Conversation with #a11y-meeting

(07:05:24) joanie: Well, since I'm in charge today, I say let's get started :-)
(07:05:30) ***joanie bangs a gavel
(07:05:38) joanie: http://live.gnome.org/Accessibility/Meetings
(07:05:47) joanie: AT-SPI/Atk API changes
(07:05:58) joanie: So, API opened a metabug
(07:06:05) joanie: https://bugzilla.gnome.org/show_bug.cgi?id=638537
(07:06:22) joanie: and has been going through existing bugs, filing new ones, etc.
(07:06:37) joanie: I've got a few on my mental to-do/to-add list as well
(07:06:49) joanie: and anyone else who has any bugs should add them
(07:06:51) ***tbsaunde says hi
(07:07:17) mgorse: I have an additio or two that I want to propose as well but need to write the list first / file a bug
(07:07:32) joanie: mgorse: awesome
(07:07:56) joanie: Personally, I think if we give this a ton of thought and prep now and then have the proposed hackfest in Spring we can make a lot of great and needed improvements
(07:08:08) joanie: Speaking of the hackfest.... 
(07:08:17) li: I saw API's work these days, several bugs has been closed
(07:08:18) li: agree
(07:08:32) joanie: API and I still need to come up with alternative dates, check with Igalia, etc.
(07:08:41) joanie: but since there are three attendees in the room
(07:08:54) joanie: late April/early May
(07:09:02) joanie: are there any weeks that are just impossible?
(07:09:22) mgorse: not for me afaik
(07:09:33) joanie: li?
(07:09:38) li: let me check
(07:10:40) joanie: There seem to be no Spanish holidays in May (how odd ;-) )
(07:10:40) li: no
(07:10:51) li: late April/early May is OK for me
(07:11:11) joanie: There are the week of the 18th
(07:11:26) joanie: and the week of the 25th is 3.0.1 release
(07:11:44) joanie: I'm leaning towards the first week of May myself, 
(07:11:49) joanie: as one of the proposed weeks
(07:12:19) joanie: if that works for mgorse and li, I'll toss that out to API
(07:12:32) joanie: so we can then toss it out to the broader group
(07:13:14) joanie: Oh, one more thing on the AT front
(07:13:30) joanie: mgorse: did you see about getting AT-SPI2 an official gnome bugzilla component?
(07:13:46) joanie: li: Is that something you can help with (or just do)? 
(07:13:49) mgorse: joanie: I just filed BGO#638794
(07:13:55) ***joanie looks
(07:14:32) joanie: cool thanks mgorse
(07:15:17) joanie: just CC'ed me and API
(07:15:19) mgorse: Also, when I was on the Open A11y call, it occurred to me that it might be good to involve Pete Brunet if we can, since it would be nice to keep ATK/AT-SPI and IA2 in synch as much as possible
(07:15:30) li: I can create a category in at-spi I think, for at-spi2
(07:15:32) ***joanie smiles tiredly
(07:16:02) joanie: li that is what I was thinking too. But maybe it makes sense as its own product?
(07:16:04) joanie: I dunno
(07:16:59) mgorse: I think a separate product would make sense, since AT-SPI has multiple components already
(07:17:31) mgorse: although I left the bug report ambiguous partly because I'm not entirely clear on what can and cannot be done with the existing product/components
(07:17:31) li: currently not only at-spi, java atk wrapper's bugs are also triaged in at-spi category
(07:17:51) joanie: should they be, or should that be split out as well?
(07:18:13) mgorse: Yeah, that seems strange for me, at least for java-atk-wrapper, since it does not touch AT-SPI
(07:19:02) joanie: so what do you think Li
(07:19:03) joanie: ?
(07:19:17) li: just checked
(07:19:32) li: I can create and delete a category in at-spi
(07:19:48) ***joanie nods
(07:19:59) li: so maybe we put at-spi bugs in one category and put at-spi2 bugs in another?
(07:20:32) li: this may be an easier way to do this
(07:20:57) joanie: I (personally) have no opinions either way on whether it be a separate product, or just a component (with good descriptions distinguishing the two)
(07:21:20) li: mgorse, I think the reason they put java atk wrapper into at-spi is it is a kind of bridge like atk-bridge
(07:21:20) joanie: but I agree li that if we just want it done, you can do it faster than the bugzilla gods :-)
(07:22:00) mgorse: Also, I'm not sure exactly how bugzilla is linked to git and GNOME module releases, so we need to make sure that is hooked up correctly
(07:22:22) mgorse: Ie, running install-module automatically creates a new bugzilla version, I think, or it can
(07:22:35) joanie: mgorse: it can, but I always do it manually
(07:22:46) joanie: and install-module just accepts that it's already there
(07:22:57) li: mgorse, yes, I think the install-module do the job
(07:22:59) joanie: I don't know what it would attempt to do if it didn't find a matching number
(07:23:15) joanie: one way to find out. Release is Monday. *grins*
(07:23:17) mgorse: Okay. Not sure if that's really an issue to worry about or not then
(07:23:45) joanie: so what's the conclusion?
(07:23:59) joanie: Try the component route or wait for the product route?
(07:24:33) mgorse: Can we edit existing component descriptions, ie, to make clear that the old ones refer to AT-SPI-CORBA?
(07:24:47) joanie: I'm nearly certain that we can mgorse
(07:24:52) li: consider the version problem, I think maybe the product route is better
(07:25:06) li: mgorse, yes
(07:25:53) li: but I can create a new component for temporary solution
(07:26:49) li: what do you think?
(07:26:59) joanie: so mgorse what if Li does that (with really clear descriptions) *and* you keep your bug open to get a new product (or at least start that discussion)?
(07:27:19) joanie: if this were like Gtk, I'd say we should hold off
(07:27:23) jmatsuzawa [~matsu@p28097-ipbffx02osakakita.osaka.ocn.ne.jp] entered the room.
(07:27:41) joanie: but it's not like tons of people are going to get confused if we start it out as a component and then it becomes a product
(07:27:47) joanie: but that's just my opinion
(07:28:24) mgorse: joanie: That seems fine with me; I think that either way is fine really as long as we can specify correct versions and it's clear to users which component/product they should select
(07:28:34) ***joanie nods
(07:28:43) joanie: sounds like we have a plan then
(07:28:53) li: several people already got confused, I have seen some at-spi2 bugs are reported in at-spi category
(07:29:18) joanie: I might be tempted to add a brief "how to tell which version you're using" blurb
(07:29:30) joanie: in both at-spi and the to-be-created at-spi2
(07:29:35) mgorse: li: I might not even have seen them then; I haven't been "watching" you or anything
(07:29:35) li: mgorse, do you think we need 3 new components or just one?
(07:30:15) li: mgorse, I reported them on freedesktop.org I think
(07:30:36) mgorse: li: I don't think it matters much, but having a few seems to make sense since that is what we have for at-spi-corba
(07:31:26) li: mgorse, OK
(07:31:52) mgorse: and they are separate modules
(07:32:48) ***joanie wonders if we're set on this front?
(07:33:08) li: I think so
(07:33:36) joanie: mgorse: yes?
(07:33:56) mgorse: yeah, I think we're set
(07:34:03) joanie: awesomesauce thanks guys!
(07:34:28) joanie: So I'll skip over the unity update on the assumption that no one here has any knowledge to do the update
(07:34:45) joanie: GNOME 3.
(07:34:57) joanie: which we're sorta already talking about
(07:35:15) joanie: but I'm still wondering about the PyGObject stuff
(07:35:24) joanie: API posted something on desktop-devel-list
(07:35:32) joanie: http://mail.gnome.org/archives/desktop-devel-list/2011-January/msg00084.html
(07:35:38) joanie: asking about it
(07:35:58) joanie: And then Colin responded here: http://mail.gnome.org/archives/desktop-devel-list/2011-January/msg00118.html
(07:36:15) joanie: saying, essentially, don't worry about it (all that much)
(07:36:34) joanie: and the final word on it was from a maintainer who is now caught in the middle: http://mail.gnome.org/archives/desktop-devel-list/2011-January/msg00119.html
(07:36:54) joanie: I was hoping API would be here to shed light on what we should conclude
(07:37:13) joanie: at least for Orca, we have funding from the Spanish government to become better GNOME 3.0 citizens
(07:37:18) joanie: including this conversion
(07:37:26) joanie: but it doesn't seem like we want to do anything yet
(07:37:35) joanie: dunno if anyone here has insight
(07:38:38) joanie: thoughts?
(07:38:42) joanie: Or wait for API?
(07:39:21) mgorse: I wonder when this pygobject hackfest will happen. Maybe best to table this for now at least
(07:39:44) joanie: I dunno, but I can google for the answer
(07:39:44) jmatsuzawa1 [~matsu@p28097-ipbffx02osakakita.osaka.ocn.ne.jp] entered the room.
(07:39:55) joanie: regardless, what you suggest (tabling it) makes sense
(07:40:18) joanie: I'm just personally a little worried as being (even more) seen as a "bad" module for not living up to the GNOME 3.0 vision :-/
(07:41:16) joanie: http://live.gnome.org/Hackfests/Python2011
(07:41:28) joanie: looks like in a couple of weeks (17-21 Jan)
(07:41:32) mgorse: Ultimately that shouldn't affect end users, since distributions make the final decision about what they ship, not GNOME, but I see what you're saying
(07:42:01) joanie: Oh I know re distros. I'm not worried about users. I'm worried more about the politics
(07:42:12) ***mgorse nods
(07:42:34) joanie: but given how soon this hackfest is, I agree with mgorse's proposal to table this topic until we see what they come up with.
(07:43:15) joanie: So the other agenda items in this area are the GtkTreeview refactoring and "other"
(07:43:21) joanie: I have nothing to add on either subject
(07:43:25) joanie: anyone else?
(07:43:37) li: I plan to look at this
(07:44:10) li: is there any web page or mail about this refactoring?
(07:44:21) joanie: lemme check. I think there was
(07:45:19) joanie: I've not found it yet
(07:45:29) joanie: I'm not as good as API in this regard :-/
(07:45:48) joanie: li Didn't API email you directly about it?
(07:46:37) li: yes
(07:47:06) li: but not many information about the refactoring
(07:47:10) joanie: gotcha
(07:47:21) joanie: Well, I'll look around
(07:47:32) joanie: as I understand it, however, it's in Gtk3 master now
(07:47:45) mgorse: The gist of it is that tree views have been internally refactored a bunch in GTK 3, and they tried to make sure that gail would still work, but someone needs to test it
(07:48:26) li: mgorse, OK
(07:48:33) joanie: And this is why and where we need more automated testing tools
(07:48:50) joanie: hopefully what Brian Nitz is working on can ultimately fulfill that role
(07:49:07) joanie: but that doesn't help us now -- unless he has something internal he can share
(07:49:13) li: I just tried to find some information which can help me understand the code change quicker
(07:49:20) ***joanie nods
(07:50:07) li: I will try to find some information later
(07:50:14) joanie: okay
(07:50:22) joanie: if all else fails li, please ping API
(07:50:33) joanie: he's been on top of this more than most of us
(07:50:34) li: joanie, sure
(07:50:55) joanie: So anything else on the broad subject of "GNOME 3.0"?
(07:51:58) joanie: If not.... Upcoming events. We've already talked about Atk Hackfest. Steve is not here to talk about Project:Possibility. So that would leave CSUN.
(07:52:28) joanie: And Eitan is not here to talk about that
(07:52:41) li: Does Eitan plan to go to CSUN?
(07:52:45) joanie: yes
(07:52:49) joanie: he is presenting 
(07:52:53) joanie: ditto for Bryen
(07:53:16) joanie: and Aleiva (whom you still haven't met) is thinking he might
(07:53:27) li: I didn't fully understand the last meeting minutes, do you plan to go? joanie
(07:53:27) joanie: mgorse: you're thinking about it or planning on it?
(07:53:42) joanie: li well, the short version of the story is this:
(07:53:58) mgorse: joanie: I think I'll be there, since we have to send someone
(07:54:13) joanie: Chris Hofstader (the newish GNU a11y guy) proposed a panel. I agreed to be on it. Chris said he thought he could fund me
(07:54:14) mgorse: we = the Mono a11y team
(07:54:21) joanie: Then Chris said, no panel
(07:54:30) joanie: and we started talking about the Atk Hackfest
(07:54:38) joanie: Then Chris said, oh, I mean panel
(07:54:39) joanie: :-)
(07:54:42) joanie: So....
(07:55:00) joanie: If there is a panel with me on it, I will fly out, do the 1 hour session and fly back 
(07:55:06) joanie: I don't plan to stay for CSUN
(07:55:28) li: so complicated :)
(07:55:35) joanie: oh, and as an aside, panel BUT no funding for me to travel
(07:55:37) joanie: lol
(07:56:03) joanie: and we don't have the budget (nor can I take the time off from my dayjob) to do CSUN and Spain
(07:56:08) joanie: so.... I dunno....
(07:56:20) joanie: that's what credit cards are for I guess :-)
(07:56:31) joanie: anyhoo.... Moving on to API's "miscellaneous time"
(07:56:31) li: :)
(07:56:34) joanie: floor is open
(07:58:53) joanie: No one has anything to add?
(07:59:19) li: I have added the components into at-spi
(07:59:25) li: on bugzilla
(07:59:26) joanie: yay! :-)
(07:59:33) mgorse: li: Great. Thanks
(07:59:59) li: :)
(08:00:07) mgorse: li: Can you set it up so that I get emailed or assigned for at-spi2 bugs?
(08:00:18) li: mgorse, sure
(08:01:49) li: mgorse@alum.wpi.edu, this mail address?
(08:03:26) li: what's your id on bugzilla?
(08:03:37) mgorse: li: mgorse@novell.com I think
(08:04:31) li: right, done
(08:04:59) joanie: li, just looking at this http://git.gnome.org/browse/gtk+/log/?qt=grep&q=treeview I think the treeview is *still* a work in progress :-)
(08:05:52) joanie: but mostly i guess it's been stable commit-wize over the past week or two
(08:06:35) li: joanie, yeah
(08:06:44) joanie: anyhoo, 6 minutes over. So officially I'm calling this meeting to a close
(08:06:49) joanie: thanks for coming
(08:06:59) joanie: hopefully next week API will be back
(08:07:00) joanie: :-)