<scribe> scribe: carmacleod

https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_search-3Fl-3D-26q-3Dis-253Aopen-2Bis-253Aissue-2Brepo-253Aw3c-252Faria-2Bcreated-253A-253E-253D2020-2D08-2D13-2Brepo-253Aw3c-252Faria-2Brepo-253Aw3c-252Faccname-2Brepo-253Aw3c-252Fcore-2Daam-26type-3DIssues&d=DwMGaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=7pxjQj6Sn7GadcQXUI8qdHzQ7owon0mR-h863iCnxZc&m=RBPG9815tQzLbjxo_vq8dGtooRnelY510P48vLSOA8A&s=1CckyZSJKRxEnCygTE4LxslQpAHW0Ir4b4RPz0u5nwE&e=

<jamesn> https://github.com/w3c/accname/issues/89

<jamesn> https://github.com/search?l=&q=is%3Aopen+is%3Apr+repo%3Aw3c%2Faria+created%3A%3E%3D2020-08-13+repo%3Aw3c%2Faria+repo%3Aw3c%2Faccname+repo%3Aw3c%2Fcore-aam&type=Issues

<jcraig> https://github.com/w3c/aria/pull/1312

<jcraig> https://github.com/w3c/accname/pull/88

Meaty topic for next week

jamesn: Bryan had something he wanted to talk about re white space in Accname
... do we need a deep dive right away, or should we gather opinions in an issue, and have a deep dive if necessary

Matt_King: can you give an example of where whitespace specs differ

<jamesn> https://github.com/w3c/accname/issues/55

bryan: it boils down to the recursion, so I need to understand (at least today) what exactly "whitespace" means

jamesn: There are 5 different definitions for whitespace in 5 different specs, and we need to pick one of them

Matt_King: is there one of these definitions that accname seems to imply?

bryan: not sure. there was an issue where newline characters broke up names

Matt_King: need to get data from browsers - are there any places where browsers break and need to be changed?

jamesn: perhaps there isn't much difference in the end. html spec includes newline, but css definition doesn't

bryan: includes weird things like non-breaking space characters
... I will go through the various specs and write down the differences in the issue
... I just need to understand what it is, so that I can make sure the algorithm is ok

jamesn: I don't think we need a deep dive for this (yet). We need a topic

carmacleod: Trees?

Matt_King: I can be available in 2 weeks

jamesn: Trees again in 2 weeks

[Why is a menuitem in a group special, but only on OS X?](https://github.com/w3c/core-aam/issues/78)

https://github.com/w3c/core-aam/issues/78

aaronlev: we were fixing another bug where if you have a button with aria-haspopup, then Chrome gives role menubutton
... correction: for menuitem that has a parent of group gets role menubutton
... only difference was on mac
... didn't make sense because menubutton always opens menu - why does a menuitem in group open menu?

jcraig: looked into this - nobody can remember why, but it's probably because menuitems in menubar are menubuttons
... discussion is scheduled for tomorrow. If nobody objects, I've recommended that this change to something that makes more sense moving forward

Matt_King: how is menuitem with aria-haspopup different in a menubar different from menuitem with aria-haspopup in a menu?

jcraig: menubars are always exposed, so multiple buttons are available on the screen at once. In menu, only one is exposed

Matt_King: I don't think it's a problem, but it's a weird inconsistency in the sense that a menubar is basically just a menu, so it supports all of the keyboard affordances that a menu does (i.e. pressing first letter goes to item, etc)

jcraig: maybe the difference is that on Windows, there are many menubars, and on Mac, there's only the one

aaronlev: there's no rule in webkit that I can see that says that a menubutton in a menubar should be exposed as menubuttons
... I saw them being exposed as menuitem
... are we getting rid of menubutton completely? or are we changing how it works?

Matt_King: if a menuitem is in a group, it should be a menuitem (and not a menubutton)

aaronlev: we all agree on that

jcraig: we agree on first issue that menuitem in group is not menubutton. Need to open a separate issue for thinking about whether menuitem in menubar is menubutton.

aaronlev: some part of confusion is that MSAA on windows has separate role for menubutton

Matt_King: looking at editor menubar example in apg with safari/VO, and for each of the menuitems in the menubar, VO says collapsed button
... also, menuitem in group is announced as collapsed button (regular button, not menubutton)

aaronlev: need to investigate
... want to point out that sometimes menuitem in menubar doesn't have a popup - it's just a button

jcraig: opening issue now

1.3 Triage

jamesn: this is all open 1.3 issues that are not assigned to a project (sorted by oldest first)

<jamesn> https://github.com/w3c/aria/issues/350

<jcraig> #78 menubutton follow-on issue filed as #80 https://github.com/w3c/core-aam/issues/80

carmacleod: Landmarks - user agents SHOULD provide landmark navigation

jamesn: make a PR and we'll see
... back to 1.3 triage
... need a 4-way slider

Matt_King: 1.4?
... sounds like more work for browsers than spec

jamesn: used for color picker

Matt_King: really only 2 dimensions, so only 2 values (even though called a 4-way slider)

jamesn: both directions have different maxes and mins

<jcraig> https://developer.mozilla.org/en-US/docs/Web/HTML/Element/input/color

jamesn: currently just use aria-valuetext with aria-roledescription - kind of a hack

jcraig: would be nice to have other examples where this would be used - not just color picker
... If you don't specify a value, the default is #000000, which is black. The value must be in seven-character hexadecimal notation, meaning the "#" character followed by two digits each representing red, green, and blue, like this: #rrggbb. If you have colors that are in any other format (such as CSS color names or CSS color functions such as rgb() or rgba()), you'll have to convert them to hexadecimal before setting the value.

Matt_King: my gut still says this is a 1.4 issue

jamesn: already moved it

<jcraig> https://github.com/w3c/aria/issues/494

jamesn: everyone please go through the 1.3 issues and if you see any that can be closed or moved to 1.4, please do!

This is scribe.perl Revision of Date 
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/MarkMccarthy:/Matt_King:/
Succeeded: s/https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_search-3Fl-3D-26q-3Dis-253Aopen-2Bis-253Aissue-2Brepo-253Aw3c-252Faria-2Bcreated-253A-253E-253D2020-2D08-2D13-2Brepo-253Aw3c-252Faria-2Brepo-253Aw3c-252Faccname-2Brepo-253Aw3c-252Fcore-2Daam-26type-3DIssues&d=DwMGaQ&c=jf_iaSHvJObTbx-siA1ZOg&r=7pxjQj6Sn7GadcQXUI8qdHzQ7owon0mR-h863iCnxZc&m=RBPG9815tQzLbjxo_vq8dGtooRnelY510P48vLSOA8A&s=1CckyZSJKRxEnCygTE4LxslQpAHW0Ir4b4RPz0u5nwE&e=//
Present: carmacleod Joanmarie_Diggs Irfan Greta MarkMccarthy harris Matt_King jamesn jcraig aaronlev bryan
Regrets: curt_bellew Scott_O Peter_K
Found Scribe: carmacleod
Inferring ScribeNick: carmacleod

WARNING: No date found!  Assuming today.  (Hint: Specify
the W3C IRC log URL, and the date will be determined from that.)
Or specify the date like this:
<dbooth> Date: 12 Sep 2002

People with action items: 

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.)