Protocols and Formats Working Group Teleconference -- 30 Jul 2015

<trackbot> Date: 30 July 2015

<richardschwerdtfeger> meeting: W3C WAI-PF ARIA Caucus

<richardschwerdtfeger> https://lists.w3.org/Archives/Public/public-pfwg/2015Jul/0088.html

<scribe> scribe: MichielBijl

Jemma: I work for UI, my first time here
... I work with Jon

MichaelC: received comments that ARIA should not focus on accessibility API's; just API's in general
... drafting a response to Charles

<clown> https://lists.w3.org/Archives/Public/public-pfwg/2015Jul/0049.html

janina: ask DPUB to update document

Heads Up Items

Rich: what I want to ask to group, is be pragmatic about; don't want it to be a emotional discussion

richardschwerdtfeger: Apple had publishers coming in that said they need this

janina: we will be very formal about this disposition
... I would propose adding the document from DPUB if we keep aria-describedat

mgylling: everything you said about publishers showing up on the 13th is true
... we need it
... we don't want to be the ones that potentially stall the aria progress

janina: we have talked about that possibility
... ARIA 1.1 is on schedule

<clown> issue-633?

<trackbot> issue-633 -- listbox and tree may contain only static items; badly need interactive widgets that can contain interactive typed items -- open

<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/633

<clown> action-1688?

<trackbot> action-1688 -- James Nurthen to Draft proposal for aria-interactive -- due 2015-10-01 -- OPEN

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1688

jamesn: I'm on vacation, so cannot push action forward

richardschwerdtfeger: let me have a look at it
... I'm gonna put on the 23rd of september

issue-397

<trackbot> issue-397 -- Need a way to specify an indeterminate set size -- open

<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/397

<clown> action-1488

<trackbot> action-1488 -- Joanmarie Diggs to Propose spec text for allowing aria-setsize="999+" format change for issue-397 -- due 2015-07-30 -- PENDINGREVIEW

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1488

MichaelC: there is note that went privately to richardschwerdtfeger and me

richardschwerdtfeger: I'm going to forward it to the list

<clown> https://rawgit.com/w3c/aria/action-1488/aria/aria.html#aria-setsize

Joseph: suffix with a + sign to indicate the setsize is at least that amount

richardschwerdtfeger: where do you map the +?
... AT's have to handle this too
... they can't announce percentage of progress
... for the + syntax, we have to change it from an integer to a string
... how big of an issue is this? (for API's)

<clown> issue-397?

<trackbot> issue-397 -- Need a way to specify an indeterminate set size -- open

<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/397

richardschwerdtfeger: we can do, no question

Joseph: you can put it in the spec, but it's at risk if we can't find mappings

richardschwerdtfeger: the apple platform does not have a object mapping

Joseph: James Craig is not against it
... is it mapped to AXValue?
... oh it's AXARIASetSize

richardschwerdtfeger: to me this comes down to an API mapping issue

If an unknown set size exceeds a known and meaningful amount, authors may append a plus sign ("+") to the value to communicate the set size is at least that amount. Otherwise, authors should set the value of aria-setsize to -1.

richardschwerdtfeger: are people okay with this?

Joseph: Cynthia said the IE team likes the setsize="-1" notation
... that to me means that they don't like the + notation

http://www.w3.org/TR/wai-aria/roles#progressbar

<richardschwerdtfeger> https://rawgit.com/w3c/aria/action-1488/aria/aria.html#progressbar

If the progressbar is describing the loading progress of a particular region of a page, the author SHOULD use aria-describedby to point to the status, and set the aria-busy attribute to true on the region until it is finished loading

jamesn: prefer to say we have at least X

richardschwerdtfeger: does anyone object to setsize="-1"?
... the resolution is that we will specify an indeterminate by setting setsize to -1

RESOLUTION: we will specify an indeterminate by setting setsize to -1

<richardschwerdtfeger> RESOLUTION: We will not be supporting set sizes 999+ for sizes of at least that size for ARIA 1.1

action-1498

<trackbot> action-1498 -- Joanmarie Diggs to Draft "term" role (related to html dfn/dt elements) in addition to the 1.0 "definition" (equiv to html dd element) -- due 2015-07-30 -- PENDINGREVIEW

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1498

richardschwerdtfeger: using the term role on a link will mean that the link role is lost

<clown> https://rawgit.com/w3c/aria/action-1498/aria/aria.html#term

<richardschwerdtfeger> [ARIA 1.1] A word or phrase with a corresponding definition. See related definition.

<richardschwerdtfeger> The term role is used to explicitly identify a word or phrase for which a definition has been provided by the author or is expected to be provided by the user.

<richardschwerdtfeger> Authors SHOULD NOT use the term role on interactive elements such as links because doing so could prevent users of assistive technologies from interacting with those elements.

Joseph: the problem is, we had a definition role, but no term role

richardschwerdtfeger: we need a link to definition in the second paragraph

Joseph: the term is interactive, and if you click it the definition shows

janina: I like that clicking the term shows the definition

richardschwerdtfeger: what we're saying is that the definition must immediately follow the term

Joseph: it's like dt and dd
... do we need a definition list role?

<clown> https://rawgit.com/w3c/aria/action-1498/aria/aria.html#definition

jongund: why not use term and term-item?

Joseph: role definition should have a aria-labelledby on it

richardschwerdtfeger: then should it be in a list? You can have it anywhere

jongund: could I use an aria-label on the definition?

richardschwerdtfeger: you cannot because you don't have a visual link between the term and the definition
... the term and the definition don't have to be side by side

fesch: what happens when you have multiple definitions pointed to the same term

Joseph: that is allowed with dt's and dd's in the HTML spec

<richardschwerdtfeger> Proposal: we accept Joanie’s text for action 1498 with the caviat that the second paragraph link the definition text with the definition role and that in the definition role the aria-labelledy SHOULD point to an element with a role of “term”

MichielBijl: mgylling: what about aria-describedBy on the term?

Joseph: you could do that

richardschwerdtfeger: the author could screw up

mgylling: I wasn't saying we should have both, just wondering

<richardschwerdtfeger> RESOLUTION: Accept Rich’s proposal and jave Joanie make the changes and close out action 1498 and the associated issue

action-1657

<trackbot> action-1657 -- Joanmarie Diggs to Make aria-level a required attribute for role=“heading” -- due 2015-06-25 -- PENDINGREVIEW

<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1657

<clown> https://rawgit.com/w3c/aria/master/aria/aria.html#heading

Where it states: "If headings are organized into a logical outline, the aria-level attribute can be used to indicate the nesting level."

richardschwerdtfeger: all we have to do, is make it a required attribute

Joseph: it's required in the characteristics table

<clown> https://rawgit.com/w3c/aria/master/aria/aria.html#heading

A heading for a section of the page.

Often, heading elements will be referenced with the aria-labelledby attribute of the section for which they serve as a heading. If headings are organized into a logical outline, the aria-level attribute is used to indicate the nesting level.

richardschwerdtfeger: resolution: accept joanie's text

RESOLUTION: accept joanie's text, and close the action and issue

issue-736

<trackbot> issue-736 -- aria-label overriding listitem and treeitem content when need to know content is valuable -- open

<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/736

richardschwerdtfeger: in other words, the content is important

<clown> https://lists.w3.org/Archives/Public/public-pfwg/2015Jul/0097.html

richardschwerdtfeger: how do you think we should approach this one?

<clown> http://clown.idrc.ocad.ca/Fluid/aria/ListItemsWithLabels.html

Joseph: I setup a use case, with and without an aria-label
... if you put a label on it, the name is changed, but the content is still available in AI2 and ??

richardschwerdtfeger: by putting an aria-label on it should the content be ignored?

Joseph: take a look at my usecase

richardschwerdtfeger: what if the additional content inside the list item is important?

Joseph: it's still available

richardschwerdtfeger: but how do users know that they should drill into the list item?

bgaraventa1979: we have the same problem with aria grid, if you have a grid cell with a list with 200+ items; you don't want all that information to be announced
... there is no limit to how much content you can stuff into a list item

richardschwerdtfeger: would you want to know if the content inside has additional value?

Joseph: if you don't put the label on, the name is taken from the content

jamesn: that is in the spec
... if it's the same content, it's an author error; not our problem, right?

bgaraventa1979: if you put aria-label on it, AT hides the content

jamesn: that is how it's supposed to be

Joseph: the AT hides the content, it's available in the a11y API's

richardschwerdtfeger: I'd like people to think about this for a week

jamesn: a tweet is a good example, you want the author and tweet text to be announced at the list item level; you don't want the reply, etc buttons announced

richardschwerdtfeger: let's take this up next week

Still says -DRAFT-

Protocols and Formats Working Group Teleconference

30 Jul 2015

Attendees

Contents

Heads Up Items

Summary of Action Items

Scribe.perl diagnostic output