RE: [w3c sml] On recognizing, handling, and constraining SML references

Hi Sandy
Thanks for the update.  A few comments that we can discuss in tomorrow's call


1.       I do not recall that there was an agreement on (2). It is not clear how 2.1.3 can be implemented and I don't see the need to talk about non-validator consumers in SML spec. Plus, I don't understand why deref() must follow the rules for non-validator consumers even though deref() is used during model validation (e.g., in Schematron constraints and SML identity constraints)

2.       targetType and targetElement constraints should be treated as satisfied and not violated on null references (this is re the table in Section 6). Rationale: These constraints should be applied only on non-null references.

Pratul

From: public-sml-request@w3.org [mailto:public-sml-request@w3.org] On Behalf Of Sandy Gao
Sent: Wednesday, September 05, 2007 9:08 AM
To: Sandy Gao
Cc: public-sml@w3.org; public-sml-request@w3.org
Subject: Re: [w3c sml] On recognizing, handling, and constraining SML references


> Team,
>
> This is to start the discussion around many reference related bugs,
> including 4658, 4673, 4682, 4683, 4780/4795, 4834, 4865, 4884, 4976.
> (4780 seems to be a duplicate of 4795.)
>
> I'm attaching both HTML and PDF. Pick your favorite format. :-)

I've updated the proposal based on discussions we had during the F2F. Pick the format (HTML vs. PDF) and the flavor (plain vs. diffed) you like.

My understanding about where we are:
- We have agreement on sections 1, 2, 4, and 5.
- We answered 2 out of 3 non-reference related questions (section 6).
- Some members are not entirely certain about section 3.
- The only other question we need to answer is "defaulted sml:ref" (in section 6).




Thanks,
Sandy Gao
XML Technologies, IBM Canada
Editor, W3C XML Schema WG<http://www.w3.org/XML/Schema/>
Member, W3C SML WG<http://www.w3.org/XML/SML/>
(1-905) 413-3255 T/L 969-3255