W3C

Accessibility Conformance Testing Teleconference

11 Feb 2021

Attendees

Present
shadi, kathyeng, trevor, Levon, Wilco, Daniel, maryjom
Regrets

Chair
Wilco
Scribe
shadi

Contents


announcement

<Wilco> MJM: going to have to step down as chair from the task force, and will step back as participant as well.

maryjom: due to circumstances, have to step down and leave the group
... will focus on Silver
... hope IBM will have replacement
... wish you all good luck

Wilco: worked since many years together on this, will miss you!

maryjom: yes, 4 years

daniel: thank you for your work!

kathyeng: thank you for your work!

trevor: appreciate everything

Wilco: down another member, close to minimum
... wondering about different approaches
... could maybe move some tasks to the community group?
... could also help streamline some of the processes
... still gathering ideas for a proposal

trevor: which kinds of tasks were you thinking of?

Wilco: mostly rule reviewing and maintenance
... requires more in-depth knowledge of WCAG
... which may need additional support from this group

trevor: so rules would go straight to AG?

Wilco: yes
... also results into a lot of overhead

trevor: largely expected that

shadi: how would that impact participation?

trevor: only have limited time
... would need to be careful if work time increases
... may not be able to participate in two groups

kathyeng: similar situation

shadi: timing of CG can be more difficult for US timezones

Wilco: was thinking of having different topics
... for the different meetings
... to meet people's interests

kathyeng: can CG join the TF?

Wilco: several participants not from W3C Member organizations

Scrollable element is keyboard accessible (Can this rule go to CFC?) - Updates were minimal to resolve issue 484: https://github.com/w3c/wcag-act/issues/484

Wilco: minor updates, mostly editorial

<Wilco> https://github.com/act-rules/act-rules.github.io/commit/1b21bc7a1a9068382917c9299c8613f32c4365fc

<Wilco> https://github.com/act-rules/act-rules.github.io/commit/aa79f5a188b6be0494ff4f3f08fd0f661df2e5dc

Wilco: tweak to assumption based on this group request
... following updated policy, can go to CFC

<Wilco> https://www.w3.org/2002/09/wbs/93339/ACTScrollableElement/results

trevor: fine going to CFC

maryjom: fine too

dmontalvo: fine with me

RESOLUTION: "Scrollable element is keyboard accessible" rule can go to CFC

Survey, Rule review of "Headers attribute specified to a cell refers to cells in the same table": https://www.w3.org/2002/09/wbs/93339/ACTCellHeader/

trevor: reads like giving a failure notice that gets ignored

Wilco: read it in the same way
... supposed to say failure of assistive technologies
... maybe wrong place

shadi: is this accessibility support?

Wilco: think just phrased wrongly
... and should go to the background
... don't need headers attribute if you have correct header markup
... not sure about fallback behavior

<Wilco> https://html.spec.whatwg.org/multipage/tables.html#algorithm-for-assigning-header-cells

Wilco: example seems correct but the assumption incorrect
... checks for hearders attribute first, then for table headers

trevor: read it the same way but needs testing
... could imagine AT having support for simple tables but not for more complex
... accessibility support testing with regular and irregular tables is needed

Wilco: question on tables constructed with ARIA attributes

trevor: expand scope to role="table"?

Wilco: wouldn't work because can't define on non-native

kathyeng: confirm situation that algorithm addresses
... IDs in the headers are not in the table
... would not be failures if IDs are valid
... is it failure of WCAG?

Wilco: would fail rule

kathyeng: but not necessarily fail WCAG?

Wilco: correct, rule is making an assumption

kathyeng: if one good ID and one bad ID, bad ID should be simply ignored

shadi: can we change expectation from "each token ..." to "at least one token ..." or "valid tokens ..."

kathyeng: seems to be similar to discussion on mapping accessible name calculation
... pointed to the algorithm in that rule
... could use similar approach for this rule too

shadi: which approach results in more accuracy?

Wilco: from my experience, this works well
... rule needs to go back and will need re-survey

RESOLUTION: rule needs to go back and will need re-survey

Summary of Action Items

Summary of Resolutions

  1. "Scrollable element is keyboard accessible" rule can go to CFC
  2. rule needs to go back and will need re-survey
[End of minutes]

Minutes manually created (not a transcript), formatted by David Booth's scribe.perl version 1.200 (CVS log)
$Date: 2021/02/11 15:45:35 $