Re: Agenda for 5 June 2014 auto-WCAG call

Dear WCAG-Auto Team

First of all, I am sorry but I can not make the call this afternoon.  Will
Thursday afternoon be the regular time?  I can make arrangements for the
next ones, no problem, but this caught me off guard this time (even though
you mentioned it at the last meeting!).  Apologies.

In terms of the assignment I took on, formalising the test criterion, while
remaining in natural language.

I do believe I have exactly the solution we need for auto-wcag.

My initial thought was "But that is what the WCAG is already...."   But I
think that is because last year I translated the RGAA from French to
English.  The RGAA is the French accessibility standard.   I went back to
look at WCAG success criteria and I admit it is not very concise in terms
of application.

Question : am I missing something?  I was in the WCAG and looking at the
success criteria as described here (for example)
http://www.w3.org/TR/WCAG20-TECHS/G14


The actual specifics of the test are (to me) quite vague (by which I mean
you have to already have a good handle on the test to understand this
page).

Now, in the RGAA there are very precise test sequences which I believe
correspond to what you are looking for.  The problem is that the RGAA is
not just a version of the WCAG, and so the checkpoints don't all match up.
  But to make my point, you have  things like this :


*******************************************
2.3 [Colors]3: Provision of a means of transmitting information other than
by color based on CSS styles

Applies to:

Any HTML element carrying a style that uses at least one of the following
CSS properties:

color
background-color
background
border-color
border
outline-color
outline
Test procedure

If one of the elements listed in the domain of application is present in
the page, continue the test; otherwise, the test is not applicable.
If the element bears information by means of color, continue; otherwise,
the test is not applicable.
If the information borne by the element is also transmitted by a means
other than color, the test is successful; otherwise, it fails.
*******************************************

As you can see the selector is quite clear (somewhat large in this case,
admittedly).

So my proposition would be to use the RGAA style (and even eventually use
the RGAA itself!).

The complete document is here, in French :
http://references.modernisation.gouv.fr/sites/default/files/RGAA-v2.2_Annexe2-Tests.pdf

Now, it might be that this same style of precise test exists already in the
WCAG, if it does, then that is what we want to use.  I couldn't find it
exactly though.   I think  Shadi might be able to help us here.  If it is
not there, then the RGAA could be the way to go.

As I mentioned I already have the translation (it actually forms part of
the referential that was included in some of Urbilog's software).

Again, very sorry to not be with you today, I will keep up with the reunion
notes.   I am free to take more actions, for example we could chose 5-10
checkpoints and I could provide the RGAA tests for all of them and we could
see if that is what we want.  I think it might be!

John