Mime-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Message-ID:  <v03110701b28b65905d08@[130.237.37.22]>
Reply-To: Mailing list for the LaTeX3 project
              <LATEX-L@URZ.UNI-HEIDELBERG.DE>
In-Reply-To:  <199812022008.PAA21912@hilbert.math.albany.edu>
Date:         Wed, 2 Dec 1998 22:53:09 +0100
From: Hans Aberg <haberg@MATEMATIK.SU.SE>
Sender: Mailing list for the LaTeX3 project
              <LATEX-L@URZ.UNI-HEIDELBERG.DE>
To: Multiple recipients of list LATEX-L
              <LATEX-L@URZ.UNI-HEIDELBERG.DE>
Subject:      Re: What is "base" LaTeX
Status: R

At 15:08 -0500 1998/12/02, William F. Hammond wrote:
>The point of my drafty draft on notation
>
>  http://www.albany.edu/~hammond/gellmu/notation [please read and comment]

On your idea to use types to improve on syntax checks, this is of course
used in languages such as Haskell <http://haskell.org/>. This is of course
way beyond the capacity of TeX.

But one should not expect that such typed objects should be able to capture
the math semantics, thereby resolving the problem: In computer languages,
typing is still merely a tool to help producing the code, but it does not
directly affect the runtime code.

If one should go deeper, one should design runtime object whose behavior is
affected by the type information. -- This is a question I interest myself
in my programming. But the time is not right now to report on that.

  Hans Aberg
                  * Email: Hans Aberg <mailto:haberg@member.ams.org>
                  * Home Page: <http://www.matematik.su.se/~haberg/>
                  * AMS member listing: <http://www.ams.org/cml/>