See also: IRC log
<trackbot> Date: 10 February 2015
<clown> agenda: this
<clown> agneda+ ACTION-1444: (Joseph) Update ATK/AT-SPI2 mapping of role presentation in tables and lists.
<scribe> scribe: joanie
<clown> action-980?
<trackbot> action-980 -- David Bolter to Define mappings for managed aria related states: aria-setsize, aria-posinset, aria-level, focused, focusable with reference to section 5.5 bullet 1 of the UAIG. -- due 2015-01-27 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/980
<davidb> alex is taking all of mine
<davidb> i suspect he'll answer before next week
The scribe is having a hard time hearing.
<davidb> DB: most of these are ia2 mapping things
JS: Looking at a note in the
action, the next step is to come up with the list of managed
states.
... This is why I think it's broader than IA2.
... Mappings are already in there.
... What we need is text that says what managed states are, why
they are necessary....
... Because the current text in 5.5 bullet 1 is too vague, as I
understand it.
<clown> http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#statePropertyMappingGeneralRules
JS: It's near to the above.
... (Reads from the text in the document)
... "Etc.?"
... In other words, what are these states.
DB: Alex is going to take this on.
JS: My only concern is that this
is not just MSAA or IA2; it's for all the platforms.
... So there has to be agreement amongst the browser
vendors.
... I'm not sure you can do that with the action, but you
should come and talk to us.
RS: I think it's probably going
to be closer to the first half of 2016.
... The group has made a lot of progress in the last month.
<clown> issue-612?
<trackbot> issue-612 -- Review ia2/atk rule in group position. should this really determine level based on aria-owns chain. see uaig: http://www.w3.org/wai/pf/aria-implementation/#mapping_additional_position and test case 69: https://www.w3.org/wai/pf/testharness/testresults?tes -- open
<trackbot> https://www.w3.org/WAI/PF/Group/track/issues/612
<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 (https://bugzilla.gnome.org/show_bug.cgi?id=727453) -- due 2015-04-07 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1409
JD: Life being life, this is not
going to make it into GNOME 3.16. Happily, release cycles are
six-months long. I'll tackle this next cycle.
... It's now due 7 April.
JS: Cynthia is not here.
JS: Cynthia is not here.
<clown> action-1546?
<trackbot> action-1546 -- David Bolter to Make sure the rowgroup mapping for msaa/ia2 is accurate -- due 2014-12-23 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1546
JS: We need mappings for the
rowgroup role for MSAA/IA2.
... David was going to look at this.
<clown> http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-rowgroup
JS: We have mappings for ATK and
AX API.
... We don't have UIA done either.
... ATK and AX API both say "not mapped."
DB: This is one of my actions
which Alex is also going to do. Nothing else to report.
... Alex is the IA2 maintainer.
JS: It currently says ROLE_SYSTEM_GROUP
RS: We have a discussion on the ARIA call to follow up on the container items.
JS: Yes, we're talking about
rowgroup, but the current mapping is ROLE_SYSTEM_GROUP for
IA2.
... That doesn't strike me as correct, but I'm not an IA2
expert.
RS: That is what passed when we did our tests, isn't it?
JS: I have a vague feeling that it didn't pass and that we then revised the test.
RS: Right because the rowgroup
wasn't exposed and now it is.
... If I remember the discussion.... David, why did you map
this?
DB: I don't remember.
RS: So we're all wondering if it
needs to stay this way.
... I know OS X doesn't map it. What does IE do?
<clown> http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-rowgroup
JS: Currently the mapping table says "group"
RS: Did that pass?
JS: I don't know.
RS: (Reads from mapping)
<clown> https://www.w3.org/WAI/PF/testharness/testcases/edit?testsuite_id=1&testcase_id=56
CS: It depends how it was
tested.
... I don't think we do anything with this.
<clown> davidb, it's "not mapped"
CS: Grids are somewhat messy. I don't have an answer right now.
JS: I found a test case. Expected
result is generally has no role.
... That's test case 56.
<clown> https://www.w3.org/WAI/PF/testharness/testresults?testsuite_id=1&testcase_id=56
JS: Some passes, some failures
(in the results)
... Test report link is above.
... (Reads results)
<clown> <div role="rowgroup>
CS: The test file uses a div?
JS: Yes.
CS: It's entirely likely that IE
is ignoring that role.
... Because "group" is the role of div.
JS: James Craig passed it on Safari with AXGroup.
(Group concludes we need to revisit this test and its results)
RS: I think we're at the point
that we need to decide what we want to do.
... If it's not going to provide anything of value, it
shouldn't map.
... We can all fix it. We can pull it out of Firefox.
Or....
... It's confusing. If you put a section role in the middle of
a grid, it's goofy.
... The problem happens if you put a label on it, though I
don't know why you would do that.
CS: It almost feels like a
property. I know it's a role, but it shows membership.
... UIA does a lot with properties, rather than roles.
JS: It corresponds to elements like thead and tbody.
<clown> http://rawgit.com/w3c/aria/master/html-aam/html-aam.html#el-tbody
JS: These map to group. See above
link.
... ATK doesn't have an accessible though. That's consistent
with the core.
... MSAA/IA2 also doesn't have a mapping, for tbody.
RS: I don't see why we need a mapping.
CS: So you're saying we shouldn't
have an accessible for this? And what would that do for tables
that have them?
... I know there are some people in certain
accessibility-related circles who find these useful.
... So we might wish to get wider feedback on this.
RS: If you tell them they aren't mapped, will they know what that means?
CS: Uncertain.
... (Reiterates need to get wider feedback)
RS: I see table being used less
and less because it doesn't tend to flow well.
... If you want to run this by WCAG we certainly can.
CS: Michael is in WCAG.
JS: Is it worth bringing up at the full ARIA meeting?
RS: We could.
JS: We can add it to the overall "group" issue and discussion.
RS: Give me an issue related to not mapping rowgroup, tbody, and thead.
<clown> issue: Discuss rowgroup, tbody, and thead in AAPIs
<trackbot> Created ISSUE-700 - Discuss rowgroup, tbody, and thead in aapis. Please complete additional details at <https://www.w3.org/WAI/PF/Group/track/issues/700/edit>.
DB: You want it taken out of the tree?
JS: Yes.
RS: If somone puts a tabindex on it, what do we want to call it?
CS: If you put that on the object
that's already in the tree, what happens?
... If you put role="rowgroup" on something, does it act like
presentation, or is it ignored?
RS: What would that hurt if it's taken out of the tree?
CS: It's weird if people think they're adding semantics (via the role), but we're taking it out of the tree/ignoring it.
RS: And we'd fallback on native host language semantics.
CS: It will still be there in the DOM and object attributes.
<clown> https://www.w3.org/WAI/PF/Group/track/issues/700
RS: Thanks for the issue, please also give me an action.
<clown> ACTION: rich to discuss issue-700 in ARIA task force meeting. [recorded in http://www.w3.org/2015/02/10-aapi-minutes.html#action01]
<trackbot> Created ACTION-1578 - Discuss issue-700 in aria task force meeting. [on Richard Schwerdtfeger - due 2015-02-17].
<clown> action-1578
<trackbot> action-1578 -- Richard Schwerdtfeger to Discuss issue-700 in aria task force meeting. -- due 2015-02-17 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1578
action-1309
<trackbot> action-1309 -- Cynthia Shelly to Determine the best place to map aria-colindex and aria-rowindex for msaa+uiexpress and uia. -- due 2015-02-05 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1309
<clown> action-1309?
<trackbot> action-1309 -- Cynthia Shelly to Determine the best place to map aria-colindex and aria-rowindex for msaa+uiexpress and uia. -- due 2015-02-05 -- OPEN
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1309
JS: This is a hot topic on the
list, where to put aria-colindex and aria-rowindex in the
markup.
... Some people say on the individual cells, some on the
rows.
... I think what this action really is, is asking what the
mapping would be.
... But maybe we're premature
JD: I know what the mapping is
for ATK/AT-SPI2 regardless of where it belongs in the
spec.
... But I don't know if this is true for others.
JS: I was hesitant because of all the discussion on the list about the ARIA property.
CS: This is kind of related to
the last one (grids and tables).
... We do have a table pattern now, so I need to look and see
if they make sense in there.
... This one we have to move out further. After CSUN.
JS: I'll put it for the 10th of March.
<clown> action-1444?
<trackbot> action-1444 -- Joseph Scheuhammer to Update atk/at-spi2 mapping for role presentation in tables and lists. -- due 2015-02-10 -- PENDINGREVIEW
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1444
<clown> action-1445?
<trackbot> action-1445 -- Cynthia Shelly to Investigate the ideal uai mapping for role presentation in tables and lists. -- due 2015-01-20 -- PENDINGREVIEW
<trackbot> https://www.w3.org/WAI/PF/Group/track/actions/1445
JS: I did some related work on this.
<clown> http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-presentation
JS: The above URL has what I
did.
... For UIA, I put "expose using the text pattern"
<clown> If the object is in the accessibility tree, expose using the text pattern. In particular, this applies to cells within a table or grid with role="presentation", and to list items within a list with role="presentation".
JS: I added a sentence about the
table and list.
... So the grid cells and list items become objects that expose
the text pattern.
CS: If they are in the
tree.
... That's confusing, because presentation takes them out of
the tree.
JS: It takes the grid out, but not the cell, right?
<clown> If the object is in the accessibility tree, expose as ROLE_SECTION. In particular, this applies to cells within a table or grid with role="presentation", and to list items within a list with role="presentation".
CS: OK, I think this is right,
though it is kind of confusing.
... It's not the wording that is confusing; it's the
concept.
JS: (Reads the text associated
with ATK)
... All the cells become ROLE_SECTION
CS: I don't think we do that
(treat them like divs, so they're all separate)
... What do you think is better?
JD: We did this deliberately for
ATK.
... The idea is that they are separate, distinct text objects
that should not be run together by the screen reader.
CS: I don't have strong feelings either way.
JS: Before they were a giant blob of text and people complained.
RS: We heard the same.
CS: Ok, then make them groups.
<clown> If the object is in the accessibility tree, expose using the GROUP role, using the text pattern.
(Group discussed above proposed text)
<clown> <table role="presentation"> …
JS: If you have the above, what happens to the table itself?
<clown> https://www.w3.org/WAI/PF/testharness/testcases?testsuite_id=1
RS: I think we have a test for this in the test suite.
CS: I'm not seeing anything
that's specifically a table.
... (Makes a quick test)
JS: We have close to 900 tests, but not a table with role="presentation" on it.
CS: I created a file. It looks like IE is removing it altogether, but not its children.
JS: Can you mail that test file to us.
CS: Sure.
<clown> http://rawgit.com/w3c/aria/master/core-aam/core-aam.html#role-map-presentation
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/(Alex)// Succeeded: s/both UIA/UIA/ Succeeded: s/not one/not a table/ Found Scribe: joanie Inferring ScribeNick: joanie Default Present: Joseph_Scheuhammer, Joanmarie_Diggs, David_Bolter, Rich_Schwerdtfeger, [Microsoft] Present: Cynthia_Shelly David_Bolter Joanmarie_Diggs Joseph_Scheuhammer Rich_Schwerdtfeger Found Date: 10 Feb 2015 Guessing minutes URL: http://www.w3.org/2015/02/10-aapi-minutes.html People with action items: rich WARNING: Input appears to use implicit continuation lines. You may need the "-implicitContinuations" option.[End of scribe.perl diagnostic output]