Re: Poetry and Maths

nholtz@civeng.carleton.ca (Neal Holtz)
From: nholtz@civeng.carleton.ca (Neal Holtz)
Message-id: <9305251515.AA06301@zonker.cascade.carleton.ca>
Subject: Re: Poetry and Maths
To: marca@ncsa.uiuc.edu (Marc Andreessen)
Date: Tue, 25 May 93 11:15:47 EDT
Cc: www-talk@nxoc01.cern.ch
In-reply-to: <9305251355.AA25709@wintermute.ncsa.uiuc.edu>; from "Marc Andreessen" at May 25, 93 8:55 am
X-Mailer: ELM [version 2.3 PL11]
Marc Andreessen writes:
> 
> "Peter Lister, Cranfield Computer Centre" writes:
> > This brought me on to mathematical expressions. There is real
> > requirement for these in technical documents, and at the moment all I
> > can do is include a bitmapped image screen dump in a document, or link
> > a dvi document containing a compiled TeX expression and hope that the
> > browsing end has xdvi. Not good when the TeX source is circa twenty
> > bytes. What's the general opinion?
> 
> The general opinion, I think, is that TeX's syntax should be used.
> That equates to a LOT of work.  Anyone feeling particularly wealthy
> today?

Not that this will be trivial, but I was wondering about the possibilities
of a \TeX{} ``server'' -- a hacked up version of TeX that would accept
a few bytes and hand you back a DVI fragment.  THere is a lot of code
around for rendering that to a bitmap.  I haven't looked at the TeX source,
but it can't be a huge job (:-).  There are of course lots of questions,
such as what kind of TeX environment to assume, how much of the DVI file
actually has to be sent, etc.  

The biggest question is -- will the performance be adequate? TeX is a bit
sluggish to get going, particularly, I suspect, if started from inetd for
each twenty byte formula.

I am very interested in this because of the potential use of HTML browsers
for building codes and structural design standards.  These documents
typically have a lot of mathematical expressions -- a typical clause
is 1/2 page long and contains a formula of some sort.

I also want the formula to be ``active'', i.e. usable in a calculator-like
fashion.  For ideas on what this might look like, see MathCAD for Windows.
I have a Master's student now looking at it -- over the next month or two
we hope to float some proposals around.  I would be pleased to hear from
anyone that has doen some similar thinking.  I might be able to supply
a bit more detailed abstract in a week or two, if anyone is interested
in more details of the proposal.  I have been toying with these ideas
for ten years or so, so I have a fair understanding about whats required
on the computation side.


-- 
Prof. Neal Holtz,  Dept. of Civil Eng.,  Carleton University,  Ottawa,  Canada
Internet: nholtz@civeng.carleton.ca  Ph: (613)788-2600x5797 Fax: (613)788-3951