- From: Richard Schwerdtfeger <schwer@us.ibm.com>
- Date: Tue, 29 Sep 2015 15:27:23 -0500
- To: PF <public-pfwg@w3.org>
- Message-ID: <OFAEB86DEE.625E1CB8-ON86257ECF.00703953-86257ECF.00705EF7@us.ibm.com>
http://www.w3.org/2015/09/24-aria-minutes.html
IRC log of aria on 2015-09-24
Timestamps are in UTC.
16:32:09 [RRSAgent]
RRSAgent has joined #aria
16:32:09 [RRSAgent]
logging to http://www.w3.org/2015/09/24-aria-irc
16:32:11 [trackbot]
RRSAgent, make logs public
16:32:11 [Zakim]
Zakim has joined #aria
16:32:13 [trackbot]
Zakim, this will be WAI_PF
16:32:13 [Zakim]
I do not see a conference matching that name scheduled within the
next hour, trackbot
16:32:14 [trackbot]
Meeting: Protocols and Formats Working Group Teleconference
16:32:14 [trackbot]
Date: 24 September 2015
16:32:14 [richardschwerdtfeger]
chair: Rich
16:32:21 [richardschwerdtfeger]
W3C WAI-PF ARIA Caucus
16:32:31 [richardschwerdtfeger]
RRSAgent, make log public
16:33:47 [jongund]
jongund has joined #aria
16:34:26 [richardschwerdtfeger]
https://lists.w3.org/Archives/Public/public-pfwg/2015Sep/0189.html
16:35:24 [janina]
present+ janina
16:35:35 [MichaelC]
regrets: Joanmarie_Diggs, Cynthia_Shelly
16:35:37 [fesch]
present+ fesch
16:35:37 [Stefan]
Stefan has joined #aria
16:36:13 [Stefan]
Scribe: Stefan
16:37:13 [Stefan]
Topic:Matt's draft
16:37:17 [Stefan]
http://www.w3.org/2015/09/17-aria-minutes.html
16:37:25 [richardschwerdtfeger]
Topic: Matt’s draft
16:37:36 [mck]
mck has joined #aria
16:38:08 [Stefan]
The issue 633 branch visible at:
http://rawgit.com/w3c/aria/mck_issue633/aria/aria.html
16:38:23 [mck]
http://rawgit.com/w3c/aria/mck_issue633/aria/aria.html#grid
16:38:49 [Stefan]
Matt: there was feedback for grid role last week
16:39:03 [Stefan]
Matt: incorporated feedback in article feed role
16:39:57 [clown]
clown has joined #aria
16:40:05 [Stefan]
Topic: 1. Issue 633: listbox and tree may contain only static items;
badly need interactive widgets that can contain interactive typed
items (Matt King)
16:40:06 [clown]
present+ Joseph_Scheuhammer
16:40:17 [Stefan]
The issue 633 branch visible at:
http://rawgit.com/w3c/aria/mck_issue633/aria/aria.html
16:40:58 [richardschwerdtfeger]
http://rawgit.com/w3c/aria/mck_issue633/aria/aria.html#articlefeed
16:42:06 [clown]
http://rawgit.com/w3c/aria/mck_issue633/aria/aria.html#roles_categorization
16:42:40 [Stefan]
Matt: articlefeed role: I rewrote the description
16:44:35 [joanie]
regrets- Joanmarie_Diggs
16:44:40 [joanie]
present+ Joanmarie_Diggs
16:45:09 [Stefan]
Joseph: short sentence needed to alert the reader
16:45:53 [Stefan]
Matt: everything ok with this approach?
16:45:59 [Stefan]
+q
16:46:11 [Stefan]
Rich: keyboard interaction pattern?
16:46:31 [Stefan]
Matt: no, it's structural
16:47:16 [Stefan]
-q
16:47:47 [Stefan]
Rich: What about changing feeds with keyboard?
16:48:26 [clown]
s/short sentence needed to alert the reader/the first paragraph of
the role description should have one sentence. If it has a second, it
is to alert the reader to a related role/
16:48:30 [Stefan]
Matt: in APG we can make some fairly strong statements
16:48:51 [Stefan]
Rich: what about mobile devices?
16:51:18 [Stefan]
Stefan: best practices in APG should cover interactive articles being
part of articlefeed
16:51:41 [Stefan]
Stefan: .. with respect to keyboard support
16:52:27 [Stefan]
Matt: when peple are using screen readers they want to read the
content, also by structure
16:53:24 [Stefan]
Rich: what about aria-feedappendcount or aria-feedprependcount
16:53:55 [Stefan]
Matt: they are not changing constantly .. author marks feed busy
16:54:25 [Stefan]
Rich: but they are changing, do you expect AT to notice the changes?
16:54:39 [Stefan]
Matt: yes they should evaluate onChange event on busy ..
16:54:58 [jamesn]
jamesn has joined #aria
16:55:10 [Stefan]
Rich: Joseph ae we firing firing change events on busy by spec?
16:55:18 [Stefan]
Joseph: thinks so
16:55:30 [clown]
http://w3c.github.io/aria/core-aam/core-aam.html#event-aria-busy
16:57:02 [Stefan]
Matt: you could do this also for indefinitely scrolling grid
16:57:26 [Stefan]
Rich: why we need feed in name?
16:57:42 [Stefan]
Matt: tried to generalize that by this way
16:58:17 [jongund]
jongund has joined #aria
16:59:33 [Stefan]
Matt: this can be even further generalized for tables fetching fdata
16:59:50 [Stefan]
Stefan: we need proper signaling for data fetching to AT
17:00:58 [joanie]
q+
17:01:38 [richardschwerdtfeger]
ack joanie
17:03:01 [Stefan]
Joanie: we have this property, no matter if it maps or not
17:04:13 [Stefan]
Matt: if you have list in acctree with n children, and property how
many children are added and removed this will be useful tied to a
state change
17:05:53 [Stefan]
Rich: feeds are a bottomless pit .. I don't see any value
17:07:37 [Stefan]
Matt: true when adding bstaff to the end, but not for the beginning
-> useful or AT to know the offset
17:07:50 [Stefan]
Rich: I can see the point
17:08:23 [Stefan]
Matt: don't limit to particular use case
17:08:37 [Stefan]
Matt: there are other use cases
17:09:30 [Stefan]
Matt: feeds are not constantly changing normally ...
17:10:19 [Stefan]
Rich: Important to know how many things have been added
17:10:59 [Stefan]
Matt: the more AT knows about it, the better, generally
17:18:45 [Stefan]
Stefan: being busy is exactly when child elements are being added or
removed .. I wonder why in some screen reader implementations you
still can navigate these regions although they are currently updated
17:19:40 [Stefan]
Matt: use of aria-busy is meant to help people to understand that
17:20:33 [Stefan]
Joseph: indicates that "significant" changes are made - what means
that?
17:21:06 [Stefan]
Rich: some of the text for aria-busy is way over the top -- tooo AT
specific
17:22:21 [clown]
http://w3c.github.io/aria/aria/aria.html#aria-busy
17:25:52 [clown]
q+ to ask what "changes to the element" mean?
17:28:37 [Stefan]
MichaelC: spec should just define not too much AT details
17:29:15 [Stefan]
Rich: giving examples is fine
17:30:12 [Stefan]
Matt: we can wipe out the educational piece
17:30:36 [Stefan]
Rich: challenge is here: is it live?
17:31:33 [richardschwerdtfeger]
q?
17:31:47 [richardschwerdtfeger]
ack clown
17:31:47 [Zakim]
clown, you wanted to ask what "changes to the element" mean?
17:32:27 [Stefan]
Matt: aria-busy is not a presentational property .. shoukld be misued
17:34:13 [Stefan]
Matt: aria-busy is just saying to AT: ignore events going on here
17:36:00 [Stefan]
MichaelC: belongs in implementation guide not in the spec
17:36:53 [Stefan]
Matt: I will start over
17:37:49 [Stefan]
Rich: I still don't know why the additional properties for feed exist
17:38:41 [Stefan]
Rich: if something is busy it means it is interactive, not dead
(static table is never busy)
17:38:48 [Stefan]
Matt: application is busy
17:38:57 [Stefan]
Matt: NOT and NEVER content
17:39:37 [Stefan]
Matt: application is busy updating content -this is the message
17:40:07 [Stefan]
Rich: you are saying any element can be live?????
17:40:27 [Stefan]
Matt: this was as and will be true forever
17:41:42 [Stefan]
Rich: aria-busy was only applied to widgets and live regions because
they are interactive and can change
17:41:43 [clown]
"For example, until a section of a page is fully initialized and
complete, an author could mark the document element as busy."
17:42:03 [Stefan]
Rich: a table is static and here saying aria-busy makes no sense in
this definition
17:44:17 [Stefan]
Joseph: aria-busy not defined means "false" right now
17:45:26 [Stefan]
Joanie: if false, I don't remove or add children, I would always
presume
17:45:43 [Stefan]
Joanie: is AT using it this way???
17:46:47 [Stefan]
Joanie: aria-busy=true on a table -> ah contents of this table are
going to change
17:47:37 [richardschwerdtfeger]
q?
17:48:13 [Stefan]
Stefan: a table is an interactive widget in the sense that on user
scrolling it can be programmed to fetch data (rows)
17:49:47 [Stefan]
Rich: Some Resolutions ahead. ....
17:50:19 [Stefan]
Rich: The feed stuff will get hairy
17:52:48 [richardschwerdtfeger]
Indicates whether an element, and its subtree, are currently being
updated.
17:52:54 [Stefan]
Joanie: Matt is talking about latest version and master, there he
uses the term "populated" .. makes more sense
17:52:56 [clown]
https://rawgit.com/w3c/aria/master/aria/aria.html#aria-busy
17:52:56 [joanie]
https://rawgit.com/w3c/aria/master/aria/aria.html#aria-busy
17:53:03 [joanie]
Indicates that a live region or widget is being populated.
17:54:40 [Stefan]
Joanie: don't you think "populated" is the right term?
17:54:52 [Stefan]
Joanie: do you think "populated" is the right term?
17:55:29 [Stefan]
Joanie: populated means "being added"
17:57:16 [Stefan]
Matt: we can reduce number of words here
17:59:27 [Stefan]
MichaelC: we also potentially want to hide abstract roles from the
spec (minimalistic view)
17:59:57 [Stefan]
MichaelC: it is a way to address different users of spec
18:01:46 [Stefan]
Joanie: do you work on action 1073?
18:02:14 [Stefan]
Rich: (to Joanie) Joanie: do you work on action 1073?
18:02:42 [Stefan]
•https://www.w3.org/WAI/PF/Group/track/actions/1073
18:05:21 [clown]
Indicates the current "selected" state of various widgets
18:05:36 [clown]
Change to "Indicates the current "selected" state of various widgets,
and whether the element is selectable"
18:05:38 [clown]
?
18:06:21 [Stefan]
Rich: we will close this as obsololete
18:07:15 [clown]
action-1361?
18:07:15 [trackbot]
action-1361 -- Matthew King to Suggest new text for the application
role -- due 2015-06-11 -- OPEN
18:07:15 [trackbot]
https://www.w3.org/WAI/PF/Group/track/actions/1361
18:07:23 [Stefan]
RESOLUTION: Close ACTION 1073 (obsolete)
18:08:19 [clown]
http://rawgit.com/w3c/aria/mck_issue633/aria/aria.html#application
18:09:02 [clown]
wow…
18:09:05 [Stefan]
rrsagent, make minutes
18:09:05 [RRSAgent]
I have made the request to generate
http://www.w3.org/2015/09/24-aria-minutes.html Stefan
Rich Schwerdtfeger
Received on Tuesday, 29 September 2015 20:30:59 UTC