Draft minutes of QA WG Telcon 22-March-2004

Draft Minutes

QA Working Group Teleconference
Monday, 22-March-2004
-- 
Scribe: Patrick

Attendees:
(PC) Patrick Curran (Sun Microsystems)           
(KD) Karl Dubost (W3C, WG co-chair)           
(DH) Dominique Hazaël-Massieux (W3C)           
(LH) Lofton Henderson (CGMO - WG co-chair)       
(LR) Lynne Rosenthal (NIST - IG co-chair)       
(AT) Andrew Thackrah (Open Group)

Regrets:

(MC) Martin Chamberlain (Microsoft)
(DD) Dimitris Dimitriadis (Ontologicon)
(MS) Mark Skall (NIST)
(SM) Sandra Martinez (NIST)
(VV) Vanitha Venkatraman (Sun Microsystems)

Absent:

Guest:
(DM) David Marston (IBM)


No new Action Items were assigned



Agenda: http://lists.w3.org/Archives/Public/www-qa-wg/2004Mar/0073.html
Previous Telcon Minutes: 
http://lists.w3.org/Archives/Public/www-qa-wg/2004Mar/0045.html


1.) roll call 11am EDT, membership

2.) routine business
        - Future telecons [0]

Handbook and outline of testgl next week

        - June f2f dates & location

Discussion of dates: Olivier has conflict June 14, Mark may have 
conflict June 8
Will meet Tuesday, Wednesday, Thursday
Week of June 7th is best
Two choices of venue: Montreal and Santa Clara
Consensus is for Santa Clara


3.) SpecLite
        - proposed outline  [1]

[LR] reviews her outline
Is this the right level?
Are these the points we want to make? (add, change, delete?)


Specifying Conformance: this is the heart of the document - everything you
need to know is here...

[PC]: two of the top-level sections (B and E) seem to address good 
practices rather than
the form/requirements for spec. Should these be embedded rather than in 
a separate section?
Yes.

[LR]: just about everything can be classified as "specify conformance"
How to structure - make it readable, not overwhelming?

[LH]: 'A' points to everything else.

[LR]: two elements: what you need to talk about performance
miscellaneous: "define the scope", "review the spec"
[PC]: these really are best practices - how to get there - rather than 
addressing
the class of product (the spec). We had similar problems with TestGL - 
we wanted to
recommend "review spec", define scope, test the tests, etc. -

[LH]: agrees B and E are different animals - unless we get a concrete 
proposal for
how to rework this, go with this structure

[LR]/[LH]: discuss value of the scope section

[LR]: agrees in principle that "process stuff" should be "merged in" to 
other sections

[LR]: do we need to define "conformance model"?
Consensus: yes
Section A.1 contains a brief description/analysis

David: do these guidelines focus on specs for conformance as opposed to 
other purposes
of specs?
[LR]: yes - and we should make this clear in title/introduction
 
[LR]: Reviews in detail

Section A: like a table of contents. Everything will end up in a 
conformance clause
Look in other sections for more detail. This section should link to a 
template.
An experienced spec-writer might learn everything needed from this section.
Could be used as a checklist.
Section A2: make this an advanced topic?
[LH]: no - this is important, and not too difficult to understand
[LR]: should it be moved elsewehere?
No: keep it where it is for now?

Section B: Discussion of specs (eg RDF, XPath) to which one doesn't 
conform directly,
but via other specs that 'include' or 'reference' them

Section C: define terms - [LR]: this should focus on conformance clause. 
[PC]/[LH]/[LR]:
Should this be applied more broadly? No need - it's in manual of style

Ran out of time. Discuss in email...
[LH] suggests we discuss in telecon two weeks from today
Should we create an issues list? Consensu: yes = after first working draft

       
4.) Adjourn

5.) Overflow (12-12:30): available.


[0] http://www.w3.org/QA/WG/#sched
[1] http://lists.w3.org/Archives/Public/www-qa-wg/2004Mar/0064.html

Received on Tuesday, 30 March 2004 16:34:16 UTC