W3C home > Mailing lists > Public > www-math@w3.org > February 2005

Re: Semantic information for math representations of physics

From: Romeo Anghelache <romeo@roua.org>
Date: Thu, 03 Feb 2005 23:51:29 +0100
Message-ID: <4202AAF1.7090104@roua.org>
To: www-math@w3.org

JB Collins wrote:
> Maybe I threw you off at the brief reference to
> TeX/LaTeX.
> My comment about linkage to TeX/LaTeX arises from the
> following consideration. I need to provide a mechanism
> to get this documentation from its author to a user.
> TeX/LaTeX is the de facto standard application (some
> would say) for documentation by those who develop,
> author, and publish mathematical models. Given that,
> it  seems to me that semantic level of representation
> of mathematical models needs to begin embedded in a
> LaTeX document even though it may end up on the web in
> XML.
> It seems to me that others are thinking along these
> lines as well. (Set me straight if you think
> otherwise). I am still exploring available tools and
> trying to define exactly what I need to build for
> myself.

My two cents:
I think semantic authoring has to quit the TeX world as soon as possible.
It is a de facto standard useful only for those activities which are 
just concerned with form, or the flat semantics of the screen or paper.
The architecture of TeX, and of the macro layers built for it, proved, 
to me at least, highly obsolete and a hindrance in developing semantic 
layers with them.
Trying to produce semantics with TeX rapidly becomes a tolerated pile of 
patches because of the interference with the publishing oriented 
original design.

For those reasons, I think a fresh semantic authoring language is 
necessary to emerge. It should look like TeX, for the authors, but 
should be rather a meta-language focusing on definitions and 
relationships between them: a natural language with machine 
understandable islands belonging to different namespaces / controlled 
vocabularies (well-defined theoretical models).
The librarianship would benefit greatly from this, along with the 
(re)users of these documents, and publishers would be allowed to compete 
on matters of form and packaging without interfering with the semantic 
needs of the authors and of those readers who can use a computer.

OmDoc is the XML, machine-friendly, image map of the *author-friendly* 
language I am thinking about.

Until such a language will exist, trying to effectively build semantics 
with TeX is a big time-waster, unless one is focusing on recovery. My 
optimistic estimate for such a try is that one will end up with 5% of 
the old TeX functionality and 95% orthogonal semantics and white hair.
The (almost) relevant rest of the TeX functionality is already present, 
today, in the MathML presentation implementations from Design Science 
and Mozilla.

Given these, best thing to do is to start designing the vocabulary you 
need to cover the models you are concerned about, their properties, 
their mathematical image, and the portions which can be fed to machines.

Along this process, the invariants of the meta-language for authoring 
these will take shape too, perhaps just on your desktop.

I'm deeply interested in this subject but I won't be able to effectively 
  do anything that matters to it unless, hopefully, I'll settle for good 
in some library, mainly for this purpose.

Romeo Anghelache

> Have I cleared up your confusion?
> Thank you for suggesting gellmu as a tool. Do you have
> a website or preferred reference?
> Regards,
> Joe Collins
> Naval Research Lab
> --- William F Hammond <hammond@csc.albany.edu> wrote:
>>JB Collins <joebmath@yahoo.com> writes:
>>>. . .
>>>Perhaps this answers your question about what I
>>>want to do, and that IS the more ambitious goal of
>>>creating a web-based standard - most likely
>>coupled to
>>>the use of TeX/LaTex, the de facto standard for
>>>technical authors.
>>I must confess confusion about exactly what you want
>>to do.
>>It may (or may not) be helpful for you to know that
>>gellmu provides a
>>bridge from LaTeX to XML in various ways.  Its
>>"article" document type
>>has (and eventually will have more) element
>>attributes for providing
>>the level of extensibility an author might want in
>>order to have
>>translation to semantic formats.  Moreover, it
>>should be possible
>>using gellmu's newcommand facility to write things
>>out concisely.
>>                                    -- Bill
> __________________________________ 
> Do you Yahoo!? 
> All your favorites on one personal page – Try My Yahoo!
> http://my.yahoo.com 
Received on Friday, 4 February 2005 08:03:28 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:27:36 UTC