RE: Minutes of 5 April Telcon

[James Tauber <jtauber@bowstreet.com>]

Sorry I got dropped off for the last 15 minutes.

I would still like to know more about the concur decision.

Is it simply that the implementation complexity outweighs the perceived
usefulness?

Concur was as simple for me to implement as any other pattern (and
considerably easier than overall non-determinism). I realise this was
because my implementation is backtracking.

James

> -----Original Message-----
> From: James Clark [mailto:jjc@jclark.com]
> Sent: Thursday, April 05, 2001 9:18 PM
> To: trex@lists.oasis-open.org
> Subject: Minutes of 5 April Telcon
> 
> 
> Minutes of first TREX telcon held at 10.30 EDT Thursday 5th April
> 
> Present
> 
> James Clark
> Don Smith
> Murata Makoto
> Mike Smith
> Josh Lubell
> Karl Best (observer)
> Eric van der Vlist 
> Mike Fitzgerald
> Fabio Arciniegas
> James Tauber
> Norm Walsh
> 
> Decisions
> 
> The following decisions were made (all unanimous).
> 
> 1. Change the charter of the group to make its objective the creation
> of a unified schema language based on TREX and RELAX.
> 
> 2. Resolve the issue "Should empty elements allow 
> whitespace?": elements
> with no declared children should allow whitespace.
> 
> 3. Resolve the issue of "Parameterized patterns": not to be added in
> the first version.
> 
> 4. Resolve the issue of "Drop concur": drop the "concur" element.
> 
> Discussion
> 
> JJC reported that he had met with Murata-san (the designer of RELAX)
> to discuss unification of TREX and RELAX, and that their conclusion
> was that they felt it was probably possible to arrive at a design that
> was satifactory to both.  There was a general feeling that unification
> was politically desirable.  It was unanimously decided to change the
> charter of the TC to make the objective be the development of a
> unified language.
>  
> On the "Should empty elements allow whitespace?" issue, the decision
> as taken that empty elements (ie elements not declared to have any
> children) should allow whitespace.
> 
> On the "difference" issue, there was some sentiment for change, but
> the right choice of name was not clear: "except" was one possibility
> that was mentioned.
> 
> On the "xml:base" issue, any decision was postponed until xml:base
> gets REC status.
> 
> On the "Add pattern to match any element?", the concern was whether
> this was important enough to be worth a special syntactic
> abbreviation. No conclusion was reached.
> 
> On the "Syntax of global attribute on attribute element", there was
> some sentiment for change, but no conclusion on a new syntax.
> 
> On the "Split include element?" issue, most people felt it was
> desirable to split, if a good pair of replacement names could be
> found.
> 
> On the "Name of grammar element" issue, most people felt that the
> current name was satisfactory, but Murata-san expressed concern over
> the conflict with the grammar element in RELAX namespaces.  The TC
> will look at RELAX namespaces and see whether there is an alternative
> to "grammar" that would be more appropriate there.
> 
> On "Parameterized patterns", the feeling was that this was on the
> wrong side of the 80/20 divide for the first version, and the decision
> was taken not to include them in the first version.
> 
> On versioning, there were arguments for and against both a namespace
> URI and a version attribute on the document element.  No conclusion
> was reached.
> 
> On the "Drop concur" issue, the decision was taken to drop the concur
> element.
> 
> The next telcon will take place in two weeks at the same time.
> 
> James
> TREX
>