W3C

Accessibility Conformance Testing Teleconference

08 Jul 2021

Attendees

Present
Aron, Shadi, KathyEng, Wilco, Trevor, ToddLibby, Daniel
Regrets

Chair
Wilco, Kathy
Scribe
Shadi

Contents


Update from WCAG 3

<Wilco> https://docs.google.com/document/d/1JVmquc7mLJaxZhypPbBhR99fEFBmc0YDn1Wz2Jcl_oY/edit?usp=sharing

WF: several of us have been working on how to merge ACT Rules with WCAG 3 Methods
... looked at "Text alternative" outcome
... specifically on "Decorative Images"
... have a new section called "Data requirements"
... listing input needed to test the outcome
... testing is exclusive to one method, so if tested in one method should not be tested in another
... turned out to be relatively straight-forward
... added "Input Aspects" from ACT Rules
... and put the test cases from ACT Rules as "Examples"
... then added "Tests", which basically comes from ACT Rules
... the mapping is different from the current ACT Rules Format
... if we decide we want to go that way, we will likely need to change that
... added section called "Background"
... has the background, accessibility support, and assumptions from ACT Rules
... finally also a tab with the "Glossary", also from ACT Rules

TB: does decorative images have percentile?

WF: yes, this has 1-4 scoring based on percentages

TB: data requirements is list of images
... these get sorted into correct and incorrect lists
... then that generates the score, ratio of correct (appropriate) vs total

WF: correct, that roughly how it works

DM: assuming decorative images is just a subset of all images

WF: yes, each image would be tested by *exactly* one method

KE: data requirements is for entire outcome or for method?

WF: for entire outcome
... outcome needs these two lists
... each method contributes to these two lists

KE: have thoughts on scoring
... maybe number of images *tested* rather than *all* images
... because cannot test all images
... is that for discussion here?

WF: outcome as currently written is all images within the view
... that conversation is out of this scope
... question was, if we can fit ACT Rules into methods
... and the answer seems to be yes
... another thing to point out
... one method would be essentially one rule
... for atomic rules, we might need a composite rule to sit between
... for example to combine all language rules

SAZ: like the approach of excluding scoring from this particular discussion
... having these granular tests allows for broader discussions on scoring and conformance models
... important is to get this basis first

KE: will we be changing the ACT Rules Format?

WF: would really like methods to qualify as ACT Rules
... if we continue this approach, then may need to make some changes
... could make certain things optional
... next steps is to wrap up this proposal and bring it to AGWG
... then figure out what would need to change if there is a go-ahead

TB: mentioned methods would map to rules
... so would methods be disjoint sets

WF: solves a problem in this case
... not sure would be the same approach in all outcomes
... would need to try out more outcomes
... but would also want to avoid every data requirement being different
... that would be complicated

SAZ: AGWG Charter runs out end October 2022
... expect re-chartering discussions in about a year from now
... expect this Charter will be to finalize standardization of WCAG 3
... so would be good to get structured testing (aka ACT) as part of the consensus model

ACT rules checks sheet

WF: "Document has heading for non-repeated content" was discussed in AGWG
... but was a heated discussion
... not clear consensus, quite some opposition
... need to see how this works out
... if it stays under Bypass Blocks
... so we're really on hold for that

AJ: have 3 approvals but can't merge

WF: yes, needs approvals from 3 members
... people with write access

AJ: but then you or Carlos need to merge?

WF: correct
... just let us know

TB: so you had a comment on the videos rule
... need to check cross-browser support for MP4

<dmontalvo> https://github.com/act-rules/act-rules.github.io/pull/1647

TL: didn't get to work on mine yet
... will get to it in the coming few days

WF: all have assignees
... looking good!

Open ACT pull requests

WF: #1658

AJ: going to final call now

WF: great!
... #1649
... can also be merged
... #1510

DM: probably outdated
... close and I'll open a new one

WF: closed

SAZ: #1548?

WF: need to meet with Aron

AJ: have some updated thoughts, will try to get to it this week

Plans for the implementation matrix

WF: been discussing in the planning group
... to get the current implementation data we have
... will be manual for now and slightly out of date
... but show the world how it could look
... and work in parallel on automation
... also migrating rules into the new design
... would like to announce in September, after the holiday period
... stay with current implementation matrix design
... and refine it as we go along

KE: talked about dividing up tasks

WF: yes, bunch of coding needed
... if anyone wants to chip in

AJ: not a developer but can try to help

TB: have spare couple of hours, depending on the effort

WF: need to figure out which bits can be handed off
... but scheduling 3 days next week to focus on this
... meeting with Hidde to learn about W3C side of mechanics
... would be good for more of us to know how this software works

SAZ: also need to do more outreach and attract involvement

KE: trying to get more participants
... people busy but more room for participation
... let us know if you have ideas on who we can contact

WF: agree!
... hope also implementation matrix will help attract participation

Summary of Action Items

Summary of Resolutions

[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.200 (CVS log)
$Date: 2021/07/09 09:06:00 $