issue #19

QAWG participants,

I strongly disagree with the proposal to reverse the current (pre-Tokyo) 
disposition of issue #19 [5].

Synopsis
-----

In April, issue 19 was resolved that per-Framework glossary sections were 
not needed.  In July, after several months of Framework spec development 
and a few publication cycles, at a teleconference with all lead editors 
present, the editors decided that in fact per-Framework definition sections 
were appropriate and necessary.  This decision was circulated to the QAWG 
and independently to "stakeholders" for comment and discussion.

The basic reasons are articulated in ref [3a] below.  To summarize:

1.) With the aggressive publication schedules, it is not possible to 
coordinate with QA Glossary (esp. for new terms).

2.) In any case, some terms may not be appropriate for the general QA Glossary.

3.) Appropriate terms may eventually migrate into QA Glossary, with an 
appropriate definition for that document.

4.) It is appropriate and desirable in a per-Framework "Definitions" 
section to supplement and expand upon the generic, terse QA Glossary 
definitions, in the style of some of the WAI guidelines documents.  This 
allows for contextual, functional, and exemplary information, that relates 
the term to the context of the particular framework document.

The per-Frm definitions should operate under constraints.

** they do not contradict QA Glossary;

** they reference any existing QA Glossary term;

** they are clear about supplementing, not replacing any QA Glossary term;

On the last, see the general caveat at the start of chapter 6, 
"Definitions" [4], in the current SpecGL draft.  Note that the title 
"Definitions" was deliberate, in order to prevent any possible confusion 
with the QA Glossary.

References & History documented
-----
[0] Resolved in April, "don't need per-Framework glossary"
[1] 
http://lists.w3.org/Archives/Public/www-qa-wg/2002Jul/0005.html  (changed 
at editor's telecon (for TestGL, but all lead editors present).
[2]  http://lists.w3.org/Archives/Public/www-qa-wg/2002Aug/0149.html 
(message to QAWG indicating changed resolution issue 19)
[3a] 3-sept message to Dominique, lead editors, glossary owners 
(unarchived, see below)
[3b] Dominique reply to [3a] (unarchived, see below), to same list
[4] http://www.w3.org/TR/2002/WD-qaframe-spec-20020826/#definitions
[5] http://www.w3.org/QA/WG/qawg-issues-html#x19

Regards,
-Lofton.

Unarchived messages amongst Frm and glossary editors...

[3a] ##### unarchived direct message, 3-sept #####
X-Sender: lofton@rockynet.com
X-Mailer: QUALCOMM Windows Eudora Version 5.1
Date: Tue, 03 Sep 2002 10:38:15 -0600
To: Dominique Hazaël-Massieux <dom@w3.org>
From: Lofton Henderson <lofton@rockynet.com>
Subject: Re: FYI : a new draft of "QA Framework - Specifications
Guidelines" published
Cc: lynne Rosenthal <lynne.rosenthal@nist.gov>,kirillg@microsoft.com,
Mark Skall <mark.skall@nist.gov>,Karl Dubost <karl@w3.org>
X-RCPT-TO: <lofton@rockynet.com>

At 04:09 PM 9/3/02 -0400, you wrote:
[...]
 > "test assertion: any sentence or equivalent assemblage of
 > words and phrases that prescribes the behavior that must be
 > obtained when a stimulus occurs under a certain set of
 > conditions. (See also QA Glossary [QA-GLOSSARY].)"
 >
 > Can a test assertion be expressed in a formal language without words
 > and phrases rather than in sentences of a human language?
Err... anyway, this definition doesn't match the one used in the
glossary. We need to fix this.
Dom,
Just to let you know, the TestGL editors plus SpecGL editors resolved that 
the GL documents may have a Definitions section, and those definitions may 
differ from the central QA Glossary. (At least for a while.) See issue #19 [1].
Reasons:
1.) it is impossible to progress the GL documents swiftly and coordinate 
with the QA Glossary. The QA Glossary typically changes fairly infrequently 
(e.g., before Montreal it was said that the 'test assertion' definition 
would be changed, and but that happened yet.) In editing SpecGL, I put the 
whole initial Definitions section in one day, and changed/added weekly 
thereafter.
2.) the appropriate definitions for the Definitions section of a GL 
document and the central glossary may not be identical. They should not be 
incompatible of course. See Issue #19, [1]. The QA Glossary will be terse, 
dictionary-like definitions (per Montreal). We want the Definitions section 
of a GL document to be free to elaborate, in the style of the WAI 
documents, on useful contextual, functional, and example verbiage. You can 
see this already in a couple of definitions in the SpecGL "Definitions" 
chapter.
I have copied all of the lead editors. Does anyone have a problem with 
this? If yes, then it is a simple matter to re-open issue #19 [1] and 
discuss it within the WG at large.
-Lofton.
[1] http://www.w3.org/QA/WG/qawg-issues-html#x19
##### end [3a] #####

[3b] ##### DHM reply to [3a] #####
Subject: Re: FYI : a new draft of "QA Framework - Specifications Guidelines"
published
From: Dominique Hazaël-Massieux <dom@w3.org>
To: Lofton Henderson <lofton@rockynet.com>
Cc: lynne Rosenthal <lynne.rosenthal@nist.gov>,
kirillg@microsoft.com, Mark Skall <mark.skall@nist.gov>,
Karl Dubost <karl@w3.org>
X-Mailer: Ximian Evolution 1.0.5
Date: 04 Sep 2002 08:47:38 -0400
X-RCPT-TO: <lofton@rockynet.com>

Le mar 03/09/2002 à 12:38, Lofton Henderson a écrit :
 > Just to let you know, the TestGL editors plus SpecGL editors resolved that
 > the GL documents may have a Definitions section, and those definitions may
 > differ from the central QA Glossary. (At least for a while.) See issue
 > #19 [1].
 >
 > Reasons:
 >
 > 1.) it is impossible to progress the GL documents swiftly and coordinate
 > with the QA Glossary. The QA Glossary typically changes fairly
 > infrequently (e.g., before Montreal it was said that the 'test assertion'
 > definition would be changed, and but that happened yet.) In editing
 > SpecGL, I put the whole initial Definitions section in one day, and
 > changed/added weekly thereafter.
 >
 > 2.) the appropriate definitions for the Definitions section of a GL
 > document and the central glossary may not be identical. They should not be
 > incompatible of course. See Issue #19, [1]. The QA Glossary will be
 > terse, dictionary-like definitions (per Montreal). We want the Definitions
 > section of a GL document to be free to elaborate, in the style of the WAI
 > documents, on useful contextual, functional, and example verbiage. You can
 > see this already in a couple of definitions in the SpecGL "Definitions"
 > chapter.
That makes sense, indeed. Maybe we should ask the Glossary editors to
check that the proposed definitions in each new TR draft match the ones
in the Glossary?
Dom
 > [1] http://www.w3.org/QA/WG/qawg-issues-html#x19
-- 
Dominique Hazaël-Massieux - http://www.w3.org/People/Dom/
W3C/INRIA
mailto:dom@w3.org
##### end [3b] #####