-
Notifications
You must be signed in to change notification settings - Fork 256
2024‐05‐31
Attendance (10): Mike Gower, Bruce Bailey, Alastair Campbell, Francis Storr, Filipo Zorzi, Scott O'Hara, Giacomo Petri, Lori Oakly, Dan Bjorge, Steve Faulkner
Regrets: Patrick Lauke
Bruce thanks group for work on Reflow and other thorny SC is benefiting WCAG2ICT drafting.
Mike complements group on closing/resolving 135 items since the beginning of the year.
Francis raised that branch cleanup was topic last week. In Patrick's regrets for today's meeting he asked that we look over recently closed PRs/issues that may have been closed a bit prematurely. Following up on Patrick's Chore: delete all branches that have been merged into main #3602, Francis has continued with similar work Chore: delete stale branches #3878. These are unmerged branches which have been untouched for years. Francis deleted 50 or so, but there many more. Francis believes the deletes are clean, but that he may have been too aggressive. Facilitators will review on their Monday coordination call.
Per usual, working from project board
Mike sent 14 (!) items for REVIEW - WCAG 2 proposed changes (due by June 10).
Expand 1.4.10 Reflow note about 200% and breakpoints, add matching note to 1.4.4 Resize Text #2630 needs review and is an item for discussion by facilitators on their coordination call.
1.2.3 / 1.2.5 Audio description - right language required? #3873 clear expectation for that language matches that of page. This can also be an issues with ALT. Conformance requirements for (CAV) for mentions same language. If Fail, under what SC? Agree that WCAG does not require translations. What about content offered in two languages, but captioning/AD only in one. Bruce would fail against 1.2.5 on its face. No one on call would pass without first significantly pushing back. There was not consensus that it part of the normative language. Dan offered Accessibility Supported as a clean way to fail. Mike will update draft response for WG. Understanding could note that AD/CC in a different language is a fail. This is a question that comes up frequently enough to be worth explicitly addressing in Understanding. A more complicated scenario is where captioning is provided for the English language version, but not other languages. Does the web page as a whole fail?
Could captions be an alternative for time-based media SC 1.2.1 with audio-only content #3870 ask if streaming text is offered for audio-only media, must it be synchronized? How would an audio-only format provide for the visual needed for captioning? If there is a streaming text, but its not synchronized does that count as an "alternative for time-based media"? With a satellite radio, there is a display which scrolls, so the audio could be captioned on a nominally audio-only device. There are also some single media examples where an option for streaming captions is not unrealistic. When streaming text is provided to meet 1.2.1, but the text stream is not in sync with audio, does that fail 1.2.2? There not a requirement for synchronization when a static text document (e.g., transcript or screen play) is provided to meet 1.2.1. Since static text would be satisfactory for 1.2.1, what is the logic then for failing stream text? Alastair suggests a better way to address the question is with a technique.
Alastair was not convinced (yet) that an audio-description in the wrong language makes it in-operable. Dan agree that it’s not as cut-and-dry as would be ideal, but thinks we would prefer to come up with a generalizable way for this type of thing to fail, and that this is the closest he can see.
Scot asks for feedback on an HTML5 issue, Considerations for providing alternative text to video poster graphics #7954. Please be encouraged to add to the discussion. Also an item for facilitators meeting.
Alastair asks the answer in the question? “Is placeholder ok even though it sucks?”. Um, it sucks therefore not ok?
Mike noted that placeholder text will usually fail contrast requirements, and probably other SC. Dan and Scott suggests making failure explicit. Placeholder text is part of the Accessible Name calculation, but low in the tree, and remains a barrier (in actual practice) for Label in Name when using Voice Recognition. Scott will revise and/or create a new issue.
Steve shared his article re-upped: placeholder the piss take label about how placeholder text is terrible.
Dan noted that with 2.1 Understanding pages, only the first horizontal navigation menu work. This is only with 2.1, version and not 2.0 or 2.2. Mike sent Kevin a note.