See also: IRC log
<pratul> Agenda at http://lists.w3.org/Archives/Public/public-sml/2008Feb/0043.html
<pratul> Jim, looks like u are the lucky winner of the scribe lottery :-)
<scribe> scribe: james lynn
<scribe> scribenick: Jim
No objections to minutes. Minutes are approved.
W3C has decided on Edinburgh, UK.
SML will decide on meeting in Edinburgh or in U.S. based on response to email that John sent.
<MSM> not many? I see 35 open issues in Bugzilla ...
<MSM> 18 labeled needsReview
<MSM> I'll type what I'm saying into irc then
<MSM> I don't disagree with Kumar's idea that we may be able to get through the issues quickly.
<MSM> But I also agree with Ginny that 22 is also a perfectly fine target date.
<MSM> I'd lean slightly toward 22 -- but if people answer Ginny's question with "yes", then another week would be useful.
Agreement to target Feb 22 as the date we send the LC to the webmaster. This gives us two weeks to come to agreement on proposals and review the changes.
It is assumed that the members of the WG will review the spec over the next two weeks, and not wait until everything is fixed, i.e. not wait until the last few days.
Agreement to mark editorial in agreement with Comment #2.
<MSM> +1 to Sandy's suggestion.
<MSM> Having it mentioned explicitly, as a reminder of a relevant fact, is useful.
<MSM> Perhaps reword the sentence to say "This can happen when the referencing elements use different schemes, or express ... different ways"
<MSM> [The sentence doesn't express a checkable / enforceable constraint. But that's true whether it's in a normative section or not. The overhead of moving it to a non-normative section seems high.]
Sandy would like the editors to make sure the sentence in 4.2.4 is clear somewhere in the spec, else add it.
<MSM> [I don't want to stand in the way, but on the whole I think 4.2.4 *is* the right place to make the observation. If it troubles people to have it not explicitly marked non-normative, then I'd put it into a non-normative Note. But I'm happy to leave it to the editors' discretion.]
Agreement to mark as editorial.
<MSM> OK
Agreement to mark as editorial.
<MSM> Is the problem (a) that the non-normative section says something untrue?
<MSM> or (b) that a normative statement has inadvertently been placed in a non-normative section?
<MSM> To understand how to fix the error, I think we need to know whether we are looking at (a) or (b).
Pratul: The way this bug was submitted, it deals only with the non-normative section 9.1.3. If we need to normatively address the case of using sml:nilref when sml:ref="false", it should be a separate bug.
<MSM> To know that, we need to know what the meaning of <x sml:ref="false" sml:nilref="true">...</x> is
<MSM> with respect, I do not agree with the claim that this bug is not about the question raised by Sandy
<MSM> Please let the minutes show my dissent from the chair's ruling on that matter.
Pratul wants to put this Bug on hold since the bug is not material to LC and there is no agreement within the WG, and move to the next bug.
Agreement to fix as per Comment #2 and mark as editorial.
<MSM> I think there are two issues with the definition: (a) it should make clear(er) that each document individually is checked, and also that the model documents as a body obey the cross-document constraints.
<MSM> and (b) 'verying that ... the docs are valid' seems not quite right -- it might be better to say that validation is the process of determioning *whether* the documents are (indidvidually and collectively) valid
Agreement to fix as per Comment #1 and Comment #2 and mark as editorial.
Agreement to mark as editorial in accordance with Comment #1. Not necessary for LC.
The group agrees to MSM's suggestion above on 5451
<MSM> [The question raised by Sandy about the definition of model validation, and the question I raised above in IRC, are raised to the best of my ability in bug 5461 http://www.w3.org/Bugs/Public/show_bug.cgi?id=5461
In reference to Comment #2, the group believes that the wording differs when talking about properties of data as compared to what a processor must do.
We should change the scope of this Bug to include both SML and SML-IF.
Agreement to move to editorial. WG should definitely review the final wording.
Still need a proposal from MSM and Sandy.
Agreement to mark as editorial and will need close review.
<MSM> [There are certainly W3C specs that use the phrase "if and only if" -- I don't know of any that spell it "iff".]
<MSM> 5461
<MSM> Sandy, if you've lost W3C access, should we paste the bug text into IRC?
<MSM> I think what I heard was "Model validation is the process of determining whether an SML model is both conforming and valid" ?
<ginny> Model validation is the process of assessing whether or not an SML model is conforming and valid. (insert reference to conformance section)
Agreement reached on wording suggested by Ginny and MSM or some combination.
Agreement to mark as editorial.
<MSM> http://www.w3.org/Bugs/Public/show_bug.cgi?id=5462
<MSM> In 4.1.2 we could change
<MSM> An element information item in an SML model instance document is
<MSM> an unresolved SML reference if and only if ...
<MSM> to
<MSM> An SML reference in an SML model instance document is
<MSM> an unresolved SML reference if and only if ...
<MSM> [I think we want this change whether we allow or we ignore the example in comment 0]
Kumar: To be consistent with the target* constraints we should allow this but ignore the sml:nilref unless we have a good reason for handling it differently.
Agreement to allow this case but ignore the sml:nilref attribute along with addition of the Warning phrase above. Mark as editorial.
<MSM> +1 to Kumar's understanding
<MSM> conforming SML-IF Document from an SML model." should say "from a valid conforming SML model" ?
The question regarding the above should be opened in a separate bug.
There was a discussion regarding the use of OPTIONAL in the sentence: It is OPTIONAL that a conforming SML-IF Consumer process all elements defined in this specification, but any element that is processed MUST be processed according to the requirements stated in the normative sections of this specification.
Kumar and Ginny questioned the use of OPTIONAL in the above sentence. Ginny wondered why we should even care about controlling the use case where a consumer is processing an SML model for its own proprietary purposes.
MSM felt that we should retain the use of OPTIONAL expressed a perspective on this based on schema experience. Ginny will open a separate bug to track this issue.
Bug 4675 can be closed.
Agreement to close Bug 4992.
Will continue to review this bug.
Agreement to accept changes and close Bug 5388.
<MSM> ciao!
rsagent, generate minutes
Last Scribe Date Member Name Regrets pending 2007-08-30 Lipton, Paul until mid-January 2007 2008-01-03 Kumar, Pandit 2008-01-10 Valentina Popescu 2008-01-17 Boucher, Jordan 2008-01-21 Gao, Sandy 2008-01-22 Wilson, Kirk 2008-01-22 Eckert, Zulah 2008-01-23 Smith, Virginia 2008-02-07 Lynn, James Exempt Arwe, John Exempt Dublish, Pratul Exempt MSM Exempt PHMinutes formatted by David Booth's scribe.perl version 1.133 (CVS log)