W3C home > Mailing lists > Public > w3c-sgml-wg@w3.org > November 1996

Re: Too late, is it? (Contains a simple new entity proposal)

From: David G. Durand <dgd@cs.bu.edu>
Date: Wed, 13 Nov 1996 10:52:00 -0500
Message-Id: <v02130502aeaf98593434@[128.148.157.46]>
To: "Henry S. Thompson" <ht@cogsci.ed.ac.uk>, W3C-SGML-WG@w3.org
At 7:21 AM 11/13/96, Henry S. Thompson wrote:
>I am (choosing my words carefully) disappointed that there has not
>been any indication of willingness on the part of the ERB to respond
>to the virtually unanimous criticism of the proposed empty element
>`solution' from the postings to this list in the week since the draft
>appeared.  The non-overrideability of the welded-in entity list has
>had a similar history.  Have I missed something, or is all discussion
>about anything other than typos now pointless?

Ate least some of us who _agree_ with the empty element decision have been
resting, as it's pointless for _us_ to re-open that argument. Given the we
have had many go-arounds on the issue, I think it's harder to reverse,
since it's not like everyone didn't get their say the first time.

However, the entity-list issue is brand-new, since it was never discussed
here, and therefore the unanimity is a much-stronger marker -- No one on
this list, but not on the ERB, seems to be willing to defend the decision.
I think that we should strengthen my previous suggestiuon and require that
_all_ general entities must be declared in the declaration subset. That
will guarantee DTD-less parsing of any document that uses general entities,
while also removing the need to make predefined entities hardwired. It's
also 8879 compliant without any need for bogus Processing Instructions.

>I note that some weasle-wording has been added to v002 which refers to
>"detecting HTML documents", but the result drives a coach-and-four
>through the carefully crafted definitions of "well-formed" and
>"valid".  That is, I take it that although documents with any of the
>elect eleven empty errors are neither well-formed nor valid, they are
>never-the-less required not to cause errors, and in fact to be
>processed `correctly' by anything claiming to be an XML application.

>This is very close to optional features in by the back door, when we
>thought they'd been squashed, and for good reason!

   I think you are definitely right.

   -- David

RE delenda est.
I am not a number. I am an undefined character.
_________________________________________
David Durand              dgd@cs.bu.edu  \  david@dynamicDiagrams.com
Boston University Computer Science        \  Sr. Analyst
http://www.cs.bu.edu/students/grads/dgd/   \  Dynamic Diagrams
--------------------------------------------\  http://dynamicDiagrams.com/
MAPA: mapping for the WWW                    \__________________________
Received on Wednesday, 13 November 1996 10:46:33 EST

This archive was generated by hypermail pre-2.1.9 : Wednesday, 24 September 2003 10:03:43 EDT