<jamesn> Meeting: ARIA Authoring Practices Task Force
<Jemma> https://github.com/w3c/aria-practices/wiki/July-30%2C-2019-Meeting
<Jemma> Topic : ARIA-APG Release 4 Publication
<Jemma> scribe:Evan
ooh I haven’t done this in a while
<Jemma> Scribe: Evan
<Jemma> we need to ping Michael regarding the publication status
<Jemma> Topic : Disabled Days in Date Picker Dialog
<Jemma> https://github.com/w3c/aria-practices/commit/ecdb87a55dceed94ce0460bafc80d99c96757f3a
<Jemma> Days from the previous or next month in the grid are disabled, e.g., are not focusable. Clicking them was generating an error.
<Jemma> The fix causes the calendar to display the previous month if a day from the previous month is clicked or the next month to display if a day from the next month is clicked.
<Sarah_Higley> the jackhammers, right? Or aggressive woodpeckers? :D
<Jemma> Is this the behavior we want?
jemma: is the fix what we want?
<Jemma> https://github.com/w3c/aria-practices/wiki/July-30%2C-2019-Meeting
<Jemma> http://w3c.github.io/aria-practices/examples/dialog-modal/datepicker-dialog.html
james: why couldn’t I use the keyboard to go to the next month?
matt: what’s the least jarring when the user clicks on a date from the previous month?
james: I think it should choose it and move to it
matt: selecting/choosing causes
the dialog to close
... should you need to select the next month, then click the
date, then close it?
james: I’ve noticed that nearly every travel website avoids this by displaying two months at once
Sarah_Higley: jQuery UI selects and closes at the same time
<Sarah_Higley> jQuery UI example link: https://jqueryui.com/datepicker/#other-months
matt: it sounds like we might want to do something else
james: I can’t find a date picker that displays dates from a different month
<Jemma> Topic : Escape behavior in combobox
jemma: james will add a comment in the PR with examples
<Jemma> Issue 1066: Combobox: Why escape key should clear current selection
<Jemma> https://github.com/w3c/aria-practices/issues/1066
<Jemma> Matt proposes that escape should not clear combobox entry
matt: this caught me off guard
because I thought escape clearing text entry was fairly
common
... if what we’ve done here isn’t consistent, we should compare
to native as well as the web
<Sarah_Higley> sorry, I'm having some trouble hearing Matt as well
jemma: usually if you want to clear the selection or input contents, you usually click the “x” button next to it
sarah: I asked around a bit and
people didn’t expect that behavior
... on some windows apps, the first escape closes the menu and
a second press clears the value
evan: the mac finder search does both at the same time
matt: we might have it written one way in the pattern and implemented differently in the example
<Jemma> Escape Clears the textbox. If the listbox is displayed, closes it.
evan: the pattern says, “Escape: Dismisses the popup if it is visible. Optionally, clears the textbox.”
sarah: if it were switched to describe two escape presses, that would make sense to me
https://w3c.github.io/aria-practices/#textbox-keyboard-interaction
<Jemma> Thanks, Evan
matt: let’s put this on our agenda for TPAC
<Jemma> https://github.com/w3c/aria/issues/1011
matt: what’s the reason for ever using group inside of list?
<Jemma> "Neither the definition of the roles list, listitem and group nor the definition of aria-level in the specification contains any indication of how nested ARIA lists are created. Is the use of group sufficient or is aria-level required? If there is an element with role=group in role=list, this could be used to group or nest the list entries...? Unfortunately, the specification makes no statement about this. Therefore I would like to know how"
matt: my intuition is that it’s
allowed but shouldn’t be done
... i would like to know if there’s a use case for doing this
(visually)
... you have three different boundaries when doing this, which
would be very difficult for screen reader users
... is this an accidental side-effect of the ontology?
... you can group items inside a listbox but you can’t nest
listboxes inside of listboxes
james: this is not a result of the ontology, which clearly outlines how it’s meant to be used
curt: it seems like the case here is that someone wants to label sub-lists and grouping them seems like the way to do it
matt: is there anybody opposed to removing support in the spec?
james: we should look for more usage examples first
matt: even if it is being used, the ability to perceive its meaning seems unreasonable
curt: you mentioned the menu
earlier, and in that case the semantics would be menu, not
list/group
... …which supports matt’s assertion
<Jemma> Topic : Guidance for aria-level
<Jemma> "Discuss type of language to use for aria-level guidance for tablist and grid."
matt: I wanted to walk through what Simon/Valerie wrote and get feedback about aria-level
<Jemma> related aria issue is https://github.com/w3c/aria/issues/915
matt: what I’m curious about is
whether aria-level has any value on tablist or grid
... james had a potential example on grid
jemma: evan and sarah will take a look at this
matt: what I’m looking for here
is what we should say about the marginal cases
... we potentially need some standard language where the value
or support is not yet well understood
matt: I was hoping to have more
people for this discussion!
... I like to take some time after each release to reflect on
what’s working, what’s not working, github tools, etc.
sarah: it would be nice to have
some sort of chat channel for active collaborations
... aside from that, everything was smooth
matt: not everyone is on the same platform, which makes it harder
curt: we’ve been using Slack, which seems to be what people are standardizing around
zoe: how would we keep Slack open for public discourse?
jemma: Slack could be more for internal work-in-progress
sarah: it’s possible to make Slack public
evan: it’s possible to use webhooks with Slack channels to connect GitHub issues
matt: that could potentially get messy but it could also be very helpful
<Sarah_Higley> I'm sorry, I have to drop. Thanks everyone!
matt: maybe we should experiment with Slack
This is scribe.perl Revision: 1.154 of Date: 2018/09/25 16:35:56 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: Irssi_ISO8601_Log_Text_Format (score 1.00) Succeeded: s/Matt proposes/comment on the issue proposes/ Succeeded: s/comment on the issue proposes/Matt proposes/ Present: jamesn evan Sarah_Higley Jemma ZoeBijl Curt_Bellew Matt_King Found Scribe: Evan Inferring ScribeNick: evan Found Scribe: Evan Inferring ScribeNick: evan Found Date: 30 Jul 2019 People with action items: WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option. 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]