15:05:22 <API> #startmeeting
15:05:22 <Services> Meeting started Thu Jan 15 15:05:22 2015 UTC.  The chair is API. Information about MeetBot at http://wiki.debian.org/MeetBot.
15:05:22 <Services> Useful Commands: #action #agreed #help #info #idea #link #topic.
15:05:34 <API> #topic Progress towards 3.16
15:05:43 <API> #info Last week I have been working on some gnome-shell bugs
15:06:13 <API> #info bug 729603 has a new patch, pending to be reviewed
15:06:13 <Services> 04Bug http://bugzilla.gnome.org/show_bug.cgi?id=729603 normal, Normal, ---, gnome-shell-maint, UNCONFIRMED, Regression: GDM user list buttons no longer have accessible labels
15:07:14 <API> #info I alto took a look on bug 738701, but my current WIP patch instead of just putting in the proper order the first selection event, it just prevent it (note the WIP)
15:07:14 <Services> 04Bug http://bugzilla.gnome.org/show_bug.cgi?id=738701 normal, Normal, ---, gnome-shell-maint, UNCONFIRMED, Please emit accessible object:state-changed:selected events after window:activate for switchers
15:07:31 * clown waves (sorry for being so late).
15:07:55 <API> #info will try to book some time this week to resume the work on that bug, and also on bug 742806
15:07:55 <Services> 04Bug http://bugzilla.gnome.org/show_bug.cgi?id=742806 normal, Normal, ---, atk-maint, UNCONFIRMED, Add new roles for math content
15:07:55 <API> done
15:07:55 <API> next one?
15:08:04 <joanie> #info Joanie proposed a few new roles so that we can support MathML better now until we have subroles sorted out (which doesn't look like it will happen by the end of this cycle).
15:08:23 <joanie> #info Joanie will proceed after review, including AT-SPI2 corresponding roles.
15:08:47 <joanie> #info Joanie just realized that she hadn't committed the AT-SPI2 patches related to the new READ_ONLY state.
15:09:20 <joanie> #info One of them hasn't been reviewed: Bug 690004.
15:09:20 <Services> 04Bug http://bugzilla.gnome.org/show_bug.cgi?id=690004 normal, Normal, ---, at-spi-maint, UNCONFIRMED, Add ATSPI_STATE_READ_ONLY
15:09:30 <joanie> #info Joanie will commit them all once fully reviewed.
15:09:39 <joanie> (assuming they are ACNed, of course)
15:09:59 <joanie> I think that's all the relevant stuff for today.
15:10:08 <joanie> (done)
15:11:33 <API> somebody else?
15:11:49 <mgorse> I don't have much to report. Asked about wl_test on wayland-devel, got "that isn't going to wayland/does stuff that clients aren't supposed to do"-type responses; trying to decide what I should be doing there
15:12:12 <API> hmm
15:12:32 <API> "does stuff that clients aren't supposed to do" are also more or less the answer we got last time we asked about that
15:12:38 <API> but ok
15:12:40 <API> mgorse, thanks
15:13:11 <mgorse> mclasen originally proposed a privileged API, which I guess would end up being something similar to wl_test, or a subset of it, but not sure if there's infrastructure for that currently
15:13:34 <API> yes,   but the problem about a privileged API
15:13:44 <API> is how to get the privilige to use it
15:14:01 <API> that was part of the "security model" that was mentioned last time I jumped there
15:14:08 <API> there==wayland-devel
15:14:22 <API> there was some people talking about that time of functionality
15:14:28 <API> s/that time/that type
15:14:46 <API> in the sense that there are stuff that breaks somehow the model of stuff that you should not do
15:15:11 <API> making stuff that in the past was easy, like a screenshot
15:15:24 <API> or a screensaver difficult
15:15:33 <API> difficult as "impossible just using vanilla wayland"
15:15:38 <mgorse> hmm; makes me wonder if it's possible to make a screenshot in Wayland
15:15:54 <mgorse> since the answer to that sounds like it would be related to a11y issues
15:16:07 <API> mgorse, well, as Im saying
15:16:26 <API> that was one use case that also had several threads at wayland-dev
15:16:55 <API> how to do that without breaking "the model"
15:17:00 <mgorse> ok
15:17:14 <API> anyway, thanks for asking
15:17:28 <API> but in relation with joanie infoes
15:17:42 <API> do you think that you will be able to book some time to take a look to ther patches?
15:17:44 <API> *her
15:17:53 <joanie> it may just be the one
15:18:09 <mgorse> I'll look after the meeting. Is it just bug 690004, or are there others?
15:18:09 <Services> 04Bug http://bugzilla.gnome.org/show_bug.cgi?id=690004 normal, Normal, ---, at-spi-maint, UNCONFIRMED, Add ATSPI_STATE_READ_ONLY
15:18:09 <joanie> the -core
15:18:09 <joanie> lemme triple check
15:18:12 <joanie> you ACNed -atk's before
15:18:40 <mgorse> ok. probably just forgot to review the other patch
15:18:44 <joanie> but don't wait on me
15:18:51 <joanie> i'll comment on the bug if there's another
15:19:25 <joanie> but since it's silent
15:19:29 <joanie> it appears it's just that one, yes
15:19:49 <joanie> and thank you mgorse
15:20:29 <jjmarin> Hi ! I've just arrive now and I have to use my tablet 'cause GNOME doesn't start :(. A have a general question,  so what is the status of a11y support in Wayland for GNOME 3.16 ? I mean  we  are there with some issues or are not even close ?
15:21:15 * clown waves at jjmarin
15:21:22 <API> jjmarin, well
15:21:26 <jjmarin> o/
15:21:29 <API> as we were mentioning
15:21:32 <API> there are still hole
15:21:34 <API> *holes
15:21:48 <API> and there are still questions on wayland itself
15:22:01 <API> what I mean is that we are still on the position that
15:22:13 <API> we are not sure how to implement some stuff  using wayland
15:22:27 <API> so no, GNOME 3.16+wayland will not be accessible
15:22:41 <API> eventually we should update the wiki page
15:22:47 <API> about the current status
15:22:48 <joanie> that's my bad
15:22:51 <API> and do some testing
15:23:00 <joanie> and on my todo list
15:23:05 <API> to at least ensure that stuff is not crashing with GNOME 3.16+wayland
15:23:24 <API> joanie detected some crashes some time ago, I think that most of then got solved
15:23:30 <API> but I think that one is still around
15:23:47 <API> jjmarin, does that answer your question?
15:23:58 <jjmarin> yes, thanks
15:24:44 <API> ok, so any other question or comment?
15:24:52 <API> option 2 is moving to next topic
15:24:53 <jjmarin> It is more LongWay than Wayland :P
15:25:11 <clown> heh
15:26:27 <API> well, it is also true that there isn't anyone for accessibility working actively on a11y+wayland, so is difficult to get any advance in that way
15:26:34 <API> so, after this ranty rant
15:26:43 <API> or sad disclosure of the reality
15:26:55 <API> moving to next topic
15:27:05 <API> #topic W3C updates
15:27:07 <API> clown,
15:27:10 <API> ?
15:27:52 <clown> I don't think there is a lot.
15:28:20 <clown> I know I made joanie happy by chaning the mapping or role "rowgroup" to "not mapped" for ATK.
15:28:22 <clown> :-)
15:28:27 * joanie nods
15:28:31 <joanie> seriously
15:28:36 <joanie> Windows is why we can't have nice things
15:29:01 <API> just Windows
15:29:07 <API> thre rest of the world is fine
15:29:08 <API> :P
15:29:16 <clown> everything else is "under discussion", or assigned actions waiting for the action to be completed.
15:30:00 <clown> Anything from your point of view, joanie?
15:30:09 <joanie> nope
15:30:34 <joanie> at least not from w3c to us
15:30:36 <API> ok, so for minutes sake, could you make and (probably mostly empty) info?
15:30:42 <API> *an
15:30:51 <joanie> the reverse direction (like the new READONLY state) is for a different meeting :)
15:31:04 <clown> #info  There are a lot of pending actions on the W3C front, and not much to report at the present time.
15:31:22 <API> clown, thanks
15:31:26 <clown> no problem.
15:31:30 <API> so I guess that from that is hard to make questions
15:31:37 <API> im moving agin
15:31:39 <API> #topic Marketing
15:31:43 <API> jjmarin, ?
15:31:53 <API> (if the tablet allows)
15:32:02 <jjmarin> I hope so
15:32:50 <jjmarin> I can write an article about a11y for the annual report, bit I don't have to much input
15:33:02 <jjmarin> I think some posts can help me
15:33:11 <jjmarin> for example about Wayland
15:33:22 <jjmarin> and the Orca refactoring work
15:33:41 <jjmarin> what do you think ?
15:34:01 * jjmarin is using a bt keyboard :-)
15:34:09 <joanie> I do need to blog more
15:34:11 <joanie> so I'll try
15:34:24 <joanie> I know not to make promises in case I cannot keep them
15:34:32 * clown joanie is too busy editing the aria spec...
15:34:37 <jjmarin> that would great for me and all your fans :-)
15:35:01 <jjmarin> It can be short
15:35:07 <API> jjmarin, so an article for a11y for the annual report is conditioned of everybody else doing a kind of 2014 summary?
15:35:43 <jjmarin> Not exactly
15:35:50 <jjmarin> but it would help
15:36:09 <jjmarin> the article will be short
15:36:36 <jjmarin> but it is difficult to know what is the important stuff done lately
15:36:49 <API> ok, so I will try to summarize what you said, but with infos
15:36:55 <API> you can tell me if Im wrong later
15:37:16 <jjmarin> ok, thanks
15:37:16 <API> #info jjmarin volunteered himself to write the a11y article for the GNOME annual report
15:38:08 <API> #info but in order to write it, it would be really helpful if the rest of the people write something (blog post, email, etc) giving some hints about the relevant stuff done during year 2014
15:38:28 <API> is that a good summary? or it is too summarized?
15:38:33 <jjmarin> API, you are good at infos :-)
15:38:43 <jjmarin> very professional
15:38:46 <API> I will take that as a "yes"
15:38:56 <API> I will also try to do that
15:39:01 <jjmarin> sure :)
15:39:23 <API> during this year I have a 3.12.1
15:39:38 <API> it is true that I didn't do the usual 3.X post, being X=14
15:39:47 <API> in any case
15:39:55 <API> questions, doubts, comments, moving?
15:40:55 <jjmarin> move on
15:41:02 <API> #topic Miscellaneous time
15:41:13 <API> something not agended to mention?
15:41:25 <API> (assuming that "agended" exists as a verb)
15:41:39 <jjmarin> hehe
15:42:49 <jjmarin> I wonder if the GNOME shell message "Oh no! Something has gone wrong." is accessible. I guess it isn't
15:43:11 <API> hmm
15:43:12 <clown> is it text, at lease?
15:43:13 <API> good question
15:43:15 <clown> *least
15:43:19 <API> but I don't have any idea
15:43:24 <jjmarin> hehe
15:43:38 <API> and I guess that it would be hard to test/reproduce
15:43:45 <API> I mean, while working on that gdm bug
15:43:55 <mgorse> Sometimes I see a frame, that orca reads, with a button to log out. I think I might've seen the "something has gone wrong" there, but I can't remember
15:44:03 <mgorse> It doesn't tell me anything about what went wrong or how to fix it
15:44:21 <API> mgorse, fwiw, the message doesn't say how to fix it
15:44:30 <API> but it is true that at least it says that something went wrong
15:44:31 <clown> nor what is wrong.
15:44:32 <jjmarin> mgorse: you're right, there is a Log out button
15:44:54 <API> well, if at least the log out button is exposed by orca, something is there
15:45:02 <clown> but it looks like enough of the system (at-spi, orca) is around to report the problem.
15:45:14 <joanie> all it says is "something has gone wrong"
15:45:19 <jjmarin> the subtext is  "A problem has occurred and the system can't recover. Please log out and try again"
15:45:29 <joanie> oh, is that it?
15:45:33 <joanie> I forget
15:45:36 <clown> right, jjmarin.  I meant report that there is a problem (but no details).
15:45:39 <joanie> but it's a useless screen for all users
15:45:59 <API> but everybody deserves the same useless screen ;)
15:46:09 <jjmarin> it is the grey screen of death
15:46:24 <API> hmm, good poing
15:46:32 <API> was the blue screen of death accessible on windows?
15:46:36 <API> *point
15:46:48 <mgorse> don't think so
15:46:48 <jjmarin> yes, exactly
15:46:50 <clown> I believe the blue screen of death is a DOS screen.
15:47:04 <mgorse> it's more or less equivalent to a kernel panic
15:47:22 <clown> not a dialog or window.  Windows is non-operative if you see the BSOD (I think).
15:47:26 <API> well, fwiw, that screen on gnome
15:47:30 <API> that grey screen of death
15:47:42 <API> is most of time a gnome-shell crash
15:47:47 <joanie> the grey screen of oh noes
15:47:52 <API> so that is the reason at-spi2 and orca can still run
15:48:07 <API> is not so low level as the classic blue screen or a kernel panic
15:48:34 <API> in any case
15:48:39 <API> as I was mentioning
15:48:49 <API> probably test that grey screen would be difficult
15:48:54 <API> while working on that gdm bug
15:48:59 <API> I found testing/reproducing it
15:49:02 <API> harder that it should
15:49:13 <API> or at least, harder that it should imho
15:50:22 <API> and at least
15:50:31 <API> gnome-shell has a gdm mode to test that screen
15:50:40 <API> there isn't a "grey screen of death" mode
15:50:49 <API> so right now, I don't have a clue about how to test that screen
15:51:12 <API> so
15:51:16 <API> opinions=
15:51:17 <API> ?
15:51:25 <jjmarin> no dead, just zombie or infected
15:51:39 <clown> what actually can cause the GSOD?  Can that be replicated?
15:51:42 <API> should we open a bug for "test dead screen", or just forget as it is a everything is failing use case?
15:51:57 <API> clown, well, that is my point
15:52:07 <API> as gnome-shell doesn't have a specific mode to go to that screen
15:52:15 <jjmarin> at least we have an accessible button
15:52:26 <API> right now I think that the way to replicate it is crash everything
15:52:35 * joanie grins
15:52:40 <API> well, lets go for a small compromise
15:52:53 <API> taking into account that the gdm patch is still waiting for review
15:53:03 <API> I was planing to ping someone (probably jasper)
15:53:22 <API> so I can ping him, and on that ping moment, ask him if there is a way to test the grey screen of death
15:53:34 <API> and if it is easy, make a little test
15:53:42 <jjmarin> +1
15:53:53 <API> as my pessimistic side is telling me that that gdm patch would need some tweaks after review
15:54:52 <API> #action API will ping gnome-shell developers for the gdm patch, will use that moment to ask about the grey screen of crash
15:55:06 <API> ok
15:55:09 <API> so almost finishing
15:55:15 <API> something else or closing meeting?
15:56:16 <clown> nothing from me.
15:56:22 <jjmarin> after talking about dead screens, everything is donw I guess :-)
15:56:48 <API> #closemeeting
15:57:00 <API> #endmeeting