W3C

- DRAFT -

Protocols and Formats Working Group Teleconference

17 Sep 2015

See also: IRC log

Attendees

Present
Joanmarie_Diggs, fesch, janina, Rich_Schwerdtfeger, MichaelC, Joseph_Scheuhammer, Bryan_Garaventa, Michiel_Bijl
Regrets
Chair
Rich
Scribe
fesch

Contents


<trackbot> Date: 17 September 2015

<richardschwerdtfeger> Meeting: W3C WAI-PF ARIA Caucus

<richardschwerdtfeger> https://lists.w3.org/Archives/Public/public-pfwg/2015Sep/0108.html

<MichaelC> +Matt

<scribe> scribe: fesch

<richardschwerdtfeger> https://lists.w3.org/Archives/Public/public-pfwg/2015Sep/0167.html

<richardschwerdtfeger> https://lists.w3.org/Archives/Public/public-pfwg/2015Sep/0108.html

Topic Issue 633

<richardschwerdtfeger> https://lists.w3.org/Archives/Public/public-pfwg/2015Sep/0167.html

Action-1724

<trackbot> Action-1724 -- Matthew King to Create a proposal to simplify grids in order to incorporate the list view concept -- due 2015-09-17 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1724

mk: have a branch for grids but have a syntax error, so missing headers
... go to grid...
... I rearranged and streamlined the description of grid, making it broader, so it is broader than tabular data
... made a basic change to the summary, second paragraph

<clown> http://rawgit.com/w3c/aria/mck_issue633/aria/aria.html

rs: there are some redundant text
... why do we need to call out the grid role here? It should be inherited

mk: look at the however...

rs: why to we have treegrid mentioned in grid?

<richardschwerdtfeger> https://rawgit.com/w3c/aria/master/aria/aria.html#grid

mk: I thought I got rid of all the erroneous mentions of treegrid

js: : I don't see treegrid in Matt's branch

jd: grid is messed up

mk: I tried to fix it all
... the screen reader developers mostly today don't support captioning table with labelby
... it is new in aira 1.1 to label a table this way... if you were building a table role you wouldn't have a caption

rs: summary was deleated
... If you want to have a visual caption, you could say you provide an aria-labeledby
... what you are saying is, if you built it on a table with a caption you would get it for free

mk: if you built a grid off a table and the table had a caption, I am assuming it (the caption) would be a label on the grid

<clown> http://rawgit.com/w3c/aria/master/html-aam/html-aam.html#el-caption

rs: because HTML and ARIA have so much overlap we should have joint testing
... I think there is a problem as there is no relationship,

js: mostly it is exposed at text...
... labeledby property is on the table

rs: so it is OK

mk: I added to be keyboard accessible the author should control focus
... added users navigating, that the elements that receive focus should be gridcells

js: sounds like it is right out of APG

rs: do we talk about owned elements?

mk: no nothing specific

rs: that was really important

js: yes, it is in there

mk: yes I clarified it
... I added rowheader and columnheader, to be clear

rs: para with accessible name - need changes ...
... want to address caption issue
... common ways to provide captions are labeledby and...
... are people happy with the rest of that text?

js: I see aria-owns from the characteristic table

mk: that is not intentional

rs: missing inherited properties...

mk: editing error causing respec to fail

js: need to read final version

mk: discusses grid text, rewrote readonly part

rs: I think we need to mention gridcell and cell could be used interchangeably (synonyms)

mk: I thought one was a descendent of the other

rs: I thought the group wanted to go to cell

<richardschwerdtfeger> http://www.w3.org/2015/09/03-aria-minutes.html

discussion of minutes...

mk: wrote two paragraphs about readonly... discussion
... hopefully the readonly paragraphs will help screen reader develops...

<clown> http://rawgit.com/w3c/aria/master/html-aam/html-aam.html#el-td

js: the current mapping for td is gridcell
... td and th are mapped to gridcell
... might be HTML has not caught up to us (don't have cell)

rs: we need to open an action against the mapping spec.

js: probably want to say if a td is in a grid map it to gridcell if in a table map to cell

mk: in UIA won't gridcell and cell map to the same thing?

cyn: we have a table pattern and a grid pattern

rs: what about th?

js: reads mapping spec on th

rs: will send a note on tds
... change the captions

mk: did you read the readonly?

rs: yes that is OK

Action-1725

<trackbot> Action-1725 -- Matthew King to Create a proposal for an article feed -- due 2015-09-17 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1725

<clown> User Agents MUST implicitly propagate the value of readonly set on a grid element to all its owned gridcell elements and expose the value in the accessibility API.

js: get rid of implicitly

mk: yes OK

<jemmajaeunku> +present JaeunJemmaKu

<clown> action-1725

<trackbot> action-1725 -- Matthew King to Create a proposal for an article feed -- due 2015-09-17 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1725

<clown> http://rawgit.com/w3c/aria/mck_issue633/aria/aria.html#itemfeed

mc: describes concept

rs: so you want stuff put on either end of the list?

mc: yes

rs: couple of questions - Is that a SHOULD?

js: capitalize the word to get the styling

rs: are people happy giving this much guidance to AT?
... if you want to do that, then bring in AT vendors

mk: I am talking to Freedom and another vender....

rs: you have an exact host language attribute you want authors to use (tabindex)
... we are looking at alternative keyboard navigation for later SVG

mk: say why they should be focusable that way and not some other way

js: what if the person is using activedesendent?

mk: it is not a widget

fe: what if someone hides something

mk: we want them to play nice
... if you have a user initiated action, we don't care about that
... only infinite scroll

rs: you want to have interactive lists,

mk: yes, will add

rs: what happens if you drill down into a list item

mc: if you are non screen reader user and using this element, this contract doesn't do anything for you...

rs: are you going to make a authoring practice for it?

mk: this doesn't prescribe how this is done
... this element is a structure not a widget

rs: if I put a link in the list item, what about putting a list in a list item, will that confuse you?

mk: no, the only time the screen reader user cares about when you navigate through a focusable item...
... doesn't matter what is in a listitem

rs: you could have a feed within a feed

mk: you could something like that, I don't think it would be a problem

rs: may want to know when you have a feed within a feed you should put labels on them
... any reason we don't call it a feed?

mk: we have another role, where the children inside are article so you can label and give a description in the item, so the description is what a screen reader should hear when scrolling through the list
... this one could be feed and the other is an articlefeed

rs: I don't think you one to be a subclass of the other, they should be a subclass of list

mk: I thought I would be able to simplify them as a child
... I don't have a strong preference whether they are siblings or descendents

rs: I would make them siblings

mk: big TODO is to define the prepends and postpends ...

<clown> http://w3c.github.io/aria/aria/aria.html#aria-relevant

js: wondering anything to do with aria-relevant?
... aria-relevant is all about additions and removals

mk: could be doing both additions and removals in the same action
... the screen reader doesn't have a signal

js: aria-relevent can do both additions and removals at the same time - all

jd: do you need of index of where it is?

mk: can represent all actions with sighed values
... so if a screen reader developer wanted to reveal that information, it would be very easy

jd: do you have a screen reader vendor that wants to know addition, subtraction and position?

mk: confusing example of adding and subtracting at the same time, but pretty easy to figure out where you are
... important thing is when aria-busy, that the screen reader not loose where they are
... user might be moving quickly through the list

jd: Orca does not use the virtual buffer....

<Zakim> clown, you wanted to ask about the relevance of aria-relevant

Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.140 (CVS log)
$Date: 2015/09/17 18:01:57 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.140  of Date: 2014-11-06 18:16:30  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Succeeded: s/treegrid is messed up/grid is messed up/
Found Scribe: fesch
Inferring ScribeNick: fesch

WARNING: No "Topic:" lines found.

Present: Joanmarie_Diggs fesch janina Rich_Schwerdtfeger MichaelC Joseph_Scheuhammer Bryan_Garaventa Michiel_Bijl
Found Date: 17 Sep 2015
Guessing minutes URL: http://www.w3.org/2015/09/17-aria-minutes.html
People with action items: 

WARNING: Input appears to use implicit continuation lines.
You may need the "-implicitContinuations" option.


WARNING: No "Topic: ..." lines found!  
Resulting HTML may have an empty (invalid) <ol>...</ol>.

Explanation: "Topic: ..." lines are used to indicate the start of 
new discussion topics or agenda items, such as:
<dbooth> Topic: Review of Amy's report


[End of scribe.perl diagnostic output]