To: abuse@panix.com
NNTP-Posting-Date: Fri, 15 Nov 2002 14:05:50 +0000 (UTC)
User-Agent: tinx 1.2.12
Xref: asuka.mt.cs.cmu.edu comp.text.tex:150816

   Around the middle of September I got some very helpful advice
   in comp.text.tex about using TeX fonts in an external program,
   and around the end of October uploaded gpl'ed mimeTeX to
   Note:  a nasty bug in its array command is fixed by replacing
   mimetex.c and mimetex.h with the two files in

Brief description:
   MimeTeX parses LaTeX math expressions, emitting either mime
   xbitmaps or gif images of them, which can be used in html
   documents in the form, e.g.,
     <img src="../cgi-bin/mimetex.cgi?f(x)=\int_{-\infty}^xe^{-t^2}dt"
     border=0 align=absmiddle>
   allowing you to embed math directly in html, which reduces
   the need for lots of external gif images and makes your html
   documents more readable and easily maintained.  An online
   demo and tutorial is at

   MimeTeX isn't primarily meant for latex2html-like tasks
   where you're maintaining native (La)TeX documents that are
   later redistributed in several formats, including html.
   Rather, mimeTeX is primarily meant to help maintain
   native html documents containing math.  In this sense
   it's a kind of "lightweight" alternative to MathML,
   with the advantage that mimeTeX preserves (La)TeX syntax.

Similar non-MathML packages, including textogif
   and gladTeX,
   require setup procedures that use TeX to help generate external
   gif (or png) images of your equations, which are later included
   in your html document as it's being rendered.  MimeTeX, as far
   as I know, is the only such non-MathML package that has its
   own built-in rendering engine, entirely independent of TeX,
   and therefore requires no setup procedure or external images
   whatsoever.  It renders realtime, on-the-fly images directly
   from your LaTeX math embedded in html documents.

   Widespread use of MathML will eventually dilute the
   population of LaTeX-aware users, muddying LaTeX's future.
   (La)TeX is more than "TeX The Program"; TeX is its syntax.
   Knuth produced a test suite that validates any program
   claiming to be TeX, so no one version of the code is
   particularly important.  It's the syntax that's important.
   (La)TeX will survive so long as a significant user
   population continues to use this syntax.
        MathML poses a threat to LaTeX's syntax in the html/xml
   market, so it's important to provide some LaTeX-compliant
   alternative.  MimeTeX is meant to be a prototype alternative.
   It's probably too small and kludgey for a final solution,
   but it demonstrates feasibility, and is full-featured
   enough to measure potential interest in LaTeX-compliant
   alternatives to MathML.
- --
John Forkosh  ( mailto:  j@f.com  where j=john and f=forkosh )

------- End of Forwarded Message
Received on Friday, 15 November 2002 09:36:45 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 20 February 2010 06:12:51 GMT