W3C home > Mailing lists > Public > public-test-infra@w3.org > April to June 2011

Re: Alternate proposal on testing requirements

From: Shadi Abou-Zahra <shadi@w3.org>
Date: Mon, 02 May 2011 16:40:00 +0200
Message-ID: <4DBEC240.8080002@w3.org>
To: James Graham <jgraham@opera.com>
CC: public-test-infra@w3.org
Hi James,

On 2.5.2011 16:24, James Graham wrote:
> On 05/02/2011 02:55 PM, Michael Cooper wrote:
>> As some on this list know, I've felt that the current testing
>> requirements page<http://www.w3.org/wiki/Testing/Requirements> didn't
>> fully capture what I envisioned. I have finally managed to bang out a
>> rough version of my view on testing requirements
>> <http://www.w3.org/wiki/Testing/Requirements/Cooper>. I've attempted to
>> incorporate what I understand of others' requirements but know I've left
>> some stuff out. I hope the organization at least is helpful, and I think
>> many of the specifics I suggest were not called out as such before.
> FWIW this seems to miss the fact that there can be > 1 test per file as
> well as > 1 files per test. In general trying to keep track of all the
> relationships is a really hard job. I don't suggest that we try to e.g.
> track all the subresources used by each test.
> In general (and not just relating to that document) though, I am
> slightly concerned that there is an element of architecture-astronautics
> creeping in. In particular I doubt there there is a significant overlap
> between the testing materials and methodologies for technologies such as
> web browsers, editors, and a11y tools. I am loathe to create
> requirements that imply any work done by this group has to address all
> those cases simultaneously rather than allowing the communities
> interested in each to develop the tools they need to solve their own
> problems. I expect the latter approach will work better for everyone
> involved; allowing rapid progress through developments of solutions that
> meet the requirments of one community well rather than doing a poor job
> trying to serve multiple communities all at once.

A spec like WAI-ARIA (actually it is a suite of specs) or WCAG rely on 
several testing methods at the same time. Also, the framework needs to 
support different W3C groups. In other words, I think we need to find 
the right balance between browser-only and astronautics design.


Shadi Abou-Zahra - http://www.w3.org/People/shadi/ |
   WAI International Program Office Activity Lead   |
  W3C Evaluation & Repair Tools Working Group Chair |
Received on Monday, 2 May 2011 14:40:25 UTC

This archive was generated by hypermail 2.4.0 : Friday, 17 January 2020 17:34:06 UTC