fix(aria-level): update to specify only level 1-6 #1873
Conversation
|
It should be considered whether ARIA headings may also have a level greater than 6 and that AT that does not transmit this is faulty. So far, there are already some browser-screenreader combinations that correctly output a level greater than 6 |
index.html
Outdated
| @@ -12021,7 +12021,7 @@ <h2>Definitions of States and Properties (all aria-* attributes)</h2> | |||
| <pdef>aria-level</pdef> | |||
| <div class="property-description"> | |||
| <p><a>Defines</a> the hierarchical level of an <a>element</a> within a structure.</p> | |||
| <p>This can be applied inside trees to tree items, to headings inside a document, to nested grids, nested tablists and to other structural items that may appear inside a container or participate in an ownership hierarchy. The value for <pref>aria-level</pref> is an integer greater than or equal to 1.</p> | |||
| <p>This can be applied inside trees to tree items, to headings inside a document, to nested grids, nested tablists and to other structural items that may appear inside a container or participate in an ownership hierarchy. The value for <pref>aria-level</pref> is an integer greater than or equal to 1. If <pref>aria-level</pref> is being used on an element with role <pref>heading</pref> then authors MUST set the value to an integer between 1 and 6.</p> | |||
There was a problem hiding this comment.
What would be the benefits of this change?
From the summary in #1050 (comment) it seems that levels 1-9 on headings are widely supported in user agents, with only Chrome not supporting higher integer values (properly). Similarly only JAWS appears to be failing with levels above 6 while Talkback struggles slightly with levels 7-9.
There was a problem hiding this comment.
@pkra my thoughts on this change were that as of right now all UA and AT support 1-6 and unless we suggest all UA and AT to support 1-9 then at the moment to ensure authors using aria-level with role heading are getting the expected experience across all UA and AT then they should use 1-6
There was a problem hiding this comment.
Thanks for clarifying. That sounds like good authoring advice (like your APG issue w3c/aria-practices#2620). From a spec perspective, it looks like pretty good support (especially with Chrome and JAWS looking like bugs), so I personally would rather file issues with user agents and AT instead of this change.
There was a problem hiding this comment.
Tend to agree. While I’m not itching to use heading level 8s anytime soon, I’m not sure why the spec should be so strict on that?
yes html only has up to level 6, but also, html doesn’t have buttons that natively convey a pressed state…
There was a problem hiding this comment.
Thanks @scottaohara for jumping in. Just want to confirm before moving forward. Does it make sense to update the spec to go to be integers from 1 and 9 and then open issues with UA and AT's so that they all support up to level 9?
There was a problem hiding this comment.
@ariellalgilmore I would wait until the WG has a chance to look at the PR next week.
|
Discussed at today's meeting: https://www.w3.org/2023/02/23-aria-minutes#t02 |
Co-authored-by: Peter Krautzberger <[email protected]>
SHA: af84654 Reason: push, by pkra Co-authored-by: github-actions[bot] <41898282+github-actions[bot]@users.noreply.github.com>
Closes #1050 Related APG Issue/PR: w3c/aria-practices#2620
Closes #1050
Related APG Issue/PR: w3c/aria-practices#2620
Diff