W3C

- DRAFT -

Accessibility Conformance Testing Teleconference

10 Aug 2023

Attendees

Present
Helen, kathy, Wilco, Suji, thbrunet, ToddL, Catherine
Regrets
Chair
SV_MEETING_CHAIR
Scribe
Suji

Contents


<scribe> scribe: Suji

ACT Standup

<Wilco> https://github.com/act-rules/act-rules.github.io/pull/2060

[Wilco]: Secondary requirements, with changes to conditionally applicable. Kathy to review it. Chose related instead of conditionally applicable.

[Helen]: Chat with Karen to help start manual test. Worked on PR.

[Tom]: Will be working this week

<kathy> https://github.com/act-rules/act-rules.github.io/pull/2093#pullrequestreview-1564841573

Kathy: PR to remove <html> and <body> tags. John suggested putting it back.

Wilco: Not convinced. Since it is applicable only to descendants of the <body> element. I am okay with it but it is not my preference.

Suji: I am okay either ways

<kathy> https://www.w3.org/WAI/standards-guidelines/act/rules/de46e4/

Helen: I like how Kathy's done it

Tom: Make an argument browser's will do or add a note it is within a html body tag

Wilco: I prefer John's solution over adding a note. I am not sure if applicability should mention the <body> element. It doesn't technically matter. Is anyone opposed to the cahnge?

Helen: Send an email to John

Wilco: Applicability is strong and should not exclude html element... John has convinced me. Let's change it the way he suggested.
... it is clearer and cleaner to have it in the examples

Suji: Connect with Kathy and work on annual review

Todd: Approving PR's

Cancel August 17th

Secondary requirements and accessibility support

Wilco: Number of rules where failing a rule in only certain scenarios like browser technology and Assistive technology. Example, auto-play audio and all do it the same way and all do it all the time

<kathy> https://github.com/act-rules/act-rules.github.io/discussions/2095

Wilco: Meta elements, using it to restrict zooming in mobile devices. Some browsers in mobile ignore it all together and varies from browser to browser. You can fail the reflow as a failure or not and mark it as a best practice or warning.
... iframes behave differently in different browsers
... empty heading elements get often ignored but not everywhere. There is inconsistency around whether we fail them or mark them as needs review

Kathy: Wilco added a discussion https://github.com/act-rules/act-rules.github.io/discussions/2095
... Didn't send it out to the community yet

Wilco: Trying to get test suites to be more consistent. We have to decide whether or not to fail the success criteria. There are also levels of Accessibility support. Testers have flexibility on things.
... so depending on the browser and technology they can determine it is a pass/fail

Tom: Minimum requirement, one combination of browser + AT. You may not know they issue exists until you test with a particular combination.

Helen: When you code, you fix based on the browser?

Tom: File bugs with browser + AT

Wilco: Browser are allowed to choose how they render

Kathy: In the previous trusted tester example, some examples weren't auto-playing now all the videos were auto-playing. So there might have been a browser update

Wilco: If it works in none, then it is not acceptable

Tom: How many different combinations to support and test? We need one minimum combination

Helen: Depending on how things are used, it may be or may not be an issue
... non-text, like an image vs a place holder image for a video with a play button
... on Android something works with Talkback but not with Switch Access. Then it is an issue.
... it should be failed as an issue due to browser

<kathy> Suji: depends on instance to instance

<kathy> ... at a bare minimum, if it works with one browser and AT, it passes

<kathy> ... mark others as best practice

<kathy> wilco: so not a failure if it works in several browsers

Todd: I do test with 2 different screen readers, mobile - voice over and talkback
... I am not sure which way I would lean on this
... I am neutral

Tom: Automated failures fail everywhere. Manual stuff is different

Wilco: In Axe, we are stricter on Accessibility support
... manual audits are not thorough

<kathy> suji: different results in Windows and Mac

Wilco: If they are secondary what is we are not failing that under WCAG

Tom: We are acquiring a recommendation

Helen: Checkpoints which fail 6 different success criterions, fixing it for one might fix it for all of them
... should we be bothered by which one it fails?

Wilco: We wanted to make sure that correct success criteria is failed. Sometimes other success criteria might fail but not required to mention

Helen: Do you list it in both and them report in both?

Wilco: Report a issue and mention the list of success criterions impacted

Helen: It might always fail for AAA requirement. It is difficult to achieve.

Wilco: Failing as best practice, we know it can be a WCAG failure. Should we be looking at it as a different type of issue?
... Pass/Fail/Can't tell

Helen: Techniques and best practices... but is not a failure

Wilco: It is a normative failure in a screen reader you didn't cover
... outside the scope of my audit
... issue with a particular browser

Helen: JAWS and Voiceover are good at guessing. NVDA doesn't.
... over-engineering to get something to work in a particular browser

Kathy: Trevor expressed concerns about A11y support in secondary requirements, we should include him in the dicussion.
... Fails with JAWS, we fail it. Because there could be other AT's that it fails with
... The SC is stricter than the rule. Keyboard trap SC listed as secondary for both. SC allows more solutions than the rules.
... trying to get it to fit best. If it is listed as secondary requirement, it may get overlooked.

Helen: When testing with screen reader, if there is a power user vs a beginner

Wilco: Turn off meta elements. Rule on scrollable elements. Patrick reached out, may be that passes if cursor is enabled.
... it is not standard behavior

Kathy: Testing with auto-play earlier, you can enable/disable auto-play setting. We told tester to enable the setting. Users would disable it.

Helen: With screen readers, auto-filled alt-text for images

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: 2023/08/10 13:58:13 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision VERSION of 2020-12-31
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00)

Default Present: Helen, kathy, Wilco, Suji, thbrunet, ToddL, Catherine
Present: Helen, kathy, Wilco, Suji, thbrunet, ToddL, Catherine
Found Scribe: Suji
Inferring ScribeNick: Suji

WARNING: No meeting chair found!
You should specify the meeting chair like this:
<dbooth> Chair: dbooth


WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: 

WARNING: IRC log location not specified!  (You can ignore this 
warning if you do not want the generated minutes to contain 
a link to the original IRC log.)


[End of scribe.perl diagnostic output]