| Time |
Nick |
Message |
| 15:03 |
meeting |
Meeting started Tue Jul 17 15:03:28 2012 UTC. The chair is manuq. Information about MeetBot at http://wiki.debian.org/MeetBot. |
| 15:03 |
|
Useful Commands: #action #agreed #help #info #idea #link #topic #endmeeting |
| 15:04 |
manuq |
hopefully this will be short |
| 15:04 |
|
so tomorrow we'll come back to design meetings |
| 15:04 |
|
and next dev meeting can be based on design desitions |
| 15:05 |
|
first topic, newcomers? |
| 15:05 |
silbe |
yeah, too bad I can't participate in the first Design Team meeting since almost forever :-/ |
| 15:05 |
manuq |
do we have anyone new in the room? |
| 15:05 |
|
silbe: we can discuss design meeting schedule at the end |
| 15:05 |
jvonau |
hi, just lurking. |
| 15:06 |
silbe |
manuq: ok |
| 15:06 |
manuq |
jvonau: are you new to sugar? sorry if I don't remember you |
| 15:06 |
jvonau |
no, with olpc-au |
| 15:07 |
manuq |
jvonau: and new to dev meetings? |
| 15:07 |
jvonau |
yes, but I read the logs. |
| 15:07 |
danielf |
jvonau: welcome on-board |
| 15:07 |
manuq |
jvonau: ok! do you want to introduce yourself? |
| 15:08 |
jvonau |
Well, been with olpc-au for awhile now and do pretty much all the stuff to put out an image for them.. |
| 15:08 |
humitos |
manuq: don't you should use the "topic" command? (just asking |
| 15:08 |
|
) |
| 15:09 |
manuq |
humitos: thanks |
| 15:09 |
|
#topic newcomers |
| 15:10 |
|
jvonau: ok, I think I've read emails from you in the sugar-devel list |
| 15:10 |
|
guess jvonau is around here before himself |
| 15:11 |
|
ok |
| 15:11 |
|
#topic follow up the shell port (manuq) |
| 15:12 |
|
from last week, the shell starts, activities can be started, the home view is visible |
| 15:13 |
|
issues: other zoom views are not displayed at the moment |
| 15:13 |
humitos |
manuq: good! |
| 15:14 |
manuq |
the journal does not start, nothing in the logs but I'm making progress |
| 15:15 |
|
I'm reporting a couple of bugs to pygobject |
| 15:15 |
|
yesterday I updated my sugar-build to the latest gtk+ from git |
| 15:16 |
|
which have some fixes from carlos garnacho from a few days ago |
| 15:16 |
|
like a fix for the big X in the search box that was affecting Browse too |
| 15:17 |
|
(and the search boxes in the zoom views of the shell) |
| 15:18 |
|
so, that's it, I hope I can do a lot more this week, to arrive to guadec with something more polished |
| 15:18 |
|
any questions? |
| 15:18 |
humitos |
manuq: no a question, but: congratulations! |
| 15:18 |
danielf |
manuq: is there any progress with the gst regressions? (speak wasn't working) |
| 15:19 |
silbe |
manuq: good luck! |
| 15:19 |
danielf |
manuq: espeak^ |
| 15:20 |
manuq |
danielf: so, espeak is not working in sugar-build, because dnarvaez is trying to keep fedora and ubuntu with the same dependencies, and ubuntu doesn't have it |
| 15:20 |
|
dnarvaez: can you correct me if I'm wrong? |
| 15:21 |
|
humitos, silbe, thanks :) |
| 15:22 |
|
hmm maybe he's not around? |
| 15:22 |
|
ok, next topic? more questions? |
| 15:22 |
danielf |
manuq: ok, I have no excuses to follow the Speak activity port, I wait to the next topic to tell more |
| 15:23 |
manuq |
#topic follow up the activities port |
| 15:23 |
|
has seen a lot of work recently |
| 15:23 |
|
danielf, please go on :) |
| 15:24 |
danielf |
About the Speak activity: I got the first Fatal Python Error of my life |
| 15:24 |
manuq |
danielf: did it hurt? :) |
| 15:25 |
|
danielf: do you have something in the logs? |
| 15:25 |
danielf |
a few, now I'll fill the activity with logs to localize the problem |
| 15:25 |
manuq |
is a segmentation fault? |
| 15:26 |
danielf |
manuq: no, the log says Fatal Python Error |
| 15:26 |
manuq |
danielf: and when does it happen? |
| 15:27 |
|
dogi has quit IRC |
| 15:28 |
danielf |
manuq: I started the port, and the activity started, Alice started to speak and in a second the activity finished |
| 15:28 |
manuq |
danielf: great, you found a bug! :) |
| 15:28 |
benzea |
sounds like attaching gdb or even running it trough valgrind might be an idea |
| 15:29 |
manuq |
benzea: yes, danielf: I'm sure someone in IRC can instruct you how to inspect it |
| 15:29 |
|
humitos? ^ |
| 15:29 |
benzea |
though, I guess it is a gstreamer binding problem? |
| 15:29 |
humitos |
manuq: what? I don't know how to use gdb |
| 15:30 |
manuq |
danielf: how are you importing gstreamer |
| 15:30 |
|
? |
| 15:30 |
danielf |
benzea: another thing to do is port the Gst code, there was no conflict, but... |
| 15:30 |
manuq |
humitos: ok, no problem |
| 15:30 |
humitos |
danielf: did you port the gst code? |
| 15:30 |
|
danielf: because I'm not sure what to do with that |
| 15:31 |
danielf |
manuq: No, there was no conflict with pygst |
| 15:31 |
|
humitos^ |
| 15:32 |
manuq |
danielf: so you are using "import pygst" |
| 15:32 |
danielf |
humitos: The Gobject Introspection is much more similar than the C libraries, there isn't docs for Python but there is the official documentation for C |
| 15:32 |
benzea |
won't that load the static pygobject bindings? |
| 15:33 |
danielf |
manuq: import gst and import pygst |
| 15:33 |
manuq |
benzea: I fear that |
| 15:33 |
benzea |
manuq: yup, it will happen ... |
| 15:33 |
humitos |
benzea: yes, I think that "import gst" load the static bindings |
| 15:34 |
benzea |
and then it is not too surprising that it fails at some point ... mostly ugly that it does not fail right away |
| 15:34 |
humitos |
benzea: I'm using "import gst" in Jukebox with Gtk3 and it's working properly |
| 15:34 |
|
benzea: danielf manuq that is something that we should discuss |
| 15:34 |
benzea |
which means, there is no point in trying to debug this further before the bindings are ported |
| 15:35 |
manuq |
so, gnome is porting to GStreamer 1.0 https://live.gnome.org/GnomeGo[…]/PortToGstreamer1 |
| 15:35 |
humitos |
http://wiki.sugarlabs.org/go/F[…]ing/Jukebox#Notes |
| 15:35 |
manuq |
here is a porting guide: https://wiki.ubuntu.com/Novacut/GStreamer1.0 |
| 15:36 |
danielf |
humitos, manuq (manueles): A little detail: My activity ran ugly, I modified some lines which toke errors in the log an then the Fatal Python Error |
| 15:37 |
benzea |
humitos: if it works in jukebox it may simply be that you got lucky ... |
| 15:37 |
humitos |
benzea: :) |
| 15:37 |
|
benzea: that's the meaning of life... lucky :) |
| 15:38 |
garycmartin |
Talking of porting, I noticed some talk of GNOME dropping GConf (it is no longer maintained), is there an official replacement? The folks in Maliit are moving to QSettings, I've never heard of that before. |
| 15:38 |
manuq |
garycmartin: yes, GSettings |
| 15:38 |
silbe |
garycmartin: the replacement used by Gnome is GSettingsd |
| 15:38 |
|
*GSettings |
| 15:39 |
humitos |
yes |
| 15:39 |
garycmartin |
Thanks, OK so I guess we may need to dance around with Maliit and find a way to support it's settings. |
| 15:40 |
manuq |
ok, danielf, can you give a try to the introspection bindings with the links provided for humitos? |
| 15:40 |
|
dogi <dogi!~dogi pool-108-20-104-235.bstnma.east.verizon.net> has joined #sugar-meeting |
| 15:40 |
manuq |
provided by |
| 15:40 |
danielf |
manuq: sure |
| 15:40 |
humitos |
manuq: danielf that would be great! even more, if you can, please document what you did and if it worked or not |
| 15:41 |
manuq |
great |
| 15:42 |
|
this is why keeping regular meetings is good :) |
| 15:42 |
|
anyone else want to raise an activity port report/issue? |
| 15:42 |
|
any other topic? |
| 15:43 |
m_anish |
manuq, i had a question |
| 15:43 |
manuq |
sure m_anish |
| 15:43 |
m_anish |
manuq, so its great to see the design meetings finally start to happen :-) hopefully they'll be regular... Are the 0.98 roadmap dates being identified... |
| 15:43 |
|
manuq, something like this http://wiki.sugarlabs.org/go/0.96/Roadmap |
| 15:45 |
|
manuq, another question. will 0.98 ship gtk2 libs alongside gtk3 ones or will it be gtk3 only? |
| 15:45 |
danielf |
m_anish: 0.98 will be compatible with Gtk2 activities |
| 15:46 |
manuq |
m_anish: thanks for pointing out, yes we should do an schedule for 0.98 soon |
| 15:46 |
|
m_anish: danielf is right |
| 15:46 |
m_anish |
danielf, ah! okay |
| 15:46 |
silbe |
m_anish: we said during the last meeting that sugar-toolkit will continued to be maintained. Maybe what you're more interested in is whether OLPC OS 12.2.0 will ship GTK2 libs? |
| 15:46 |
m_anish |
manuq, okay, thanks |
| 15:46 |
|
silbe, yep, that too! :-) |
| 15:46 |
|
manuq, danielf any word on that? ^^? |
| 15:46 |
|
i knw its more an OLPC question |
| 15:47 |
manuq |
so I'm not the one to ask about 12.2.0 plans, but as it will be based on 0.98, I think it will have gtk2 compatibility |
| 15:47 |
m_anish |
manuq, okay |
| 15:48 |
manuq |
ok! |
| 15:49 |
|
we are close to success keeping this meeting to 1 hr timeframe! |
| 15:49 |
|
anything else? design meeting schedule? |
| 15:50 |
silbe |
manuq: will the GTK3 port for 0.98 leave anything unported? Intro screen, tests, whatever? I.e. do we retain any dependency on GTK2 libraries? |
| 15:51 |
|
cjl has quit IRC |
| 15:51 |
|
cjl <cjl!~chatzillac-98-204-202-184.hsd1.md.comcast.net> has joined #sugar-meeting |
| 15:52 |
manuq |
silbe: well I'm not touching sugar-datastore |
| 15:52 |
silbe |
manuq: no problem, sugar-datastore doesn't have any UI code. |
| 15:53 |
manuq |
silbe: so, the idea is to completly port the shell |
| 15:53 |
silbe |
manuq: ok, great. thanks! |
| 15:53 |
manuq |
silbe: we are not using any compatibility layer, so the port is all at once |
| 15:54 |
silbe |
manuq: perfect. Are you going directly to GSettings or will that be a separate step? |
| 15:54 |
m_anish |
silbe, i had that question as well :-) |
| 15:55 |
manuq |
silbe: I'm keeping GConf by now, GSettings should be a separate step, if keeping GConf doesn't block us |
| 15:55 |
silbe |
manuq: fine with me. |
| 15:56 |
m_anish |
yep, makes sense |
| 15:56 |
manuq |
silbe, m_anish, great if you agree, I was in doubt |
| 15:57 |
silbe |
manuq: If GConf stays around long enough and still works properly, it's good engineering to do it in a separate iteration. The GTK3 port is complex enough already. |
| 15:57 |
m_anish |
manuq, sorry for stretching the meeting... another general question I had, in general builds for the XO laptop have gnome alongside. I wanted to know whether there were any efforts in integrating the settings used by sugar and gnome |
| 15:58 |
manuq |
silbe: yeah |
| 15:58 |
m_anish |
for example, with dsd's NM 0.9 patch i believe the network settings between sugar and gnome are shared |
| 15:58 |
manuq |
m_anish: no problem stretching the meeting :) |
| 15:59 |
|
m_anish: we can discuss that, or wasn't discussed already? |
| 16:00 |
m_anish |
manuq, for example ... http://dev.laptop.org.au/issues/705 |
| 16:01 |
|
manuq, i guess there would be time to submit patches along these lines as they probably won't affect UI ... but it will help immensely if points of integrating settings between gnome and sugar could be identified and worked upon |
| 16:01 |
|
manuq, anyway, i guess i just wanted to raise a point, i dont have an actionable item on it right now |
| 16:01 |
|
manuq, feel free to move to the next topic(s) |
| 16:02 |
manuq |
m_anish: to me, sounds like a good idea, should be proceeded as a feature page as any other contribution to sugar |
| 16:02 |
m_anish |
manuq, okay |
| 16:02 |
manuq |
I have no more topics |
| 16:03 |
|
countdown? |
| 16:03 |
m_anish |
+1 |
| 16:03 |
manuq |
5 |
| 16:04 |
silbe |
from my side only the Design Team meeting schedule is left |
| 16:04 |
manuq |
silbe: yes, please |
| 16:05 |
silbe |
Both as a Sugar maintainer and as a party interested in the Sugar design I'd like to take part in the Design Team meetings. |
| 16:06 |
manuq |
garycmartin and others, can we try to move the design meeting? what about monday 15 utc? works for me if we keep it to 1 hr |
| 16:06 |
silbe |
Unfortunately I'm unavailable on Wednesdays and Thursday and the meeting is currently scheduled for Wed (at least for this week, but the idea probably was to keep that time slot). |
| 16:06 |
garycmartin |
In my email reply to silbe (on the mail list): "We're currently tied into Sugar/OLPC meetings Monday, Tuesday, Friday (around the ~15:00 UTC time slot which seems to be good for many folks time zones). Let's review a regular time slot for the next meeting once I've seen who makes it this week. We could possibly try an earlier meeting on Mondays say at 14:30UTC as Daniel's build release/triage meetings are not until 16:00 UTC." |
| 16:07 |
manuq |
garycmartin: good |
| 16:07 |
silbe |
I have a different meeting on Mon 14 UTC, but if necessary the Design Team meeting will trump it. |
| 16:07 |
|
m_anish too, FWIW |
| 16:07 |
manuq |
garycmartin: so we will make it tomorrow, and maybe we can move to Mondays 14:30 UTC for the next weeks |
| 16:08 |
m_anish |
silbe, yep, design meetings would trump it |
| 16:08 |
garycmartin |
The design meetings used to be run on Mondays during the last cycle. And before that Sundays. It's always been tough to find a time/day that is good for different time zones. |
| 16:08 |
m_anish |
we could move the other meeting around, but i think being able to have regular attendance in design meetings is going to be imp |
| 16:09 |
silbe |
garycmartin: +1, it's an impossible problem to solve in general. :-/ |
| 16:10 |
manuq |
says +1 to Monday 14:30 UTC |
| 16:11 |
m_anish |
says +1 to that time as well |
| 16:11 |
silbe |
+1 to Mon 14:30 UTC |
| 16:11 |
garycmartin |
silbe: My hope is that the IRC meeting are minimal in design decisions. Realtime design meetings (via text) are really not a good use of time. We should try an have 95% of the discussion via email on the mail list, the wiki for more static items, and bugs.SL.org for actionable items. The realtime meeting should just be about pinging folks on what progress they made for the previous week. |
| 16:12 |
silbe |
garycmartin: if that works it would be great. It didn't work in the past, but the circumstances are different now so we should just try again. |
| 16:13 |
garycmartin |
OK, can try the next but one meeting on a Monday. |
| 16:14 |
silbe |
garycmartin: Do I parse this correctly to mean that there will be a meeting tomorrow, but all following meetings will be Mon 14:30 UTC? |
| 16:14 |
garycmartin |
silbe: If it doesn't work like that I'll probably need to drop out of the Design meetings as happened last time :( I'm not a fast typer and like to think things over. Realtime is really bad for making good design decisions. |
| 16:15 |
|
silbe: Yes, Wednesday meeting tomorrow as per email. It's just about touching base on those agenda items and kicking of email threads (and wiki, bugs) as needed. |
| 16:16 |
m_anish |
garycmartin, looking fwd to tomorrow... Atleast from the dextrose builds, there are many useful features we want to push upstream :-) |
| 16:16 |
silbe |
garycmartin: +1. For me the synchronous communication (i.e. IRC) is more for cases where we're not understanding each other (which often benefits from a short RTT), rather than where we (apparently) disagree (at first). |
| 16:17 |
|
garycmartin: ok. I'll just follow up via email where necessary then. |
| 16:18 |
manuq |
garycmartin: +1, let's keep focus in the meeting for wrap up progress, topics and action items, and keep discussions in the mailing list, actions in the bug tracker, specs and mockups in the wiki |
| 16:19 |
garycmartin |
silbe: Yes, the plan is that decision intentions are emailed out to the lists for discussion for those than can't make the meeting. |
| 16:19 |
silbe |
That's it from my side. Does anyone else have anything left to discuss? |
| 16:19 |
m_anish |
nope :-) done here! |
| 16:20 |
|
4? |
| 16:20 |
manuq |
ok! now.. |
| 16:20 |
|
5 |
| 16:20 |
m_anish |
4 |
| 16:20 |
silbe |
3 |
| 16:20 |
garycmartin |
2 |
| 16:20 |
manuq |
... |
| 16:20 |
danielf |
1, 0 |
| 16:20 |
manuq |
#endmeeting |
| 16:20 |
meeting |
Meeting ended Tue Jul 17 16:20:45 2012 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot. (v 0.1.4) |
| 16:20 |
|
Minutes: http://meeting.sugarlabs.org/s[…]-17T15:03:28.html |
| 16:20 |
|
Log: http://meeting.sugarlabs.org/s[…]12-07-17T15:03:28 |