RE: repository of ICS / testing mailing list.

This makes very good sense to me.

I neglected to include reference to the proposal in the agenda for
tomorrow but we should discuss it.

Just to focus on the banal for a moment, I propose

No idea whether W3C has done anything like this before ...


> -----Original Message-----
> From: []
> Behalf Of Eduardo Casais
> Sent: 18 June 2009 18:57
> To:
> Subject: CTG: repository of ICS / testing mailing list.
> I propose an addition to the "Conformance" section of the CTG.
> The CTG stipulate that conformant proxy deployments must formalize
> their compliance
> through an ICS. However,
> a) There is no independent organization in charge of testing proxy
> deployments,
> certifying their conformance and establishing an ICS. These tasks are
> left to the
> proxy operators themselves, which are therefore self-certifiers.
> b) The W3C does not intend to validate an ICS after the fact. Rather,
> the community
> of service providers and application developers is supposed to do so
> through testing
> interfaces provided by the proxy operators.
> c) Conformance to the CTG implies the production of an ICS, but the
> guidelines are
> mute as to how and where an ICS is to be made available.
> The following text is to be included in a new section 3.5 "Repository
> of conformance
> information".
> "The World-Wide-Web Consortium sets up and maintains a public mailing
> list to
> disseminate and store information about the conformance of
> transformation proxy
> deployments against the present guidelines. The mailing list, which
> be moderated,
> fulfills the following purposes:
> a) An operator of a transformation proxy publishes
>    1.   the ICS and revisions thereof corresponding to its proxy
> deployment
> 	(possibly as a URI pointing to a public site of the operator
> where the
> 	relevant documents can be accessed);
>    2.   information about the testing interface to its proxy (possibly
> via a URI
> 	pointing to a site with the complete information about
> configuration
> 	parameters, conditions of access, etc);
>    3.   the announcement of discontinuation of a proxy and the
> retraction of its
> 	associated ICS.
> b) An end-user, a service provider or an application developer may
> publish
>    1.	a review of a published ICS with respect to its correctness,
> completeness
> 	and intelligibility;
>    2.   results of testing a transformation proxy against the
> guidelines through
> 	an operator's testing interface (possibly as a URI pointing to a
> site
> 	containing the complete description of the results, a ZIP
> with the
> 	resources used for testing, etc);
>    3.   an assessment of the consistency of a proxy deployment against
> the guidelines
> 	on the basis of tests or experience reports.
> The mailing list is located at[to be
> The following text is to be inserted into section 5 "Testing":
> "Information about the availability of a testing interface MUST be
> published in the
> mailing list given in section 3.5."
> The text in section 3.4 must be adjusted as follows:
> "A Transformation Deployment that wishes to claim conformance MUST
> available
> in the mailing list specified in section 3.4 a conformance statement
> Conformance
> Statement) that specifies the reasons for non-compliance with any
> clauses containing
> the key words should and should not.
> Retractions of conformance statements MUST be announced in the
> aforementioned
> mailing list."
> Setting up such a mailing list would be advantageous in several ways:
> a) It makes it easy for users, developers and service providers to
> retrieve
> conformance declarations: rather than scouring the WWW for each
> operator's documents,
> they can find the information in one place (at least, all the relevant
> URI in one
> place).
> b) It makes it easy for operators to discharge their duty of
> information: the mailing
> list constitutes an officially sanctioned channel to release ICS and
> associated data
> about proxy deployments.
> c) It increases transparency in the mobile market: the centralized
> publication of
> ICS and test results makes it clear which operators are actually
> abiding to the
> guidelines and where there are problems. From this perspective, such a
> mailing list
> would go some way to self-policing the environment of transformation
> proxies.
> d) It helps resolve problems with proxy deployments: all parties can
> learn about
> issues with specific proxy deployments; operators using the same
> products know then
> what to correct, developers and service providers what possible
> workarounds to rely
> upon.
> e) It facilitates the maintenance of the CTG: the W3C can follow
> contributions to
> the mailing list to identify parts of the guidelines that must be made
> more precise
> or extended to take into account aspects not dealt with in their
> current version.
> E.Casais

Received on Monday, 22 June 2009 14:32:01 UTC