W3C home > Mailing lists > Public > www-math@w3.org > December 1998

Re: Review of WWW Mathematics Browsers, MathML and HTML.

From: William F. Hammond <hammond@csc.albany.edu>
Date: Mon, 21 Dec 1998 09:57:38 -0500 (EST)
Message-Id: <199812211457.JAA29201@hilbert.math.albany.edu>
To: davidc@nag.co.uk
Cc: hutch@psfc.mit.edu, www-amaya@w3.org, www-math@w3.org
David writes:

: > It goes without saying 
: 
: why? 
: 
: > that I see no place for character entities in
: > authoring dtd's.
: 
: You use characters, so why not character entities (which are on the
: same level from an XML/SGML viewpoint).

In a few words: it's more flexible.

Yes, I confess that I do use characters.

I have decided that there are only 62 characters, namely
[0-9][A-Z][a-z], that I completely trust in every situation.  Every
other character has special "command" or "mark-up" meaning in at least
one presentation format that I know.  (For example, I see /bin/sh code
as a possible presentation format.)  Therefore, every other character
has a name in my authoring DTD.  As an author I sometimes choose not
to use all of these names in every document that I produce, depending
on the list of presentation formats that I plan for that document.

For \alpha ---> <alpha/> at the authoring level I offer these reasons:

1.  I want to avoid having the documents that I write
    be dependent on public standards that have not stood the
    test of time.

2.  I may have a presentation format in my list that does not
    encompass "&alpha;".

3.  With some types of sgml/xml processing, e.g., sgmlspl/sgmlspm,
    it is more difficult to provide non-default handling
    for "&alpha;" than for "<alpha/>".

4.  I may decide that I do not want to rely on a browser to
    provide the string "alpha" in red when it cannot handle it.

                                   -- Bill
Received on Monday, 21 December 1998 09:57:45 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Saturday, 20 February 2010 06:12:48 GMT