[18:00] <s-fox> o/
[19:00] <jono> #startmeeting
[19:00] <meetingology> Meeting started Thu Dec 13 19:00:23 2012 UTC.  The chair is jono. Information about MeetBot at http://wiki.ubuntu.com/meetingology.
[19:00] <meetingology> Available commands: #accept #accepted #action #agree #agreed #chair #commands #endmeeting #endvote #halp #help #idea #info #link #lurk #meetingname #meetingtopic #nick #progress #rejected #replay #restrictlogs #save #startmeeting #subtopic #topic #unchair #undo #unlurk #vote #voters #votesrequired
[19:00] <jono> hey everyone, and welcome to the Ubuntu Accomplishments meeting!
[19:00] <jono> who is here for the meeting?
[19:00] <s-fox> o/
[19:00] <JasnaBencic> o/
[19:00] <cwayne> \o
[19:00] <cielak> o/
[19:01] <jono> \o/
[19:01] <jono> :-)
[19:01] <jono> mhall119, are you here too?
[19:01] <jono> so unfortunately I have to leave in 30m for a call, but I have a few agenda topics to raise
[19:01] <jono> is it OK if I raise these first?
[19:01] <cielak> sure
[19:01] <jono> awesome :-)
[19:02] <jono> so the first is regarding the server deployment
[19:02] <jono> so mhall119 has been taking care of getting the validation, admin, and web gallery deployment instructions in place
[19:02] <jono> one thing we need to decide on is how we push changes to IS for deployment
[19:02] <jono> typically they prefer to simply pull a branch for a deployment
[19:02] <jono> as such, I wanted to propose the following:
[19:03] <jono>  - we will have the staging and production servers set up
[19:03] <jono>  - staging pulls from trunk for each of the server-related projects
[19:03] <cielak> will we have a separate project for the admin dashboard?
[19:03] <jono>  - when we cut a new release of a project, trunk is branched to a version number (as we do with the -daemon and -viewer projects), and that version is pushed to production
[19:04] <jono> cielak, right now the admin is just a branch, but I will set up a project
[19:04] <mhall119> jono: I'm here
[19:04] <cielak> okay
[19:04] <jono> does that workflow sound OK to you folks?
[19:04] <cielak> seems fine to me
[19:04] <jono> this way we can then just file an RT with a branch pull for an update
[19:04] <s-fox> sounds like a good idea to me
[19:04] <jono> mhall119, do you think IS will be happy with that?
[19:04] <mhall119> jono: why would staging pull from trunk?
[19:05] <jono> mhall119, as that is where we develop the projects
[19:05] <jono> the idea being that you only push to trunk when your code works :-)
[19:05] <jono> so trunk is sacred
[19:05] <mhall119> so this is feature staging, not deployment staging
[19:05] <cielak> so the projects that are going to have their 'production' branches are the validation server, the dashboard and comunity-accomplishments?
[19:05] <jono> mhall119, right
[19:05] <mhall119> ok
[19:05] <jono> cielak, right, so the only weird one is UCA
[19:06] <jono> we can have production branches for validation, admin, and gallery
[19:06] <jono> actually I guess we can just do the same for UCA
[19:06] <jono> my thinking here though is that the server won't pull UCA from trunk and instead pull it from the releases PPA
[19:06] <jono> as the accoms live in /usr/share
[19:07] <cielak> the production one? or the staging one?
[19:07] <jono> cielak, for production
[19:07] <cielak> right
[19:07] <cielak> this is fine with me
[19:07] <jono> cool
[19:07] <jono> mhall119, sound OK with you?
[19:08] <mhall119> jono: yeah, we'd want to pull from something close to what users are using
[19:08] <mhall119> so PPA instead of trunk
[19:08] <jono> sounds good
[19:08] <jono> yep
[19:08] <jono> OK, I will document this on a wiki page
[19:08] <jono> and then we can point IS at it
[19:09] <jono> to ensure it is what they expect
[19:09] <jono> so the other topic I wanted to discuss was getting ready for the release in 13.04
[19:09] <jono> last night I was chatting to Janos about the gallery
[19:09] <jono> unfortunately he can't be in this meeting
[19:10] <jono> he has created a milestone and suggested we file bugs against the milestone that we need fixing in preparation for the release
[19:10] <jono> I wanted to suggest we do the same across our different projects
[19:10] <jono> I know we have many of these bugs already
[19:10] <jono> cielak, how are things looking with the daemon/viewer in terms of quality?
[19:11] <cielak> we have not messed up anything, but new features that were added have not yet been thoroughly tested
[19:11] <jono> cielak, so I think we are going to want to put in place a more rigerous testing plan for this next release
[19:11] <cielak> as for the daemon and the viewer, I tend to assign bugs to 0.4 milestones
[19:11] <jono> sounds good
[19:11] <jono> did we decide on a 0.4 release date?
[19:11] <jono> I seem to remeber feb
[19:11] <cielak> 15 feb
[19:12] <jono> cool
[19:12] <jono> this makes sense
[19:12] <jono> so I think if we can shoot to have the server deployed by then and release 0.4
[19:12] <jono> and then why don't we focus on getting our accoms and any other bug fixed after that
[19:12] <jono> so the release after 0.4 will be 1.0
[19:12] <jono> and will be the release in 13.04
[19:12] <jono> make sense?
[19:13] <cielak> so you mean we do not care much about bug fixing for 0.4, but for 1.0 instead??
[19:13] <jono> my thinking is that 0.4 is all about bug fixing
[19:13] <jono> I personally think we should lock down features
[19:13] <jono> but to focus on the server, daemon and viewer mainly for 0.4
[19:13] <jono> and then we focus on UCA for 1.0 and any other bugs
[19:14] <cielak> okay, let it be this way
[19:14] <jono> cielak, is that OK with you?
[19:14] <cielak> there are two more major features I'd like to implement in 0.4
[19:14] <jono> oh?
[19:14] <cielak> so just wanted to ensure I understand the plan well ;)
[19:14] <jono> :-)
[19:15] <jono> which features, cielak?
[19:15] <cielak> first is getting the daemon to run scripts in paralell
[19:15] <cielak> this will improve the performance, as most of time scripts are just waiting for servers to respond
[19:16] <cielak> this is kind of experimental, I will investigate how that changes resource usage and execution time
[19:16] <jono> cielak, sounds cool
[19:16] <cielak> and the other thing is support for custom validation servers for third-party collection
[19:16] <jono> cielak, right
[19:16] <cielak> just to make our platform fully expandable
[19:17] <jono> yup
[19:17] <jono> cielak, are you going to work on both of those features?
[19:17] <cielak> I should have some more time to work on that during the winter break, and I am really excited to work on them, but if anyone wants, help is very welcome ;)
[19:17] <cielak> plus there is the thing of making the daemon a DBus service
[19:18] <jono> cielak, right
[19:18] <jono> cielak, would you mind emailing didrocks and discussing the dbus work there?
[19:18] <jono> I know you have not had as much IRC time
[19:19] <jono> I am hoping the dbus thing won't be a blocker for inclusion in the USC
[19:19] <cielak> I will, but first I will reproduce all the problems, to clarify what I had troubles with
[19:19] <jono> sounds great
[19:19] <jono> thanks, cielak :-)
[19:20] <cielak> just being busy recently, but that should change soon
[19:20] <jono> ok, I will take some time later to put together a release schedule with these dates
[19:20] <jono> cielak, awesome
[19:20] <jono> we are not far off our 1.0 goal :-)
[19:20] <jono> this is going to be *awesome*
[19:20] <jono> :-)
[19:20] <jono> it will be super cool to see people's trophies on the web
[19:20] <jono> and shared on Twitter :-)
[19:20] <cielak> indeed!
[19:21] <jono> ok, those were my agenda items
[19:21] <jono> any other topics?
[19:21] <s-fox> i have none.
[19:22] <cielak> I have one
[19:22] <JasnaBencic> a lot of things I don't understand but this is great :) .... cielak should I still focus on 10.04 or new versions of Ubuntu so I can catch up with you guys easier?
[19:22] <cielak> JasnaBencic: we never really supporter 10.04
[19:22] <jono> JasnaBencic, , 12.04 is our minimum version
[19:22] <jono> cielak, what topic?
[19:22] <cielak> I expect you may be successful running Accomplishments on 10.04, but we can't guarantee that
[19:23] <cielak> I wanted to ask if anyone has recently followed the Accomplishments Writing Guide
[19:23] <cielak> I don't remember us updating it frequently
[19:23] <cielak> so it might be wise to check if it's not outdated
[19:23] <doctormon> Sorry late, is meeting still happening?
[19:23] <cielak> doctormon: yes! :)
[19:24] <jono> cielak, I haven't checked into it recently
[19:24] <jono> this is something we should check into, particularly after 0.4
[19:24] <jono> doctormon, still going  :-)
[19:24] <s-fox> Zilvador,   did you follow the guide?
[19:24] <jono> although I have to run in a few mins
[19:25] <cielak> so this is a thing that certainly should land on our todo list
[19:25] <doctormon> I'm here to tend to the code submitted for gtk-shelves; message to jono might have been lost in the pipes.
[19:26] <Zilvador> s-fox, I did. And I am planning to go through it soon to update it.
[19:26] <s-fox> ^ cielak  :)
[19:26] <cielak> doctormon: I would be interested in examining your code
[19:26] <jono> doctormon, did you file a MP?
[19:26] <cielak> Zilvador: awesome! you are very welcome to improve it :)
[19:26] <jono> ok, sorry folks, I need to run to this call
[19:26] <Zilvador> :)
[19:26] <jono> thanks for joining everyone!
[19:27] <jono> feel free to keep discussing
[19:27] <cielak> thanks jono! see you :)
[19:27] <doctormon> jono: No, because it's not production code.
[19:27] <jono> just end the meeting when you need to
[19:27] <jono> doctormon, ok
[19:27] <cielak> doctormon: you did a simple proof-of-concept GTK app, right?
[19:27] <JasnaBencic> thank you jono
[19:27] <doctormon> cielak: lp:~doctormo/junk/gtk-shelves
[19:27] <jono> doctormon, have a screenshot?
[19:28] <doctormon> Took a while to root through all the gtk bolocks though. The gtk devs were unhappy about it, *shrug* but it works.
[19:28] <doctormon> jono: You've seen the screenshot already, the difference is now you can scroll.
[19:28] <doctormon> (without it messing up)
[19:28] <jono> doctormon, can you link us again?
[19:28] <mhall119> doctormon: have you tested it on various GTK themes?  I had one way of doing it that worked on some, but not others
[19:29] <cielak> doctormon: testing out, this looks fine, but I wonder what would happen if the items would have variable height
[19:29] <doctormon> cielak: Bad things
[19:30] <doctormon> mhall119: Yes, this works by painting, not styles.
[19:30] <doctormon> jono: Image is gone on imagebin, cielak can you generate a screenshot for jono?
[19:30] <cielak> oops, the accomplishments viewer has items of different heights... is there a chance your trick could be applied in such case too?
[19:31] <doctormon> cielak: No, you'd have to find the height of the row and change the image accordingly. That's a lot of code.
[19:31] <mhall119> doctormon: cool
[19:31] <doctormon> Recommend you change the viewer habbit.
[19:31] <doctormon> (actually what I have is control over the icon display, it paints them all as a fixed size anyway)
[19:32] <cielak> doctormon: the point is that the different height is not because icon size, but because the accompishment title varies, and long titles (or long translations) result in higher rows
[19:32] <doctormon> I took the titles out, too much trouble.
[19:32] <mfisch> cielak: sorry I had another meeting
[19:33]  * davidcalle has just arrived
[19:33] <cielak> yeah, but there is no way we could not display the title in the viewer
[19:33] <cielak> hello davidcalle :)
[19:33]  * davidcalle waves :)
[19:33] <doctormon> But it is possible to calculate the pango paint size of the text and consider that in the painting. But we're talking about a very different order of magnatude on the code there.
[19:33] <cielak> there is a screenshot of doctormon's experimental branch: http://i.imgur.com/f6fAY.png
[19:34] <doctormon> For instance, repeating each row is done by the context, we'd have to manually do all that.
[19:34] <cielak> doctormon: yes, I expect so
[19:34] <cielak> this would also significantly affect performance
[19:35] <doctormon> Probably. There are ways to show the title though. have a fixed number of letters and use an elipse. Do a hover over screen with title and other info.
[19:35]  * janos_ these icons in the shot are not so good... simpler would be better
[19:36] <doctormon> Useful once we get the icon-level and branding destinctions in there. Maybe the trophies will each be unique enough.
[19:36] <cielak> a hover would make browsing difficult
[19:36] <doctormon> cielak: It's not for browsing, it's for admiring. Browsing is when you want to view the achievements you don't have yet. No?
[19:36] <janos_> is the UI searchable with ctrl-F or something?
[19:36] <cielak> a fixed number of letters is't a perfect solution too, though, look at how varied the titles are
[19:36] <janos_> i think titles are important
[19:37] <cielak> janos_: the trunk/dailies contain already a search feature
[19:37] <cielak> doctormon: so that would make sense only once we have tons of easily distinguishable icons for trophies
[19:38] <janos_> cielak: thanks i just meant that to highlight the importance of titles (searchability)
[19:38] <doctormon> We could afix the position and give each more space, 3 lines height each.
[19:38] <janos_> but i do see doctormon's point about admiration
[19:38] <janos_> can a one-line title be vertical aligned in the middle?
[19:39] <cielak> a fixed 3 lines height seems like a kind of a good solution
[19:39] <doctormon> janos_: It would be if we painted it ;-)
[19:39] <cielak> yeah, if painted manually, we can do everything
[19:39] <cielak> using a custom, not theme-dependent font+size would be significant
[19:40] <doctormon> So, we don't use the listview as just a painting canvas. We hijack the icon's own paint mechanism to do that side of things.
[19:40] <doctormon> So it's basically 3 hijacks of different widget/parts.
[19:40] <cielak> but drawing the text manually has one disadvantage - the more we move away from standard GTK, the more accesibility features we may loose
[19:41] <doctormon> cielak: In this case not, we're not painting the text onto the iconview widget.
[19:41] <cielak> but onto it's embeded label?
[19:41] <doctormon> cielak: But the item's own internal draw method. The text is still in the item's data store, the item still has a bounding box. It's all good.
[19:41] <cielak> aah
[19:41] <cielak> that's fine indeed
[19:42] <doctormon> Now don't ask me about drag and drop :-P
[19:42] <doctormon> It might work ;-)
[19:42] <cielak> it should... I guess, but we never supported it anyway
[19:42] <cielak> I don't see any use of dragging the trophies, neither within nor outsite the window
[19:43] <cielak> doctormon: and I expect this would not interfere with GtkTreeModelFilters we use?
[19:43] <cielak> silly me, of course it wouldn't
[19:45] <doctormon> :-)
[19:45] <cielak> my opinion is that it would be worth to work on that in a viewer's branch
[19:46] <doctormon> I've tried hacking on the viewers code, but it's deps are far too high for devel
[19:46] <cielak> what do you mean?
[19:46] <doctormon> (I can't get it to run without a server, dbus and all that other guff running)
[19:46] <cielak> doctormon: you will need dbus and the accomplishments-daemon
[19:47] <cielak> server shouldn't be needed, though
[19:47] <doctormon> Exactly why do I need the daemon for to test a viewer? Is there no test daemon with pre-set data?
[19:47] <cielak> the viewer is just the frontend for the daemon
[19:48] <cielak> it does not know at all how accomplishments work, how to process their files etc
[19:48] <cielak> it asks the daemon to do whatever is needed
[19:48] <cielak> a good thing about that is that it keeps the viewer much less complicated for you to hack
[19:48] <cielak> but there is no way you could run it without the daemon
[19:49] <cielak> anyway, awesome work with the branch :) It would be great if you could reproduce these effects in the viewer, but if not I guess I could work on that, your code seems clear to me
[19:50] <cielak> in case of problems with getting the daemon to run you are welcome in #ubuntu-accomplishments
[19:50]  * gepatino waves. sorry I'm late
[19:50] <cielak> hi gepatino, good to see you :)
[19:51] <cielak> okay, there are 10 more minutes time left, so if anyone wants to discuss anything else, please go ahead!
[19:52] <gepatino> is there something to talk about the web gallery?
[19:52] <cielak> janos_: ^
[19:53] <doctormon> thanks guys!
[19:53] <mhall119> the only problem I had deploying the gallery was that the samples/users.json is out of date
[19:53] <mhall119> and wasn't setting the share id/folder properly
[19:53] <mhall119> probably made for an older version of the data model
[19:56] <jono> cielak, do you have a screenshot of doctormon's branch?
[19:56] <cielak> jono: http://i.imgur.com/f6fAY.png
[19:56] <jono> cielak, cool
[19:57] <jono> would be cool if the trophies could have a little shadow too
[19:57] <cielak> I think this is to be done in the accomplishment icon
[19:57] <jono> cielak, so is he going to work on a production branch?
[19:58] <cielak> yes, it seems so
[19:58] <jono> cool
[19:58] <mhall119> we should change all the trophies to ponies
[19:59] <cielak> mhall119: +1 for a new ponies-related accomplishments collection
[20:00] <cielak> right, we're running out of time and it seems there is nothing more to discuss
[20:00] <cielak> thanks everyone for joining us!
[20:02] <gepatino> mhall119, could you send a mail with this info to the list? I'll try to take a look at it
[20:03] <gepatino> I mean the issue with the samples/users.json file, not the ponies icons :)
[20:05]  * janos_ reading up
[20:07] <JasnaBencic> meeting done? cielak you'll get a lot of questions from me via mail :)  I'm very happy that I came here despite I didn't understand a lot :)
[20:08] <cielak> JasnaBencic: sure. you are also welcome to ask us all via the mailing list we use
[20:08] <JasnaBencic> ok
[20:09] <janos_> mhall119: oh the samples/users.json was just for an initial proof of concept demo
[20:11] <janos_> mhall119, gepatino: yup, samples/users.json is just for developers, should not be in prod. I will update the readme
[20:11] <gepatino> ok
[20:12] <janos_> oh i see the meeting is over :) heading over to the other channel then, bye all
[20:17] <mhall119> ok