W3C home > Mailing lists > Public > www-qa-wg@w3.org > September 2003

Re: [QA Ops] QA CR and WG charters

From: Dominique HazaŽl-Massieux <dom@w3.org>
Date: 18 Sep 2003 11:56:50 +0200
To: Karl Dubost <karl@w3.org>
Cc: www-qa-wg@w3.org
Message-Id: <1063879011.19335.160.camel@stratustier>
Le mer 17/09/2003 ŗ 15:17, Karl Dubost a ťcrit :
> A comment has been made by the OWL WG, which raises a fair enough 
> point. If they try to implement the QA Ops GL, they may have to modify 
> their charter to comply with the checkpoints:
> 	CP 1.1 Where it's explained in the Examples and Techniques. Amend a 
> charter for an existing WG.
> 	CP 1.2, 1.3, 1.4, 1.5, 2.1, 2.2, 2.3 too.

Err, amending the charter is only one of the techniques... The
checkpoint speaks about "documenting its QA commitment level in some
consensus record", and the E&T mentions recording the commitmemnt during
a teleconference for instance.

> - What's happening if they modify their charter to try the QA Framework 
> Ops GL and notice later on, that the CP has disappeared.

The OpsGL won't change in the next 6 months! Again, if they don't feel
confident in the stability of the opsGL (which are in CR and are thus
supposed to be fairly stable), they can take the decision during a
regular meeting, record it in their minutes, and come back to this at a
later stage if they don't think that decision was beneficial.

More generally, the real point is: if they take the decision without
thinking it's beneficial for them, then there is a probably a bug in
taking the decision at first...

> - What's happening if they notice that they will engage themselves in a 
> process they don't want now.

Well, they shouldn't get involved in it, then, and tell us why they
don't want it now.

> - Is there room for a thought experience during our CR phase. So WG 
> trying to do like if they were making it real, but not really. For 
> example write a mockup charter, like if they had to comply and report 
> what are the problems.

I don't think that count as implementation. The charter is really not a
requirement from the OpsGL, and we have spent a fair amount of time
making sure it would not be.

Dominique HazaŽl-Massieux - http://www.w3.org/People/Dom/

Received on Thursday, 18 September 2003 05:56:53 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:14:31 UTC