Path: utzoo!utgpu!jarvis.csri.toronto.edu!rutgers!tut.cis.ohio-state.edu!pt.cs.cmu.edu!cadre!sean
From: sean@cadre.dsl.PITTSBURGH.EDU (Sean McLinden)
Newsgroups: comp.text
Subject: Re: LaTeX, SGML, ODA
Message-ID: <3280@cadre.dsl.PITTSBURGH.EDU>
Date: 21 Aug 89 02:22:56 GMT
References: <859@cgch.UUCP> <1215@gmdzi.UUCP>
Reply-To: sean@cadre.dsl.pitt.edu (Sean McLinden)
Organization: Decision Systems Lab., Univ. of Pittsburgh, PA.
Lines: 48

In article <1215@gmdzi.UUCP: thomas@gmdzi.UUCP (Thomas Gordon) writes:
:From article <859@cgch.UUCP:, by wscd@cgch.UUCP (Dirk Schwarz-Hertzner):
:: Is there a newsgroup discussing SGML and/or ODA/ODIF?  
:: Does anyone have a "grammar" for LaTeX that is suiatable for a basis of
:: a SGML description? I admit that in LaTeX nearly everything is allowed 
:: everywhere. But maybe such a grammar to some extent describes what is not
:: allowed.
:: 
:
:The German Research Net (DFN) distributes an SGML-based set of document
:types (paper, report, ...) together with programs for translating
:such documents into LaTeX and nroff/troff.  The structure of the
:paper and report document types is very similar to the comparable
:LaTeX document types.    The system is called Daphne.   If there
:is any interest, I'll dig out DFN's address.

I, for one, would be interested. In answer to the first question, I
suspect that there will be, someday, enough interest in a newsgroup such
as comp.protocols.iso.sgml, bt first, enough people will have had to see
a real world implementation to discuss it.

In terms of ODA, there is an ODA toolkit available from Carnegie-Mellonn
University that is a set of C library routines. This was developed as
part of the NSF Expres project with ODA selected as a mapping between
various markup and page description languages. I suppose that there are
actual programs that use this library, but they didn't come with my
distribution tape.

I have been told of an ODA implementation based on ISODE 5.0 done by
some group in Norway, but I don't know much more about it or its
availability. There was also a Swedish implementation of an SGML parser
but it was too expensive for my (University) blood. I don't know if
either of these support a mapping but it ought to be easy to map from
TeX to almost anything.

Digital Equipment Corporation has been making noises about their extensions
to ODA (called CDA for "Compound Document Architecture"), being incorporated
into the ODA standard but I suspect that IBM might be obstructionist so
whether or not this will be available outside a DEC platform is unclear.

All of this information is nth hand, so if anyone knows otherwise, or
anyone would like to put some of this software into the public domain
we'd all appreciate it.


Sean McLinden
Decision Systems Laboratory
University of Pittsburgh