Web   ·   Wiki   ·   Activities   ·   Blog   ·   Lists   ·   Chat   ·   Meeting   ·   Bugs   ·   Git   ·   Translate   ·   Archive   ·   People   ·   Donate

#sugar-meeting meeting, 2009-06-16 10:15:00

Minutes | Index | Today     Channels | Search | Join

All times shown according to UTC.

Time Nick Message
10:15 BryanWB yay!
10:15 mchua Gah, sorry guys; I thought I could make Sandeep chair so that he could end that last meeting.
10:15 BryanWB #TOPIC Webified
10:16 dogi thx mchua
10:16 BryanWB mchua: yeah tks
10:16 lucian_ i've polished SSB creation a bit, now bundles get installed properly and creating an SSB works from inside an existing SSB quite well
10:16 dogi BryanWB, she is admin of meeting too ... like bernie
10:16 homunq_ wonders if you can fool meeting bot with /nick ....
10:16 bernie mchua: you didn't follow your own notes on keeping meetings short and focused, eh?
10:16 :-)
10:17 BryanWB lucian_: any issues u r having trouble resolving?
10:17 bernie dogi: am I?  where does the meeting bot live?
10:17 BryanWB bernie: let's stay on topic ;)
10:17 dogi http://vig.laptop.org/wiki/ind[…]p/Machine:meeting
10:17 lucian_ BryanWB: I'm not sure how to persist session data
10:17 dogi vig.sugarlabs.org ???
10:17 BryanWB lucian_: u mean via the journal?
10:18 lucian_ there are things in get_activity_data() that would need to be in the bundle of created SSBs
10:18 BryanWB: no, that works
10:18 BryanWB: since I used Browse, it interacts quite well with the Journal
10:18 BryanWB lucian_: um, i made a hack a while ago but not sure it is the best way
10:18 lucian_ my problem is that not everything in get_activity_data() would need to be saved
10:19 s/saved/included in the SSB/
10:19 tomeu goes to vig.sugarlabs.org
10:19 BryanWB lucian_: ok, basically w/ an .xo bundle u can stick the session data somewhere and then have it copied into    isolation/#/uid_to_home_dir/ w/ the shell script that launches it
10:19 lucian_ something like that
10:19 but i'm worried that it would grow too large
10:20 if there's a big gecko profile in the session data, for example
10:20 BryanWB lucian_: It can easily grow too large, esp. if u pick up cruft in the gecko profile
10:20 bernie dogi: oh, yeah, I'm an admin! thanks
10:20 BryanWB lucian_: can easily be 50 MB
10:20 bemasc lucian_: what's in the profile that you don't want? Presumably just caches, right?
10:20 lucian_ bemasc: at least the caches, yes
10:21 perhaps private data as well
10:21 BryanWB lucian_: geck must have naming convention crap that u can use to exclude it
10:21 dogi wants to vig.sugarlabs.org cname of vig.laptop.org ... :P
10:21 lucian_ BryanWB: yes, i'll have to look into that some more
10:21 dogi and cu later ...
10:21 bemasc lucian_: ok.  I believe there are commands in Firefox to drop those things.  I bet you can access them through hulahop.
10:21 lucian_ anyway, I've also made Browse aware of being an SSB or not
10:21 BryanWB lucian_: the hard part will be automating what part of the profile should persist and what shouldn't
10:21 lucian_ since the changes are quite small, I figured it's the best way
10:22 BryanWB: yes
10:22 BryanWB lucian_: can u give an example of what should persist?
10:22 lucian_ bookmarks, history, greasemonkey (after I get it to run), gears
10:22 i think that's about it
10:22 silbe sdziallas: sdziallas@buildslave-fedora-11-32bit.sugarlabs.org (2001:1418:16e:1:5652:ff:fe26:c5e5) created, your ssh key from soas@bs2 added, root rights (sudo) granted
10:22 BryanWB lucian_: that should be straightforward to store
10:23 lucian_ BryanWB: i'm not sure though, i may be missing some vital stuff that gecko stores somehow
10:23 silbe sorry :-/
10:23 bemasc ok, it sounds like lucian_ is making great progress and has a path forward.
10:23 BryanWB lucian_: the documentation for mozilla is great though, u should be able to find out relatively quickly
10:23 lucian_ yep
10:23 BryanWB bemasc: agreed, onto next one
10:24 lucian_ i'll also need some design tips
10:24 BryanWB lucian_: like what?
10:24 lucian_ but there's a meeting scheduled on thursday to help me with that, among other things
10:24 BryanWB lucian_: opyadav at #olenepal is a mega design guru
10:24 lucian_ BryanWB: like what to call/where to place things like 'create ssb', 'add userscript', etc.
10:24 mchua lucian_: the mozilla folks are extremely friendly esp. for student participation, and design... drop https://cs.senecac.on.ca/~david.humphrey/ a line.
10:25 lucian_ mchua, BryanWB: thanks, i'll look into that
10:25 BryanWB lucian_: he might be able to give u tips if u send him an e-mail om _at_ olenepal dot org
10:25 Webified done?
10:26 going once . . . going twice . . .
10:26 lucian_ BryanWB: pretty much, yes
10:26 BryanWB #TOPIC Groupthink
10:26 bemasc: how things?
10:26 bemasc So I posted a new version of my demo.  I basically totally rewrote the algorithmic core.  It's actually a lot shorter now, and somewhat faster.
10:27 It's still not fast enough to write an essay, but it's fast enough for nontrivial uses, so I think it's time for me to move on to another feature.
10:27 I might come back and try to improve performance later.  I believe it should be possible to use a truly "fast" algorithm, but it's a lot of work.
10:28 BryanWB anyone have an idea for bemasc?
10:28 bemasc The next feature on my roadmap is persistence (i.e. save/load with the Journal).  That's a critical feature, seeing as asynchronous collab is a big part of the project's motivation.
10:28 BryanWB bemasc: i would love to see a simple version of paint use groupthink
10:29 bemasc: sounds like u r on track
10:29 bemasc BryanWB: I agree, though that's not part of my GSoC proposal (the proposal specifically states that I will develop a shared text editing system).
10:29 walterbender bemasc: I think you should focus on what you are doing...
10:29 making it simple and robust
10:30 BryanWB walterbender: +1
10:30 and let others add features later
10:30 bemasc Persistence is not so easy.  My original plan for persistence was to use D-Bus's own serialization format, the one Telepathy uses to send bits over the wire, to save files to the Journal.
10:30 walterbender FYI, we'll be testing it again tomorrow in the collaboration testing session
10:31 homunq_ and?
10:31 bemasc Unfortunately, the functions for doing this are sort of buggy/untested.  I've gotten one patch into upstream so far, but it still doesn't work.
10:31 Also, even if it does work, it will require a some new version of libdbus.
10:31 and dbus-python.
10:32 So... maybe I should be trying to come up with a fallback serialization method.
10:32 BryanWB bemasc: i like json, if only because it is the only serialization method i can grok ;)
10:32 s/method/format/
10:33 bemasc BryanWB: hmm.  I'll see if json fits into this square hole.  Pickle doesn't, unfortunately.
10:34 The other issue I have run into is that the Journal does not really remember activity sharing parameters correctly.
10:34 BryanWB bemasc: have u addressed this w/ tomeu or erikos?
10:35 bemasc So if two people start on a shared document, turn off their computers, go home, work on the document offline, and then come back to school the next day, there's no way for them to merge those two sessions.
10:35 I filed a bug at dev.sugarlabs.org
10:35 I'll probably have to fix it myself, though I don't really even know what component it's in.
10:35 aa bemasc: do you remember the #?
10:36 bemasc http://dev.sugarlabs.org/ticket/935
10:36 aa thx :)
10:37 BryanWB bemasc: i think we have 25 mins left, do u mind if we wrap up your section and move on to the next project?
10:37 bemasc next!
10:38 BryanWB #TOPIC  Materializing Sweetness
10:38 aa: where vamsi?
10:38 homunq_ I'm sorry, I have to go now.
10:38 aa no idea, I pinged him
10:38 BryanWB homunq_: enjoy your trip
10:39 homunq_ silbe, I am going to Nicaragua. I'll try to make our meeting thursday, but no guarantees at all.
10:39 BryanWB #TOPIC  journal and silbe
10:39 silbe homunq_: no problem
10:39 aa homunq_: take care!
10:39 silbe homunq_: enjoy your stay
10:39 lucian_ homunq_: bonne voyage!
10:40 silbe not much progress this week as i'm busy working on the slides for my seminar talk tomorrow. started working on the VCS backend module, though.
10:40 subzero homunq_: cia
10:40 ciao*
10:40 walterbender silbe got distracted by Turtle Art :)
10:40 BryanWB walterbender: i find it very distracting too ;)
10:40 silbe creating repo and initial entry is working fine, working on getting a list of versions now
10:40 walterbender: dumped it for Open Office :-P
10:41 BryanWB silbe: need anything from humble band of Sugaristas or should we move on to next project?
10:42 silbe BryanWB: let's move on, i hope to have more to show on thursday
10:42 BryanWB silbe: cool
10:42 #TOPIC Karma
10:42 pretends to be unawares
10:42 subzero: how is progress?
10:42 subzero link: http://karma.sugarlabs.org/quadrilaterals
10:43 thats the quadrilaterals activity
10:43 I had troubles with buttons, so I created my own version (Karma version ):)
10:43 It's almost complete
10:43 BryanWB subzero: looks great, what have u figured out and what problems are u still having?
10:44 subzero well, I'm using masterHandler for mouse, so it calls all the lsiteners
10:44 problems: it runs slow on the xo
10:45 I'm going to test it with epiphany
10:45 BryanWB subzero: when it tried it out today it was quite slow w/ Firefox 3.5, have u tried it out yourself since we talked last?
10:46 subzero not yet, I'm in the university, the xo is in my room
10:46 BryanWB subzero: yeah, don't take my work for it, try it yourself
10:46 subzero BryanWB: I'm wondering how to reduce the amount of required ram
10:46 BryanWB: ok
10:47 BryanWB subzero: also, when I ran FF the buttons were huge, i hadn't adjusted it to the screen resolution
10:47 subzero: perhaps that could have affected it
10:47 subzero mmm, damn...
10:47 BryanWB subzero: so 1) finishing activity 2) performance 3) then what?
10:47 subzero I'll test it today
10:48 BryanWB subzero: have u slept? in the last 24 hours? i have been talking to u for the last 14 hours
10:48 aa iwikiwi: welcome!
10:48 BryanWB iwikiwi: welcome!
10:48 subzero wiriting the other componets? textfield, drawCircle, and so on
10:48 BryanWB: nop
10:48 iwikiwi aa, BryanWB: thanks
10:48 BryanWB subzero: ha ha
10:48 lucian_ iwikiwi: hello, mate!
10:48 iwikiwi lucian_: heya!
10:49 BryanWB subzero: great progress, have anything else before we move on?
10:49 subzero BryanWB: I lost time using the XO, so... I tried to do what I say I'll do
10:49 BryanWB: nope
10:50 BryanWB #TOPIC printing
10:50 aa ok, good news is iwikiwi got his XO ;)
10:50 iwikiwi also, I'm working on a print 'module' atm, the skeleton and settings are there, need to add features
10:51 BryanWB iwikiwi: how much will u have to hack moodle?
10:51 iwikiwi BryanWB: This is whats troubling, I cant gauge moodle properly
10:51 BryanWB iwikiwi: how so?
10:52 iwikiwi Its really huge, and I cant figure out a few things without live support, which is becoming harder to find
10:52 BryanWB iwikiwi: yeah, the moodle community seems to communicate primarily thru forums and not chat
10:52 aa iwikiwi: config files are imported
10:53 and global $CFG is used throughout
10:53 lucian subzero: about drawCircle, have you tried cake.js ?
10:54 sorry, wrong channel
10:54 BryanWB iwikiwi: perhaps the moodle gsoc participants could help?
10:54 iwikiwi aa: yeah, I know. I found a skeleton to work on. I am working on that atm. But since maybe I started work on it for only 10 hrs now, I am feeling it akward.
10:54 BryanWB: well the thing is actually i took help from them and ended up building a assignment plugin. Which had been a moodle devs suggestion
10:55 and also I had took a 3 and half days break this week due to family issues, I am a bit lagging on this weeks progress other than getting XS to work good for me
10:56 BryanWB iwikiwi: for me the XS often feels like a timesink and can be very frustrating
10:56 iwikiwi my dev env had been specifically moodle and apache till now, but martin had adviced i work on xs
10:57 BryanWB iwikiwi: u should use the xs, but don't let ejabberd problems slow u down at all, that is where time disappears
10:58 iwikiwi yep, thankfully I have it set up well now. All thanks to finding dabeb online
10:58 and matching times with him
10:58 BryanWB we are running out of time folks , i believe marketing meeting starts soon
10:58 iwikiwi: anything else?
10:58 aa: have anything to add?
10:59 aa nope, its been a slow week :)
10:59 iwikiwi well, I'm worried about my progress with moodle, thats about it
10:59 highly worried
10:59 BryanWB iwikiwi: keep at it and u will figure it out
10:59 aa iwikiwi: count on me, as always
10:59 bemasc iwikiwi: even if you can't get moodle working, if USB printers work that is a big win.
11:00 iwikiwi bemasc: about that, I was thinking I focus more on the sugar side and do moodle as a follow up.
11:00 but I think that will again bring debates and arguments
11:01 aa the moodle part shouldnt be that hard, it true however that moodle is... not very intuitive
11:01 bemasc iwikiwi: you and aa are in charge.  You don't have to listen to anyone else.
11:01 anyway, we're out of time
11:01 iwikiwi okay
11:01 BryanWB aa: moodle sometimes feels like a big flat mess
11:01 at least to me
11:01 sorry mlxsa
11:01 algo mas? ma od? shto novikovo? Haita?
11:01 aa with sometimes -> always
11:01 BryanWB anything more?
11:02 SeanDaly gar
11:02 BryanWB #endmeeting

Minutes | Index | Today     Channels | Search | Join

Powered by ilbot/Modified.
Webmaster