14:34:33 <API> #startmeeting
14:34:33 <tota11y> Meeting started Thu Jun 16 14:34:33 2011 UTC.  The chair is API. Information about MeetBot at http://wiki.debian.org/MeetBot.
14:34:33 <tota11y> Useful Commands: #action #agreed #help #info #idea #link #topic.
14:34:45 <API> agenda on the topic
14:35:11 <API> #topic Testing Distro
14:35:25 <API> so clown jhernandez|afk
14:35:25 <API> hmm
14:35:26 <API> afk
14:35:27 <API> :P
14:35:39 <API> well, clown jhernandez|afk any update, I guess that clown can start
14:35:55 * clown gathering my thoughts
14:36:26 <clown> last week, I provided a patch to orca to upgrade from using GConf to GSettings ...
14:36:49 <clown> ... that, at least, initialized the orca <-> gs-mag communication ...
14:37:02 <clown> ... joanie committed a modified version of that patch ...
14:37:20 <clown> ... but it revealed another bug, which I am currently fixing ...
14:37:31 <jhernandez|eee> clown: ok
14:37:40 <clown> ... once that fix is done, I want that version of orca on the testing distro.
14:37:44 <clown> done
14:38:20 <jhernandez|eee> I know about first bug, not second
14:38:26 * clown looks it up.
14:38:35 <clown> https://bugzilla.gnome.org/show_bug.cgi?id=652500
14:38:35 <tota11y> 04Bug 652500: major, Normal, ---, clown, ASSIGNED, DBusException seen when Orca is running with gnome-shell mag
14:38:55 * clown love tota11y
14:39:03 <jhernandez|eee> so, If you want, I can update orca's package with these patches
14:39:13 <joanie> jhernandez|eee: or I could roll a tarball
14:39:23 <jhernandez|eee> and  to generate a new spin
14:39:24 <joanie> because I've made a bunch of other mag related changes
14:39:27 <jhernandez|eee> joanie: of course
14:39:32 <jhernandez|eee> I know
14:39:35 <joanie> jhernandez|eee: or you could roll one ;-)
14:39:35 <jhernandez|eee> so, make sense
14:39:42 <joanie> but I'm volunteering if you're busy
14:39:54 <jhernandez|eee> joanie: nice
14:40:13 <jhernandez|eee> so
14:40:33 <API> ok, people, and now try to sort out all those data
14:40:40 <jhernandez|eee> please, to all a11y haackers, if they want new features included?
14:40:40 <joanie> :-)
14:40:40 <API> and prove pretty #info
14:40:43 <API> please :P
14:40:59 <clown> API, I'll try to summarize my summary with info's
14:41:00 <jhernandez|eee> update the wiki properly
14:41:07 <jhernandez|eee> clown: ok
14:41:20 <clown> #info provided a patch to orca to upgrade from using GConf to GSettings
14:41:40 <clown> #info that fixed orca <-> gs-mag communication
14:42:13 <clown> #info had the effect of revealing another bug, a dbus exception
14:42:24 <clown> #info which I am currently fixing
14:42:25 <clown> done
14:42:47 * clown passes conch to others
14:42:55 <jhernandez|eee> IMHO, this spin will not be so 'public' as others, and I'm thinking that if we start with the  new testing spins plans, this spin will be our last only-focused one
14:43:15 <jhernandez|eee> JFYI
14:43:18 <jhernandez|eee> :]
14:43:39 <API> jhernandez|eee, and this leads to next subitem in this topic
14:43:40 <API> :P
14:43:44 <joanie> not done
14:43:45 <jhernandez|eee> ok
14:43:48 <API> ups
14:43:53 <API> ok, please joanie go on
14:44:05 <joanie> #Action Joseph will provide a patch to fix bug 652500
14:44:05 <tota11y> 04Bug https://bugzilla.gnome.org/show_bug.cgi?id=652500 major, Normal, ---, clown, ASSIGNED, DBusException seen when Orca is running with gnome-shell mag
14:44:23 <joanie> #Action Joanie will roll a new tarball from master once she has Joseph's patch
14:45:00 <joanie> #Action Javi will spin a new (and possibly the last accessibility-specific) spin of the testing distro once he has the new tarball from Joanie and any other required changes requested by our team.
14:45:04 <joanie> (done)
14:45:28 <jhernandez|eee> joanie: thanks!
14:45:42 * joanie wonders if the bot commands are case-sensitive and hopes not :-/
14:45:44 <joanie> oh well
14:45:45 <mgorse> Hi all. Sorry I'm late.
14:46:09 <API> ok
14:46:11 <API> so as I said
14:46:14 <API> jhernandez|eee, and this leads to next subitem in this topic
14:46:33 <API> jhernandez|eee, any update about conversations with crozat/peterS?
14:47:11 <jhernandez|eee> not so much, but,
14:48:25 <jhernandez|eee> I wanted to hear a11y feelings about the possibility of only to provide our own tarballs instead of tarballss ad-hoc
14:48:43 <joanie> jhernandez|eee: I don't get the question
14:48:53 <clown> what are "ad hoc" tarballs?
14:48:57 * joanie nods
14:49:06 <jhernandez|eee> don't know how will I manage this thing with not a11y mods
14:49:09 <jhernandez|eee> well
14:49:30 <jhernandez|eee> sorry, maybe ad-hoc isn't the right word
14:49:57 <jhernandez|eee> we are providing some code not included in master
14:50:11 <jhernandez|eee> don't know if we actually have this situation
14:50:28 <jhernandez|eee> don't know if all API changes were pushed to gnome-shell master, for example
14:50:32 <API> well, yes, one of the ideas of the distro are try some patches not committed yet
14:50:43 <jhernandez|eee> yes
14:51:00 <clown> noting that patches for brightness/contrast effects in gnome-shell are not committed to master.
14:51:06 <API> so you mean that crozat and peters only want tarballs with changes on the master?
14:51:37 <joanie> (or are you just wondering how much of a pain it will be when everyone says, "and apply this patch and that patch Javi!"?)
14:52:04 <jhernandez|eee> API totally not, but it couyld be
14:52:13 <jhernandez|eee> joanie: yes
14:52:15 <joanie> :-)
14:52:18 <joanie> so....
14:52:36 <joanie> why not make a 'no patches' rule?
14:52:50 <clown> jhernandez|eee:  would it help if we provide tarballs where patches have already been applied?
14:52:50 <joanie> i.e. you include from master (or whatever) OR the provided tarball
14:52:55 <API> well, in our case we mostly were using the testing distro to test quickly some patches that were still reviewed ...
14:53:12 <API> joanie, I'm somewhat lost now
14:53:12 <jhernandez|eee> joanie: maybe we can keep a wiki for that, but I'll need some help for manage this if this causes lot of work
14:53:28 <joanie> jhernandez|eee: tell me what needs managing?
14:53:36 <API> what means this "not patches rule "
14:53:50 <API> ?
14:53:51 <joanie> API tarballs only
14:54:01 <clown> API, I think we don't want to force jhernandez|eee to apply the patches himself.  Right?
14:54:02 <joanie> i.e. if clown wants a patched version of Orca
14:54:08 <joanie> Joanie rolls a tarball with that patch
14:54:10 <API> and those tarballs with things from the master
14:54:17 <API> or they can also have custom patches?
14:54:23 <jhernandez|eee> this is my reason, if lot of people starts to providing patches, not sure if I can manage all these changes
14:54:27 <API> custom patches == patches being reviewwed
14:54:42 <joanie> jhernandez|eee: but here's what I'm trying to sort out:
14:55:01 <joanie> Is the problem not having packages OR is the problem applying patches?
14:55:06 <joanie> i.e. are tarballs a pain?
14:55:29 * clown waves at korn
14:55:33 <jhernandez|eee> joanie: I think that if there's no lot of people sending patches, it will be easy
14:55:44 <joanie> jhernandez|eee: that wasn't what I asked you. :-)
14:55:50 <jhernandez|eee> so, maybe we can start as we're doing
14:55:53 <joanie> are tarballs a problem?
14:56:12 <jhernandez|eee> not tarballs
14:56:23 <joanie> and to be clear: patches ARE the problem, yes?
14:56:25 <korn> Hi clown...
14:56:25 <jhernandez|eee> in fact, tarballs are the best for me
14:56:30 <joanie> awesome
14:56:46 <joanie> so my proposal is this: Javi does NOT include patches. Period.
14:56:53 <joanie> IF someone has a patch they want to include
14:57:01 <API> ok, but those tarballs can be custom tarballs, right?
14:57:05 <joanie> that someone needs to sort out how to get a tarball created
14:57:08 <joanie> API: yes
14:57:11 <API> ok
14:57:12 <jhernandez|eee> API and joanie yes
14:57:14 <jhernandez|eee> ok
14:57:14 <jhernandez|eee> ok
14:57:17 <joanie> i.e. in clown's case, I would roll a tarball
14:57:27 <joanie> jhernandez|eee: Does that address your concerns? (really)
14:57:36 <joanie> if it doesn't we'll find a different option
14:57:41 * clown notes that there is a #agreed "command".
14:57:43 <jhernandez|eee> ok
14:57:59 * API thinks that clown suggestion is a good idea
14:58:12 <joanie> clown's suggestion?
14:58:17 * joanie scrolls back
14:58:20 <clown> joanie: use #agreed
14:58:23 <joanie> oh
14:58:23 <joanie> lol
14:58:24 <API> yep that one
14:58:32 <API> so, now that we debated this
14:58:34 <joanie> I thought we meant with tarballs v patches
14:58:48 <API> jhernandez|eee, could you summarize all these with some #info and #agreed?
14:58:56 <API> or there is something not discussed yet?
14:59:04 <joanie> (and write in sentences and use proper names?)
14:59:05 <jhernandez|eee> API, ok I'll try
15:00:22 <API> jhernandez|eee, thakns
15:00:25 <API> thanks in fact
15:00:36 * clown alas joanie, "all commands are case sensitive" (http://meetbot.debian.net/Manual.html#commands)
15:01:23 <joanie> clown: no worries. I will clean up the minutes. Thanks for looking it up!
15:01:37 <joanie> or I'll hack our meetbot :-P
15:01:42 <clown> whoops, I totally misread that.  it says "INsensitive"
15:01:52 <joanie> lol
15:02:08 <jhernandez|eee> #info Javi is afraid about how to manage new code into new spins and he asked a11y team if they're all agree with only provide tarballs, not isolated patches.
15:02:26 <jhernandez|eee> ok?
15:02:37 <jhernandez|eee> how should I use agreed?
15:02:40 <joanie> jhernandez|eee: perfect!
15:02:57 <joanie> according to the docs only API can
15:02:58 <clown> #agreed  the a11y team will provide only tarballs to Javi
15:03:03 <jhernandez|eee> ok
15:03:04 * clown oh...
15:03:09 <jhernandez|eee> ok
15:03:22 <API> ok, so anything else in this topic=?
15:03:25 <jhernandez|eee> clown: thanks
15:03:31 <jhernandez|eee> not from my side
15:03:33 <joanie> API: yeah, you need to do the agreed it seems
15:03:35 <clown> API, you need to do the agreed thing
15:03:39 <joanie> http://meetbot.debian.net/Manual.html#commands
15:03:42 <API> ah ok
15:03:50 <API> #agreed  the a11y team will provide only tarballs to Javi
15:03:56 <jhernandez|eee> I hope next week we can talk about next spins
15:04:05 * clown that sounds strangely familiar...
15:04:13 <API> we are not in a jurry
15:04:15 <API> hurry
15:04:21 <API> we are doing a lot of things
15:04:23 * clown never wants to be in a jury
15:04:24 <jhernandez|eee> API: I know
15:04:32 <API> so, as this point is finished
15:04:34 <API> lets move
15:04:48 <API> #topic Gtk+ Deprecations
15:04:55 <API> https://mail.gnome.org/archives/desktop-devel-list/2011-June/msg00078.html
15:05:21 <API> #info some gtk widgets were deprecated: GtkHBox, GtkVBox
15:05:52 <API> #info this affects current gail implementation, and in fact, there are some warnings related to that
15:06:14 <API> anyway, I don't know if this should be included on the gail-to-gtk move
15:06:16 <joanie> #info Joanie opened bugs against Accerciser and Orca to do the needed changes, but this likely affects most of our team's modules.
15:06:21 <API> or should be solved before that
15:06:52 <API> hmm, strange that jjardon didn't created a metabug to solve that on the platform, as he usually does :P
15:07:16 <alibezz> API: should we use -UGTK_DISABLE_DEPRECATED not to affect things so much?
15:07:48 <API> alibezz, well, but thats a temporal solution
15:07:58 <alibezz> API: yes
15:08:03 <alibezz> what i mean is:
15:08:14 <API> I mean that this would allow us to use them during some time, but in the end tose widgets will be removed
15:08:21 <alibezz> will things break if we don't present a solution to that?
15:08:37 <API> alibezz, eventually yes
15:08:39 <alibezz> yes: the best is to really solve things.
15:08:47 <alibezz> but while we don get there
15:08:50 <API> as I said at a moment in the future those widgets would be removed
15:08:55 <alibezz> do we NEED to use this flag?
15:09:05 <API> alibezz, temporaly ;)
15:09:12 <alibezz> good to know :)
15:09:26 <alibezz> as joanie said, there is a bug for that in accerciser
15:09:29 <API> but it is a good thing to add to minutes
15:09:40 <joanie> oh, the numbers?
15:09:45 <joanie> if so one sec
15:09:48 <API> #info meanwhile this is not solved, modules can use  -UGTK_DISABLE_DEPRECATED
15:09:58 <API> what numbers?
15:10:15 <joanie> bug 652486
15:10:15 <tota11y> 04Bug https://bugzilla.gnome.org/show_bug.cgi?id=652486 normal, Normal, ---, accerciser-devel, NEW, Gtk+ H/V Subclass Deprecations
15:10:31 * alibezz hopes to send a patch for that (accerciser one)
15:10:32 <joanie> and bug 652485
15:10:32 <tota11y> 04Bug https://bugzilla.gnome.org/show_bug.cgi?id=652485 normal, Normal, ---, orca-maint, NEW, Gtk+ H/V Subclass Deprecations
15:10:54 <jhernandez|eee> joanie: thanks for opening the bug ;)
15:11:11 <alibezz> just for the record: I intend to work with accerciser bugs after finishing the mallard docs to learn more
15:11:21 <alibezz> I intend to fix this one and some others, too.
15:11:27 <jhernandez|eee> alibezz: we can talk about this laater
15:11:29 <alibezz> joanie even suggested it to me
15:11:42 <jhernandez|eee> since accerciser's still using gtk2
15:11:59 <alibezz> jhernandez|eee: awesome
15:12:02 <joanie> well, I think even in gtk2 you can use the alternatives
15:12:14 <joanie> but like jhernandez|eee said, topic for a different place
15:12:17 <alibezz> yes
15:12:24 <jhernandez|eee> joanie: yes, but, I'm porting to gtk3
15:12:46 <jhernandez|eee> and this will be done at the same time
15:12:51 <joanie> jhernandez|eee: gotcha
15:13:07 <jhernandez|eee> k
15:13:11 <jhernandez|eee> :]
15:13:17 <alibezz> oooooh makes more sense.
15:13:24 <alibezz> this may be just a gtk+ 3 issue
15:13:40 <alibezz> so maybe it doesn make sense in accerciser current branch
15:14:02 <alibezz> good point I hadnt noticed. thanks.
15:14:07 <jhernandez|eee> we can do that, but, we're going to port to gtk ASAP
15:14:14 <jhernandez|eee> gtk3
15:14:17 <jhernandez|eee> :]
15:14:23 <joanie> so API moving on?
15:14:30 <API> well, it makese sense
15:14:33 * jhernandez|eee nods
15:14:45 <API> jhernandez|eee, alibezz do you want to summarize on a #info something of what you said?
15:15:31 <alibezz> API: yes
15:16:13 <alibezz> #info after porting to gtk3, jhernandez may count on alibezz's help to fix bug 652485
15:16:13 <tota11y> 04Bug https://bugzilla.gnome.org/show_bug.cgi?id=652485 normal, Normal, ---, orca-maint, NEW, Gtk+ H/V Subclass Deprecations
15:16:13 <jhernandez|eee> I was checking last joanie's #info
15:16:49 <API> ok, so moving on
15:16:52 <jhernandez|eee> alibezz: ;)
15:17:08 <API> on the agenda is also included others?
15:17:17 <API> well, glib also make some deprecations,
15:17:33 <API> but not sure if they would affect final apps or just base libraries
15:17:41 <API> so not sure if it is worth to talk about them here
15:17:54 <joanie> API I just added others because it sounds like the gtk folks are up to all sorts of changes which we might not be aware of but should be
15:18:00 <API> so lets move to next topic
15:18:15 <API> joanie, well if we are not aware, we can't talk about them :P
15:18:28 <joanie> API but you might be aware of them
15:18:44 <API> well, in the case of gtk
15:18:48 <API> is just those
15:19:07 <API> at least is the one affecting the gnome 3.1.2 release that I'm doing
15:19:20 <API> the big problem is with G_CONST_RETURN deprecation
15:19:28 <API> and as I said, not sure if they would affect final apps
15:19:32 <API> so lets wait
15:19:33 * joanie nods
15:20:03 <API> #info, there are other deprecations, like in glib, but for the moment not clear if this will affect final apps or just base libraries
15:20:18 <API> #info we will wait, and talk about it again in the future
15:20:20 <API> so moving
15:20:35 <API> #topic Action item updates
15:21:21 <API> anything to add here?
15:21:22 <API> joanie, ?
15:21:44 <joanie> Well, I'd like updates on Mike's and Javi's especially
15:21:51 <joanie> mgorse's is:
15:21:56 <joanie> Modify atk-bridge not to listen for atk signals and send out events unless an AT is listening. (This will likely involve modifying at-spi2-core as well, to add a way of tracking active ATs.)
15:22:10 <joanie> jhernandez|eee's is:
15:22:14 <joanie> Investigate Matthias' statement that when he tries to run Accerciser in jhbuild on his system, it "locks up solid after a few minutes every single time."
15:22:42 <jhernandez|eee> my turn
15:23:12 <jhernandez|eee> I was investigating this, and I only had a only solid lock up
15:23:25 <jhernandez|eee> I son't know if you're experimenting this
15:23:59 <jhernandez|eee> but from my side, my only clue is this
15:24:01 <jhernandez|eee> ** (accerciser:29792): WARNING **: at-spi: Got no message at atspi-accessible.c line 794
15:24:09 <jhernandez|eee> and then it locked up
15:24:40 <jhernandez|eee> I need to trace from where this problem is
15:24:45 <joanie> I wonder if Matthias is getting that same error.
15:25:10 <jhernandez|eee> Mathhias said to me that he didn't get anhy errors
15:25:17 <mgorse> I think that means it tried to make a call and didn't get a reply
15:25:22 <jhernandez|eee> so, don't really know if it's the  same issue
15:25:45 <alibezz> did you make sure you're using the same jhbuild versions?
15:25:57 <jhernandez|eee> mgorse: It could be a at-spi-dbus fall?
15:26:19 <jhernandez|eee> mgorse: I'm using all from scratch, and in fedora 15, as Matthias
15:26:53 <jhernandez|eee> s/mgorse/alibezz in last comment
15:26:54 <jhernandez|eee> :P
15:27:19 <jhernandez|eee> from scratch, I mean, masters
15:27:31 <joanie> Given the time I'm not sure if we should sort out the exact problem here.
15:27:52 <joanie> I think what's important is that jhernandez|eee is working on it. (yay!)
15:27:56 * clown could go longer if needed
15:28:04 <jhernandez|eee> joanie: totally agree, I was only exposing my info about this
15:28:17 <joanie> jhernandez|eee: I know. :-)
15:28:26 <joanie> if you could info a summary it would be awesome
15:28:42 <joanie> and then later in #a11y perhaps we could all talk about it further?
15:28:50 <API> oookkk
15:28:57 <API> so, after all those bunch of lines ...
15:29:06 <API> can I ask for come pretty #info ? ;)
15:29:22 * joanie grins
15:29:23 <jhernandez|eee> should I give a #info about this?
15:29:29 * joanie laughs
15:29:33 <joanie> yes please jhernandez|eee
15:32:13 * clown just checking that I'm still connected (awfully quit in here).
15:32:20 * clown "quiet"
15:32:23 <jhernandez|eee> #info Javi's working on Matthias' reported issue. Javi had only a solid lock up with jhbuild, but he needs to investigate more about this, because isn't clear in how to make reproducible
15:32:50 <jhernandez|eee> #info we have a clue to start from
15:32:56 <jhernandez|eee> done
15:33:19 <joanie> thanks jhernandez|eee!
15:33:22 * jhernandez|eee apologies for to be a little slow
15:33:24 <jhernandez|eee> :S
15:33:25 <jhernandez|eee> :P
15:33:31 <jhernandez|eee> joanie: ;)
15:34:00 <joanie> mgorse: how are things coming on modifying the bridge to be quiet when no AT is listening?
15:34:55 <mgorse> joanie: I still haven't done that. Sorry. Been working on some refactoring that fregl needs. I'll look into it once I have that code in a branch.
15:35:47 <API> joanie, mgorse could you write that as a #info?
15:35:57 <joanie> both of us?
15:35:59 <joanie> ;-)
15:36:27 <API> well, joanie made the question, mgorse the answer
15:36:33 <API> so there are two #info here ;)
15:36:37 <mgorse> #info mgorse hasn't yet added the "do nothing if no one is listening" to atk-bridge that he said he'd have done for 3.1.3
15:37:11 <joanie> thanks mgorse
15:37:23 <joanie> miscellaneous time or end of meeting API?
15:37:48 <API> no, lets wait for jhernandez|eee #info
15:37:55 <API> and after that ending meeting
15:38:01 <joanie> API: but he did it
15:38:02 <API> so today, no miscellaneous time
15:38:04 <joanie> (right?)
15:38:06 <jhernandez|eee> mm
15:38:14 <jhernandez|eee> I did it
15:38:15 <API> ah, I was waiting for a second part :P
15:38:16 <API> ok
15:38:22 <API> so, 10 minutes over time
15:38:29 <API> sorry, no miscellaneous time
15:38:36 <API> http://www.sadtrombone.com/
15:38:40 <API> #endmeeting