Draft Minutes - Telcon 10/28/02

QA Working Group Teleconference
Monday, 28-October-2002
--
Scribe: Peter Fawcett (RealNetworks)

Attendees:
(KD) Karl Dubost (W3C, WG co-chair)
(PF) Peter Fawcett (RealNetworks)
(KG) Kirill Gavrylyuk (Microsoft) (late)
(LH) Lofton Henderson (CGMO - WG co-chair)
(JM) Jack Morrison (Sun) - permanent regret
(LR) Lynne Rosenthal (NIST - IG co-chair)
(MS) Mark Skall (NIST)
(AT) Andrew Thackrah (Open Group)

Guest:
(DM) David Marston

Regrets:
(dd) Dimitris Dimitriadis (Ontologicon)
(DH) Dominique Hazael-Massieux (W3C - Webmaster)
(SM) Sandra Martinez (NIST)

Absent: None

Summary of New Action Items:
A-2002-10-28-1: Peter to write up definitions of withdrawn and 
deprecated. 11/4/02.
A-2002-10-28-2: Andrew to clean up language of Guideline 1, 2nd 
paragraph. 11/4/02.
A-2002-10-28-3: Lynn to draft language on single TOC section. 11/6/02.
A-2002-10-28-4: David to look at Guideline 6, work on language. 
Address triple @'s. 11/2/02
A-2002-10-28-5: Mark - Change language on reference to ICS so that it 
must be referenced or
	there must be language as to why it is not applicable.

Agenda: http://lists.w3.org/Archives/Public/www-qa-wg/2002Oct/0133.html
Previous Telcon Minutes: 
http://lists.w3.org/Archives/Public/www-qa-wg/2002Oct/0125.html

Minutes:

1.) roll call


2.) SpecGL miscellaneous
          - any problems w/ issue clarifications [1], [2]
          - is Issue #51 closed (de facto)?
          - etc

No objections to clarifications of issues list have been raised in 
the mail list and none were raised on this telcon.
Since no one objects, this has been closed.

Issue 51 has been affectively withdrawn.
	- need definitions of withdrawn/deprecated.
	but. SpecGL now does address such features. Can we declare 
closed and point to the resolution from the
face to face.
	- Yes, consider closed but add definitions

A-2002-10-28-1: Action item for Peter - definitions of 
withdrawn/deprecated. Work with Lynn on language.
	Guideline 7 has language on deprecated.
	Due: 11/4/02.


3.) Spec Guidelines new drafts [3], [4]
          - new Intro -- overview/general comments?
          - issues highlighted in LR, DHM email [5], [6]
          - visit/discuss other "@@@" issue points

Spec Guidelines new drafts:
Go through triple '@'s that need explanation/discussion.

Guideline 1 has language from both Lynn and Andrew that Lofton 
basically combined. The 2nd paragraph is hard to understand and reads 
poorly In particular the last sentence needs some clarification on 
use cases and systems.

A-2002-10-28-2: Action Item to Andrew Clean up language of Guideline 
1 2nd paragraph with Lynn.
	Due 11/4/02..

  Checkpoint 3.1 - A specification must include Table of contents.

Lofton - no matter how you divide up a document you need to provide 
an easy way to navigate
the document and find information. A table of contents is not 
required in all documents but some way break up the information and 
make information easy to find should be a requirement.

David - There needs to be a way to navigate to conformance 
information with out having to
simply read the document from end to end.

Lynne noted that there are multiple TOC checkpoints for multiple 
Guidelines. Can all these TOC checkpoints be combined into one?

A-2002-10-28-3: Action Item to Lynn to draft language on single TOC 
section rather than multiple as it is
now. Due 11/6/02


3.5 - If profiles are chosen, address rules for profiles.
Question, What is a testable rule.

Should be resolved based on language else ware in the document.  See 
Lofton's mail to
list.

'Rule' should be replaced with 'requirement' as we have definition of 
'testable requirement'

Lofton suggested (in mail) 'contradict' or 'redefine' instead of 'clash'.

Group agrees.

6.2 - Does the rational really justify the checkpoint

Lofton - bigger issue is that guideline 6 has become catch all of 
checkpoints on conformance
that don't fit elsewhere.

Based on Mail discussion between Lofton and David.

David to take a look at Guideline 6 (checkpoint 2 and 3).

A-2002-10-28-4: Action Item to David to look at Guideline 6, work on 
language. Address triple @'s.
11/2/02


12.1/12.2 - How can an ICS be normative, how.

Was discussed last week. The way this currently reads, it could be 
interpreted to mean that the
normative ics is only possible ics.

should be re-written to state minimal requirements for ics or to make 
it normative to have one.

A-2002-10-28-5: Action Item - Mark - Spec much include reference to 
ICS or include language as to why it is not
applicable.

14.1 - Test Assertions:
Will take too long...

1.3 - Make this statement stronger?

Directive to Dom. We all agree, Yes... but how?

2.2 - Clarify language on "which is not a requirement' Needs to be 
clarified or removed. It
may have an intended purpose but no one is positive what was meant.

Mark - pre announces his regrets for next week.

---
Adjourned
-- 

Received on Thursday, 31 October 2002 13:35:25 UTC