W3C

Accessibility Conformance Testing Teleconference

13 Sep 2018

Attendees

Present
Alistair, Anne, Jey, MaryJo, Romain, Shadi, Kathy, SteinErik, Trevor, Charu
Regrets

Chair
MaryJo
Scribe
shadi

Contents


Survey of pull request 242: Aspects Under Test

<maryjom> https://github.com/w3c/wcag-act/issues/233

<maryjom> https://services.w3.org/htmldiff?doc1=https%3A%2F%2Fw3c.github.io%2Fwcag-act%2Fact-rules-format.html&doc2=https://api.csswg.org/bikeshed/?url=https://raw.githubusercontent.com/w3c/wcag-act/7723b92d89d69478349526538ba0026c0d8d0fa9/act-rules-format.bs#input-aspects

<maryjom> https://www.w3.org/2002/09/wbs/93339/ACT27AUG2018/results#xQ4

MJM: start with Shadi's comment

SAZ: changes only editorial ... re-ordered sentences for better flow

RDT: comment not directly on the pull request
... just realized that the sentence is vague
... and that combined with MUST requirement raises concern
... not sure how the sentences constitutes a conformance criteria

<maryjom> https://www.w3.org/2002/09/wbs/93339/ACT27AUG2018/results

[[Each aspect MUST identify a distinct part of a test subject, which does not overlap with another aspect of the same rule.]]

RDT: not only wording issue but more conceptual
... what does "no overlap" mean?
... DOM tree could overlap with the Accessibility tree (eg. accessible name in the DOM could come from an Accessibility tree)

AG: aspect is badly defined
... also that definition does not mean anything ti me

[[An aspect is a distinct part of the [Test Subject](#output-test-subject) or its underlying implementation.]]

AG: don't understand that definition
... we have definitions built on definitions that are vague and convoluted

RDT: agree, definition is a little loose
... have to think about what we want to say a little clearer
... maybe say that only one aspect of the test subject is tested
... but not sure what that means

AG: to me the test subject should be the aspect

RDT: think test subject could be a page
... but that can be built using several resources

AG: but at the end of the day you are testing the trees
... when you test a car, doesn't make sense to say testing the engine

RDT: issue comes with more complex tests, where you have to look at several parts/trees

ATN: in several rules looking at DOM tree and CSS styling

AG: i'm ok with the concept, just needs better defining

MK: we have multiple issues here
... #257 aspects under test
... just want to point out that we are discussing a different section

<romain> +1, the previous discussion is independent from #257/#242

MJM: agree, separate issue
... suggest we open an issue on that

<MoeKraft> https://github.com/w3c/wcag-act/issues/257

MK: we already have #257 on that

SES: discussions seems abstract to me, and can't follow it
... want some suggestions
... suggestions should help move the discussions along

MK: this topic is about moving the sub-sections to an external WG Note
... #257 discusses the definition of aspects
... can have it there

<maryjom> https://services.w3.org/htmldiff?doc1=https%3A%2F%2Fw3c.github.io%2Fwcag-act%2Fact-rules-format.html&doc2=https://api.csswg.org/bikeshed/?url=https://raw.githubusercontent.com/w3c/wcag-act/7723b92d89d69478349526538ba0026c0d8d0fa9/act-rules-format.bs#input-aspects

MJM: can move on, RDT agrees to discuss it in #257

ATN: there is disagreement on the content of the example
... if that will really make it as an ACT Rule
... not getting sufficient agreement
... but there is a Failure technique for that

AG: but can still write a rule on that

ATN: just want to flag that there is disagreement on that example

AG: I feel confident about that
... we have a test for that already

<MoeKraft> https://api.csswg.org/bikeshed/?url=https://raw.githubusercontent.com/w3c/wcag-act/7723b92d89d69478349526538ba0026c0d8d0fa9/act-rules-format.bs#input-aspects

MK: after moving the sub-sections out of the document
... think the examples may actually need cleaning up

ATN: i'll take a look if I can fix those

MK: maybe the styling or formatting
... i'll take a closer look

MJM: after that fix, we can then merge

Survey of pull request 247: Test/Aggregation Definition

<maryjom> https://github.com/w3c/wcag-act/issues/241

<maryjom> https://services.w3.org/htmldiff?doc1=https%3A%2F%2Fw3c.github.io%2Fwcag-act%2Fact-rules-format.html&doc2=https://api.csswg.org/bikeshed/?url=https://raw.githubusercontent.com/w3c/wcag-act/82dd23f1d2e5494025791ae1a2a876b08d7f1162/act-rules-format.bs#atomic-rules-list

<maryjom> https://www.w3.org/2002/09/wbs/93339/ACT27AUG2018/results#xQ1

SAZ: not objecting, agree with the content and intent
... just some sentences and definitions need to be better stated, I think

https://github.com/w3c/wcag-act/issues/241#issuecomment-420722141

CP: duplicate list item

MJM: editorial and can be simply removed

https://github.com/w3c/wcag-act/pull/247#issuecomment-420816040

KE: suggestion to add clarity

MJM: makes sense

KE: not sure how we want to refer to rule IDs
... made a proposed edit to refer to rule IDs

MJM: makes sense

MK: many conflicts now between the different pull requests
... suggest to accept this pull request, then open new ones for Shadi's and Kathy's edits
... this way keep the code clean

MJM: Kathy can you open a new issue?

KE: do you want one for each or one for both?

MK: one for both

MJM: an issue, not a pull request

Aspects Under Test

MJM: anyone have specific suggestions now?
... otherwise we'll defer

SAZ: if Alistair wants this removed, need to clearly explain why

AG: is the document available in its entirety

MK: will marge the changes and provide a link

AG: find GitHub very difficult
... can read document and respond to email

Next Steps

SAZ: several steps needed for CR publication
... review by WG is expected to be more stringent
... also need to negotaite exit criteria
... will be difficult to publish before TPAC if we don't have the final draft by the first week of october

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.152 (CVS log)
$Date: 2018/09/19 11:07:23 $