Re: ***MATH***

Marcus E. Hennecke (marcush@crc.ricoh.com)
Tue, 16 Jul 1996 13:33:22 -0700 (PDT)


Date: Tue, 16 Jul 1996 13:33:22 -0700 (PDT)
From: "Marcus E. Hennecke" <marcush@crc.ricoh.com>
Message-Id: <199607162033.NAA02363@cougar.crc.ricoh.com>
To: www-html@w3.org, hammond@csc.albany.edu
Subject: Re: ***MATH***

On Tue, 16 Jul 1996 15:45:59 -0400 (EDT), "William F. Hammond" <hammond@csc.albany.edu> wrote:
> On Tue, 16 Jul 1996 12:45:31 -0400
> "Marcus E. Hennecke" <marcush@crc.ricoh.com> wrote to www-html@w3.org
> : Not true. The problem with LaTeX's way to write math is that the computer
> : can't make sense out of it. It is not structural. For example:
> : 
> : 	\int f(x)xdx
> : 
> : How would the computer know that x is the integration variable?  . . .
> 
> LaTeX is used to mark up *mathematical text*.
> 
> It is not a computer-algebra-system language nor is it intended to
> be markup for such.

Um, yes. That's exactly my point.

> For mathematical text "\int f(x) dx" is fine.

Only if it is being displayed. Text-to-speech renderers will have
problems.

> We could easily overburden HTML-MATH into oblivion by imposing a
> requirement that it be a markup powerful enough to service both (1)
> computer-algebra-system needs (presumably filterable to whatever CAS)
> and (2) mathematical text needs.

I am not suggesting that the whole of Mathematica be incorporated into
browsers, just the notation and the ability to render formulas. I don't
see how this places any more burden on browsers than a LaTeX formula.

>  For example, use (1) would require a
> means of dealing with "typing", whereas for (2) formal typing is not
> required.  To achieve (1) would be a huge undertaking.

I guess I don't quite understand what you mean. I am looking at the
proposal by Ping and find that it, too defines a notation that a computer
would be able to make sense of. It seems it's not that hard to do.

> Let's not get confused about what HTML-MATH should be.

For one thing I don't think it should be a copy of LaTeX. It is not
device independent enough. For example, it requires the author to take
care of proper line breaking. If an equation is longer than one line,
you are forced to used the eqnarray environment and break the equation
by hand. This is not acceptable on the Web. And in order to do line
breaking, the computer must know a little bit about the meaning of the
terms and symbols in the equation.

> What I have seen at http://www.w3.org/pub/WWW/MarkUp/Math/WG/
> suggests that the efforts are pointed in the right direction.

At least there is now a &dd; entity so that the integration variable
of an integral can be specified. The important thing is that we don't
end up with a bunch of elements that are essentially layout commands
but with something that the browser can make sense of. LaTeX is not
it.

Marcus
--
Marcus E. Hennecke
marcush@crc.ricoh.com        http://www.crc.ricoh.com/~marcush/