Meeting minutes
Introductions
<chlane> scribe; chlane
chlane: I said I'd scribe today :) unless you really wwant to
<chlane> ok
New Issue Triage
https://
https://
scott: figcaption is not up to date in html-aam for UIA
scott: I'll probably make a PR updating
<scotto> https://
scotto: scoped is used in reference in header, footer and soon aside, when then map to implicit landmark roles or become generic
scotto: spectranaut pointed out "scoped" is not a defined term
scotto: I want to get some more information on similar terms, if we should use it, or if we should just define it
jamesn: css defines something called "scoped" which might actually work!
<jamesn> https://
jamesn: they defined it as an exported definition, so if we mean what they have defined scoped as, we should use it
<melsumner> do we mean something other than what computer science means?
<jamesn> https://
jamesn: utility called "xref" which you can use to search for references
jamesn: and how to reference it when writing spec
New PR Triage
https://
jamesn: editorial, peter reviewed, is that enough scott?
scotto: its a single word sounds good
https://
scotto: needs minor review
scotto: trying to close carolyns old simpler issues
jamesn: reviewers chris lane and juanita (who needs to be added to the github org) and melanie
https://
jamesn: reviewers: melanie and james and valerie
https://
scotto: another html-aam from caroyln, if you review please read linked discussion
jamesn: sounds like something aaron or cynthia should review?
cyns: me and aaron are probably too swamped right now
cyns: if it can wait I can do it in a couple weeks
scotto: it can wait
reviewers: james, siri and cyns
<Zakim> jcraig, you wanted to say (metaphorically "say") sorry I have a meeting conflict, but if there are review assignments or other actions for me, please mention them here with jcraig in the notes and I'll review later. Thanks all.
Should we meet next week? CSUN / axe-con
jamesn: some members are going to be at CSUN and others at axe-con
jamesn: if you will be there, speak up and you can all meet!
jamesn: maybe we will need to move the deep dive next week
melsumner: I agree we should push
jamesn: what about hte meeting next week, should we skip?
scotto: the internet can wait
jamesn: next meeting will be the 24th of march, and the next deep dive will be then as well
jamesn: if there is any more details to put into the agenda, melsumner please email me
jamesn: day lights saving is this weekend btw
Deep Dive planning
jamesn: we now have one for the 24th, any one want to propose one for 31st of march?
jamesn: if not, people have time
Handling Author Errors: form & region roles @cookiecrook to review agendabot]
jamesn: jcraig is down to review
this is a nudge to jcraig
please review you are the last outstanding :)
Should form-with-name map to IA2_ROLE_LANDMARK? - can we close?
jamesn: scott thinks nothing needs to be updated at this point?
scotto: yeah I made a PR for this, but I do have one more question on the issue
scotto: I need clarification from jcraig
jamesn: we need jcraig !
Figure out what to do with the DPUB mappings overlap with the ARIA Core spec, in the context of newer implementation differences - publishing responded - what is next?
jamesn: another jcraig thing. we were waiting for the DPUB working group to get back to us and they made a comment.
jamesn: aaron can you read the comment?
aaronlev: I'll take a look
scotto: I'm sort of familiar with this... it seems like a lot of these roles could be solved with aria-roledescription
jamesn: seems like they think all things are equally important
jamesn: can I leave this with aaronlev and jcraig ?
aaronlev: I don't have time for it, I support the idea, I don't see the urgency
aaronlev: btw we are looking for a full time standards person at google
jamesn: I'm going to move the agenda and deep dive, because we just aren't ready to move anything forward here
Support aria-description - how can we move this forward?
bryan: I'm not sure how to consolidate everything that has been discussed into one simple PR
jamesn: melsumner can you help with responding to jcraig comments?
melsumner: we are actually waiting on jcraig who thinks he can resolve some of it, and I can also look into what I can resolve
When is hidden content taken into calculation of name and description? - any updates?
jamesn: any updates to this accname issue?
aaronlev: I'm waiting for jacobo to respond to the comments I left on his PR, I pinged him yesterday
<aaronlev> https://
aaronlev: spectranaut can you help move this one forward?
yes
<aaronlev> That's the current PR for the accname improvements for invisible/hidden content
aaronlev: it's a very long project, there is a lot of nuance
jamesn: I'm concerned about how hard it is in chromium, will it be hard to get into firefox and webkit?
aaronlev: it's harder to pave the road than drive down it
Should SPACE to be allowed to activate the <a> element?
melsumner: specifically, I am proposing that browsers implement these space bar to active to an <a> element for links the same way a space bar will active a button
<Jaunita_George> +1!!
melsumner: what I have done into this issue is explained how I ran into this from a use case scenario, I have since heard other people have the same problem
melsumner: I also looked through all the specs and can't see anywhere something that says "what should space bar do"
melsumner: which is weird because more other keys are covered
melsumner: maybe I missed something?
melsumner: what should a space bar do when a link <a> is focused
melsumner: jaws-test commented that neither the html or aria specs says how elements should be operated, it is currently left up to the browser
melsumner: where does this request belong??
melsumner: probably this group should decide philosophically, but I'm not sure where it should go finally
jamesn: I agree we should define this, but ARIA can't define it, and I am not sure where it should go
jamesn: this is a fine place to track and discuss for now
<Jaunita_George> +1 to cyns
cyns: it probably belongs in html
<MarkMcCarthy> +1
cyns: I talked to one person at html and I didn't get hard pushback. I'm happy to back channel it a little
everyone is very excited about focus on links is set by default
on mac now
jamesn tested it and actually it doesn't work
cyns: eventually, not now, we should discuss that different operating systems have different keyboard behavior
scotto: why does anything need to change here? why do all links need to change because of confusing cause by making a link look like a button? if a link looks like a link, it's long standing behavior that space doesn't active the link? why should we change it at the element link?
scotto: it seems a large swing into the other direction to fix a small number of links
scotto: for example, when navigating a page, space bar is useful
scotto: jaws-test brings up a good point, if you want to style a link ot look like a button, then add a spacebar event listener
jamesn: maybe we could have an attribute on a link instead of having to write a javascript?
melsumner: if you are using tab to focus on an interactive element
melsumner: in my mind, there are things that are forms and things that are not forms, so, links are not a form so it doesn't respond to space bar
melsumner: is this a bug? was it done on purpose?
cyns: keyboard interaction was thought about in the early days of implementing the internet. my impression is that it was accidental
sarah_higley: I'm curious how much this is actually an issue. how much this is presenting a problem to keyboard users?
sarah_higley: I feel like this is all speculation until there is more user research
<MarkMcCarthy> +1 sarah_higley - good point
sarah_higley: because it is such a long standing behavior
aaronlev: what happens if you are in a form
aaronlev: enter does the default action in the form - so what happens if you have a link in a form?
aaronlev: this doesn't seem like an aria working group thing
jamesn: I were going to echo -- is this REALLY a user problem?
jamesn: we think it is a problem because we get reports in audits
melsumner: I see this as a minor inconvenience, if space bar doesn't work, I'll try enter
<sarah_higley> @aaron enter does not submit a form when focus is on a link
<sarah_higley> at least not on Windows :D
melsumner: the element doesn't meet users expectation, but, users know you could probably also press enter.
<siri> +1
<Zakim> cyns, you wanted to say I've seen in usability, but it wasn't the thing being tested
melsumner: the way sarah_higley put it hit home -- content heavy websites that use anchor tabs, and you can scroll the content through anchor tabs. I just tried in aform, and it had a link in it, when you press enter in a link, it did not submit form,
cyns: I've seen this come up in usability as a side issue a handful of times
jamesn: people who try enter on a button if space doesn't -- sometime enter does something unexpected, and enters not the button you want (default button on mac?)
scotto: if we have to do this for links, because the visual style confusing people, what about buttons that look like links?
<siri> lol
jamesn: is this something that could be discussed in open ui?
<siri> cancel button
jamesn: a link that functions like a button/looks like a button
cyns: why do this instead of html?
jamesn: it's easier?
cyns: what about an attribute in html?
jamesn: oh that is fine to
cyns: I'll drive this discussion forward with html
scotto: I don't know about open ui
scotto: I think we should just go directly to html
cyns: assign it ot me
RSSAgent, make minutes