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

Re: levels, modules and profiles. AI-20030414

From: Lofton Henderson <lofton@rockynet.com>
Date: Mon, 03 Nov 2003 17:43:11 -0700
Message-Id: <>
To: Karl Dubost <karl@w3.org>
Cc: www-qa-wg@w3.org


At 10:25 AM 11/3/03 -0500, Karl Dubost wrote:
>Le Mercredi, 29 octo 2003, à 11:28 America/Montreal, Dominique 
>Hazaël-Massieux a écrit :
>>Le mer 29/10/2003 à 17:08, Karl Dubost a écrit :
>>>Le mercredi, 29 oct 2003, à 08:52 America/Montreal, Dominique
>>>Hazaël-Massieux a écrit :
>>>>AI-20030414-4:  will report on how the terms "levels", "profiles", and
>>>>"modules" are
>>>>used in W3C (deadline:  2003-04-23)
>>>done for a long time, was a mail in the mailing list. Can't find it.
>>Actually, I don't remember having seen it... Are you sure you sent it?
>>(FWIW, I have closed the AI, but I would appreciate if you can find a
>>reference anyway).
>To read

Yes, I searched and found this when after I saw your "can't find it".  But 
I thought I must have the wrong thing, as the solution pre-dates the action 
item by 6 months!

>In short, right now at W3C, each WG have different concepts of what's a 
>profile a module or a level and sometimes, for example, levels are used 
>for versioning.
>So I wonder if we should create a table with 3 columns
>                 | Module | Level | Profiles | <- QA WG definitions
>Spec A                                                            <- how 
>it's called by
>-------------------------------------    others WGs.
>Spec B
>Spec C
>Like it will be a long work to do, do you think it's valuable?

IMO, something like this would be very valuable.  I think that the data in 
your table validates the need to clearly identify the separate concepts and 
try to move towards common terminology.

We won't be able to change existing published Recs, but optimally we can 
influence future work.  And we can perhaps start to foster awareness of the 
common concepts that lie behind the finished Recs of the different WGs, 
even where they are obscured by differently used terminology.

I think that this would be a terrific contribution to SpecET, and would 
help to resolve the several Last Call questions that we got about the 
distinction between the concepts.

Btw, much of the complexity in your table is now being faced by SVG.  SVG 
1.2 is a huge body of functionality, and there is pressure to partition it 
along many axes (corresponding to many discrete and distinct application 
sectors).  I am now only half way through a series of lengthy mail threads 
from the past two months (when I was mostly gone) where this very issue is 
being wrestled.  You can get a taste of the scope of the problem here 


Perhaps if you undertake the "long work", I could help and contribute.  And 
maybe it could even bring some benefit, in the short term, to WGs with 
complex problems.

Received on Monday, 3 November 2003 19:46:02 UTC

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