Time |
Nick |
Message |
21:03 |
meeting |
Meeting started Mon Jul 20 21:02:59 2015 UTC. The chair is tch__. Information about MeetBot at http://wiki.debian.org/MeetBot. |
21:03 |
|
Useful Commands: #action #agreed #help #info #idea #link #topic #endmeeting |
21:03 |
tch__ |
Hello eveyone |
21:03 |
gonzalo__ |
plarke, every one help |
21:03 |
samdroid |
g'day |
21:03 |
gonzalo__ |
hi samdroid :) |
21:04 |
samdroid |
so we start the meeting? |
21:04 |
gonzalo__ |
yes |
21:04 |
tch__ |
well, basically, this first meeting is an "informal" one :) but the topics are: 1. what do we and deployments want for 0.108 and 2. when do we want it? (sounds like this http://www.quickmeme.com/What-[…]e-Want/?upcoming) |
21:05 |
gonzalo__ |
:) |
21:05 |
tch__ |
I would also like to start talking about the previous cycle, |
21:06 |
|
does anybody wants to make comments? (regarding the cycle itself, e.g, was it too long, too short, whatever) |
21:06 |
walterbender |
seems like from the tech POV you guys did a great job |
21:06 |
samdroid |
it was pretty nice |
21:07 |
gonzalo__ |
honestly, we achieved more than i expected |
21:07 |
tch__ |
walterbender, I feel though, we had _very_ little feedback from the deployments though, except for german who helped us with the tests, |
21:07 |
walterbender |
yes |
21:07 |
samdroid |
+1 |
21:07 |
walterbender |
I was about raise that point |
21:07 |
gonzalo__ |
tch__, deployments are usually slow to provide feedback |
21:07 |
walterbender |
I feel a bit at sea in terms of deployment needs/concerns |
21:08 |
|
for planning the next release |
21:08 |
Quozl` |
i also feel there is no feedback from deployments, as if they have abandoned the development process. |
21:08 |
walterbender |
so I suggest we look at what would help with long-term maintenance |
21:09 |
|
they presumably want stability |
21:09 |
|
and we presumably want to lessen the support burden |
21:09 |
Quozl` |
i suspect they have achieved stability by rejecting all problem reports from users. |
21:09 |
tch__ |
walterbender, in my experience I am sure they want stability and a nice browser, |
21:09 |
samdroid |
;) |
21:09 |
|
*:) |
21:10 |
walterbender |
to a large extent stability these days is keeping current with our upstream |
21:10 |
Quozl` |
i agree, and possibly that their browser concerns often trump any other problems. |
21:10 |
samdroid |
good luck with a nice browser |
21:10 |
|
WebKit2 we need new apis upstream (or to rewrite the whole history system) |
21:11 |
|
and if we were to go CEF we're going to uncharted (and unpackaged?) terrirory |
21:11 |
|
*CEF/servo |
21:11 |
Quozl` |
leaves |
21:11 |
gonzalo__ |
bye Quozl` |
21:11 |
|
samdroid, what is CEF/servo? |
21:12 |
walterbender |
back in 5 |
21:12 |
samdroid |
gonzalo__: cef=chromiume emdeded framework. servo is mozillas new browser and it uses the cef api |
21:12 |
|
gonzalo__: cef is probs a better browser - google spends more on browsers than apple does... now at least |
21:13 |
tch__ |
samdroid, what the relation with webkit? does it use wk or is a new thing? |
21:13 |
samdroid |
blink - the webkit fork |
21:13 |
tch__ |
samdroid, ah, yeah |
21:14 |
samdroid |
it's something to look into - but it would definatly end up with c code in browse |
21:14 |
gonzalo__ |
samdroid, well, that is not compatible with old linux (like F18) nor with new right now |
21:14 |
samdroid |
gonzalo__: wk2? |
21:14 |
Quozl` |
i think the webkit2 browse is something to concentrate on, and the telepathy api. bye. |
21:14 |
tch__ |
samdroid, hmm, we probably need to evaluate its pros |
21:14 |
gonzalo__ |
blink and the new google stuff |
21:14 |
tch__ |
Quozl`, thanks for your 2cents :) |
21:15 |
samdroid |
webkit 2 pros: gi.repository bindings |
21:15 |
tch__ |
samdroid, thats a big noe |
21:15 |
|
s/noe/one |
21:16 |
samdroid |
webkit2 -: no history api |
21:17 |
tch__ |
samdroid, I think the users big concer are related to whether or not they can display their websites |
21:17 |
samdroid |
ok |
21:17 |
gonzalo__ |
samdroid, yes, but other than that, we are pretty close |
21:18 |
tch__ |
samdroid, gonzalo__ is there any work related to webkit2 in browse already? |
21:18 |
samdroid |
yes |
21:18 |
gonzalo__ |
tch__, yes |
21:18 |
samdroid |
it works ok |
21:18 |
|
but not histroy/persistance |
21:18 |
gonzalo__ |
we have a branch webkit2, and is working |
21:18 |
samdroid |
or the edit menu |
21:19 |
gonzalo__ |
that are the main issues |
21:19 |
tch__ |
samdroid, is that a regression in webkit? or we need to adapt browse? |
21:19 |
gonzalo__ |
and the problem is, the webkit2 removed api that we use |
21:19 |
tch__ |
gonzalo__, ouch, |
21:19 |
|
gonzalo__, removed from webkit or the bindings? |
21:19 |
gonzalo__ |
samdroid.maybe we could get the history using javascript? |
21:19 |
tch__ |
gonzalo__, is it intentional? |
21:19 |
gonzalo__ |
tch__, yes |
21:20 |
samdroid |
gonzalo__: we can get the history 100% fine |
21:20 |
gonzalo__ |
tch__, samdroid already filled a bug |
21:20 |
samdroid |
but we can not set the history |
21:20 |
gonzalo__ |
samdroid, true |
21:20 |
samdroid |
ok so maybe I'll chase that one up |
21:20 |
gonzalo__ |
i forgotten that |
21:20 |
tch__ |
samdroid, ok, so this things is already moving.. |
21:20 |
|
gonzalo__, ^ |
21:20 |
samdroid |
CEF doesn't seem to have a history api at 1st look - not seeing a set history func |
21:20 |
|
yeah |
21:21 |
tch__ |
samdroid, gonzalo__ last question regarding webkit2, any chance it would work on f18? |
21:21 |
|
or we would need to maintain a webkit1 fork for f18? |
21:22 |
gonzalo__ |
tch__, i think will be better get the webkit1 on f18 |
21:22 |
tch__ |
gonzalo__, ok |
21:23 |
|
another important topic is collaboration, |
21:23 |
samdroid^ |
+1 |
21:23 |
tch__ |
has anyone looked into collaboration stack recently/ |
21:23 |
|
? |
21:24 |
|
basically, telepathy dropped Tubes channel... which is what we use for collab |
21:24 |
|
they dropped it a long time ago, btw, 2011/12 |
21:24 |
|
I haven't looked into the details, but it does not seems to be a trivial work |
21:25 |
|
the only workaround for newer distros is to use an older version of telepathy gabble, |
21:26 |
|
but that can't be considered a solution since newer telepathy is used by other DE apps |
21:27 |
|
putting package maintainers on a problematic situation |
21:27 |
samdroid |
so is the plan to port it to the new telepathy api version? |
21:28 |
tch__ |
samdroid, that would be ideal, but first we need to understand whats in telepathy now, to see if its possible... |
21:28 |
samdroid |
ok |
21:28 |
tch__ |
samdroid, another route could be to find a different tech, |
21:28 |
samdroid |
yep |
21:29 |
tch__ |
samdroid, but in any case, I won't be easy... specially because there are many activities to adapt |
21:29 |
samdroid |
(I have no idea how collab stack works) |
21:29 |
tch__ |
I am not even sure how many, |
21:29 |
samdroid |
oh, yeah they all use tp directly |
21:29 |
tch__ |
samdroid, I know very little too, just the basics |
21:30 |
samdroid |
maybe from a sugar3 api sense we shouold add an abstraction over whatever api we use to make it more indipendant of the system |
21:31 |
tch__ |
samdroid, assuming we can find a proper replacement for what was dropped, yes |
21:32 |
samdroid |
ok |
21:32 |
tch__ |
samdroid, gonzalo__ walterbender moving on... have you tried sugar on fedora 22+? |
21:32 |
samdroid |
looks nice on fedora 24(?) rawhide |
21:33 |
|
but treeviews zebra stripes have regressed :P |
21:33 |
tch__ |
samdroid, thanks to you and gonzalo__ ;) |
21:33 |
|
satellit_e has quit IRC |
21:33 |
tch__ |
samdroid, I have noticed most of the changes are related to artwork gtk themes |
21:33 |
|
and not so much to the actual code aPI |
21:33 |
|
s/aPI/API |
21:34 |
gonzalo__ |
tch__, yeah |
21:35 |
tch__ |
samdroid, gonzalo__ I believe that is a good thing.. being optimistic, I would say gtk API its pretty stable recently, |
21:35 |
|
luckily for us, because we haven't really done much about that, |
21:36 |
|
I think samdroid made some adjustments recently, and the other issue I can recall was solved with the custom bindings for clipboard |
21:36 |
|
in SugarExt |
21:36 |
samdroid |
that's a nice hack |
21:37 |
tch__ |
glorious 20 lines of C code... but I haven't checked that on f22+ |
21:37 |
|
i know its works on f18-20 |
21:37 |
samdroid |
I don't think it's broken |
21:37 |
|
I was testing dranging jobjects to the cb |
21:37 |
|
and that works (with cb patches) |
21:38 |
tch__ |
samdroid, make sure to try with journal entries containing images, etc |
21:38 |
samdroid |
tch__: quick collab question: Aren't there 2 collab transports? olpcmesh and telepathy? or are they both telepathy? |
21:39 |
tch__ |
samdroid, If there is a olpcmesh module/package, I don't know anything about it... the only mesh I know of the the physical network stuff |
21:39 |
|
samdroid, which was dropped from xo1.5+ |
21:40 |
samdroid |
oh |
21:40 |
|
ok |
21:40 |
tch__ |
maybe walterbender has more info about that |
21:41 |
|
so, we briefly mentioned webkit2, telepathy, gtk (which sounds pretty critical stuff to me), but is there anything else you guys want to mention? |
21:41 |
|
ideally, |
21:42 |
|
this talk is about what needs to be done, we then decide what can be done :) |
21:42 |
samdroid |
getting rid of gtk menus looks like it could happen for 108 |
21:42 |
gonzalo__ |
tch__, we can talk about what we _want_ do, other that _need_ do |
21:43 |
tch__ |
gonzalo__, well, ideally we should start by what needs to be done, and then see what can actually be done |
21:43 |
gonzalo__ |
samdroid, yes, but we need maintain for compatibility (some activities use the menues). |
21:44 |
samdroid |
gonzalo__: ya, and combo boxes |
21:44 |
tch__ |
gonzalo__, in other words, we should at least mention these things |
21:44 |
samdroid |
gonzalo__: but we can move them over |
21:44 |
|
gonzalo__: and then we get less shell bugs :) Hopefully :) |
21:44 |
tch__ |
samdroid, have you experimented with that? |
21:44 |
samdroid |
yep |
21:44 |
tch__ |
samdroid, is a big change? |
21:44 |
samdroid |
nap |
21:44 |
gonzalo__ |
samdroid, yes, but i don't want break activities that didn't changed because we change the tolkit |
21:45 |
samdroid |
gonzalo__: I wasn't suggesting we delete to old code |
21:45 |
|
tch__: I've submitted a patch |
21:45 |
gonzalo__ |
ok, just to be sure |
21:45 |
samdroid |
http://wiki.sugarlabs.org/go/F[…]s/Replace_GtkMenu |
21:45 |
tch__ |
samdroid, gonzalo__ , thats another thing we need to evaluate, |
21:45 |
gonzalo__ |
tch__, the change is mostly on sugar, not in the toolkit |
21:46 |
|
tch__, maybe we can add deprecation warnings in the toolkit only |
21:46 |
tch__ |
gonzalo__, I see, |
21:46 |
samdroid |
tch__: ...? |
21:46 |
gonzalo__ |
tch__, sugar use the menu api on the journal and in the clipboard only |
21:47 |
tch__ |
ok, |
21:47 |
gonzalo__ |
tch__, then is doable replace that submenu by a modal window |
21:47 |
|
tch__, and have a few advantages |
21:47 |
samdroid |
search! |
21:47 |
gonzalo__ |
* si more usable with touch |
21:47 |
|
s/si/is/ |
21:48 |
|
* we have a few bugs in the menues interaction |
21:48 |
samdroid |
gtkwindow palettes look more like proper palettes |
21:48 |
gonzalo__ |
yes |
21:48 |
samdroid |
proper gap rendering |
21:48 |
|
(except when the window manager puts shadows all over the place) |
21:49 |
tch__ |
something I was thinking about is how much of these things actually improve the user experience of user (not saying that improving maintainability and shell stability is a bad thing though) |
21:49 |
|
but that takes us back to the feedback topic, which I think can't be solved in a developers team meeting |
21:49 |
gonzalo__ |
tch__, i think we can detail the improvements in the feature page |
21:50 |
|
samdroid, i would like work with you in this feature |
21:50 |
samdroid |
ok |
21:51 |
tch__ |
nice, something concrete :) |
21:51 |
|
hehe, |
21:51 |
gonzalo__ |
samdroid, would be, you would work, and i will review, test and criticize :) |
21:51 |
|
haha |
21:51 |
samdroid |
that's fine by me :) |
21:51 |
tch__ |
moving on... |
21:51 |
gonzalo__ |
samdroid, tch__, the other think would be land the journal patches |
21:52 |
|
thing |
21:52 |
samdroid |
#action gonzalo__ work with samdroid on the modals |
21:52 |
gonzalo__ |
i think we need test deeply |
21:52 |
samdroid |
#action look more into wk2 browse |
21:52 |
|
+1 |
21:52 |
tch__ |
in the early meetings for 106 cycle, we said we were ok to keep time-base releases, 6 months cycles, |
21:52 |
gonzalo__ |
maybe prepare testing rpms for german |
21:53 |
tch__ |
samdroid, gonzalo__ walterbender are we still ok with that? |
21:53 |
samdroid |
with the journal patches? |
21:53 |
|
tch__: i suppose |
21:53 |
tch__ |
well, I am, but is a good exercise to ask, |
21:54 |
samdroid |
gonzalo__: +1 though, let's merge those journal patches early and then get them really tested |
21:54 |
gonzalo__ |
tch__, i am ok with 6 months |
21:54 |
tch__ |
samdroid, gonzalo__ walterbender , if its 6 months, then the next release would be in january 2016 |
21:54 |
gonzalo__ |
tch__, but we would think in call 0.108 Sugar 1 (or 3) and call it stable |
21:55 |
tch__ |
gonzalo__, it could be, I think that is a decision charged with philosophical energies though he, |
21:56 |
gonzalo__ |
yep, |
21:56 |
tch__ |
gonzalo__, samdroid, walterbender regarding the time, think january is ok for the final release? |
21:56 |
samdroid |
nice timing with the whole campaign around reelecting slobs :) |
21:56 |
tch__ |
I was thinking about GCI, in case we participate it would be troublesome, |
21:57 |
samdroid |
It always is :) |
21:57 |
gonzalo__ |
tch__, when is GCI? |
21:57 |
samdroid |
but we can hack more on activities during gci |
21:57 |
tch__ |
gonzalo__, usually is Dec 1 |
21:57 |
gonzalo__ |
samdroid, +1 |
21:57 |
tch__ |
if 108 is released before or at the same of GCI, then it gives more flexibility, |
21:58 |
|
thats the pro, the con is we have 1 month less :P |
21:58 |
|
flexibility for the GCI topics |
21:58 |
|
samdroid, I don't mind, working on activities is always good |
21:58 |
samdroid |
tch__: will we have collab done for that release in that cycle? |
21:59 |
|
tch__: 'cause fixing collab is kinda the big feature for this cycle |
21:59 |
tch__ |
samdroid, we don't know yet, but if it is, that would be a popular topic for GCI |
21:59 |
gonzalo__ |
tch__, if we want stability, we can add gci stuff at the end |
21:59 |
|
can't |
21:59 |
|
my fingers! |
21:59 |
tch__ |
gonzalo__, yes, that why I am suggesting to release it before, |
21:59 |
samdroid |
tch__: so maybe we should choose the times later when we have an idea of the collab stuff? |
22:00 |
tch__ |
and by flexibility I meant that we usually ignore GCI stuff, because we are close to release |
22:00 |
samdroid |
:_) |
22:00 |
|
*:) |
22:00 |
tch__ |
is something we can consult with our pillows ;) |
22:00 |
gonzalo__ |
tch__, +1 |
22:00 |
samdroid |
count down to end meeting? |
22:01 |
gonzalo__ |
i need go, any other issue? |
22:01 |
tch__ |
I think we have mentioned the important topics, |
22:01 |
|
3 |
22:01 |
samdroid |
2 |
22:01 |
gonzalo__ |
1 |
22:01 |
tch__ |
#endmeeting |
22:01 |
meeting |
Meeting ended Mon Jul 20 22:01:36 2015 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot. (v 0.1.4) |
22:01 |
gonzalo__ |
we can follow on the ml |
22:01 |
meeting |
Minutes: http://meeting.sugarlabs.org/s[…]-20T21:02:59.html |
22:01 |
|
Log: http://meeting.sugarlabs.org/s[…]15-07-20T21:02:59 |