See also: IRC log
<scribe> scribe: MichielBijl
<jamesn> https://www.w3.org/Bugs/Public/show_bug.cgi?id=29469
Contents of bug report: Get rid of choices or in 1st para description.
3rd paragraph - put the instruction b4 the explanation.
JN: Anybody have comments?
MB: Last paragraph reads “See the examples for recommended ways of designing around this limitation. (Editor's note: examples are not yet coded.)” what sort of code examples do we need?
JN: Do we normally refer code examples in description? Seems weird.
MB: No don't think so. I'd just remove it.
JN: I'm okay with last paragraph, seems like general advice, but might be more important for listbox.
MB: Agreed.
JN: Move on to keyboard interaction
*something about tabindex*
JN: First bullet should not be under keyboard interaction.
<JF> +1
Bikir: change to just read “widget should be in the taborder”
+1
<annabbott> +1
JN: Same with second bullet. First sentence is all we need.
+1
*all cheer and shout in agreement*
“When a listbox receives focus, visual focus is set on an option inside the listbox. If browser focus is set to the container with role listbox then aria-activedescendant must be used to refer to the option with visual focus.”
JN: Remove that bullet
MB: Ignore me
JN: visual focus ≠
selecting
... Why is this one so long?
... Is it useful being this long?
JK: It's good to understand all
Birkir: it's broken down more
MB: listbox is equal to select element with either rows or multiple set right?
JN: Yes.
... we need to clarify difference between select and focus
Birkir: yeah we do
JN: how do we do that?
... should that be part of the description?
MB: Yeah
Birkir: yes
JF: *issue with
aria-select=none*
... testing combobox example
... if you move focus it states it's not selected
... question is, combobox is like a single select, should you
be setting aria-select true on focus?
<jamesn> "Single-selection containers where the currently focused item is not selected. The selection normally follows the focus, and is managed by the user agent."
<jamesn> "Multiple-selection containers. Authors SHOULD ensure that any selectable descendant of a container in which the aria-multiselectable attribute is true specifies a value of either true or false for the aria-selected attribute."
w3c.github.io/aria/aria/aria.html
<jamesn> " If no DOM element in the widget is explicitly marked as selected, assistive technologies may convey implicit selection which follows the keyboard focus of the managed focus widget."
JG: What does that say for the
author?
... single select don't bother with it?
JN: single select do the same thing; aria-selected on focus
JG: shouldn't matter from APG point
JN: Agreed
... That's what the states and properties says
... should clarify what we mean by aria-selected
Birkir: once you've entered a list you can't deselect
MB: example for html select
http://dir.rawr.eu/select.html
... to confirm, you cannot clear your selection in that example
on the Mac.
<annabbott> AA: drop the bullet & text for context menu
MB: are those even a thing?
JN: yeah, a rare thing, but they
are a thing
... dropping it sounds reasonable as it's a general thing
(context menu key combo)
Jemma: lots of duplication in the multiple key interaction
MB: focus on removing duplication
AA: add general key strokes (up/down, home/end)
Birkir: agreed
JN: make home/end optional?
Birkir: same as type ahead (strongly recommended)
MB: +1
<jamesn> https://www.w3.org/Bugs/Public/show_bug.cgi?id=29469
JN: what about note?
MB: meaningless
JN: drop note
MB: fourth bullet is vague
... does it mean they're not present in the DOM, or not
descendants.
JN: posinset only used if not in DOM
Jemma: what is the context of this situation
JN: if not all children of the listbox are loaded (due to dynamic loading)
MB: “If not all options are currently in the DOM, all options should have their aria-setsize and aria-posinset set accordingly.”
Birkir: want to merge the second and third bullet
MB: +1
<annabbott> AA: +1
<jamesn> "Each entry in the listbox should have a role option and should be a DOM child of listbox or referenced in the listbox by aria-owns."
Birkir: should “selected list item” be option?
AA: we use different things in other bullets
MB: whatever is used, it should be consistent
AA: should it be “selected list option”?
JN: add note for editor to clean up?
*all agree*
MB: there should be something in there to state the thing is a multiselect
*all agree*
JN: aria-multiselectable, and all children/entries/options/items need aria-selected=true|false
JG: are we getting rid of control+page up/down
JN: I hope so
MB: me too
JN: could be optional, but prefer removing them
https://www.youtube.com/watch?v=iRmX-GvFeXI
MB: should we push this to next week?
JN: should we have multiple ways to do this?
MB: I could get on board with that.
JN: activate on focus could be bad for performance
MB: could circumvent with delay
JN: let's review the description
AA: doesn't give you the hierarchy
MB: +1
https://github.com/bkardell/common-panel
spec: https://specs.webplatform.org/#panels-and-panelsets
Err, actual spec: https://specs.webplatform.org/common-panel/bkardell/gh-pages/
JF: we should look at it closely and take cues from it
JN: we're trying to make this clearer right now
dismis zakim
https://github.com/w3c/aria/issues/226
https://github.com/w3c/aria/issues/221
JF: yes, task force should be two words
Okay, thanks John!
This is scribe.perl Revision: 1.144 of Date: 2015/11/17 08:39:34 Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/ Guessing input format: RRSAgent_Text_Format (score 1.00) Succeeded: s/JF/JG/ Succeeded: s/JF/JG/ Succeeded: s/ctrol/control/ Found Scribe: MichielBijl Inferring ScribeNick: MichielBijl Default Present: AnnAbbott, jaeunjemmaku, MichielBijl, JamesNurthen, JF, JonGunderson, Birkir Present: AnnAbbott jaeunjemmaku MichielBijl JamesNurthen JF JonGunderson Birkir birkir Regrets: LeonieWatson BryanGaraventa IanPouncey MattKing Got date from IRC log name: 22 Feb 2016 Guessing minutes URL: http://www.w3.org/2016/02/22-aria-apg-minutes.html People with action items: WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]