- From: Joanmarie Diggs <jdiggs@igalia.com>
- Date: Tue, 23 Sep 2014 16:19:10 -0400
- To: "wai-xtech@w3.org" <wai-xtech@w3.org>
Link: http://www.w3.org/2014/09/23-aapi-minutes.html
Plain text follows:
[1]W3C
[1] http://www.w3.org/
- DRAFT -
Protocols and Formats Working Group Teleconference
23 Sep 2014
See also: [2]IRC log
[2] http://www.w3.org/2014/09/23-aapi-irc
Attendees
Present
Joseph_Scheuhammer, Joanmarie_Diggs, Bryan_Garaventa,
+1.416.848.aaaa, +1.416.848.aabb, David_Bolter
Regrets
Cynthia_Shelly
Chair
Joseph_Scheuhammer
Scribe
joanie
Contents
* [3]Topics
1. [4]News: Heart-beat working draft in time for TPAC.
2. [5]News: ACTION-1506 Stand-alone document for
accessible name computation.
3. [6]ACTION-1396: (Cynthia) MSAA+UIAExpress mapping of
grid role.
4. [7]ACTION-1311: (Joseph) Clarify that presentation
role should preserve separation of nodes within
tables/lists.
5. [8]ACTION-1372: (Cynthia) Create a test case, if there
is none, to test the rowgroup role.
6. [9]ACTION-1373: (Cynthia) Create a test case for when
aria-posinset and aria-setsize are provided explicitly
on some but not all elements.
7. [10]ACTION-1409 : (Joanie) Progress on ATK API for
position, level, etc.
8. [11]ACTION-1445: (Cynthia) Investigate the ideal uai
mapping for role presentation in tables and lists
9. [12]Processing
issues/actions:https://www.w3.org/WAI/PF/Group/track/p
roducts/23
* [13]Summary of Action Items
__________________________________________________________
<trackbot> Date: 23 September 2014
<clown> agenda: this
scribnick: joanie
<scribe> scribenick: joanie
News: Heart-beat working draft in time for TPAC.
JS: At last week's editors' meeting, we decided to try for a
heart-beat draft in time for TPAC (31 October?).
... There is a set of edits I'd like to get done for that for
the core mapping guide.
... I'd ask others to look through the issues and actions to
see if there's anything they'd like to include and raise the
profile so that it gets done.
... The deadline is two weeks before TPAC to get it through the
W3C publishing process.
... So I'm going to say, give me what you want done by 10
October.
DB: I want to ping Marco and Alex about this. Do you want to
send out an email?
JS: I could, but my reluctance is that there are people on the
list who are not in this group. They might provide ARIA spec
suggestions.
... Ok, I'll be explicit about what we are looking for (mapping
guide) and send an email.
DB: (Asks about others)
JS: The group is also interested in ARIA, HTML, and SVG spec
stuff as well.
DB: I'm concerned about this going to far before implementation
feedback.
JS: This is a working draft; it provides the direction we're
heading.
... SVG is going to be brand new. It will provide what the
accessibility mappings are for certain SVG elements.
... People can then react.
... This doesn't involve testing.
DB: There's an investment of people working on it.
JS: (Looks for document)
<clown>
[14]http://rawgit.com/w3c/aria/master/svg/svg-implementation.ht
ml
[14] http://rawgit.com/w3c/aria/master/svg/svg-implementation.html
DB: This is fairly large already.
JS: He started with the core mapping guide and started removing
stuff.
... But I don't think he's removed enough.
<clown>
[15]http://rawgit.com/w3c/aria/master/svg/svg-implementation.ht
ml#mapping_role_table
[15]
http://rawgit.com/w3c/aria/master/svg/svg-implementation.html#mapping_role_table
DB: The above is probably the more interesting part.
(Discussion about mapping of div)
JS: He was looking for the most generic role you can give it.
... Almost all are none or group.
... We put name and description in its own document. But he
still has name and description in here.
... Nothing special about group position in SVG.
DB: Yet.
JS: The SVG documentation tells you how to override core.
... There's something similar for HTML.
News: ACTION-1506 Stand-alone document for accessible name
computation.
<clown>
[16]http://rawgit.com/w3c/aria/master/accessiblename/accName.ht
ml
[16] http://rawgit.com/w3c/aria/master/accessiblename/accName.html
JS: I have made at least 10 drafts or so of this. Above is the
latest.
... It currently takes the name and description computation out
of the core mapping guide and sticking that stuff in its own
document.
<clown>
[17]http://rawgit.com/w3c/aria/master/accessiblename/accName.ht
ml#mapping_additional_nd_te
[17]
http://rawgit.com/w3c/aria/master/accessiblename/accName.html#mapping_additional_nd_te
JS: I have cleaned up the actual algorithm which is above.
... Wherever possible I substituted a role or Accessibility API
term instead of the HTML term. Or added the SVG term.
<clown>
[18]http://rawgit.com/w3c/aria/master/accessiblename/accName.ht
ml#nameMappingTable
[18]
http://rawgit.com/w3c/aria/master/accessiblename/accName.html#nameMappingTable
JS: I'd like to get your opinion on the above mapping table.
... I have more or less finished the accessible name row; I've
not finished the accessible description role.
DB: It would be good for Alex Surkov to review this document.
JS: I want to further improve this text to make it more
algorithmic and less "chatty."
DB: Has James commented on it yet?
JS: When we talked about this two weeks ago (?), he said "I
have every confidence in you."
... I've made James an editor of this document. Rich and Steve
Faulkner will also be editors.
... Rich for SVG and Steve for HTML.
DB: I feel like Alex Surkov has contributed to this, but I
guess an editor is something different.
JS: (Explains the nature of being an editor)
... Cynthia is not here, so....
ACTION-1396: (Cynthia) MSAA+UIAExpress mapping of grid role.
JS: Ditto
<clown> action-1492?
<trackbot> action-1492 -- David Bolter to Investigate issue 671
and propose a solution that is clear for widgets -- due
2014-08-05 -- OPEN
<trackbot>
[19]https://www.w3.org/WAI/PF/Group/track/actions/1492
[19] https://www.w3.org/WAI/PF/Group/track/actions/1492
DB: Not yet done. Other issues came up unexpectedly.
JS: Moving due date back another week.
DB: Sure.
ACTION-1311: (Joseph) Clarify that presentation role should preserve
separation of nodes within tables/lists.
<clown> action-1311?
<trackbot> action-1311 -- Joseph Scheuhammer to Clarify that
presentation role should preserve separation of nodes within
tables/lists; work out in UAIG 1.1 and with
implementors/vendors how to resolve this. -- due 2014-08-26 --
OPEN
<trackbot>
[20]https://www.w3.org/WAI/PF/Group/track/actions/1311
[20] https://www.w3.org/WAI/PF/Group/track/actions/1311
JS: This is the one I need advice on from everyone.
<clown> <table role="presentation" …>
JS: At one point, Firefox changed how they handle tables with
role="presentation" so that all the text in the table became
one object in the accessibility tree.
... Matt King raised a concern about the huge text blob.
<clown> IA2_ROLE_TEXTFRAME
JS: Firefox changed so that each td element became a textframe.
... We need to clarify this in the mapping guide.
... When the presentation role is applied to tables or lists,
we need to separate the elements to avoid a single text blob.
... I'm struggling with the wording.
DB: The best way for me to help is write an email.
JS: I'd like to include this in time for the heartbeat release
prior to TPAC.
JD: Is there draft text already that you are asking us to
review?
JS: No. I've given up. I am asking the implementors to provide
the clarification/text.
<clown> I'm looking to the experts to propose the text.
[21]https://bugzilla.mozilla.org/show_bug.cgi?id=1013584
[21] https://bugzilla.mozilla.org/show_bug.cgi?id=1013584
JD: Looking at the above bug, 1) Div should be used (i.e.
ATK_ROLE_SECTION) not text. This is explained in the
conversation I had in that bug with Alex Surkov.
... Beyond that, in my opening report I state:
"Expected results: The child elements would each be exposed as
if they were div elements."
JD: How much (more) clarification do you need, Joseph?
JS: And the next question is where does that text go?
... I guess the mapping of the presentation role? Let me
see....
<clown>
[22]http://rawgit.com/w3c/aria/master/implementation/aria-imple
mentation.html#role-map-presentation
[22]
http://rawgit.com/w3c/aria/master/implementation/aria-implementation.html#role-map-presentation
JS: No. I don't think it works there. Looking at the above.
<clown> <img role="presentation" ..>
<clown> <img alt="" ..>
JD: I think that table you point to IS the appropriate place to
put this text. However:
... I do not think that ROLE_PANEL is the right mapping. A
panel tends to be a thing (yes, generic) which contains
widgets.
... role="presentation" is not a thing that contains widgets.
Therefore, treating all things role="presentation" as if they
were div elements would be perfect
... On ATK/AT-SPI2, div is mapped to ROLE_SECTION.
<clown> issue-677
<trackbot> issue-677 -- ATK/AT-SPI mapping of
role="presentation" to ROLE_PANEL -- open
<trackbot> [23]https://www.w3.org/WAI/PF/Group/track/issues/677
[23] https://www.w3.org/WAI/PF/Group/track/issues/677
JS: Ok, I just made the above issue to address the mapping
change.
<davidb> DB: (via IRC) I'd want to keep IA2 close to ATK so if
we change this we should also consider IA2_ROLE_SECTION
<clown> "Expected results: The child elements would each be
exposed as separate accessible objects."
<davidb> JD:"The child elements would each be exposed as
separate accessible objects".... of type blah blah over here
JD: To clarify, there should be a pointer to the
potentially-platform-dependent role.
<clown> "In the case of tables and lists, the child elements
would each be exposed as separate accessible objects as per the
different AAPI role mappings"
JD: i.e. if ATK wants ROLE_SECTION and IA2 ROLE_TEXTFRAME
JS: The above language will be added to the role="presentation"
mapping.
DB: The case we're talking about is only if there's text in the
table, right?
<clown> <table role="presentation">… <td><p>Some text <img
src=".."></p></td> …</table>
<davidb> JD: if there was other stuff in the TD it can still be
in the div/section (DB is trying to paraphrase here)
<davidb> DB: ah ok.
<clown> becomes: "<div><p>Some text <img src=".."></p></div>
<davidb> TD -> DIV
ACTION-1372: (Cynthia) Create a test case, if there is none, to test
the rowgroup role.
JD: Cynthia is not here.
ACTION-1373: (Cynthia) Create a test case for when aria-posinset and
aria-setsize are provided explicitly on some but not all elements.
JD: Cynthia is not here.
ACTION-1409 : (Joanie) Progress on ATK API for position, level, etc.
<clown> action-1409?
<trackbot> action-1409 -- Joanmarie Diggs to Monitor progress
of bug against atk in gnome's bugzilla requesting new position,
level, and other api be added
([24]https://bugzilla.gnome.org/show_bug.cgi?id=727453) -- due
2014-08-05 -- OPEN
[24] https://bugzilla.gnome.org/show_bug.cgi?id=727453)
<trackbot>
[25]https://www.w3.org/WAI/PF/Group/track/actions/1409
[25] https://www.w3.org/WAI/PF/Group/track/actions/1409
JD: This is on hold for at least a month or so because GNOME
3.14 was just released. This new API will need to go into
3.15/3.16.
... I will likely do an initial implementation of the API
myself and sanity check it via Gtk and/or WebKitGtk to be sure
that I didn't get it wrong.
... And if it seems sane, try to get the ATK and AT-SPI2
maintainers to accept it.
ACTION-1445: (Cynthia) Investigate the ideal uai mapping for role
presentation in tables and lists
JD: Cynthia is not here.
Processing
issues/actions:[26]https://www.w3.org/WAI/PF/Group/track/products/23
[26] https://www.w3.org/WAI/PF/Group/track/products/23
JS: End of meeting.
Summary of Action Items
[End of minutes]
__________________________________________________________
Minutes formatted by David Booth's [27]scribe.perl version
1.138 ([28]CVS log)
$Date: 2014-09-23 20:08:43 $
__________________________________________________________
[27] http://dev.w3.org/cvsweb/~checkout~/2002/scribe/scribedoc.htm
[28] http://dev.w3.org/cvsweb/2002/scribe/
Received on Tuesday, 23 September 2014 20:19:57 UTC