16:06:24 #startmeeting 16:06:24 Meeting started Thu Sep 19 16:06:24 2013 CET. The chair is joanie. Information about MeetBot at http://wiki.debian.org/MeetBot. 16:06:24 Useful Commands: #action #agreed #help #info #idea #link #topic. 16:06:30 #topic GSoC updates 16:06:36 floor is yours magpie2 and clown 16:06:37 clown? 16:06:46 wanna go first? 16:07:00 sure 16:07:43 #info Magdalen has made progress on the UI additions for the universal access panel for the system settings (gnome-control-center). 16:08:00 actually, that's all I have. You can probably expand on that, magpie2 16:08:12 #action Magdalen will file a bug about the ui 16:08:35 and attach you glade file? 16:08:39 *your 16:08:48 good idea 16:08:55 we should probably (constructively) mention the current state w.r.t. that last-minute change 16:09:08 #info A. Day said it would be unlikely my proposal for the ui would be accepted into 3.10 but he seemed to like the look. Magdalen hopes to find more out when filing the bug 16:09:31 minute change joanie? 16:09:32 yes, mark it as "work in progress". 16:09:38 ? 16:09:57 was that "?" for me, joanie? 16:10:01 sorry 16:10:03 still waking up 16:10:12 magpie2: what did you mean by "minute change"? 16:10:36 aday told me he thinks we can do it for 3.12 16:11:07 #info A. Day was asking about the GSettings and what they are for. He wanted to speak to joanie about that Magdalen suggests Joseph could talk to him as well to help clarify that 16:11:43 same jjmarin but i am not sure why 3.11 is not a good bet 16:11:48 sure. I can probably pirate some things from an email I sent to the mentors explaining what the different modes meant. 16:11:56 magpie2: 3.11 is 3.12 :) 16:11:56 unless bigger changes might be in the subtext 16:12:04 just like 3.9 is 3.10 16:12:12 3.11 is the beta release of 3.12 16:12:12 oh 16:12:20 odd numbers mean unstable. Once it's blessed for stable it becomes even 16:12:21 not the official release 16:12:22 cool thanks for clarifying 16:12:47 my ui doesn't work yet 16:12:56 magpie2: it wasn't expected to btw :) 16:13:03 not expected magpie2 16:13:10 * clown there's an echo in here 16:13:24 but between clown and joanie's advice the only reason for that is time rather than anything 16:13:31 sorry let me info that 16:13:39 right. time is our enemy 16:13:46 * clown my, how pithy 16:15:38 * joanie is waiting for the info 16:16:20 #info Joseph told Magdalen she has delivered the coding deliverables by completing a suggested glade ui prototype but Magdalen is going to try to get it working with the GSettings since Joanie and Joseph have given good guidance to help get that done. However she is started at university this week and has a lot of time constraints so it may have to be fully working after GSoC in case the time runs out. 16:16:28 was that clear? 16:16:31 i'm not sure 16:16:34 * joanie reads 16:16:44 anyway I better summerise the code of the project? 16:16:45 yes it is clear to me 16:16:50 the deliverable codes 16:16:59 with links? 16:17:05 you could 16:17:23 #info the deliverable codes completed thus far are the following links 16:17:43 anyone who wants to help this bit feel free i'm starting with the pyatspi 16:18:36 #link https://git.gnome.org/browse/pyatspi2/commit/?id=820712e21fb3797cd48c01f94499f502567ee73d 16:18:54 #link https://git.gnome.org/browse/pyatspi2/commit/?id=192638bfd0eeb446c14d5be6247fc474ee664ba0 16:19:13 #link https://git.gnome.org/browse/pyatspi2/commit/?id=2c38851a74cf3fe6a634449f6f681d8758d0ea7a 16:19:36 thee's a bug in review too 16:19:59 magpie2: were those commits from a patch attached to some bug in bugzilla? 16:20:20 or, from patches 16:20:22 yes 16:20:36 #link https://bugzilla.gnome.org/show_bug.cgi?id=707693 16:20:36 04Bug 707693: normal, Normal, ---, at-spi-maint, UNCONFIRMED, caret.py example needs to be improved. 16:20:51 for GSoC code submission, I advise taking those patches, and tar'ing them all up and submitting the tarball to GSoC. 16:21:01 * clown looks at the reqirements 16:21:04 ok thank you clown 16:21:21 "All students will need to add files, which can include source files, a single .diff file, multiple .diff files, binary files, etc., to their project in Melange. " 16:21:22 can you action that to save time while i grab the rest of the links? 16:21:32 sure. 16:22:23 #action for GSoC code submission, Magdalen will take all of her submitted patches that were committed from gnome's bugzilla, tar them up, and sent the tarball to GSoC. 16:22:28 magpie2: I'm sorry; I forgot to review that patch. Thanks for the reminder. 16:22:55 did that action make sense to everyone? 16:23:02 I think so clown 16:23:12 I can only speak for myself however :) 16:23:22 GSoC says "diff" files, but I think patches are essentially diff files. 16:23:35 #link https://bugzilla.gnome.org/show_bug.cgi?id=705652 16:23:35 04Bug 705652: normal, Normal, ---, gsettings-desktop-schemas-maint, RESOLVED FIXED, Add GSettings for focus and caret tracking in GNOME Shell 16:24:02 #link GSoC code submission guidelines: http://www.google-melange.com/gsoc/document/show/gsoc_program/google/gsoc2013/codeguidelines 16:24:51 #link https://git.gnome.org/browse/gnome-shell/commit/js/ui/magnifier.js?id=9d8f30f955fcd15a509a19310c0e946f13eca3a6 16:25:26 Should I include misc contributions that happen during summer? 16:25:41 joanie, do you think there are any intellectual property issues here — giving the code to Google? 16:25:55 I do not clown 16:25:58 i have another g-s patch for their jhbuild examples i did but have still to patch 16:26:02 it's within the license 16:26:04 i guess it's publically avaialbe under GPL. 16:26:09 exactly 16:26:10 *available 16:26:21 clown they allow students the rights the their code i think 16:26:42 it still must be free and open source 16:26:43 sure, but what rights to they get by getting a copy? 16:26:54 I can get a link for that if needed 16:26:55 right. It is goverend by the GPL. 16:26:56 magpie2: what about the big fix (the tracking implementation bug) 16:27:02 you but the small ones 16:27:06 s/but/put/ 16:27:08 oh 16:27:12 silly 16:27:13 :) 16:27:18 https://git.gnome.org/browse/gnome-shell/commit/js/ui/magnifier.js?id=9d8f30f955fcd15a509a19310c0e946f13eca3a6 16:27:20 don't want to forget that one ;-) 16:27:21 gsettings too 16:27:32 sorry 16:27:37 #link https://git.gnome.org/browse/gnome-shell/commit/js/ui/magnifier.js?id=9d8f30f955fcd15a509a19310c0e946f13eca3a6 16:27:40 and the one your are going to do for the UI 16:28:02 the glade file patch. 16:28:04 i have not even put that on github :/ 16:28:23 ok 16:28:27 #info the code for the ui is not up on github or git.gnome 16:28:31 you will get to it, I imagine. 16:28:57 #action Magdalen will do the github today in case the ui is not fully woking before mo 16:29:04 mon for the bug 16:29:35 #info correction: in case the ui is not fully working for the gnome-control-center ui bug Magdalen is filing today 16:30:01 sorry i should have put that as one info 16:30:12 * magpie2 getting the gsettings patch 16:30:21 magpie2: even if the code isn't working, it's worthwhile putting the new ui in the bug as a indication of the direction we are going, for the designers to have a look at. 16:30:45 i was just going to put the picture until such a time i got it working? 16:31:06 you could that too. Or, a picture and the .ui file 16:31:13 becuase the code is not complete underneith but it does'nt affect the look 16:31:35 Yes, but the designers might want to play with it in glade. 16:31:47 they can't do that with a picture. 16:32:17 but, it's up to you. that's just my advice. 16:32:37 #action Magdalen will maybe add the code link to my github and say that it is not complete and upload the pictures for today and try to get a patch ready before mon 16:32:50 /my/her 16:34:10 * joanie waits to see if there is anything else 16:34:16 #agreed we made a very good timeline for gsoc and the mentors should be proud 16:34:26 gsettings 16:34:57 https://git.gnome.org/browse/gsettings-desktop-schemas/commit/schemas?id=3e7aaf2f0fe2f0c98a01cbdad0a74bc079390f89 16:35:01 and bug 16:35:12 and a couple others but they can wait if we are running over 16:35:37 https://bugzilla.gnome.org/show_bug.cgi?id=705652 16:35:37 04Bug 705652: normal, Normal, ---, gsettings-desktop-schemas-maint, RESOLVED FIXED, Add GSettings for focus and caret tracking in GNOME Shell 16:36:29 #action Magdalen will also put the ui snaps on the wiki and make sure all work completed during the GSoC is properly referenced to make things easier 16:36:38 clown 16:36:42 yes? 16:36:45 you actually make a good point though 16:36:55 * clown wonders never cease 16:37:13 will it be a problem for google that you initially wrote the tracker code and i adapted it or are they expecting stuff like that? 16:38:14 it's too late now to think about that i guess but i am suddenly fretting that could make it fail? 16:38:28 well, Jasper made changes too, didn't he? 16:38:36 i mean this is open source work right 16:38:46 I don't think that matters. You did work on it. Your ideas are in there. 16:38:48 nothing is from scratch for most projects 16:39:17 ok yeah common sense tells me you are right 16:39:31 oh 16:39:33 1 16:39:48 1? 16:41:25 * clown waiting for magpie2 16:41:31 #action Magdalen is to finish a 'final GSoC' blog contribution to the planet feed and she will base that around the ui work and include the things she learned from Joseph and Joanie recently, in that post so it explains about about the accessibility control center ui and how that should work. 16:42:43 #info Magdalen thinks the Admin, Mentors and the G-S team (especially Jasper) have been really, very amazing and she is delighted to have been able to learn from their expertise this summer. 16:42:53 it was an honour :-) 16:43:13 soooo... I know we're going long, but I have one other thing 16:43:15 16:08:55 <@joanie> we should probably (constructively) mention the current state w.r.t. that last-minute change 16:43:24 yes 16:43:46 i was going to get to that but then I got carried away with the happy stuff 16:43:47 mgorse: any updates on the freeze problem? 16:44:07 let me get the links 16:44:13 it's important 16:44:20 #info there was a commit to the magnifier code that disabled the initialization of atspi on the condition that the magnifier was not used. 16:44:22 If you mean whether I have a patch finished and sent to the release-team, then, unfortunately, no 16:44:56 #info apparently, the desktop freezes for 30 sec when atspi is running if the Display control panel is accessed. 16:45:09 * mgorse waits 16:45:11 mgorse: one of the things I want to do this week is help you with that actually 16:45:29 magpie2: let's first get the status in the meeting minutes 16:45:35 #info there was much discussion about whether this commt should be reverted, while mgorse is working on fixing atspi. 16:45:45 I sort of left thinking things have been resolved but the patch is not reverted i see 16:45:48 that's all I've got in terms of my understanding. mgorse? 16:46:20 ok 16:47:19 #info gnome-shell can deadlock if a process calls it via D-Bus while gnome-shell is trying to make an AT-SPI call to that other process via D-Bus 16:48:23 #info Magdalen found out that atspi had a bug yesterday after finding the default atspi initialisatin had been removed yesterday and told joanie and mgorse. Mike is infoing his progress 16:48:47 #info it is exasperated by a hack made to AT-SPI a while ago that makes it not time out on a call for up to 15 seconds fo ran app that immediately started 16:49:44 #info Magdalen Jasper Mike, Joseph, Giovanni and Joanie discussed this yesterday 16:49:49 #info mgorse is working on a patch that makes AT-SPI calls enter a main loop, rather than just dispatching D-Bus messages, and thinks that this is safe, but any change of that sort seems to have a bit of risk 16:50:13 okay; done, I think 16:50:21 mgorse 16:50:51 can you please tell us your action for it and more about the 'risk' 16:51:14 I plan on making a freeze break request once I have a patch working 16:51:26 mostly I'm worried that there is some case that I hadn't thought of 16:51:42 like the one that Giovani found was something that I hadn't thought about 16:51:48 #info if the patch is not fixed there is a danger the feature will not be working in 3.10 as planned. 16:52:04 "the feature" which feature? 16:52:26 The other option, I think, would be to change those timeout values. They are configurable. But that would just reduce the delay, not eliminate it 16:52:28 mgorse: we will worry about the current problem and if another one crops up deal with that but i can help you there as much as you need 16:52:59 even if it's just dumping stuff and running tests while you look at code to speed things up 16:53:08 Anyway, yeah, I plan on submitting a freeze break request, so I thin kit's up to the release team whether it goes in 16:53:26 my ui stuff will take but 10 mins so im free 16:54:24 did gnome-shell say what their plan was? 16:54:48 maybe we should ask for a Jasper comment? 16:54:55 magpie2: not during this meeting 16:55:02 it sounds like things are moving forward 16:55:13 and all we can do is progress one step at a time 16:55:17 i was worried about the suggestion to remove the tracker 16:55:18 mgorse: thanks for looking into this! 16:55:35 yes, mgorse, thanks. 16:55:42 okkkkkaaaaaay 16:55:48 i think that would be a terrible thing 16:55:52 50 minutes of gsoc updates 16:55:57 in a 60 minute meeting ;) 16:55:59 oh 16:56:01 sorry 16:56:03 mind you, it needed to be said 16:56:06 all of it 16:56:07 yes it did 16:56:11 but I think now is time to move on 16:56:22 #topic Other 3.10 updates/issues 16:56:31 sorry about that i just sort of left yesterday thinking it was sorted and that was clearly wrong 16:56:41 me me me 16:56:44 #info In general we seem to have caught all the accessibility-related regressions 16:56:51 ok 16:56:56 * magpie2 listening 16:56:57 (modulo the atspi freeze in gnome-shell) 16:57:24 #info But Frederic Peters caught a gui issue https://people.gnome.org/~fpeters/a11y-zoom-prefs.png 16:57:33 #action Joanie will investigate this today 16:57:42 another 3.10 stuff? 16:57:47 * joanie yields the floor 16:57:52 i have a one 16:58:06 magpie2: ok 16:59:52 magpie2: ok == what is it? 16:59:53 #info Magdalen has been rethinking her post-gsoc TODO goals and following the conversations yesterday she would like to try to learn something from mgorse and try to make sure all the atspi codes that need to be in javascript for gnome-shell are working as they should and asks for Mike's thought on that possibility 16:59:55 oh 16:59:57 thanks 16:59:58 :) 17:00:07 magpie2: that is not a 3.10 issue 17:00:21 it is until atspi is fixed isn't it? 17:00:25 maybe not sorry 17:00:31 what issue is it? 17:00:32 magpie2: that was discussed last topic 17:00:35 I think the online account of the gnome-initial-setup is still no keybnav, if you don't press Next the first time, you won't move of this screen 17:00:43 no i mean the methods 17:00:45 until we have mgorse's patch, we need to wait 17:00:48 the full api 17:01:08 jjmarin: can you verify that and file it? 17:01:23 having tested with a fully updated jhbuld environment 17:01:28 i want to make sure the API is checked 17:01:36 joanie: sure. I'm doing now :-) 17:01:40 thanks!! 17:02:09 i can help with that too if needed jjmarin 17:02:20 magpie2: you need to focus on your work 17:02:27 jjmarin can do this 17:02:30 and/or I can 17:02:40 magpie2: yes, I can do it :-) 17:02:48 thanks anyway :-) 17:02:58 ok i will finish my timetable 17:03:13 should we move on? 17:03:16 so, other "issues" (aka bugs that were introduced this cycle that are not yet fixed) 17:03:27 magpie2: when we are ready to move on, I'll move us on 17:03:29 Probably. The ones we don't know about :) 17:03:35 mgorse: hehe 17:03:36 only the GSoC atpsi one that i know of 17:03:51 oh also people need to have an up to date 17:04:06 magpie2: is that an "issue" 17:04:06 atspi because it's a runtime dependencyy 17:04:34 anyhoo, I'm going to move on 17:04:36 so some people with atspi that is older will see the bug i was suffering around GUADEC times with running it 17:04:46 #topic "Boston" Summit and the possibility of Wayland Accessibility Hacking 17:05:33 #info If you are planning on coming to the summit, you need to add yourself to the wiki page: https://wiki.gnome.org/Montreal2013/Participants 17:05:52 if you need sponsorship, you need to fill out the table https://wiki.gnome.org/Montreal2013/travelsponsorship 17:05:58 ok I can't afford that 17:05:58 #info if you need sponsorship, you need to fill out the table https://wiki.gnome.org/Montreal2013/travelsponsorship 17:06:07 magpie2: you are not doing wayland a11y hacking 17:06:33 anything else specific to this topic that needs to be stated? 17:06:36 after the gsoc i want to for click lock 17:06:55 and i'm going to talk to ged and francesco about mousetweaks 17:07:00 sure 17:07:05 sounds like a plan :) 17:07:12 #topic W3C updates 17:07:14 clown: ? 17:07:26 info'ing... 17:08:16 #info No news specifically from W3C, but in ARIA-related news, Joseph will giving a talk at the jQuery accessibility summit Oct 10-11 17:08:22 yay! 17:08:23 #link http://www.deque.com/deque-partners-jquery-create-accessibility-summit 17:08:31 done. 17:08:37 I actually have something too ;) 17:08:56 #info Joanie has requested, and appears very like to get, "Invited Expert" status 17:09:06 cool! 17:09:12 that's awesome 17:09:20 sound good !!! 17:09:30 For what committee? That should be useful, anyhow 17:09:34 #info Joanie will be looking at current "mappings" for HTML and ARIA with the aim of adding new API to ATK/AT-SPI2 where needed for better a11y support 17:09:37 what happens to univited experts? :) 17:09:46 they are ignored 17:09:51 by the W3C 17:09:54 and problems result 17:09:58 ;) 17:10:15 clown: I expect you and I will be doing some coordination 17:10:21 lol well they have you and joseph now so it'll be ok 17:10:23 once I get going 17:10:28 perhaps. I don't actually work on HTML (yet). 17:10:36 but you do on ARIA 17:10:42 and when we have new roles, states, etc. 17:10:55 that will mean changes to W3C documents 17:10:57 and mappings 17:11:01 and best practices 17:11:02 etc. 17:11:03 yes. Will you be focussing on HTML5? 17:11:07 also the verification tests 17:11:11 I will be focusing on it all 17:11:22 your head will explode! :-) 17:11:26 nah 17:11:32 ok, so moving on 17:11:36 I'm going to be asking you two a lot of questions about this later on 17:11:38 #topic Marketing 17:11:56 #info The release notes for GNOME 3.10 are done, and now it is in the translation phase 17:12:01 yay! 17:12:04 wohoo 17:12:06 #info Unfortunately, the magnifier's focus and caret tracking notes have been dropped because the lack of GUI 17:12:12 :/ 17:12:16 boo 17:12:18 #info The accessibility bits has been consolidated with the rest of the notes, instead of having a separated section 17:12:29 you can check the release notes in https://help.gnome.org/misc/release-notes/3.10/ 17:12:31 they weren't in the g-s ones already 17:12:36 gnome / 3.10 17:13:01 that's all 17:13:06 is it a subrelease when it's 3.9.91? 17:13:08 great thanks! 17:13:25 no, just 3.10 17:13:32 #topic Miscellaneous Time 17:13:34 :-) 17:13:44 they sent the release note with the subreleases though 17:13:44 we're over the meeting end time 17:13:49 anything that must be said? 17:13:53 yes 17:14:10 the mgorse work isn't that 3.10 issue rather than a GSoC one? 17:14:19 magpie2: it's both 17:14:22 so why repeat it? 17:14:35 jjmarin, I'm being asked for a username/password. 17:14:48 gnome / 3.10 17:14:56 "ask release-team@gnome.org for the password" 17:14:59 oh, thanks. 17:15:05 it's a secret 17:15:07 :) 17:15:10 well i'm not sure it was put in the right spot but that's ok 17:15:31 magpie2: the topic even began with "Other" 17:15:35 not working, jjmarin 17:15:40 anyhooo 17:15:42 #endmeeting