Attachment '20100930.txt'
Download 1 Sep 30 09:18:40 * Now talking on #a11y
2 Sep 30 09:18:52 <API> hi, sorry the delay, traffic jam
3 Sep 30 09:19:11 <API> aleiva, has the meeting started?
4 Sep 30 09:19:12 <mgorse> Hi
5 Sep 30 09:19:25 <API> mgorse, hi
6 Sep 30 09:19:33 <fer> we were waiting for you api :)
7 Sep 30 09:19:52 * API fells guilty
8 Sep 30 09:20:43 <API> well, although AEGIS things are mostly finished, we can start with it
9 Sep 30 09:20:49 <API> bnitz, you here?
10 Sep 30 09:21:20 * slee (~chatzilla@stevelee.gotadsl.co.uk) has joined #a11y
11 Sep 30 09:23:39 <API> well, it seems that no
12 Sep 30 09:24:10 <API> just to comment that we will probably rearrange the testing session
13 Sep 30 09:24:14 <API> http://live.gnome.org/Accessibility/HackfestAEGIS2010
14 Sep 30 09:24:27 <API> to start later that day
15 Sep 30 09:25:04 <API> btw, aleiva you here?
16 Sep 30 09:25:40 <API> he already made the hotel book, probably it would be a good idea to upload that on the wiki, so people could know his
17 Sep 30 09:25:45 <API> assigned mate
18 Sep 30 09:26:11 <fer> no aleiva either...
19 Sep 30 09:26:12 <mgorse> Are the rooms booked under our names, so we can just go to the desk at the hotel and give our name and we get the room?
20 Sep 30 09:26:17 <API> probably roommate is a better word
21 Sep 30 09:26:35 <API> mgorse, yes afaik
22 Sep 30 09:26:53 <API> ie, joanie is already flying to seville
23 Sep 30 09:27:34 <mgorse> Oh, yeah, right--that's why she's not here :)
24 Sep 30 09:28:04 <API> yeah, as Bryen, that he is already on Spain, although afaik te started with Barcelona
25 Sep 30 09:28:36 <API> well, lets move, ask about the hackfest if you have any question
26 Sep 30 09:28:50 <API> so, next in agenda would be GNOME 2.32 updates
27 Sep 30 09:29:06 <API> fer, recently I read a mail related to problems with gnome-mag
28 Sep 30 09:29:12 <API> did you read it?
29 Sep 30 09:29:20 <fer> yeah, I just got an anwer from Li
30 Sep 30 09:29:26 <fer> it's on the stable branch
31 Sep 30 09:29:36 <fer> so I should take a look at it for gnome 2.32
32 Sep 30 09:29:52 <fer> (which I totally ignored because I was not happy with such a release)
33 Sep 30 09:30:17 <API> fer, you mean gnome-mag 2.32 release or just gnome 2.32 release in general?
34 Sep 30 09:30:36 <fer> well, gnome-mag from stable branch is working fine over my stock gnome 2.30
35 Sep 30 09:30:51 <fer> (0.16.1)
36 Sep 30 09:32:05 <fer> although Li reported the bug to be present on gnome 2.30 branch
37 Sep 30 09:32:14 <fer> but I'll definitively take a look at it
38 Sep 30 09:32:43 <API> ok,
39 Sep 30 09:33:20 <API> well, someone wants to comment something else about gnome 2.32 ?
40 Sep 30 09:35:37 <API> ...
41 Sep 30 09:35:55 <API> so lets move to gnome 3.0, something forgotten the last weeks
42 Sep 30 09:36:33 <API> fer, afaik, there will be a lot of changes on gnome-mag on gnome 3.0 with respect to this gnome 2.32 release?
43 Sep 30 09:36:44 <fer> yup
44 Sep 30 09:37:07 <fer> dbus vs. bonobo and cairo drawing vs. old gtk2 gdk drawing
45 Sep 30 09:38:04 <API> gnome-mag used a lot of gtk2 gdk drawing?
46 Sep 30 09:41:01 <fer> yeah, and in a very bad way
47 Sep 30 09:41:08 <API> bad way?
48 Sep 30 09:41:14 <fer> very unoptimized
49 Sep 30 09:41:35 <fer> things like copy the cursor pixmap in every drawing instead of having it cached
50 Sep 30 09:43:28 <API> so you want to make the move to cairo and optimize the drawing in one shot?
51 Sep 30 09:44:07 <fer> not me, Benjamin Otte :)
52 Sep 30 09:44:42 <fer> who is the one screaming how bad the gdk drawing based code was :)
53 Sep 30 09:45:40 <API> fer, so Benjamin Otte will work on gnome-mag also?
54 Sep 30 09:46:17 * tomeu wants him to fix wnck first
55 Sep 30 09:46:22 <fer> He did a initial bunch of stuff, but I still need to poke him to get it finished
56 Sep 30 09:46:41 <fer> haha, it looks like he is _the_ guy these days
57 Sep 30 09:47:30 <tomeu> he has stolen the spotlight from mezcalero :p
58 Sep 30 09:48:09 <API> ok, thanks fer
59 Sep 30 09:48:21 <API> I guess that this is enough gnome-mag for today ;)
60 Sep 30 09:48:54 <API> so the other star thinking on gnome3 is at-spi2
61 Sep 30 09:49:02 <API> mgorse, something to say about it?
62 Sep 30 09:50:09 <mgorse> Mostly there are performance issues to be worked out afaik
63 Sep 30 09:50:46 <mgorse> I've been working on trying to have it use dbus connections, which should help some. Made some progress on that yesterday (but need to figure out if dbus-glib is stil actually being maintained, since I fixed a bug in it)
64 Sep 30 09:51:00 <mgorse> s/dbus connections/direct dbus connections/
65 Sep 30 09:51:26 <mgorse> so I'll need to see how much of a difference that makes once it's al working
66 Sep 30 09:52:12 <API> just to add a anecdote, have you seen michael meeks post about dbus performance?
67 Sep 30 09:52:28 <API> this post was a reaction about someone adding kernel support
68 Sep 30 09:52:37 <API> in order to improve dbus performance
69 Sep 30 09:53:04 <mgorse> Not sure. Is that something that was written a few years ago when people were evaluating dbus?
70 Sep 30 09:53:20 * msanchez (~msanchez@44.24.60.213.dynamic.mundo-r.com) has joined #a11y
71 Sep 30 09:53:22 <mgorse> I'm also wondering how performant gdbus is compared to libdbus
72 Sep 30 09:53:40 <tomeu> hmm, gdbus is not using libdbus?
73 Sep 30 09:53:49 <mgorse> (and someone also filed bug against at-spi2 since it is using dbus-glib)
74 Sep 30 09:54:00 <tomeu> that would mean the client side of the protocol was reimplemented in gdbus
75 Sep 30 09:54:28 <mgorse> tomeu: afaik it's a complete reimplementation
76 Sep 30 09:55:11 <API> mgorse, so gdbus and libdbus are two different implementations of dbus?
77 Sep 30 09:55:25 <mgorse> API: I believe so
78 Sep 30 09:55:43 <API> mgorse, there are any reason of this reimplementation?
79 Sep 30 09:55:45 <tomeu> looks like
80 Sep 30 09:55:45 <API> I mean
81 Sep 30 09:55:49 <API> something like
82 Sep 30 09:55:54 <tomeu> gvariant made easy reimplementing the marshalling
83 Sep 30 09:55:56 <API> different approachs
84 Sep 30 09:56:05 <API> or "old implementation is bad"
85 Sep 30 09:56:09 <API> or whatever
86 Sep 30 09:56:27 <mgorse> Apparently it was designed to be thread-safe
87 Sep 30 09:56:30 <API> tomeu, they made a dbus reimplementation just to use gvariant?
88 Sep 30 09:56:36 <tomeu> I guess desrt wanted to reuse gio and gvariant
89 Sep 30 09:56:54 <tomeu> API: well, there was consensus that dbus-glib wasn't the way forward
90 Sep 30 09:57:03 <tomeu> so they needed a new API in glib anyway
91 Sep 30 09:57:20 <tomeu> as they had already gio and gvariant, using libdbus would duplicate quite a bit of that
92 Sep 30 09:57:23 <API> yeah, but as you said before
93 Sep 30 09:57:25 <API> <tomeu> hmm, gdbus is not using libdbus?
94 Sep 30 09:57:50 <tomeu> yeah, just checked the source and it's not using it
95 Sep 30 09:57:56 <tomeu> only for the tests
96 Sep 30 09:59:04 <tomeu> sorry for the confusion
97 Sep 30 09:59:08 <API> mgorse, about michael meeks post, is not something written years ago
98 Sep 30 09:59:10 <API> http://people.gnome.org/~michael/blog/2010-09-17.html
99 Sep 30 09:59:21 <API> it also includes a link to the other post
100 Sep 30 09:59:23 <API> tomeu, np
101 Sep 30 09:59:24 <mgorse> API: Okay; I'll look. Thanks
102 Sep 30 09:59:36 <API> well, we are almost out of time
103 Sep 30 09:59:44 <API> so lets move to "miscellaneous"
104 Sep 30 09:59:58 <API> anyone want to comment anything else?
105 Sep 30 10:00:36 <fer> ah
106 Sep 30 10:00:43 <fer> I have a question regarding aegis
107 Sep 30 10:01:19 <fer> I'm getting the GNOME braille embossed materials for AEGIS today or tomorrow
108 Sep 30 10:01:32 <fer> but I will not arrive to seville until Wed
109 Sep 30 10:01:48 <fer> so, should I bring them with myself or send them before to someone?
110 Sep 30 10:02:28 <API> fer, well Bryen made a strange question the other day, related to try to found out
111 Sep 30 10:02:38 <API> alternative braille printers
112 Sep 30 10:02:59 <API> but peter korn told him that it would be fine to add the material on usb sticks
113 Sep 30 10:03:06 <API> anyway, about your question
114 Sep 30 10:03:17 <API> I guess that you could send it to emergy facilities
115 Sep 30 10:03:27 <fer> well, that was because we were not sure if we could get ONCE printing the material at that time
116 Sep 30 10:03:32 <API> after all the event box will be sent there
117 Sep 30 10:03:39 <API> fer, ah ok, it was not related
118 Sep 30 10:03:45 <fer> ok, I'll coordinate with aleiva later
119 Sep 30 10:03:56 <API> btw, other good question would be ask aleiva if the event box is already on seville
120 Sep 30 10:04:01 <fer> yeah, I mailied both bryen and joanie, but got no answers
121 Sep 30 10:04:13 <API> i don't no the date
122 Sep 30 10:04:21 <API> fer, well, joanie is on flying
123 Sep 30 10:04:35 <API> and Bryen is hard to get in contact when he is in a foreign country
124 Sep 30 10:04:48 <fer> he he
125 Sep 30 10:05:38 * ruoso has quit (Read error: 145 (Connection timed out))
126 Sep 30 10:07:23 <API> fer, ok thanks
127 Sep 30 10:07:29 <API> 5 min over the time
128 Sep 30 10:07:38 <API> I think that we can conclude the meeting
129 Sep 30 10:07:42 <API> again, sorry for arrive late
130 Sep 30 10:07:53 <fer> thanks to you!
131 Sep 30 10:07:59 <fer> see you then in Seville next week!
132 Sep 30 10:08:04 <API> mgorse, tomeu fer thanks individually, as seems that you were the only guys here ;)
133 Sep 30 10:08:41 * tomeu was just the peanuts gallery
134 Sep 30 10:12:02 * slee has quit (ChatZilla 0.9.86 [Firefox 3.5.13/20100914130356])
135 Sep 30 10:15:01 * ruoso (~ruoso@201009122149.user.veloxzone.com.br) has joined #a11y
136 Sep 30 10:16:20 * andre (~andre@g1.blanicka25.net) has joined #a11y
137 Sep 30 10:23:18 <aleiva> uops
138 Sep 30 10:23:20 <aleiva> dorry
139 Sep 30 10:23:24 <aleiva> sorry*
140 Sep 30 10:23:40 <aleiva> I was travelling this week and I forget about the meeting
141 Sep 30 10:27:06 <aleiva> for the record, all the hotel stuff is done
142 Sep 30 10:29:12 <fer> am I the only "single"?
143 Sep 30 10:31:06 <API> aleiva, np
144 Sep 30 10:36:02 <aleiva> API: :-)
Attached Files
To refer to attachments on a page, use attachment:filename, as shown below in the list of files. Do NOT use the URL of the [get] link, since this is subject to change and can break easily.You are not allowed to attach a file to this page.