– DRAFT –
ARIA WG

<jamesn> https://w3c.github.io/aria/#enumeration-example

[New Issue Triage](https://bit.ly/3BGPi8N)

skipping the Author test case issues... agenda topic later

[New PR Triage](https://bit.ly/2VhW9EH)

skipping the Author test case PRs...

[Meaty topic for Aug 19](https://bit.ly/3BLXoga)

no deep dive fo the next 2 weeks

topic needed from August 19?

Bryan: aria-hidden?

jamesn: only if progress made on the proposals

Bryan: I plan to have spec text proposal before then

jcraig: Group can commit to a date for deep dive for HIDDEN once the spec text proposal is in. Need to give implementors enough time to review and discuss internally before the meeting.

StefanS: Issue 559 is in the list of potential deep dive topics

https://github.com/w3c/aria/issues/559

another is the grid role issue https://github.com/w3c/aria/issues/1324

StefanS: I feel no consensus within group what grid is limited to... leaving open the problem authoring practice for layout grids... IMO should limit to interactive "data" grid. discourage layout grid.

jamesn: we punted 559 (SR "call-to-action" issue) to 1.4... August 19th or 26th seem reasonable for either.

StefanS: grid #1324 is higher priority for me

jamesn: August 19th: hidden #57

jamesn: August 26th grid #1324

<jamesn> sgenda?

WG Member availability for meetings Aug 5 & Aug 12

<jaunita_george> 12th

<harris> don't care

Jaunita: I can offer to chair

jamesn: Michael?

MichaelC: I'm off that week.

jamesn: I think we'd need a staff contact (MichaelC) or existing Co-Chair (JamesN or Joanie) to Chair the meeting

joanie: 12th it is

no meeting next week

Authoring test updates

<Jemma> I am interested in the session too. please count me in.

<spectranaut> I'm also interested!

<jamesn> https://github.com/w3c/aria/labels/Author%20Tests

Authoring test updates

jamesn: these are all the issues cyns logged for "authors MUST" tests

existing PRs for 9 of the issues

<jamesn> https://github.com/w3c/aria/pull/1582/commits/784f43cf5b3cf06f5c120c7e46239ab88c97100d

there are 7 that have no PR or assignee... Can we triage and assign today? Any volunteers?

Here's an example to show simplicity

Valerie: Curious to help with test writing, but know very little about it.

Harris: me too

jamesn: Joanie can help

joanie: I will share a link to the how-to write-up

harris: I'll wrap up several of these

<WIlliamT_NFCU> Which ones are high priority?

[side discussion about gh-pages scripting issue]

<Jemma> I was able to access wiki last week.

joanie: I can provide a link to the Test Writing how-to write-up, but the script is broken, so it is inncomplete

MichaelC: I will work to fix that script now.

<joanie> https://github.com/w3c/aria/tree/main/validator-tests

joanie: This link brings you to the ReadMe, but several of the links will break.... that should be resolved soon.

MichaelC: expect to be resolved today, perhaps within a few minutes.

WIlliamT_NFCU: What's the assignment process for test case authoring?

jamesn: assign the GH issue to yourself.

jamesn: William is a member of the W3C org, but not the ARIA team

<jaunita_george> jaunitageorge1

<WIlliamT_NFCU> WilliamTennisNFCU github id

please share GH username here in IRC to be added to the team

Jaunita is not an org member yet

<jaunita_george> jaunita-george

Any of these https://github.com/search?l=&q=is:open+is:issue+repo:w3c/aria+created:%3E%3D2021-07-22+repo:w3c/aria+repo:w3c/accname+repo:w3c/core-aam&type=Issues

take assignment of any without an assignee or existing PR

[When is hidden content taken into calculation of name and description?](https://github.com/w3c/accname/issues/57)

deep dive on this topic August 19th. Bryan's PR expected by next week.

https://w3c.github.io/aria/#enumeration-example

joannie: will give background info with example url.

<joanie> el.ariaBusy = "true";

<joanie> el.ariaBusy; // "true"

<joanie> el.removeAttribute("aria-busy");

<joanie> el.ariaBusy; // "false"

<joanie> el.setAttribute("aria-busy", "busy");

<joanie> el.ariaBusy; // "false"

<jamesn> https://w3c.github.io/aria/#example-15

<Zakim> jcraig, you wanted to mention the normative test in §6.3

joanne: I am wondering whether we miss normative text in spec or
… user agents...

<jcraig> https://w3c.github.io/aria/#enumerated-attribute-values

<jcraig> When the ARIA attribute definition includes a table enumerating the attribute's allowed values, that attribute is an enumerated attribute. Each value in the table is a keyword for the attribute, mapping to a state of the same name. When the values table notates one of the values as "(default)", that default value is the missing value default and invalid value default for the attribute.

jcraig: the spec is correct in that it is normative but we can adjust implementation and more.
… we can also make this normative be less normative by the editor..

we can see whether test case is out of compliance or not.

jamesn: we can make this more explicit

joannie: I am hearing that example 16 can reflect the value aria pressed true..

jcraig: https://w3c.github.io/aria/#example-16

talking about "assignemtn of invalid "foo"'

joannie: some aria property value is role specific - aria orientation
… user agent are also expected to evaluate this in regards to role?

jcraig: there is an issue was filed for this specific case.

joannie: third questions is
… my question to Jamesn what are the implication of this in terms of getting aria 1.2

jcraig: I could not find the issue regarding role speific default value yet.

joannie: default value of aria read only is false
… but false value can be changed by author.

then what value aria read only should return?
… if attribute is missing or attribute does not make sense, I think null value should be returned.

jcraig: you can set the value and return the value. but nothing has not done in terms of implementation.
… there is no behavior associated with returned value as james said.

cyn: ie. you added invalid value like checked for div, does it reflected?

jcraig: yes

cyn: that can be the same case for aria.

jcraig: that is the point I am getting at.

<cyns> I need to drop

joannie: how about validation aspect of that example?

jcraig: explaning the example for Jonnie's question

required value foo

jcraig: there is validation in some context from the example.

if we test it against div, with foo value, foo value will be returned.

jcraig: we can file the issue in the spec to make this assigned to me

and different user agents ..

Joanie: will file role dependent value to be reflectable

in user agent implementation

<Jemma> s/a validate/validation/

Action: Joanie to file Issue that normative text in 6.3 should be more explicit (e.g. more that one "User Agent MUST"… will be added). Follow-on may include new test case issues, and linked tracking issue in each User Agent bug tracking system (e.g. Bugzilla)

<trackbot> Created ACTION-2160 - File issue that normative text in 6.3 should be more explicit (e.g. more that one "user agent must"… will be added). follow-on may include new test case issues, and linked tracking issue in each user agent bug tracking system (e.g. bugzilla) [on Joanmarie Diggs - due 2021-08-05].

<joanie> https://github.com/w3c/aria/issues/1584

Action: Joanie to file issue that role-specific default (e.g. aria-orientation on scrollbar) should be addressed as reflectable in the IDL or in normative prose somehow

<trackbot> Created ACTION-2161 - File issue that role-specific default (e.g. aria-orientation on scrollbar) should be addressed as reflectable in the idl or in normative prose somehow [on Joanmarie Diggs - due 2021-08-05].