« Previous day | Index | Today | Next day » Channels | Search | Join
All times shown according to UTC.
Time | Nick | Message |
---|---|---|
00:11 | lucian <lucian!~lucian![]() |
|
00:26 | mchua is now known as mchua_afk | |
01:07 | dirakx1 <dirakx1!rafael![]() |
|
01:07 | dirakx has quit IRC | |
01:21 | tabs has quit IRC | |
01:33 | lucian has quit IRC | |
04:43 | lucian <lucian!~lucian![]() |
|
05:06 | lucian has quit IRC | |
05:43 | tch has quit IRC | |
06:00 | satellit__ has quit IRC | |
06:32 | yama has quit IRC | |
06:38 | yama <yama!~yama![]() |
|
06:38 | yama has quit IRC | |
06:38 | yama <yama!~yama![]() |
|
07:55 | dogi <dogi!~nemo![]() |
|
09:08 | scorche has quit IRC | |
09:08 | scorche` <scorche`!~scorche![]() |
|
09:16 | m_anish_` <m_anish_`!~anish![]() |
|
09:16 | dirakx1 has quit IRC | |
09:16 | dfarning_afk has quit IRC | |
09:16 | m_anish_afk has quit IRC | |
09:17 | dfarning_afk <dfarning_afk!~dfarning![]() |
|
09:24 | dirakx <dirakx!rafael![]() |
|
09:32 | silbe <silbe!~silbe![]() |
|
11:15 | icarito <icarito!~icaro![]() |
|
11:32 | walterbender <walterbender!~webchat![]() |
|
11:57 | icarito has quit IRC | |
12:09 | nubae has quit IRC | |
12:23 | dirakx has quit IRC | |
12:36 | icarito <icarito!~icaro![]() |
|
12:36 | dirakx <dirakx!rafael![]() |
|
12:40 | icarito has quit IRC | |
12:46 | dirakx has quit IRC | |
12:48 | bernie | silbe, m_anish_`, dfarning_afk: meeting today? |
12:52 | silbe | bernie: I don't know any reason why we shouldn't meet |
12:54 | bernie | silbe: too few people? :-) |
12:54 | icarito <icarito!~icaro![]() |
|
12:55 | silbe | bernie: it's another hour till the meeting, so some of us might just not be online yet |
12:56 | bernie | silbe: ah... damn... i did not consider the daylight saving |
12:56 | -> back to sleep | |
12:58 | icaritox <icaritox!~icaro![]() |
|
12:58 | icarito has quit IRC | |
13:00 | icarito <icarito!~icaro![]() |
|
13:01 | silbe | g'night ;) |
13:03 | icaritox has quit IRC | |
13:05 | icarito has quit IRC | |
13:21 | icaritox <icaritox!~icaro![]() |
|
13:31 | icaritox has quit IRC | |
13:44 | icarito <icarito!~icaro![]() |
|
14:00 | icarito has quit IRC | |
14:04 | m_anish_` is now known as m_anish | |
14:04 | m_anish | bernie, i'm here |
14:05 | silbe | too |
14:05 | m_anish | silbe, hola! |
14:05 | silbe | moin moin :) |
14:07 | m_anish | anyone else arnd? :) |
14:08 | bernie | m_anish: hey! |
14:08 | ok... who shall chair? | |
14:08 | silbe: ping | |
14:08 | m_anish | bernie, silbe, i've uploaded the build to sunjammer. I'm taking nice screenshots and updating the wiki |
14:09 | bernie, you? | |
14:09 | silbe | m_anish: the DX2 build? |
14:09 | m_anish | silbe, yah |
14:09 | bernie | m_anish: did you restart the version numbering of DX2 from 0? i guess it's too late to change it now, but some users said that they were confused by the OS numbers restarting every time |
14:10 | m_anish | bernie, yah, its os85dx |
14:10 | silbe | nice. were there any changes to the dextrose repo in this week? |
14:10 | m_anish | it'll take me an age to upload a new build :/ |
14:10 | silbe, nope, other than olpc-bootanim (tch) | |
14:11 | bernie | m_anish: pls upload all files generated by o-o-b... not just the img + crc. that's another request i got some time ago. |
14:11 | silbe | m_anish: ok, good. |
14:11 | dirakx <dirakx!rafael![]() |
|
14:11 | m_anish | bernie, there's a tree file, about ~300mb, need that as well? |
14:11 | silbe | m_anish: +1 to tree file. That's what I use for testing. |
14:11 | dirakx | Hi all. ;). |
14:12 | silbe | dirakx: hi! |
14:12 | m_anish | silbe, ok |
14:13 | bernie | m_anish: are you also uploading the matching xo1.5 images? |
14:13 | m_anish | bernie, yah, just managed to build it, not upload, will do that |
14:14 | bernie | m_anish: it will take you a while at paraguayan network speeds :-) |
14:14 | m_anish | bernie, internet has been very sucky lately, and 3G takes ages to upload, and office is just to slow |
14:14 | i'll do it overnight | |
14:14 | bernie | m_anish: i *think* the 3G modems have a monthly transfer limit, after which they become super slow |
14:14 | s/modems/contracts/ | |
14:15 | m_anish | until a few days ago, robbie was overloaded, uptime was ~15, then they restarted it, and i was able to make the build |
14:15 | anyway, startmeeting, ? | |
14:15 | bernie | m_anish: ah yeah |
14:15 | m_anish: wanna chair? | |
14:15 | m_anish | ok |
14:15 | #startmeeting | |
14:15 | meeting | Meeting started Sat Mar 26 14:15:55 2011 UTC. The chair is m_anish. Information about MeetBot at http://wiki.debian.org/MeetBot. |
14:15 | Useful Commands: #action #agreed #help #info #idea #link #topic #endmeeting | |
14:16 | m_anish | #topic status updates |
14:16 | silbe, you wanna start :)? | |
14:17 | silbe | m_anish: not much to mention. Did some reviews upstream and worked a bit on the DX3 image, but more work than result. :-/ |
14:18 | bernie | m_anish: not much from me as well... lasered my eye and worked for rms |
14:18 | dirakx | ;) |
14:18 | bernie | dirakx: ah, didn't notice you! |
14:18 | m_anish | bernie, ok, both eyes functional? |
14:18 | silbe, ok | |
14:19 | dirakx | for me. various activities new releases and pootle documentation now complete Service/translate in wiki. |
14:19 | bernie: ;) | |
14:19 | bernie | m_anish: (side note) after we have all the images uploaded and mildly tested, we can proceed update the wiki to move the dx1 info to its own subpages and put the dx2 releases on the main page! |
14:19 | m_anish | silbe, i'd probably be free to start working on dx3 by apr 1 |
14:19 | silbe | I added a serial console to my XO-1.5 again to diagnose random crashes (probably suspend/resume related), but did it incorrectly (a few millimetres off so the adapter wouldn't fit) so I have to rework it. :-/ |
14:19 | bernie | m_anish: my vision in the operated eye is still blurry... it's expected to improve by next week |
14:20 | silbe | m_anish: great! |
14:20 | m_anish | silbe, ok |
14:20 | bernie | dirakx: thanks! |
14:20 | m_anish | dirakx, can you update on ur + icarito's progress w/ activities? |
14:21 | dirakx | m_anish: icarito has been working on a) new toolbars for various activitires...and b)A new activity for Peru.. |
14:22 | silbe | BTW: Simon has done some work w.r.t. Telepathy bugs, but AFAICT the major ones that cause Traceback spam in the logs haven't been fixed yet. |
14:22 | dirakx | oh he is not around to tell us more. |
14:24 | m_anish | dirakx, ok, what's the major items on your todo list for the next week or so? |
14:25 | dirakx | m_anish: next week I have to ''order'' and review all core strings for pootle on sugar...and working on the high/medium priority list of activities. |
14:25 | m_anish: Icarito has to work on those also. | |
14:25 | acording to our work distribution. | |
14:26 | m_anish: just to notice that .PY people can change that list according to situation.. | |
14:26 | m_anish | dirakx, great. do you have a very rough plan atm of what we'd be able to do within the next 2-3 months activity wise, and whether you'd need more ppl to test? |
14:27 | bernie | dirakx: is pootle behaving well? |
14:27 | m_anish | dirakx, yep, i wasn't really able to discuss it with them much and I don't wanna just fwd it, otw they won't look at it |
14:27 | dirakx | bernie: Still has some issues..:(, it's not updating POTs well. |
14:28 | m_anish: tsting has to centered in those activities..I mean the activities that .PY use. | |
14:28 | bernie: I'm consulting with Saya the cause of that. | |
14:29 | bernie | dirakx: what version are we running now? |
14:29 | dirakx | m_anish: patched 2.0.1/2.0.5 |
14:29 | bernie | dirakx: ok. does it solve erikos' bug? |
14:29 | dirakx | oh bernie |
14:30 | bernie: nop. we found that is a xgettext bug btw. | |
14:30 | silbe: help in debug. | |
14:30 | bernie | dirakx: ah, so not in pootle? |
14:31 | dirakx | bernie: afaik nop. |
14:32 | bernie | m_anish: is the microformat updater working from aslo? |
14:32 | tch <tch!~tch![]() |
|
14:32 | m_anish | bernie, been working for a while now |
14:32 | dirakx | pootle was very broken but we are getting there to fix it. |
14:33 | silbe | dirakx: ? |
14:33 | tch | hello guys :) |
14:33 | silbe | tch: hey! |
14:33 | dirakx | silbe: we were talking about record's xgettext bug.. |
14:33 | tch: hello :) | |
14:33 | m_anish | tch, late :P |
14:34 | tch | current topic? :) |
14:34 | m_anish | (status updates) |
14:34 | silbe | dirakx: that reminds me: have you forwarded your report to the maintainer? there was some reply indicating that he might not have gotten a copy. |
14:35 | dirakx | silbe: gosh not yet..I'll forward it. |
14:35 | silbe | dirakx: thx |
14:35 | bernie | tch: hola |
14:36 | m_anish: ugh, i just figured out that we had indeed already released some early dx2 alphas and the latest one was labeled os439dx... | |
14:37 | m_anish | bernie, yah :/ |
14:37 | bernie | m_anish: if now we go back to os85dx, people will be super confused |
14:37 | silbe | does anyone (except buildbot) still use the "subuntu" VM on treehouse? I'm having a rather hard time getting it to boot :-/ |
14:37 | m_anish | if its imp, i'll change it to os500 or something :-) |
14:37 | bernie | m_anish: sure, why not... but you'd have to reupload the whole thing :-( |
14:38 | m_anish: maybe you could build on the dextrose vm? we still have it | |
14:38 | m_anish | bernie, i'm dreading that already |
14:38 | bernie, we do? i only checked that sunjammer is ubuntu and jita is f14 | |
14:39 | tch | m_anish: I could reupload it from home, faster than the pyedu office at least |
14:40 | bernie | m_anish: i'll give you an account in no time |
14:40 | m_anish | tch, i'll try to build from here, but i might have to be present at the office to get it to work, the h/w infrastructure has been a lot slower/flakey lately |
14:41 | bernie, sounds good | |
14:41 | tch | yeah, I think car|os is still tweaking the VM setup |
14:42 | m_anish | tch, time for our status updates. ;) you wanna do the honors? |
14:43 | tch | sure |
14:43 | no meeting bot? | |
14:44 | m_anish | tch, its whirring, go ahead |
14:47 | tch | Ok. Well, basically this week we been struggling with the infrastructure problems trying to get the international build and extending the dx services. |
14:47 | bernie | m_anish: ssh anish![]() |
14:48 | tch | Anish tried all weekend to get these builds done, while finishing the update services we need for the offline schools |
14:48 | bernie | m_anish: jita should also work to build dx2 images... i've been building them on my laptop which is f14 x86_64 |
14:48 | m_anish | bernie, works, thanks |
14:49 | bernie | tch: you can also have an account on dextrose.sugarlabs.org, or jita.sugarlabs.org... whichever you prefer :) |
14:49 | tch | and myself I simply extended the feedback server to be able to send the reports to a mailing list, so we can follow the reports |
14:49 | bernie | ah, no... building requires root... so better doing it on a machine that doesn't also run production services |
14:49 | tch: nice | |
14:50 | tch | it wll be useful to to keep track of the feedback reports :) |
14:51 | Guess the next step should be creating tickets automatically haha (??) | |
14:51 | the important thing is that we already have a way to access these reports. | |
14:52 | m_anish | tch, great work. I'm workin on the rpms. |
14:52 | bernie | tch: are there many reports? |
14:52 | dirakx | tch: are these possible to see from outside ?. just to check if there are some activity wise. |
14:52 | bernie | m_anish: SMParrish and I have already been building images on dextrose.sugarlabs.org in the past... should be straightforward |
14:53 | m_anish | bernie, ok |
14:53 | tch | well, do the maths, dx2 50 testers and we have like 5-6 (??) reports, is not bad |
14:53 | bernie | m_anish: i just gave you root access, btw |
14:53 | m_anish | bernie, ok |
14:54 | tch | dirakx: is a private mailing list, but we could forward reports with paraguay educa permission |
14:54 | dirakx | would be grreat. |
14:54 | great* | |
14:55 | if possible. | |
14:56 | bernie | tch, m_anish: can someone subscribe me pls? |
14:56 | tch | m_anish: anything else I am missing? |
14:56 | m_anish | bernie, dirakx ACK |
14:56 | tch, standby | |
14:56 | dirakx | m_anish: thx |
14:57 | tch | bernie: yeah we can, just need to ask roberto or pacita for permission. Thats actualy the idea behind the ML ;);) so we can discuss the issues there |
14:58 | m_anish: ^ | |
14:58 | m_anish | tch, nope, Its good, the todo's I have (still) are (1) browse homepage change via update, (2) builds (3) making a small doc/wiki abt the features on dx-2 |
14:58 | tch, +1 | |
15:00 | ok, what else do we need to discuss? | |
15:00 | tch | visa details? |
15:00 | haha | |
15:01 | m_anish | #topic visa |
15:01 | bernie | tch: k |
15:01 | m_anish | tch, can we call pflores or somebody up in uy re: the invitations |
15:01 | tch | yup, we will need one for m_anish and aleksey probably |
15:02 | bernie | ah, guys... i still have no funding for the plane ticket... |
15:03 | i'm not sure i want to spend >$1200 from my pocket to attend a conference, even if I'm *VERY* interested... | |
15:03 | m_anish | bernie, i learnt in the last few weeks that its cheaper to fly from US-Ind than it is for US-Py, i had intentions of coming to US :/ |
15:03 | dirakx | :(. I also want to go..but economics are not o.k yet. |
15:04 | m_anish | anyway, we're done for the meeting? |
15:04 | tch | can't you ask the fsf for sponsorship, so you can give a talk about free software in education? which is the HOT TOPIC in latin america right now |
15:04 | bernie: ^ | |
15:04 | dirakx | that would be an option, bernie. |
15:05 | tch | people from the ministry.of.education.py is planning to go |
15:05 | bernie | tch: the fsf has a policy of not sending off employees to conferences. at least, this is what the previous ED told me. |
15:06 | dirakx | tch: great. |
15:06 | bernie | tch: rms is our traveling agent :-) |
15:06 | m_anish: there are some strange games in os85dx :) | |
15:07 | m_anish | bernie, all on py's request |
15:07 | :) | |
15:07 | tch | bernie: change the rules! |
15:07 | m_anish | i removed gc, tp from the buids |
15:07 | bernie | m_anish: didn't you build the image off the intl ini file? |
15:08 | m_anish | bernie, yep, i thought it might make sense to keep activities consistent? |
15:08 | bernie, then we'll receive testing feedback on the activities that py uses as well | |
15:09 | bernie | <sarcastic comment on the eduteam suppressed> |
15:09 | :-) | |
15:11 | m_anish | :-) |
15:11 | i guess we can endmeeting now | |
15:11 | 3 | |
15:11 | bernie | m_anish: it makes sense to help .py in this way, but at the same time the intl dextrose build is supposed to showcase dextrose for other deployments too |
15:12 | m_anish | bernie, we can package some more activities, but keep the set py uses |
15:12 | bernie | m_anish: +1 |
15:12 | dirakx | m_anish: +1 |
15:12 | m_anish | bernie, and have the updater url have some more installable |
15:12 | bernie | m_anish: space isn't an issue except for gcompris and crap like that |
15:12 | tch | lol.. |
15:13 | m_anish | bernie, dirakx i'll bug you in a while on this then |
15:13 | dirakx | m_anish: yep np. |
15:13 | m_anish | oops i meant just dirakx |
15:13 | bernie | wow, has gcompris been dropped? can't believe my eyes!! |
15:13 | m_anish | bernie, just from intl build, (don't get too happy) |
15:13 | bernie | haha :) |
15:13 | m_anish | it didn't make sense to increase the size by 150 mb just bec of one activity |
15:14 | tch | m_anish: we should insist in not providing it with the build, and only by the act updater |
15:14 | m_anish | tch, you mean for py? |
15:14 | tch | yup |
15:14 | m_anish | tch, i'm afraid i tried that already |
15:14 | they told me to keep another build without gc handy if they start running out of space on badly corrupted flash drives | |
15:15 | bernie | m_anish: the tetris mat activity comes up in spanish... it probably doesn't have i18n |
15:15 | m_anish | tch, i have to take their word in this that its actually being used, however big it is |
15:15 | bernie, bug report :) | |
15:15 | bernie | tch: +1 |
15:16 | m_anish: when i asked, they told me "kids like it". and i replied: "kids also like doom, shall we put it in?" | |
15:18 | endmeeting? | |
15:18 | m_anish | bernie, tch i beg to differ here. I think we'll be stepping on toes if we start insisting on removing activities, i made them understand that the build will be a lot bigger because of just GC |
15:18 | and TP | |
15:18 | i can't go fighting with them :-) | |
15:19 | 2 | |
15:19 | 1 | |
15:19 | 1 | |
15:19 | #endmeeting | |
15:19 | meeting | Meeting ended Sat Mar 26 15:19:28 2011 UTC. Information about MeetBot at http://wiki.debian.org/MeetBot. (v 0.1.4) |
15:19 | Minutes: http://meeting.sugarlabs.org/s[…]-26T14:15:55.html | |
15:19 | Log: http://meeting.sugarlabs.org/s[…]11-03-26T14:15:55 | |
15:19 | tch | guess we are too professionals to switch versions :,( |
15:19 | dirakx | thanks all. |
15:19 | bernie | m_anish: yes, you're right. i think it will take them some time to see the *consequences* of making the build bigger. now they might be thinking "who cares?". |
15:19 | tch | lol, |
15:20 | bernie | m_anish: did they make gcompris protected too? |
15:20 | m_anish | bernie, well if it helps they always have the option of removing it |
15:20 | bernie, not protected anymore | |
15:20 | silbe, dirakx, thanks! | |
15:21 | dirakx | m_anish: ty |
15:21 | m_anish | tch, bernie i guess we're still talking :-) |
15:22 | silbe | m_anish: thanks for what? |
15:22 | m_anish | silbe, meeting :) |
15:22 | silbe | m_anish: ah. yw. :) |
15:23 | m_anish | bernie, i haven't been able to reproduce the coredump btw :/ |
15:24 | bernie | #realendmeeting |
15:24 | :-) | |
15:24 | m_anish: if it's really really rare... then it doesn't annoy users that much | |
15:25 | m_anish: our goal is not to make dextrose perfect, just a lot better than the previous version :) | |
15:26 | m_anish | bernie, agree, just check if you see the coredump after using the build for a while. I think I had seen it on one more machine a while ago, but I don't have that (and it could be unrelated) |
15:28 | bernie | m_anish: btw, coredumps are normally turned off on linux systems... i think i've turned them on in dextrose to debug that nasty problem with sugar crashing on downloads. have you ever seen it? |
15:30 | silbe | bernie: I've had Sugar crash some times while the Journal was active - not (just) download related :-/ |
15:30 | m_anish | bernie, hmm does it go something like... you start a download through browse, then internet connectivity gets broken -> you reconnect -> try to resume the downloading journal entry -> sugar crashes. i think i may have seen something on those lines... |
15:31 | silbe | m_anish: if you find any recipe to reproduce the crash, that would be fine. It's always been totally random for me :( |
15:32 | m_anish | silbe, bernie there was another 'design' question. Folks using the xo are very reliant on using browse (and by extension google services) on the laptop. for example, they use google image search, blogger, etc a lot, However, google keeps on making newer avatars of these portals clunkier and slower, but we don't really have fast(er) hardware to complement that. |
15:32 | silbe, yea :/ | |
15:32 | silbe, but that was the vague sequence of actions | |
15:33 | silbe, and the xo-1 h/w is here to stay for a while i guess | |
15:33 | bernie, ^ | |
15:33 | silbe | m_anish: I still don't grok why people do that. It's slower, more expensive, produces more CO2, requires a permanent internet connection, etc. Where's the benefit? |
15:35 | m_anish: FWIW, I don't think there's much if anything we can do to improve performance. The browser engine and the network connection are the bottlenecks, not the tiny bits of code to make the browser look like a Sugar activity. | |
15:35 | m_anish | silbe, its very inefficient b.w. and/or productivity wise considering how much the XO is built around optimizing things |
15:36 | bernie | m_anish, silbe: i think that crash has a small probability of happening when sugar is doing dbus communication... this is why it happens frequently when you browse the journal or download a file |
15:36 | silbe | m_anish: a local proxy configured to selectively block flash ads, javascript snippets etc. might be worth a try, but it's a high maintenance overhead option. |
15:36 | bernie | m_anish, silbe: the crash happens in the garbage collector of python, probably due to some memory corruption that occurred earlier... so it's really hard to find |
15:36 | m_anish | silbe, i guess i see a problem here. 2 yrs from now, google(and internet ingeneral) will be clunkier yet, and we'll still have xo-1's around |
15:37 | bernie | m_anish, silbe: the python maintainer of fedora told me that in f15 the bug will be covered because they built cpython with asserts disabled :-) |
15:37 | m_anish: maybe we should do that too... rebuild the python rpm with debug disabled... it might even make sugar a little faster. | |
15:37 | m_anish | silbe, maybe load mobile versions of websites (i knw its not ideal) |
15:37 | but that's the equivalent processing power on a xo-1 i guess? | |
15:38 | silbe | m_anish: for XO-1 there's only one answer: do as much as possible locally. Even for XO-1.5 I recommend that, but it's not crucial. |
15:38 | lucian <lucian!~lucian![]() |
|
15:39 | m_anish | bernie, yep |
15:39 | silbe | bernie: I don't think that's really a good idea. Best case Sugar keeps working with corrupted data. Chances are it just crashes a short time later. |
15:40 | m_anish: not quite. Your average 2010 smart phone should have roughly twice the computing power and comparable to twice the memory of the XO-1. | |
15:40 | m_anish | silbe, i think it might be easier said than done, (staying local) |
15:41 | silbe, what this basically implies that the browser on the xo-1 is already half dead, and it will be completely dead in another 1-2 years :/ | |
15:42 | silbe | m_anish: it's mostly a social issue. We still have all the tools from before the Cloud era and even some new ones (e.g. sup as MUA). |
15:42 | bernie | silbe: then maybe we should compile cpython with *MORE* debug stuff in it and hope to see the crash during a download of a big file or somethong |
15:42 | thing | |
15:43 | btw, the bug is this one: http://bugs.sugarlabs.org/ticket/2064 | |
15:43 | silbe | m_anish: more or less so, yep. Blame the mainstream mentality. Most web site owners don't even care for non-IE browsers. |
15:43 | bernie: +1 | |
15:43 | bernie | silbe: does someone _still_ care about IE?? |
15:44 | lucian | silbe: i doubt that's entirely true |
15:44 | m_anish | bernie, haha, its still the dominant browser, isn't it? |
15:44 | bernie, i know a few govt portals in ind who design only for IE, it sucks :/ | |
15:44 | lucian | is surprised that is even legal |
15:45 | silbe | bernie: and we should move forward with splitting Sugar (shell) into separate processes to narrow it down some more (though the Journal already is a good candidate). |
15:46 | lucian | points at gnome-shell |
15:47 | silbe | lucian: ok, granted, any claim saying "most" for such a large number of different entities should raise red flags. :) Let's just say "a surprisingly and annoyingly high number of sites". |
15:47 | m_anish | silbe, i guess this is worth discussing over m-l, see if someone has a few bright ideas? |
15:47 | silbe | m_anish: context Browse or crash? |
15:48 | m_anish | silbe, i guess we both agree there is a problem here |
15:48 | silbe | m_anish: it's hard not to agree |
15:49 | m_anish | lucian, money : gravitational force :: legality : general relativity ;-) |
15:49 | silbe, context=browse | |
15:49 | silbe, i guess so | |
15:50 | silbe, bernie can you tell from experience whether browse was a lot smoother 2-3 years ago? | |
15:50 | silbe | m_anish: the problem partly is that the XO-1 hardware is so outdated that even most of the phones that are capable of running a web browser have more resources than the XO-1. |
15:50 | m_anish | silbe, +1, but that isn't gonna change, and we want to be able to provide the same level of services via browse as before |
15:50 | silbe | m_anish: it has always been slow on XO-1 for me. Haven't noticed any change. |
15:51 | m_anish: one idea might be to play with compressed swap to relieve the memory pressure a bit. Not sure if it would help or make things even worse. | |
15:51 | m_anish | silbe, i've noticed it to be painfully slow in classrooms in caacupe (without any core dumps) |
15:52 | lucian | silbe: make things worse, i'd guess. that's what happens on my G1 |
15:53 | silbe | m_anish: but I think the only solution that's going to make a major difference (as opposed to a few percent speed-up) is going local. |
15:53 | lucian | well, webkit *is* better on both cpu and memory usage |
15:53 | (and don't say firefox is efficient until you've actually proven it) | |
15:53 | if the xo-1 got a recent enough fedora, it could run the upcoming Surf | |
15:53 | i guess | |
15:54 | silbe | lucian: I always said Mozilla is inefficient, and I have proven it. Well, compared to a text browser that is. ;) |
15:55 | lucian | silbe: i wasn't necessarily talking to you. whenever webkit was mentioned, people jumped to defend xulrunner's efficiency |
15:55 | then when i proved it was a hog several times, no one bothered to admit they were wrong :) | |
15:55 | silbe | lucian: but I'm really curious how Surf will compare with Browse in terms of performance. |
15:55 | <g> | |
15:55 | lucian | silbe: it's already faster, just incomplete |
15:56 | bernie | m_anish: IE is still dominant, but not for long: http://marketshare.hitslink.co[…]d=0&qptimeframe=M |
15:56 | lucian | with py-gi and some time, it'll catch up in terms of features |
15:58 | bernie | m_anish: Browse on the XO? it's always been quite sluggish in my memory, but nowadays i see users trying to use it with maps.google.com or gmail.com, which makes it even more painful. |
15:58 | lucian | is there any chance of getting a recent enough OS on it to have py-gi? |
15:59 | bernie | m_anish: i remember that many years ago we fixed a problem with scaled bitmaps in xulrunner... they were taking up huge amounts of memory because xulrunner was keeping 3-4 copies of the same image in memory |
15:59 | lucian: dx3 will be based on f14. is that recent enough? | |
15:59 | silbe: when can we have a test image of dx3? | |
16:00 | silbe | bernie: I'll try to get one out today. |
16:00 | m_anish | bernie, i would've suggested using mobile versions of websites, but that is a quickfix and will result in big changes in the way some websites look n feel :/ |
16:02 | bernie | m_anish: here IE seems even below 50%: http://en.wikipedia.org/wiki/U[…]e_of_web_browsers |
16:03 | lucian | bernie: i haven't checked, but probably yes |
16:03 | bernie | m_anish: this one is even more amazing: http://en.wikipedia.org/wiki/F[…]er_share_crop.png |
16:03 | lucian | bernie: i think it just reflects my own experience |
16:04 | i very rarely see anyone use IE | |
16:04 | either firefox or chrome | |
16:04 | silbe | bernie: don't trust any statistics you haven't forged yourself ;) |
16:04 | lucian | several forged statistics showed IE to be weak in EU :) |
16:05 | bernie: even if f14 doesn't have quite new enough py-gi and webkit/libsoup, i'm sure they can be added much more easily | |
16:06 | m_anish | lucian, most ppl here I see use either IE on win7 or safari on apple, most people back in Ind are stuck on ie6/7 on old computers |
16:06 | lucian | m_anish: that's sad. and odd to me |
16:06 | has to go | |
16:07 | m_anish | lucian, ok chau! |
16:08 | bernie | lucian: i think there are 2 ways to count browsers. The majority of occasional computer users are probably still using IE, but the majority of people who use the internet intensively have switched |
16:09 | lucian: if one wants to know the market share just for targeting consumer with ads, they'd better target the IE users because it's more likely that they'd be vulnerable to simple marketing tricks | |
16:12 | silbe | lucian: cu |
16:14 | m_anish | - afk for a while |
16:16 | m_anish is now known as m_anish_afk | |
16:25 | m_anish_afk is now known as m_anish | |
17:21 | icarito <icarito!~icaro![]() |
|
17:55 | valhalla has quit IRC | |
17:56 | valhalla <valhalla!~valhalla![]() |
|
18:22 | walterbender has quit IRC | |
18:22 | dirakx has quit IRC | |
18:35 | scorche` is now known as scorche | |
18:36 | yevlempy <yevlempy!~yevlempy![]() |
|
19:03 | m_anish is now known as m_anish_afk | |
19:10 | dirakx <dirakx!rafael![]() |
|
19:16 | scorche has quit IRC | |
19:19 | tch has left #sugar-meeting | |
19:19 | yevlempy has quit IRC | |
19:51 | scorche <scorche!~scorche![]() |
|
19:53 | icarito has quit IRC | |
20:07 | icarito <icarito!~icaro![]() |
|
21:01 | yama` <yama`!~yama![]() |
|
21:01 | yama` has quit IRC | |
21:01 | yama` <yama`!~yama![]() |
|
21:03 | yama has quit IRC | |
21:11 | lucian has quit IRC | |
21:17 | icarito has quit IRC | |
21:27 | lucian <lucian!~lucian![]() |
|
21:31 | icarito <icarito!~icaro![]() |
|
21:36 | lucian has quit IRC | |
21:51 | icarito has quit IRC | |
22:06 | silbe has quit IRC | |
22:27 | dirakx has quit IRC | |
22:32 | scorche has quit IRC | |
22:38 | scorche <scorche!~scorche![]() |
|
23:14 | dirakx <dirakx!~rafael![]() |
|
23:18 | icarito <icarito!~icaro![]() |
|
23:32 | dirakx has quit IRC | |
23:35 | dirakx <dirakx!~rafael![]() |
|
23:37 | icarito has quit IRC |
« Previous day | Index | Today | Next day » Channels | Search | Join