Minutes for Teleconference on 12 June 2014

Automated WCAG Monitoring Community Group
Teleconference 2014-06-12, 16:00 - 17:00 h (CET)
Attendees:
Hanno Lans
John Hicks
Kamyar Rasta
Wilco Fiers

ITEM 1  Formal language for test creation

John: Firstly, RGAA (a web accessibility guideline in French) [1] might
be a good model to follow as WCAG is not precise enough. There is
specific language RGAA which breaks tests down in two parts: selectors
and test methodology. The test methodology is broken down into binary
test so there is yes and no for each level.
Secondly, maybe RGAA is not formal enough also and we might turn it to a
conditional language in pseudocode.
As next step I like take 4-5 tests and formalize them using RGAA.

Wilco: the format is nice and similar to our approach but it is better
to enhance the plain text a bit to make it more readable.

John: we can format it or even add something like have especial way of
put in that in a XML like or having look at how EARL does it when it
describes the result language.

John: Will send the English translation.

John: I will look to see if there is any description of RGAA. But you
could also use the examples to inspire the description.

Wilco: Is it open is this?
John: Yes, I did the translation personally two years ago. And we don't
mind.

Wilco: we can see if we can take test steps are in here, and move them
in to the format that we are proposing to use.

kamyar: look at [4] to see how concise is this from respective of
developer. What are the ambiguities there.

ITEM 2: Creating test descriptions with Quail

Hanno updates: a quail test example here [2]. This is a script to detect
if a table is data table or layout table. This is an import test because
it is used by other tests.

Wilco: So we could treat this as a selector. Is this data table or not.

Hanno: the first selector is table and then we have this especial
selector to refine that selector.

Wilco: the next step is to see if we can take the code and text there
and write it down in the format that John mentioned.
John: Introducing headings like what we see in [2] is good idea because
it tells you what the step is about. One thing I can do it to put
heading before each test to describe what the test is.

Hanno: not all quail tests are documented in this way [2].

Wilco: We can potentially use some the quail code and document them in
this way. That would benefits both our project and quail.

Hanno: Issue list of quail [3]. we mainly focusing on failure and
techniques. We try to connect failure to techniques and techniques to
each others.

Hanno: will provide a simple list of quail tests for the community group
as test description. Will focus on something like four simple success
criteria.

ITEM 3: Planning test development

Wilco: setup a wiki page contains different SC and assign name to them.
For each SC that person is going to write different tests.

References

[1]
http://references.modernisation.gouv.fr/sites/default/files/RGAA-v2.2_Annexe2-Tests.pdf
[2] https://www.w3.org/community/auto-wcag/wiki/IsDataTable
[3]
https://github.com/quailjs/quail/issues?direction=desc&labels=success+criterion&page=1&sort=updated&state=open
[4] http://lists.w3.org/Archives/Public/public-auto-wcag/2014Jun/0006.html

Summary of action items:

https://www.w3.org/community/auto-wcag/wiki/Action_items

kamyar: look at example that John gave [4] as email and find any
potential problem from developer's perspective.

Hanno: Tests of Quail, the list of SCs that are automatically testable
and working on some of the script descriptions.

John: Send translated version of the proposal (RGAA) and see how well
they are implementable in some of WCAG's success criteria. Find out if
there is any description for RGAA.

Wilco: create a proposal on how to format new WAM language for the test.
And create the list of SC where people can assign their name.

Best Regards
Kamyar

Received on Tuesday, 24 June 2014 16:28:20 UTC