Final Minutes of QA WG Teleconference, 2003-1-22

QA Working Group Teleconference Minutes
Wednesday, January 22,2003
--
Scribe: Sandra I. Martinez

Attendees:
(dd) Dimitris Dimitriadis (Ontologicon)
(DH) Dominique Hazaël-Massieux (W3C)
(LH) Lofton Henderson (CGMO - WG co-chair)
(SM) Sandra Martinez (NIST)
(LR) Lynne Rosenthal (NIST - IG co-chair)
(DM)David Marston

Regrets: Mark Skall (NIST)
         Kirill Gavrylyuk (Microsoft)
         Peter Fawcett (RealNetworks)
         Andrew Thackrah (Open Group)


Absent: (KD) Karl Dubost (W3C, WG co-chair)
  (JR) JohnRobert Gardner (Sun)

Summary of New Action Items:
AI-2003-0122-1: David to draft rationale for CP 8.3 by next telcon.
AI-2003-0122-2: Dominique to formulate text for issue AR-023 on CP 8.4 by 
next telecon.


Agenda: http://lists.w3.org/Archives/Public/www-qa-wg/2003Jan/0101.html

Previous Telcon Minutes : 
http://lists.w3.org/Archives/Public/www-qa-wg/2003Jan/0147.html

Previous Face to Face Minutes: http://www.w3.org/QA/2003/01/f2f-minutes

Minutes:

Spec Guidelines:
GL1 (Scope) issue proposal:

(DH) :Initiated the discussion addressing an e-mail sent to the working 
group to accept or reject Lynne’s proposal on the rewrite of GL 1 and asked 
for comments on the rewording. Lofton provided some recommendations and 
comments on the rewrite.

(LR): I agree with Lofton’s recommendation to move the statement about 
identifying the applicable classes of products from CK 1.1 to a discussion 
paragraph with a forward reference to CP 2.1 “Identify all classes of Product”.

(LR)  Lofton also requested clarification on the verbiage utilized to 
rewrite CP1.2 specifically what is meant to say with “…what is in or out of 
scope” and also what exactly needs to be satisfied, examples or use cases. 
I will rather let each working group decides, either one will work.

(LH): In that case, put that in the discussion.

(DH)  Lynne’s changes are accepted.

Remaining AR issue closure:

(DH): In the other e-mail I provided the status of Alex’s comments, 
including proposal for the remaining open issues. David sent comments on 
the controversial issues and Lofton added editorial suggestion to which I 
agree.

(LH): Suggested to spend 35 or 40 minutes on the controversial issues.

(DH): Lets address AR-014, do we want to allow other sources of definitions 
for conformance terms? In my proposal I recommended that if the conformance 
terms must be defined it should be done either by reference or by including 
the definition in the text.

(LR): Agree with Dom’s proposal and Alex suggestion.

(DH): No objections? Lets move on to AR-022. There is a lack of rational 
for 8.3.

(DM): GL 3 provides the general principles on what we want to do as part of 
the conformance strategy. The interaction among DoVs is not provided in the 
CPs.
CP8.3 should include this information.

(DH): Could you draft the rationale?

(DM): To include also the interaction among DoV’s?

(DH):  No, only the rationale for Last Call.

(DH): On AR-023. Alex’s comment is true, it is hard to test.

(LR): This issue was reworked in Tokyo.

(DH): We agreed to make it testable.

(LH): The reworked we did after Tokyo, missed the key aspect; a uniform 
policy for collection of discretionary choices.

(DM): We need to address other angles like the consolidation issue, to 
provide   sufficient high level instead of details, it is hard to get the 
consolidation.

(LH): I agree with Alex suggestion about been able to determine identical 
conditions.

(DH): It should be more precise.

(LH): You must address the policy and define it; it should be a global 
policy to enforce people to explicitly consider what they are doing. XSL is 
a good example.

(DH): A policy statement should be added to the checkpoint, adding a 
testable aspect in this draft.

(DM): Would you provide the verbiage?

(DH): I will try to come up with some verbiage to discuss offline via e-mail.

(Dm): Ok.

(DH): On AR-010, there is not a strong agreement on this issue.

(DM): This issue deals with the interaction between GL 3 and GL 10. GL 3 
looks at the conformance policy, and GL10 requires that you communicate the 
whole policy and each DoV aspect.

(DH): Does the conformance clause explain the policy?

(DM):  It is a spiral instead of a circular situation. When you express the 
conformance information in a conformance clause, you also need to convey 
the flavor of conformance. GL 2-9 are conveying the policy, GL3 is a 
catch-all for policy, GL10 is about writing down the decisions made in GL2-9.

(LR): Can we recheck what is in GL3 now? 3.1-Universal requirements for 
minimum functionality, 3.2-strict conformance requirements has moved, 
3.3-distinguish requirements from product specific extra feature has been 
deleted, 3.4-special conformance terms, the only thing that do not appear 
in the conformance clause and 3.5- justify the DoVs.

(DM): Yes. Justifying the DoVs and interaction are split among 2-9.

(DH): Are normative references part of the conformance clause?

(DM): No.

(DM): Should we remove references to GL 10 from GL 3?

(LR): Agree! No problem taking that out.  Remove last paragraph from GL3.

(DH): Agree! How should we handle this issue?

(LR): We should respond by presenting the rewrite on GL 3 and explaining 
that his suggestion will be taken into consideration by the next revision.

(LH): I support Lynne’s proposal.

(DH): Moving on to the easier issues. Any comments?

(LR): I agree with you proposals.

(LH): When are we going to have working versions that reflects everything. 
February 10 is schedule for Last Call to be published.

(DH): They should be ready by the next telecon, February 3.

(LH): Pubs rules will be use as basis for resolution to go to Last Call, 
not to TR at that point, just ready to go with minor adjustments. So, on 
February 3 we will be addressing resolutions for LC.

Adjourn at 12:00 p.m.

Sandra I. Martinez


Sandra I. Martinez
National Institute of Standards and Technology
100 Bureau Drive, Stop 8970,
Gaithersburg, Md. 20899

(301) 975-3579
sandra.martinez@nist.gov

Received on Friday, 31 January 2003 08:05:36 UTC