Meeting minutes
New Issue Triage
spectranaut: https://
MichaelC: it came to the attention of the APA for horizontal review. It is an ARIA issue which has details. HTML AAM editors needs to look into this issue.
scotto: this is something similar that we did for HTML attribute
MichaelC: we do not need to rush to this for 1.2. Perhaps Scot has an idea
scotto: I would not consider it as high priority. need to figure out the scope and priority
MichaelC: may be we do not need to track this.
scotto: I am subscribe to it.
MichaelC: in this case we can close this issue
scotto: I'll leave it open as of now
https://
scotto: The reason I filed this issue that john has filed an issue that input file should not be a button by default in mapping.
scotto: I did find some issues about how it was being rendered.
spectranaut: do we need to add it in our agenda item?
scotto: yeap
https://
spectranaut: this is an agenda item for today
https://
spectranaut: a11y mapping on microsoft
jcraig: including ff and chrome on windows
spectranaut: we do not have milestone for this one.
scotto: creating a milestone on coreAAM makes sense
spectranaut: https://
scotto: if I understand, 1.3 is full so this may need to be pushed for 1.4
spectranaut: agreed
New PR Triage
https://
spectranaut: it needs reviews
scotto: we would be fine mapping the address to the group.
jcraig: you can add me as reviewer.
Deep Dive planning
spectranaut: Aaron, do you want to discuss dialog when you are available?
aaronlev: we can schedule it for the 28th
Tidying up section "Characteristics of Roles"
tracking issue for <dialog> & role=dialog
Columnheader roles do not work with the abbr attribute
scotto: this is an issue that I took over. I cameup with some proposal to resolve this and to provide some potential functionality in ARIA to associate some cells
scotto: I created a proposed text in the bottom of the ticket
scotto: if people like this idea, then i can make a PR to this
scotto: I have only partial testing results.
scotto: its getting specified in aria that we can use as an outline for expected behavior
jcraig: why do we want to use aria-label instead of header alternative?
scotto: abbr attr is supposed to provide a shorter text, if we use aria-label, there is always a conflict with various AT
<Zakim> jcraig, you wanted to say column header is not "children presentational: true" https://
spectranaut: is it for 1.3?
scotto: would not push for 1.3 anymore
scotto: creating this proposal for cell headers, right in HTML you can specify a number of column or row header. I propose that we write that up.
scotto: curious if we can specify it somewhere else?
jcraig: Looked attr name which is very long
<chlane> scribe/ chlane
<chlane> jcraig: Is this use case more broad than table cell headers?
<chlane> heading or fieldset alternative
<chlane> how to make a little more useful
<chlane> can we live with aria-label
<jcraig> chlane: I think this solves the problem re: buttons in header cells
<jcraig> chlane: we haven't been solving this with aria-label or aria-labelledby...
<chlane> aria label on col header
<chlane> either content is announced
<chlane> or subtree is overwritten
<chlane> subtree is overwittern
<chlane> jcraig: presentational true behavior
<chlane> scotto: subtree content is respected or the opposite is true
<chlane> scotto: see jcraig 's point
<chlane> scotto: be specific on how label works on headers
<chlane> scotto: could this be more useful on fieldset, legend, table caption
<chlane> Siri: it should not replace acc name
<chlane> should convey row header col header?
<chlane> ctrl-arrow
<chlane> this is what will be read
<chlane> I want accname, not another value?
<chlane> Siri: read whole content if you want to?
<chlane> scotto: intent to have the alt string read when navigating cells of table
<chlane> it should not change accname
<chlane> it should allow for an alternative to be announced
<chlane> would not be part of accName
<jcraig> <th aria-headeralternative="foo">Foo <button>sort</button></th>
<chlane> jcraig: when you land on header you want it to say foo
<chlane> but as navigating accross columns
<chlane> it is including the sort button
<chlane> scotto: that is one way this could be helpful
<chlane> aria-actions could be used as an alternative
<chlane> the use case more towards the code in the issue
<chlane> leave sort button out
<chlane> when the text of that string was 'my activities(all logged activity)'
<chlane> I don't meed to hear 'all logged activity' all the time
<chlane> should just hear 'my activities', 'my dates'
<chlane> as I read columns
<chlane> jcraig: in context where it is more appropriate to be terse
<chlane> scotto: I think this goes to how list items allow for aria-label
<chlane> intent was to allow for a terse representation of <li>
<chlane> someone should still be able to read the <li> text content
<chlane> more extended use case
<chlane> beyond col headers
<chlane> jcraig: whatever bike shed name we need a number of examples
<chlane> of where its useful
<chlane> on colheader it makes sense to read the whole thing
<chlane> but in cell use terse
<chlane> static list item, how are we going to call out when to use terse
<chlane> aria-label and aria-terse-label
<chlane> how does AT determine which to read/
<chlane> ?
<chlane> short or long
<chlane> scotto: may not agree with what I just said about list items
<chlane> fieldset legend
<chlane> long lengend for radio group
<chlane> navigating into cells, hear terse
<chlane> choose fav color
<chlane> jcraig: good one
<chlane> Voiceover summary
<chlane> reads position and structure
<chlane> Siri: shortcut, if different from col header users might not get same info
<chlane> scotto: it can be abused
<chlane> jcraig: hard to enforce
<chlane> fieldset legemd
<chlane> favorite could be shortened to 'fav'
<chlane> hard to enforce, validator rule
<chlane> scotto: authoring guidance
<chlane> scotto: have to finish testing on how the html attributes are not working
<chlane> spectranaut: list of general cases summarized
<chlane> scotto: further comments on cell headers are needed
<chlane> zakim. next agendum
Internationalization Wide Review: "Whitespace characters" underspecified
<chlane> https://