14:10:40 <API> #startmeeting
14:10:41 <Services> Meeting started Thu Apr 10 14:10:40 2014 UTC.  The chair is API. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:10:41 <Services> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:11:02 <API> #topic Issues related to 3.12
14:11:33 <API> #info A bug on gjs made the magnifier crash gnome-shell when caret/focus tracking was active
14:11:53 <API> bug 727824
14:11:54 <Services> 04Bug http://bugzilla.gnome.org/show_bug.cgi?id=727824 critical, Normal, ---, gjs-maint, NEW, Crash when enabling Zoom a11y feature
14:12:26 <API> #info there is a patch and small test there, and it is already moved to gjs component (initially was reported at gnome-shell)
14:12:56 <API> #info it has critical:3.12.1 labels, so hopefully it will be fixed for 3.12.1
14:13:03 <API> done
14:13:38 <clown> cool.
14:13:40 <jjmarin> magnifier heartbleed :-)
14:13:54 <clown> gjs/introspection goof.
14:14:26 <mgorse> #info mgorse thinks he knows how to fix the leaking socket directories and plans on fixing that for 12.1, along with having the registry check that apps are still alive when children are quried.
14:14:26 <API> well, just one more thing
14:14:26 <mgorse> done
14:15:02 <jjmarin> mgorse: great !
14:15:29 <API> #info there is a bug to improve pdf accessibility: expose all pdf pages instead of just the current one, useful for "sayall" navigation
14:15:39 <API> bug 724965
14:15:40 <Services> 04Bug http://bugzilla.gnome.org/show_bug.cgi?id=724965 normal, Normal, ---, evince-maint, UNCONFIRMED, Implement a means to get the accessible text of the next/previous page
14:16:04 <API> #info patches were reviewed and updated, second review to confirm if they should be committed is pending
14:16:13 <API> #info with luck, that will be also part of 3.12.1
14:16:19 <API> and done (this time is for real)
14:16:40 <joanie> #info Joanie has been fixing minor bugs in Orca reported by testing users. They will go into 3.12.1
14:16:59 <joanie> #info Joanie is also working on some addition pdf stuff that she hopes can go into 3.12.1
14:17:02 <joanie> done
14:17:11 <joanie> s/addition/additional/
14:18:34 <API> ok, so anything else on this topic?
14:18:35 <API> moving?
14:19:01 <jjmarin> +1 from me
14:19:14 <clown> +1
14:19:22 <API> #topic Plans for 3.14 : Wayland
14:19:39 <API> #info discussion about security of the previous weeks
14:19:50 <API> #info have moved from the security technical point of view
14:20:13 <API> #info to how to deal with users and the new (hypothetical) security related dialogs
14:20:41 <API> #info Allan Day used as example that it would not be a good idea replicate those dialogs on Android
14:20:58 <API> #info that in the end all users say "ok" to "do whatever you want with my device"
14:21:35 <API> #info in summary: not a lot changed on this side
14:21:46 <API> and done
14:22:00 <API> although I was just reading those emails that lead to desktop-devel-list
14:22:12 <API> later I will take a look to wayland-dev too
14:23:10 <API> questions, comments, doubts?
14:23:21 <jjmarin> I think they want to go for a "More secure with less "security"" approach http://www.superlectures.com/guadec2013/more-secure-with-less-security
14:24:21 <jjmarin> done :-)
14:24:44 <API> well, Jasper also mentioned about
14:25:01 <API> ahving a more integrated solution without so many dialogs to
14:25:06 <API> provide "approved" workaround
14:25:28 <jjmarin> exactly, avoid the user to make decision about security as long as possible
14:26:06 <API> in any case, that is the summary
14:26:14 <API> we will keep an eye on that
14:26:28 <API> #topic Plans for 3.14: Other stuff?
14:26:42 <API> joanie, ?
14:26:50 <joanie> not from me
14:26:53 <clown> #info https://wiki.gnome.org/ThreePointThirteen/Features/
14:27:01 <clown> seems thin
14:27:05 <API> I mean, this is in the agenda, but personally I don't have too much to add since last week
14:27:12 <API> clown, well, as we mentioned last week
14:27:24 <API> probably this time we will not propose new a11y features
14:27:29 <API> as finishing the already proposed
14:27:33 <API> plus
14:27:38 <joanie> API: it's on the agenda in case someone has other stuff :)
14:27:42 <clown> right, I remember now.  okay.
14:27:52 <API> push to use the new API that deprecate old stuff (mostly on atk - at-spi)
14:27:55 <API> plus wayland
14:28:00 <API> is enough cake for all
14:28:13 <clown> right, lots of basic things to do.
14:28:32 <clown> "replace the foundation, but make sure the house remains standing".
14:29:02 <API> said the plumber
14:29:10 <API> so, then moving to next topic
14:29:18 <API> #topic W3C updates
14:29:19 <clown> the plumber is working on the foundation?  uh oh.
14:29:20 <API> clown, ?
14:29:53 <clown> #info The main new thing under consideration is to add  a "label" role.
14:30:31 <clown> #info Still not decided, but tending towards, if we do add a label role, the markup must also include a label-for attribute.
14:31:29 <clown> #info For example, <div role="label"  aria-labelfor="foo">This is the label text</div> … <span role="textfield" id="foo"></span>
14:31:53 <clown> #info the textfield in that example is labelledby by the label div.
14:32:33 <clown> #info The rationale is to provide aria support for what html already has for the table caption element and the figure figcaption element.
14:32:52 <clown> questions?  joanie, did I miss anything?
14:33:00 <joanie> sorry, multiple pings :-/
14:33:04 * joanie reads
14:33:22 <joanie> clown: I think that covers it
14:33:39 <joanie> the battle over localized role names is stalled ;)
14:33:54 <API> just a small question
14:33:57 <API> on that example
14:34:00 <API> <clown> #info For example, <div role="label"  aria-labelfor="foo">This is the label text</div> … <span role="textfield" id="foo"></span>
14:34:06 <API> right now there is not role label
14:34:13 <API> for for that one
14:34:24 <API> is there any suitable role for that "label"
14:34:26 <API> ?
14:34:41 <API> I mean, for elements that has aria-labelfor
14:34:45 <clown> API, right now there is no ARIA role for label.  There is an ATK label role, and IA2 label role.
14:34:52 <API> without the role "label", what role you have?
14:35:22 <clown> and, there is a <label> element in html that is mapped to, eg. ATK's label role.
14:35:30 <API> ok, so in that case is implicitly a role, because it has the tag aria-labelfor
14:35:36 <clown> Re:  without the role "label", what role you have?
14:35:46 <API> well, but as you are saying
14:35:55 <API> you don't need to use the label element
14:36:04 <API> I was thinking from the pov of the implementor
14:36:06 <API> ie webkit
14:36:16 <clown> If you have this:  <span role="textfield" aria-labelledby="foo"></span>  … <span id="foo">This is the label text</span>
14:36:29 <API> that would need to expose as role==ATK_LABEL if the element has aria-labelfor
14:36:37 <clown> Then the accessible for the textfield has an accessiblename of "This is the label text".
14:36:56 <clown> the span, I think, is simply expose as text.
14:37:39 <clown> yes, that's right.  The actual label text takes on whatever role it "normally" has.  For example, it might be a <h1> element, in which case it has a header role.
14:37:59 <clown> So:  <span role="textfield" aria-labelledby="foo"></span>  … <h1 id="foo">This is the label text</h1>
14:38:40 <clown> The textfield accessilbe has an acc name of "This is the label text", and there is a header accessilbe whose text is "This is the label text".
14:38:53 <clown> is that understandable?
14:39:10 <API> yes, thanks for the explanation
14:39:53 <clown> but, are you saying that "arai-labelfor" is enough?  That one doesn't really need to add role="label", in the first example?
14:39:56 <clown> Hmmm....
14:40:01 <clown> you have a point.
14:40:03 <API> no, the opposite
14:40:09 <API> I'm saying that having a role=label
14:40:17 <API> would allow to avoid the need of heuristics
14:40:29 <API> like "if i have this tag, that is not about roles, then my role is"
14:40:42 <API> I mean that with the role label is more straighforward
14:40:52 <API> "with role=label then my role is label"
14:40:56 <API> said captain obvious
14:41:06 <joanie> heh
14:41:09 <clown> gotcha.
14:41:12 <joanie> sounds like a familiar chat
14:41:14 <joanie> ;)
14:41:17 <clown> yup
14:41:25 * joanie is in the "pro-role-label" camp
14:41:29 <joanie> :)
14:41:36 <API> ok
14:41:38 <clown> joanie and API think alike alot
14:41:41 <API> so having said so?
14:41:44 <joanie> hivemind
14:41:50 <API> more questions already mentioned on the thread? new ones?
14:42:02 <clown> before we leave this subject...
14:42:21 <clown> I see an action for me on the meeting page:
14:42:28 <jjmarin> label is a heuristic itself :-)
14:42:29 <clown> "Take Piñeiro's (role-extension) email and submit to the ARIA working group on behalf of the gnome a11y group"
14:42:39 <clown> I think that's defunct?
14:42:45 <joanie> i think so too
14:42:55 <joanie> feel free to edit the wiki to remove it
14:43:06 <clown> okay.
14:43:13 <API> yes, last week conclusion was
14:43:18 <joanie> and emit the object:state-changed signal
14:43:20 <joanie> ;)
14:43:29 <API> "piñeiro will not add noise to the discussion; will add something if needed"
14:43:30 * joanie is hacking too hard
14:43:31 * clown is not a AT-SPI event source
14:43:46 <joanie> clown: but you're changing its state to defunct
14:43:48 <joanie> never mind
14:43:53 * API is moving to next topic
14:43:57 <API> #topic Marketing
14:44:02 <API> jmarin?
14:44:05 <API> jjmarin, ?
14:44:10 <jjmarin> today is short
14:44:21 <jjmarin> #info The AT-SPI entry for the wikipedia is W.I.P. https://en.wikipedia.org/wiki/AT-SPI
14:44:32 <jjmarin> At least one wikipedian (or I think he is a wikipedian) is reviewing the new entries and, by now, he hasn't removed them :-)
14:44:39 <jjmarin> done !
14:45:08 <jjmarin> questions or whatever ?
14:45:56 <API> no from my side
14:46:04 <clown> not from me.
14:47:32 <jjmarin> and joanie and mgorse are hacking :-)
14:47:32 <API> before crickets arrive, I will move to next topic
14:47:37 <API> jjmarin, thanks btw
14:47:41 <jjmarin> np
14:47:44 <API> #topic miscellaneous time
14:47:49 * clown arrrgh locusts!
14:47:53 <API> so something not scheduled to mentiond on this meeting?
14:48:30 <API> something else that is not about our love for fire?
14:48:32 <jjmarin> apart from clown's locust plague, no :-)
14:48:48 * clown whoops, they were actually crickets.  my bad.
14:48:55 <joanie> don't start the apocalypse without me
14:49:18 <clown> object:state-changed:apocalyspe-started
14:49:27 <joanie> no
14:49:32 <joanie> object:state-changed:busy
14:49:43 <jjmarin> GNOME 3.12 will be the Waylandypse
14:49:48 <joanie> on the accessible of ROLE_APOCALYPSE
14:49:50 <clown> lol
14:49:51 <jjmarin> s/12/14
14:50:26 <joanie> whose name is Waylandypse
14:50:34 <clown> is that because the plumbers are working on the foundation?
14:50:47 <jjmarin> :-)
14:51:18 <joanie> so meeting adjourned?
14:51:36 <clown> if so, at least Wayland won't be leaky.
14:51:50 <API> there are also bad plumbers
14:51:52 <API> see Mario
14:51:56 <API> tons of games
14:52:02 <clown> Luigi!
14:52:08 <API> and they are still full of turtles and mushrooms
14:52:12 <API> coff coff
14:52:17 <jjmarin> don't mention Luigi and Wario :-)
14:52:22 <API> having said so, I will close the meeting
14:52:24 <clown> Waluigi!
14:52:29 <API> thanks everybody for coming
14:52:35 <clown> thanks API
14:52:37 <API> #endmeeting