**** BEGIN LOGGING AT Sun Feb 03 02:59:59 2013 Feb 03 07:27:47 hello everyone Feb 03 07:29:15 does anyone know how to set the default zoom of microb? i dont see it on the about:config. Feb 03 08:00:35 MrPingu: with headset problems i like to restart bluetoothd and pulseaudio Feb 03 08:00:57 oh, the ones with a cable? just pulse, maybe Feb 03 09:20:30 kerio: thanks, yes the wired, I already rebooted but next time i will restart PA (I could have thought of that myself) Feb 03 11:02:06 hm, rmo doesn't work at all now, does it? Feb 03 11:02:29 oh, it does Feb 03 11:02:36 kerio: very very slow (unusable) Feb 03 11:03:47 openssh in maemo is asking for a password but its not taking the root passwd i set it to Feb 03 11:04:42 are you sure you're logging in as root? Feb 03 11:10:23 hmm... Feb 03 11:12:01 go it Feb 03 11:12:04 thanks Feb 03 11:12:50 where is the sources /etc/apt/sources.list ? Feb 03 11:13:04 i can't parse your question Feb 03 11:13:09 which sources? Feb 03 11:13:24 apt sources for n900 Feb 03 11:13:37 the repositories, then Feb 03 11:13:48 yea Feb 03 11:13:53 /etc/sources.list.d/ Feb 03 11:13:54 HAM puts his repos in /etc/apt/sources.list.d/hildon-application-manager.list Feb 03 11:14:09 it will also use every other repo in /etc/apt/sources.list and /etc/apt/sources.list.d/* Feb 03 11:14:19 mmm ham *drowls* Feb 03 11:14:34 ah ha Feb 03 11:14:43 where can i get fapman Feb 03 11:14:46 ~fapman Feb 03 11:14:47 from memory, fapman is Faster Application Manager, a bad package manager that causes problems, don't use it, ever Feb 03 11:15:06 ham is slow Feb 03 11:15:22 i never use fapman with unstable repos Feb 03 11:15:27 else i hadn't problems Feb 03 11:15:35 I'd also point out that "apt-get update; apt-get upgrade" can potentially mess up things :) Feb 03 11:15:36 and fapman is all-around awful Feb 03 11:15:41 ShadowJK: MEH Feb 03 11:16:14 i wanted upgrade to 2.32 kernel b/z thats what i used on my desktop Feb 03 11:16:37 use* Feb 03 11:16:42 fapman is better if you want to install a large number of apps and don't know the apt for it Feb 03 11:17:11 stable apps Feb 03 11:17:14 like themes Feb 03 11:17:16 fremantle only supports 2.6.28 anyway Feb 03 11:21:25 maemo extra repo is very very slow Feb 03 11:21:46 iluminator105: they're working on it Feb 03 11:22:40 is this one the mirror but its not much better either Feb 03 11:22:44 http://maemo.merlin1991.at Feb 03 11:34:43 accessing my n900 from ssh is the bomb! Feb 03 11:34:57 scp Feb 03 11:35:09 remote detonation by ssh? Feb 03 11:35:18 good idea Feb 03 11:35:23 attach some c4 to the usb port Feb 03 11:35:31 :P Feb 03 11:35:48 better than having a spam sms blow you up too early Feb 03 11:36:04 sudo c4detonate now Feb 03 11:36:23 Are you sure? Press Y to continue... Feb 03 11:37:04 why? Feb 03 11:37:35 y Feb 03 11:38:12 wasn't c4 used for warming up cans of beans or something in vietnam or korea Feb 03 11:38:55 i dont suggest attaching C4 to cans of beans, unless you want a big mess Feb 03 11:40:13 LauRoman, yeah I seem to recall it could be used in other ways besides blowing stuff up Feb 03 11:40:40 i heard it that somwhere but can't remember where Feb 03 11:42:01 When ignited with a flame rather than detonated with a primary explosive, C4 just burns, so American soldiers during the Vietnam War era would sometimes use small amounts of it as a fuel for heating rations.[citation needed] However, burning C4 produces poisonous fumes and should be avoided.[5] Feb 03 11:42:06 (from wikipedia) Feb 03 11:43:05 hehe, also: "Michael Herr in Dispatches, his book about the Vietnam War, relates that a soldier would occasionally ingest C4 from a Claymore mine in order to cause temporary illness so that he would be sent on sick leave." Feb 03 11:47:50 what is screen res of n900? Feb 03 11:49:33 can you upgrade the memory of n900 to 1gig Feb 03 11:49:48 no Feb 03 11:50:19 well in theory yes .. but i doubt you want some expensive investment in tools for that, and some failed tries Feb 03 11:52:15 how do you stream video from n900 Feb 03 11:53:51 iluminator105: I'm not aware of any software to do that simply Feb 03 11:54:12 I have used mencoder and net cat Feb 03 11:54:21 if you mean stream the camera Feb 03 11:54:38 hmm.. do you have a write up on it Feb 03 11:54:58 no Feb 03 11:54:58 eating c4 sounds stupid Feb 03 11:55:21 i doubt you've got enough leeway between sick leave and death Feb 03 11:55:37 theres also a dlna server for maemo Feb 03 11:57:11 i still dont know how many pixels the n900 is Feb 03 11:57:28 i want to create a wallpaper for it the optimal size Feb 03 11:58:17 horizontal vertical or different per desktop? Feb 03 11:58:27 800x480 Feb 03 11:59:31 or 3600x480 or 2400x480 or 1600x480 :) Feb 03 11:59:38 3200 Feb 03 12:01:15 cehteh: I have not found any source for the 1g ram, when I looked Feb 03 12:01:27 cehteh: or, indeed, the 512 Feb 03 12:01:58 yes .. i saied expensive tools and "in theory" .. build your own fab :) Feb 03 12:09:27 Got a spare dozen billion dollars? :) Feb 03 12:13:40 what the best phone to upgrade from an n900 then Feb 03 12:14:11 another n900, usually Feb 03 12:14:12 cehteh: it's not _that_ bad Feb 03 12:14:41 cehteh: you certainly could source the chips. Feb 03 12:15:11 just getting them in ones may not be an option Feb 03 12:15:50 did jolla announced some phones for the european market yet? Feb 03 12:16:17 hardware wise i'd like my wifes samsung note 2 .. but i cant stand the android Feb 03 12:16:34 not for any market Feb 03 12:16:51 maemo, meego or sailfish on that .. i would imagine thats awesome Feb 03 12:16:56 I wish someone would put up a nice is for the n4 Feb 03 12:17:41 i like nokia 920 but not att Feb 03 12:20:25 I wish that someone would make a phone other than a N900 with a physical keyboard that doesn't suck Feb 03 12:21:51 heretic Feb 03 12:22:24 I have been using my nexus seven w lot. Feb 03 12:23:04 to the point that that last sentence and this one are tuprd blind. Feb 03 12:23:45 (it's easy to tell when SpeedEvil switched to N7 for irc, from all the weird spelling correction) Feb 03 12:23:52 though I approach the same speed as on the n900, it is always more stressful, and there is never certainty Feb 03 12:24:18 with the n900, I could 100% of the time tell I had hit the wrong key Feb 03 12:24:27 without looking Feb 03 12:24:44 I get tired (like, physically) quickly when typing on transformer osk Feb 03 12:25:59 well.. i think the best keyboard was the n810 one's... but the size is bigger Feb 03 12:27:04 I think I like the N900 keyboard better, the key shape and travel are better Feb 03 12:28:40 cehteh: OMAP3430 doesn't support 1024MiB RAM. Sorry. It does support 512MiB RAM, but... Feb 03 12:28:49 nokia 920 with meego of course Feb 03 12:29:58 cehteh: ...you won't like it, you need to chips stacked on SoC. You create "tower" of SoC, RAM, RAM Feb 03 12:30:28 WizardNumberNext: there are, or were, four bank chips in the right footprint Feb 03 12:30:54 for 1924 Feb 03 12:30:59 1024 Feb 03 12:32:21 cehteh: obviously you need to research subject as memory controllers can use some organizations and cannot use other, so for instance it might support 256Mx8 and 128mx16, but definitely won't support 512Mx4. There is yet another organization, which you have to take in account Feb 03 12:32:50 SpeedEvil: from what t I might be wrong Feb 03 12:33:03 and it is not about banks, but ranks Feb 03 12:33:41 pretty much every RAM chip released after SDR100 spec was released is 4-bank chip Feb 03 12:33:55 s/bank/die/ Feb 03 12:34:10 SpeedEvil: now you make sense Feb 03 12:34:13 ;) Feb 03 12:34:45 but I am not sure, OMAP3430 can do four chips Feb 03 12:34:50 for sure two Feb 03 12:35:46 and theorycally should be able to do 4 chips as it have 2 chip select lines Feb 03 12:36:18 s/theorycally/theoretically/ Feb 03 12:36:18 WizardNumberNext meant: and theoretically should be able to do 4 chips as it have 2 chip select lines Feb 03 12:37:12 still doesn't look right Feb 03 12:38:26 SpeedEvil: I would be very much interested in such chip. 1GiB sounds very interesting\ Feb 03 12:38:41 and then I would hunt for 3730 Feb 03 12:39:34 3730 is order able from digikey IIRC Feb 03 12:40:24 actually 4-die RAM chip would increase speed, if OMAP would be interleaving memory throughout dies Feb 03 12:40:45 then there would be 16 banks - maximum perfomance possible Feb 03 12:42:05 that is reason why I want to modify ACPI tables for my desktop to change voltage and multiplier for CPU in C'n'Q Feb 03 12:50:32 so... when will i be able to buy a n900 with 512mb of ram? Feb 03 12:53:56 kerio: never Feb 03 12:54:04 ;_; Feb 03 12:54:19 dont say never .. Feb 03 12:54:19 feels bad man Feb 03 12:54:26 kerio: the GTA-04 basically is a N900 with 512 RAM :) Feb 03 12:54:35 if he gives a lot of money someone will do it Feb 03 12:54:53 but with smaller screen, without keyboard & very expensive :) Feb 03 12:54:58 here my 10k$ phone :) Feb 03 12:55:00 indeed Feb 03 12:55:34 well, N950 Feb 03 12:55:35 kerio: would you like fries with that? Feb 03 12:56:37 jacekowski: true N950 is very close to N900 hardware-wise, shame it doesn't have keyboard, neither sd-slot Feb 03 12:57:14 N950 does have a keyboard Feb 03 12:57:22 and microsd slot Feb 03 12:57:24 and everything Feb 03 12:57:40 sorry - I missed that one - true it does have keyboard Feb 03 12:58:06 for some very strange reason I was internally N950 and externally n9 Feb 03 12:58:38 jacekowski: only problem is - it cost a lot, if it appears on market Feb 03 12:59:01 that's other problem Feb 03 12:59:07 we don't care about such small details Feb 03 12:59:19 n950 have only 16GiB eMMC? Feb 03 12:59:30 no idea Feb 03 12:59:42 if notes are tearing your pockets ;) Feb 03 13:00:26 getting back on ACPI-C'n'Q problem Feb 03 13:01:10 1 Feb 03 13:01:17 2 Feb 03 13:28:31 Estel_: cfmradio has no icon :( Feb 03 13:29:38 which is weird, because there's a cfmradio.png in /usr/share/icons/hicolor/scalable/hildon/ Feb 03 13:33:56 kerio, because scalable isn't good place to put icon into Feb 03 13:34:07 move it to 64x64 or smth like that Feb 03 13:34:34 Pali afaik injection drivers doesnt work with ad-hoc mode Feb 03 13:35:02 I strongly suggest to not ship it instead of vanilla drivers, if it doesnt have that (or other possible regressions) fixed Feb 03 13:35:44 nusse, I never use that funny enabler to install cssu Feb 03 13:35:58 you can do it without it, just be sure that they will yell at you ;) Feb 03 13:36:20 just keep in mind that you need to ensure yourself, that everything is enabled as it should Feb 03 13:37:34 topro, I use fapman (FAM) and I recommend it by heart - this factoid ~fam is quite lame, was written by someone with personal problems about apt Feb 03 13:37:35 without the enabler, autoremove won't work properly Feb 03 13:37:47 fapman is just frontend for fam Feb 03 13:37:52 my autoremove works ok :P Feb 03 13:38:05 enabler is just funny sript Feb 03 13:38:09 and fapman fails to run certain preinst/postinst scripts Feb 03 13:38:18 you can mimic it's doings if want to Feb 03 13:38:26 yes you can Feb 03 13:38:29 kerio, as much fails as apt fails to do so Feb 03 13:38:43 (=never in practice, outside repeated urban legends) Feb 03 13:38:58 fapman = apt-get and apt-cache with some buttons Feb 03 13:39:08 fapman uses its own repo list, its own cache Feb 03 13:39:16 yea, that's even better Feb 03 13:39:24 and doesn't give you nearly enough control as you'd get by using apt Feb 03 13:39:26 as you can keep separate lists Feb 03 13:39:35 of course, not single frontend gives it Feb 03 13:39:43 still, it's faster for many things ;) Feb 03 13:40:09 if I would use apt for browsing programs in repos by keywords, it would take ages Feb 03 13:40:36 btw, I keep sync between apt and fapman repos lists, but there was a time when, for debugging purposes, I kept them separate. very handy Feb 03 13:41:11 anyway, topro, using fapman from very beginning of my N900 experience, it never, ever, caused any problem for me Feb 03 13:41:34 choice is your, but don't alwayw blindly follow factoids/opinions from irc here, as many times, they're biased as hell Feb 03 13:41:57 Maybe you never had issues with autoremove, 9/10 it works Feb 03 13:42:15 just few months ago some prominent (now) member of new Maemo administration, tried to convince everyone that kernel-power is evil and should never, ever, be installed by anyone Feb 03 13:42:23 Then suddenly, and god knows why autoremove fscks everything Feb 03 13:42:25 MrPingu, who forces you to autoremove? Feb 03 13:42:40 Estel_: Nobody, just saying Feb 03 13:42:42 fam have options to enable it, doesnt force it Feb 03 13:42:54 MrPingu, sure, but it's like apt-get autoremove Feb 03 13:42:59 it'*s just a frontend :P Feb 03 13:43:05 saying it's evil in itself is silly Feb 03 13:43:25 but that's expected, considering who've written factoid about fapman here :P Feb 03 13:43:40 btw autoremove isn't evil too, it just follows dependencies Feb 03 13:43:59 if dependencies are fckd, it may fckd things, but dependencies doesnmat get fckd by itself Feb 03 13:44:26 it's quite logic, despite what urban myths tend to assume Feb 03 13:44:29 i wrote that factoid Feb 03 13:44:35 s/logic/logical/ Feb 03 13:44:36 Estel_ meant: it's quite logical, despite what urban myths tend to assume Feb 03 13:44:39 seriously? Feb 03 13:44:53 so I was wrong at "considering who have written it" Feb 03 13:45:00 and one of the things that the cssu enabler does is vaguely try to fix the problem with the stock metapackage Feb 03 13:45:04 anyway, you were indoctrinated :P Feb 03 13:45:14 marking every one of its dependencies as manually installed Feb 03 13:45:24 I know :) Feb 03 13:45:33 which may create even more mess in future Feb 03 13:45:43 and, fun fact, i don't have ops here Feb 03 13:45:47 but cssu think that it will be eternall, most of the times Feb 03 13:45:49 so i have no obligation to not ignore you Feb 03 13:46:06 lol? if you feel offended, then take my deepest excuses Feb 03 13:46:18 although I don't necessary see point that offended you Feb 03 13:46:30 anyway, it's free irc-country, they tend to say Feb 03 13:46:48 s/irc-country/irc-node/ Feb 03 13:46:48 Estel_ meant: anyway, it's free irc-node, they tend to say Feb 03 13:46:58 It's that fapman has autoremove enabled by default... Not to mention I use fapman too, it has a nice gui for apt especially searching is superior in fapman Feb 03 13:47:15 MrPingu, exactly same reasons Feb 03 13:47:27 ops, missed autoremove by default, I disabled it on first use it seems Feb 03 13:47:45 well, sp factoid should state "always disable autoremove" instead "never use it" Feb 03 13:48:27 it's just that kerio spends too much time with DocScrutinizer05 here, and become authoritary biased against certain things, instead of following True Free Way Mantra :P Feb 03 13:49:36 is it possible to tether over usb? Feb 03 13:49:39 n900 Feb 03 13:49:43 sure Feb 03 13:49:48 qtmobilehotspot is the name Feb 03 13:49:51 ty Feb 03 13:50:04 Well, he is free to do so ;) Feb 03 13:50:15 damo22: mobilehotspot can do that, apparently Feb 03 13:50:21 there's no builtin UI to do so Feb 03 13:50:35 MrPingu, sure, have you not noticed little pun/irony in my writing about mantra? :P Feb 03 13:50:46 however, you can use the n900 as a modem Feb 03 13:50:46 of course there is ui Feb 03 13:50:55 qtmobilehotspot have ui Feb 03 13:51:32 you can tether N900 connections, be it gprs, wifi (acting like usb wifi card with routing functionality) or whatever Feb 03 13:51:39 it's really handy Feb 03 13:51:53 thats what i need Feb 03 13:51:59 then i can ditch my iphone Feb 03 13:51:59 You missed the word "builtin" while I missed your pun :P Feb 03 13:52:34 MrPingu, you're right, I took "builtin" as "builtin mobilehotspot" Feb 03 13:52:48 which is logic behind his sentence, but probably, not what he meant :P Feb 03 13:53:07 damo22, ditch it and buy replacement N900, just in case Feb 03 13:53:28 you can tether via wifi, via usb, via ethernet (using correct dongle) Feb 03 13:53:32 you mean buy 2x n900? Feb 03 13:53:37 sure Feb 03 13:53:45 maybe i will Feb 03 13:53:47 I got 8 of them :P Feb 03 13:54:00 bit overkill Feb 03 13:54:04 although 3 are rotary, sold or given to someone and replaced by others Feb 03 13:54:06 well Feb 03 13:54:23 u can give me one Feb 03 13:54:26 if you like Feb 03 13:54:28 :P Feb 03 13:54:32 if you become my girlfriend, why not Feb 03 13:54:41 er no Feb 03 13:54:47 thought so :) Feb 03 13:55:01 I wanna go far, but not that far :P Feb 03 13:55:18 so as a bonus to being your gf she gets a n900? Feb 03 13:55:38 Pali afaik injection drivers doesnt work with ad-hoc mode Feb 03 13:55:57 Estel_, why? Feb 03 13:56:02 I wouldn't suggest shipping it instead of vanilla driveres without regressions like that fixed Feb 03 13:56:09 well, enable injection, and try ad-hoc Feb 03 13:56:13 nice lovely reboot Feb 03 13:56:20 instant one, IIRC Feb 03 13:56:49 Estel_: surely you have more likeable qualities than the required offering of a n900 :P Feb 03 13:56:56 Estel_, I'm not going to replace original drivers Feb 03 13:57:23 Yes but wasn't that fixed in v0.2? Feb 03 13:58:07 well my n900 is perfect, except that the repo is slooooow Feb 03 13:58:15 :( Feb 03 13:58:33 Anyway, but I remember reports that ad-hoc makes it crash. Never confirmes as I don't use it Feb 03 13:58:43 damo22, N900 is like doctor's who sonic screwdriver Feb 03 13:58:45 ~mirrors Feb 03 13:58:46 i heard mirror is http://maemo-archive.wedrop.it/ http://maemo.merlin1991.at/apt-mirror/ http://talk.maemo.org/showthread.php?p=1315143#post1315143 Feb 03 13:58:57 very sexy thing Feb 03 13:59:19 don't you know who get screws, erm, screwdrivers from him? :P Feb 03 13:59:42 Pali, last time I checked, it rebooted on joining ad-hoc Feb 03 13:59:48 don't have one handy around now Feb 03 14:00:09 Estel_, now I see, it rebooted my device too Feb 03 14:00:10 damo22: I meant this: http://talk.maemo.org/showthread.php?p=1314588 Feb 03 14:00:13 but better check it before shipping vanilla as default Feb 03 14:00:16 Pali, yea Feb 03 14:00:19 topro: ( choice is your, but don't alwayw blindly follow factoids/opinions from irc here, as many times, they're biased as hell) a *very* wise advice, though in a recursive way OP probably hadn't intended to suggest Feb 03 14:00:20 I have oops backtrace in mtd2 Feb 03 14:00:24 screwing is fun, with a screwdriver of course Feb 03 14:01:11 lxp: ping Feb 03 14:01:12 Pali, nice, I would be glad if someone would be able to fix it Feb 03 14:01:23 Pali: pong Feb 03 14:01:35 We all would be glad :) Feb 03 14:01:45 P_p Feb 03 14:01:56 lxp, your packet injection driver cause kernel panic when trying to connet co adhoc network Feb 03 14:02:05 I have backtrace in mtd2ro Feb 03 14:02:10 can you look at it? Feb 03 14:02:33 ~hen Feb 03 14:02:34 [hen] hostmode-easy-now, or http://talk.maemo.org/showthread.php?t=65232 Feb 03 14:03:04 lxp, here is log: http://pastebin.com/8WiYUdGd Feb 03 14:03:25 sorry i have no time to look into it right now Feb 03 14:03:29 damo22, nowadsays, hen is most useful to debug hard to enumerate devices Feb 03 14:03:48 usbmode from pali is more handy as everyday tool to use hostmode Feb 03 14:03:58 btw Pali Feb 03 14:04:09 increasing sleep from 1 to 3 seconds after enumerate Feb 03 14:04:15 in usbmode Feb 03 14:04:18 helps a lot Feb 03 14:04:31 to get less false speed switched or unable to enumerate Feb 03 14:05:06 here is second part of oops: http://pastebin.com/w7SpSZRS Feb 03 14:05:26 also, many times, it doesnmt even *try* highspeed where it should, but it's probably hostmode legacy, hen seems to fail highspeed many times too, for reasons unknown Feb 03 14:05:30 other times it just works Feb 03 14:06:22 hoever, in case of usbmode, when it works it works, and when not, it straight ahead jumps into testing fullspeed, without even trying high speed Feb 03 14:06:25 nfc why Feb 03 14:08:28 Pali ^ Feb 03 14:08:42 Pali: i put the ad-hoc kernel panics on my todo for wl1251 but i can't tell when i have time to look into it Feb 03 14:08:58 lxp, ok Feb 03 14:09:13 lxp, I think that both logs can be usefull for you Feb 03 14:09:31 also I rebased your patches on top of 3.8 kernel Feb 03 14:09:32 * jonwil wonders if anyone will ever care about his wl1251-cal replacement work... Feb 03 14:10:21 jonwil, I looked at it yesterday Feb 03 14:10:27 jonwil: you rewrote wl1251-cal or did you replace the entire driver interface with something better? Feb 03 14:10:30 <_looping_> hi, is there a way to use youtube on N810? Feb 03 14:10:31 jonwil, I care :P Feb 03 14:10:37 I cloned wl1251-cal Feb 03 14:10:50 someone else can use it as a base to make the driver interface better Feb 03 14:10:56 lxp, if you have time, can you look at this patch? https://gitorious.org/linux-n900/linux-n900/commit/b44e1f6366c2d8ac45b39baabb5d9a2a5bfb3db1 Feb 03 14:11:14 this only caused problems when I rebased all your patch series Feb 03 14:11:22 jonwil: there is another approach in the upstream driver if i recall correctly it just pulls wl1251-nvs.bin from filesystem Feb 03 14:11:26 and I do not know if I fixed conflicts correctly Feb 03 14:11:50 <_looping_> please, is there a way to use youtube on N810? Feb 03 14:11:51 my thing does it properly including correct-ish region selection Feb 03 14:11:56 so it uses the correct channels Feb 03 14:12:27 lxp, in upstream wl1271 driver there is firmware code which load nvs data with mac address Feb 03 14:12:29 jonwil: >:c Feb 03 14:13:00 jonwil: use *all* the channels! _ò/ Feb 03 14:13:05 so I think that correct solution is to write similar code and create nvs file which will contains both wl1251-nvs and mac address Feb 03 14:13:41 Pali: do you mean the patch produces conflicts while rebasing or do you mean it doesn't work after you rebased it? Feb 03 14:14:16 lxp, rebasing produced conflicts Feb 03 14:14:27 I see no reason you couldn't either A.Write kernel level code to read from CAL and then copy what wl1251-cal does into kernel space or B.Write program that gets run once and spits out a disk file matching what would be read from CAL and sent to card which then gets read by the kernel driver on bootup Feb 03 14:14:37 jonwil, I second using all channels Feb 03 14:14:41 all was easy to fix (in upstream kernel they renamed files), but only that one was hard Feb 03 14:14:48 Either way, I have done my part by figuring out what wl1251-cal does :) Feb 03 14:14:53 jonwil, at least leave easily reachable option to unlock all channels Feb 03 14:15:02 the rest is up to people who know wl1251 driver more than I do :) Feb 03 14:15:12 jonwil: i think the best option would be to remove the wl1251-cal driver interface but we should keep wl1251-cal programm Feb 03 14:15:27 jonwil: wl1251-cal should be called by the firmware subsystem Feb 03 14:15:42 well I defer to the experts here :) Feb 03 14:15:59 jonwil: so when wl1251 driver requests the wl1251-nvs.bin firmware it is dynamically created by wl1251-cal Feb 03 14:16:26 jonwil, I think ideal solution would be: create script which read mac address from CAL and nvs file, and generate NVS firmware file (like in wl1271). then add wl1271 NVS parser code to wl1251 driver Feb 03 14:16:28 you would need to adapt my wl1251-cal clone to make that possible Feb 03 14:16:52 the upstream wl1251 maintainer told me that the driver cal interface is deprecated and will never be accepted in the upstream kernel Feb 03 14:17:07 and region change can be done via CRDA Feb 03 14:17:15 ok, well do whatever is needed :) Feb 03 14:17:24 the info required is out there Feb 03 14:17:47 Pali, also check cssu channel backscroll Feb 03 14:18:00 I reported crash that might be related to either hildon-home or ke-recv Feb 03 14:18:00 Estel_, I know, see: http://talk.maemo.org/showpost.php?p=1319983&postcount=166 Feb 03 14:18:06 if the latter, it apply to you Feb 03 14:18:13 checking it, thanks Feb 03 14:18:26 i have what's possibly a dumb question... Feb 03 14:18:35 Estel_, which sleep timeout in usbmode is needed to change? Feb 03 14:18:36 ah, isbmode Feb 03 14:18:39 look at TMO Feb 03 14:18:40 3th Feb 03 14:18:42 can't we make it so that it always uses the EU region? Feb 03 14:18:45 from 1 to 3 Feb 03 14:18:51 after usb_enum? Feb 03 14:19:01 it's the broadest one Feb 03 14:19:02 yes Feb 03 14:19:08 ok Feb 03 14:19:14 japan is broadest Feb 03 14:19:18 kerio, not eu Feb 03 14:19:37 Estel_: japan has power restrictions, and the hw can't do channels 12, 13 and 14 iirc Feb 03 14:19:44 it can Feb 03 14:19:46 I use 13 Feb 03 14:19:50 daily Feb 03 14:20:01 hm, i wonder who told me that Feb 03 14:20:02 japan has power restrictions? 0_0 Feb 03 14:20:19 it was the same person, mind you Feb 03 14:20:23 since when? everyone have it, but japan got it highest Feb 03 14:20:41 no, new zeland and japan is most liberate Feb 03 14:21:02 btw my router uses channel 13 and N900 works with it nicely Feb 03 14:21:21 at least no one i neighbourhood polluting it, except for interlapping from 11 Feb 03 14:21:50 and if it was the same person i think it was, who told you that, then, oh well. You're not ignoring me anymore? :) Feb 03 14:22:20 losing the track of the discussion is more annoying than you Feb 03 14:22:30 >:( Feb 03 14:22:36 well, so backscroll top my question what offended you ;) Feb 03 14:22:45 s/to0/to/ Feb 03 14:23:04 Este_, usbmode package updated Feb 03 14:23:16 Pali: increased the timeout a bit? Feb 03 14:23:34 after usb_enum from 1 to 3 Feb 03 14:23:48 Pali, thanks, will test it with slowest (as per enumerate time required) Feb 03 14:23:50 btw, as of now the only closed blobs required for MeeGo/Mer/Nemo on N900 are WiFi firmware, GPS, bluetooth firmware, PowerVR and maybe camera Feb 03 14:24:10 camera firmware is open Feb 03 14:24:17 is that all? Feb 03 14:24:28 but you do not want to see source code of camera firmware Feb 03 14:24:31 Camera blob is not open (omap3cam) Feb 03 14:24:48 jonwil, omap3cam not Feb 03 14:24:57 but can be replaced with new v4l2 library Feb 03 14:25:36 so yeah most important missing piece is powervr userspace blobs Feb 03 14:25:55 isn't fcam open too? Feb 03 14:26:04 hmmm yeah fcam is open Feb 03 14:26:10 jonwil, it will be never open Feb 03 14:26:10 what i love about this device is that i can still ssh to it while the screen is off Feb 03 14:26:15 and it will never be REd Feb 03 14:26:17 never say never Feb 03 14:26:23 3d Feb 03 14:26:30 Look at what the Noveau guys did for NVIDIA cards Feb 03 14:26:43 And look at what people of various sorts have been doing for other ARM GPUs Feb 03 14:27:15 and look at what happened for all TI/PVR people which wanted to write OSS driver Feb 03 14:27:51 jonwil: noveau "works" Feb 03 14:28:01 Noveau works just great on my Gentoo box Feb 03 14:28:16 Although to be fair said box only has a GeForce MX in it Feb 03 14:28:20 :P Feb 03 14:28:21 RE effort has been going nowhere since it started Feb 03 14:28:56 Skry, can you test if on 3.8 is working Ad-hoc wifi mode? Feb 03 14:28:57 jonwil: noveau is still far from matching nvidia drivers Feb 03 14:29:04 jonwil: feature and performance wise Feb 03 14:29:06 and nouveau is, well, nouveau, for example it's unusable with my 7600GT which can be classified as legacy card already Feb 03 14:29:28 Skry, packet injection driver patches on kernel-power causing kernel oops Feb 03 14:29:42 If I had more knowledge of linux graphics (and maybe a more recent/more usable copy of IDA) I would have a go at PowerVR bits myself (2D specifically) Feb 03 14:30:14 Pali: you should really make a local repository of your unreleased software Feb 03 14:30:21 as far as i know, no modern card has 2d acceleration Feb 03 14:30:30 kerio, of what? Feb 03 14:30:38 it's all done using 3d hardware Feb 03 14:30:40 usbmode, bme replacement Feb 03 14:30:44 battery applet Feb 03 14:30:49 kerio, look at gitorious Feb 03 14:30:58 i meant an apt repository Feb 03 14:31:08 N900 SGX can most definatly run and load GUI without a GL stack Feb 03 14:31:22 kerio, battery applet is in cssu devel Feb 03 14:31:28 is it? :o Feb 03 14:31:43 I will create some bme replacement repository later Feb 03 14:32:17 jonwil: it's all software hack apparently Feb 03 14:32:59 jonwil: 3d hardware is perfectly capable of doing all the stuff required for 2d with minimal software help Feb 03 14:33:09 jonwil: or firmware Feb 03 14:33:10 jacekowski: so, when you play a 2d game that simulates 3d... Feb 03 14:34:01 and well, modern desktop OSes and maemo does that as well Feb 03 14:34:11 "2d" desktop is using 3d compositing Feb 03 14:35:00 that's why real 2d hardware is not really needed anymore, because 3d hardware is perfectly capable of generating 2d graphics Feb 03 14:38:24 well I can say for sure that my N900 is not using the 3D hardware right now to render its desktop (otherwise it would have loaded libglesv2.so into memory somewhere and it has not) Feb 03 14:38:51 i though maemo uses compositing Feb 03 14:38:56 h-d uses compositing, yes Feb 03 14:38:57 as in hildon and stuff Feb 03 14:39:30 oh wait my fault Feb 03 14:39:33 its GLES not gles Feb 03 14:39:43 and yes hildon-desktop does have the library loaded Feb 03 14:40:03 jonwil: http://powervr.gnu.org.ve/doku.php, now, start working :) Feb 03 14:41:33 cant do anything N900 at all until my gentoo box finishes emerging pyqt and then pykde and then the other 100 packages it needs to emerge before I can actually get into X and use it properly again Feb 03 14:42:17 you don't need pyqt, pykde or kde for that matter to get into X Feb 03 14:42:25 so poor excuse :) Feb 03 14:43:55 its not that I need those, its that I cant boot x because I need and I cant emerge until after phqt and pykde finish emerging Feb 03 14:44:03 because thats the order emerge has chosen to emerge things Feb 03 14:44:14 i.e. I have to wait for this emerge pass to finish before I can boot x Feb 03 14:44:43 gentoo sucks Feb 03 14:44:52 i've wasted years of my life on it Feb 03 14:44:58 me too, can't say I miss it Feb 03 14:45:09 i've compiled oo.o on 850Mhz duron with 256MB of ram Feb 03 14:45:28 jacekowski: has it finished yet? :D Feb 03 14:45:33 :) Feb 03 14:45:54 if you compile oo.o on 850MHz, then you've got bigger problems than gentoo Feb 03 14:46:28 * jonwil is glad he doesn't have OO.o on his box at all Feb 03 14:46:42 open orifices are bad to have on your box Feb 03 14:46:57 that was time when a lot of hardcore gentoo users used oo.o-bin ebuilds Feb 03 14:53:18 Gentoo seems like a great way to stress test older hardware, if it can handle a full emerge of kde* it can handle anything you can throw at it :) Feb 03 14:54:17 why depreciated oo.o instead of libreoffice? Feb 03 15:06:51 * ShadowJK would assume libreoffice didn't exist yet Feb 03 15:07:49 this ooo/lo thing is damn annoying. Just as I had managed to make people who whine at me for Word use ooo writer, i have to explain the libreoffice is same thing Feb 03 15:08:07 "but it looks totally different! How do I print!? Save!? Open!?" Feb 03 15:09:59 press print/save/open Feb 03 15:10:12 tell them to use latex Feb 03 15:16:43 they changed the icons, so normal users used with ooo can't find the right buttons anymore Feb 03 15:17:15 Now they're back to relearning from scratch Feb 03 15:17:34 because normal users' brains work like that Feb 03 15:45:25 ShadowJK: just install an office 2000 theme Feb 03 16:01:46 DocScrutinizer51: distraction! Feb 03 16:05:00 kerio: what's up? Feb 03 16:06:53 meh, not much Feb 03 16:07:09 rmo seems to be even slower Feb 03 16:13:42 i like how they changed star office's name and hoped people would forget that star office sucks Feb 03 16:13:59 188.117.59.205 => 91.224.25.xxx 303kb 163kb 51.4kb Feb 03 16:14:00 <= 8.28kb 4.02kb 1.40kb Feb 03 16:14:02 188.117.59.205 => 88.65.81.xxx 5.83kb 158kb 60.5kb Feb 03 16:14:03 <= 0b 4.67kb 1.76kb Feb 03 16:14:05 188.117.59.205 => 109.230.157.xxx 29.1kb 148kb 59.4kb Feb 03 16:14:06 <= 256b 4.06kb 1.63kb Feb 03 16:14:39 the 3 cols are 4s/10s/40s iirc Feb 03 16:14:56 those are the fastest connections Feb 03 16:16:32 perhaps the amount of connections is increasing because the old ones are slow? Feb 03 16:19:13 DocScrutinizer05: unique IPs? Feb 03 16:19:50 sorry I'm busy with outher stuff Feb 03 16:20:18 root@stage:~# netstat -tn|wc -l Feb 03 16:20:20 4070 Feb 03 17:37:47 heh, impressive Feb 03 17:40:12 what? Feb 03 17:47:30 4000 Feb 03 17:49:07 * ShadowJK has 8ms ping to rmo Feb 03 17:51:31 * RST38h tried updating tonight, failed Feb 03 17:53:00 maybe I would be able to run 'apt-get update', then Feb 03 17:54:28 358ms min here, but it is UMTS at moment Feb 03 17:58:31 but at the moment I have 3.4ms ping to my server, which I consider rubbish, as desktop have rawly 0.1ms ping to server Feb 03 18:00:52 from my bouncer, i have a 2ms ping to holmes.freenode.net :D Feb 03 18:02:01 do you have professional router or something Feb 03 18:02:28 I naver have seen anything smaller the 38ms from std router Feb 03 18:03:04 but I have seen 2ms to google, when I hooked myself directly to internet w/o stupid rubbish called router Feb 03 18:03:05 i said my bouncer Feb 03 18:03:17 it's not on my home connection :) Feb 03 18:03:29 true Feb 03 18:03:29 it's on a friend's server, i assume in the same datacenter as the UK freenode server Feb 03 18:03:51 but meh, i've got 40ms to google Feb 03 18:04:06 2ms is still a lot Feb 03 18:04:57 man, connect some monster to virgin (modem, not this ad, which they give out now) and see how fast everything appears Feb 03 18:05:42 I have been connected directly for over month and I recognized true speed of virgin Feb 03 18:05:52 so... get a better router Feb 03 18:06:28 something to put openwrt/tomato on, or, if you're comfortable with closed-source software, something from mikrotik/routerboard (i heard they're really, really good) Feb 03 18:06:52 there's no good router, if you used AMD Athlon 64 X2 6400+ with 4GiB RAM as router previously Feb 03 18:07:07 yes there is Feb 03 18:07:26 these blanket statements are often inaccurate Feb 03 18:07:54 a 300MHz arm cpu is enough to push packets around Feb 03 18:08:09 did you ever have seen 2x1GHz with 4GiB DDR2-800 RAM (and I am not speaking about something, what would worth more then both my computers together) Feb 03 18:08:46 i've seen my laptop, i guess Feb 03 18:08:47 but routers have limited amount of RAM - max I ever have seen is 128MiB Feb 03 18:09:04 i think mine has 64 Feb 03 18:09:21 yep, with 46 free at the moment Feb 03 18:09:27 343 active connections Feb 03 18:09:37 now I have AMD Phenom II X4 965BE, 8GiB DDR2-1100 Feb 03 18:09:55 i'm very interested in your hardware configuration Feb 03 18:10:05 I was handling 2700 connections without even smallest slow-down Feb 03 18:10:18 duh Feb 03 18:10:31 ^that is server only^ Feb 03 18:10:46 i doubt i could sustain 3000 connections with my adsl :( Feb 03 18:10:54 that's why it is not 1090T Feb 03 18:11:17 it depends on many factors Feb 03 18:11:23 not only adls Feb 03 18:11:44 i'm not entirely certain my adsl modem is any good either Feb 03 18:11:59 for one thing, it heats up more than my router Feb 03 18:12:11 with black-box rubbish router it was max 64 connections and already dropping outeven my e-mail connections were overloading this rubbish Feb 03 18:12:45 and that is just 5 accounts, so I do noit understand Feb 03 18:16:58 I've done thousands of connections through a 486 with 32M ram :) Feb 03 18:17:17 magician! Feb 03 18:17:36 no, it's normal Feb 03 18:17:46 using uber powerhog for server is quite... Feb 03 18:17:54 well, waste of power, to say at least Feb 03 18:18:03 stock settings for Linux Kernel would be around 256 for that amount of mem Feb 03 18:18:14 wrt54gl with 32 ram and 200 mhz cpu easily handles building with 300 machines Feb 03 18:18:37 4ms to rmo is sheevaplug - switch - router - switch - isp Feb 03 18:18:37 32mb variant (wrt54g with correct revision) could handle up to almost anything Feb 03 18:18:39 yes, when they are idle Feb 03 18:18:46 nope Feb 03 18:18:53 machines from various private users Feb 03 18:19:09 really routing isn't that resource heavy, if you don't use crappy software Feb 03 18:19:19 kerio, btw, openwrt support FOSS firmware too Feb 03 18:19:22 look - you will run out of memory handling single connection per client Feb 03 18:19:26 one can use blobs or open rewrite Feb 03 18:19:31 WizardNumberNext, nope. Feb 03 18:19:59 man packets take memory and you have no alternative for memory on router like this Feb 03 18:20:18 and routing takes its memory as well Feb 03 18:20:28 are you trying to convince me that raw packets are swapped into memory? :) Feb 03 18:20:28 32M in my 486 could run atleast 1000 simultaneous connections through nat :) Feb 03 18:20:39 ShadowJK, exactly Feb 03 18:20:53 arp, mac, ip, routing - it adds up for memory starvation on 32MiB device Feb 03 18:21:05 WizardNumberNext, trust me (tm) memory isn't used for buffering packets in those devices :P Feb 03 18:21:23 you're kidding, it uses next to nothing Feb 03 18:21:31 ok, dropping packets, you still have arp, mac, ip, routing Feb 03 18:21:54 well, I'm pretty sure that cpu power would be bootleneck, not memory Feb 03 18:22:15 considering how many things of that type you can put there. And not every arp needs to sit in memory Feb 03 18:22:18 just counting raw values - 6+4*2, 14, 4+4 Feb 03 18:22:19 ya, my WRT54GL is limited by its cpu Feb 03 18:22:25 actually 4+4+4 Feb 03 18:22:36 anyway, practice seems to prove that handling 256 active connection on 16MB wrt54g*is no problem Feb 03 18:22:50 Estel_: my linksys isn't supported by openwrt properly :( Feb 03 18:22:54 every apr which is used have to sit in memory Feb 03 18:22:57 kerio, sad face :( Feb 03 18:23:07 and i'm worried about testing things because i don't have a serial port for it, so i could brick it Feb 03 18:23:15 anyway, counting like that, you will end with millions of clients Feb 03 18:23:21 cpu is bootleneck Feb 03 18:23:26 in extreme cases Feb 03 18:23:31 you don't need to know that many arp addresses.. arp of your upstream router, arp of lan clients Feb 03 18:23:33 in normal cases, nothing is bootleneck Feb 03 18:23:46 BTW, real bootleneck - G standard is limited to ~54Mbs :P Feb 03 18:23:56 ShadowJK, exactly Feb 03 18:24:07 Estel_; well 20-ish ppayload Feb 03 18:24:11 ShadowJK: I agree - only those, which you use and they stay there for some time Feb 03 18:24:18 don't remember default now Feb 03 18:24:33 I meant through ethernet the WRT54GL is also limited in bandwidth Feb 03 18:24:46 anyway, packets still go through memory and CPU as there is no other route for them Feb 03 18:24:47 WizardNumberNext, take no offense, you seems like nice guy, but constantly, you're trying to impress people here with some things that you have limited understanding of. Again, no offense Feb 03 18:25:10 using some powerhog for router purposes won't, definitely, impress people on maemo.org, wrong target group Feb 03 18:25:23 connection tracking can be done in 16 bytes per connection :) Feb 03 18:25:24 we're using 256MB RAM device to make wonders, so you know :P Feb 03 18:25:37 Though it'd be smarter to use a bit more ram Feb 03 18:25:39 Estel_: http://routerboard.com/RB751G-2HnD is really, really nice Feb 03 18:25:44 Estel_: I am too lazy right now to have a look again, Linux Kernel documentation states acurate overhead for routing Feb 03 18:25:44 ~32*1024 Feb 03 18:25:45 32768 Feb 03 18:26:01 ~32768/16 Feb 03 18:26:02 2048 Feb 03 18:26:04 sadly, it uses a closed OS :( Feb 03 18:26:15 mhm Feb 03 18:26:28 no idea, I stopped on wrtl54g*, as I'm allergic to N wifi standard Feb 03 18:26:36 aka Ultimate Neighbourhood Wifi Jammer Feb 03 18:26:40 wat Feb 03 18:26:45 nobody ever uses 5GHz Feb 03 18:26:47 Estel_; lol Feb 03 18:26:52 have look on Buffalo - there's few of them with 128 or 256MiB RAM and half of that flash Feb 03 18:27:17 bufallo are pretty decent, reportedly - to run debian on them :) Feb 03 18:27:23 Estel_; it'll only do 40MHz if it detects neighbouring channels are free Feb 03 18:27:30 otherwise, 256 ram there is funny Feb 03 18:27:33 ShadowJK: or if you force it to do so >:D Feb 03 18:27:39 ShadowJK, if it respects documentation :P Feb 03 18:27:51 Estel_: I am going to buy one, as there is room to run something nmormal (e.g. debian as you said) Feb 03 18:27:52 + what kerio said Feb 03 18:28:03 Estel_: just use the 5ghz band, nobody is ever there Feb 03 18:28:15 nothing supports 5GHz also :) Feb 03 18:28:20 btw running debian on router is kinda PoF - as there is no reason to not use openwrt, for example, using less resources Feb 03 18:28:33 ShadowJK, 5GHz is awesomely free... Feb 03 18:28:37 so empty there Feb 03 18:28:42 would love to use it Feb 03 18:28:42 I can make debian use even less Feb 03 18:29:01 especially, because I understand what I am doing in term of configuring kernel Feb 03 18:29:05 WizardNumberNext, no, you can't, as using same effort you can make openwrt use even less :P Feb 03 18:29:30 Estel_: you are quite right - it is better approach Feb 03 18:29:32 try that debian with 2.4 kernel, which is fully supported and enough for openwrt Feb 03 18:29:41 now compare resource usage of 2.4 kernel and recent ones Feb 03 18:29:45 and power savings Feb 03 18:30:05 btw i like debian on routers with 16 MB, it's fun Feb 03 18:30:14 but I haven't found any real use for it Feb 03 18:30:16 2.4 would be free here, but I'm using all of it Feb 03 18:30:19 so switched back to openwrt Feb 03 18:30:25 ShadowJK, lol Feb 03 18:30:30 + bluetooth Feb 03 18:30:55 fun fact - in last month I forget to take my device with me from workshop Feb 03 18:30:59 actually I never configured anything else then 2.6 Feb 03 18:31:04 and i had headplug and bh-214 on me Feb 03 18:31:08 mind 3.x is still 2.6 Feb 03 18:31:12 sound was perfect to ~90 meters Feb 03 18:31:24 via bluetooth 0_p without any mods Feb 03 18:31:32 * ShadowJK used to run 2.2 on that 486, but upgraded to 2.4 at some point Feb 03 18:31:52 * WizardNumberNext used to run 2.2 on i586 Feb 03 18:32:08 btw WizardNumberNext, you're n900 user, yes? Feb 03 18:32:15 actually almost any flawor of i586 Feb 03 18:32:17 Estel_; yeah it's not bad unobstructed Feb 03 18:32:30 Estelk_: yes Feb 03 18:32:42 ShadowJK, I remember some pays using cheap DIY antennas, getting 2.2 mile on bluetooth sniffing Feb 03 18:32:56 WizardNumberNext, are you using dual-scud or any bigger battery? Feb 03 18:33:08 wow, nice thing - I would put it on mine computer Feb 03 18:33:26 no, just stock old bl-5j (1320mAh) Feb 03 18:33:30 WizardNumberNext, better put it on N900, it have hirose u.fl internal connectors for wifi and bluetooth antenna Feb 03 18:33:50 my prototype custom cover is at stage of adding single connector extendedd to all N900 internal hirose's Feb 03 18:33:53 Estel_: maybe one day, whenb I would have another one to play with Feb 03 18:34:13 I would suggest buying 1500 mAh battery like polarcell or scud, then Feb 03 18:34:18 mixing old and new battery is bad idea Feb 03 18:34:20 for parallel Feb 03 18:34:49 never actually looked into details of dual-scud - how is made? Feb 03 18:35:08 I mean physically and/or connections Feb 03 18:35:12 kerio, do you have some factoid for battery thread? Feb 03 18:35:18 nope Feb 03 18:35:20 WizardNumberNext, sure, just a second Feb 03 18:35:27 but there's a fairly decent battery on aliexpress Feb 03 18:35:35 vi____1 reported a calibration of 1600mAh Feb 03 18:35:40 i bought the same model, got 1400 Feb 03 18:35:53 I am interested in averages Feb 03 18:36:14 ~dual-scud is http://talk.maemo.org/showthread.php?t=65568&highlight=3000+mAh+battery Feb 03 18:36:14 okay, Estel_ Feb 03 18:36:17 WizardNumberNext: connect two batteries in parallel Feb 03 18:36:20 ~dual-scud Feb 03 18:36:20 dual-scud is probably http://talk.maemo.org/showthread.php?t=65568&highlight=3000+mAh+battery Feb 03 18:36:29 Estel_: hrmpf, that's annoying Feb 03 18:36:36 ? Feb 03 18:36:40 because one got 1600, one 1400, so average is 1500 and you can get worse or better, so it is lottery Feb 03 18:36:41 infobot: no, dual-scud is http://talk.maemo.org/showthread.php?t=65568 Feb 03 18:36:41 kerio: okay Feb 03 18:36:51 oh, sorry Feb 03 18:36:57 haven't noticed that Feb 03 18:37:04 ~keriosnack Feb 03 18:37:14 ~snack kerio Feb 03 18:37:15 WizardNumberNext: i'd get a polarcell, they seem to all be the same Feb 03 18:37:15 * infobot throws kerio a doggy treat Feb 03 18:37:24 * Estel_ chuckled Feb 03 18:37:53 why dual-scud and not dual-polarcell? :P Feb 03 18:37:57 well, relative capacity of polarcell can be 1700 Feb 03 18:38:05 average is 1500, just like scud Feb 03 18:38:10 maybe they even use same cells Feb 03 18:38:18 I heard polarcell is fairly good Feb 03 18:38:20 kerio, historical reasons :P Feb 03 18:38:24 same scud Feb 03 18:38:37 i'm worried about the long-term for these batteries, though Feb 03 18:38:43 at least the polarcell has a brand name Feb 03 18:38:55 well, next step will be 7000mAh battery Feb 03 18:38:55 this one has some chinese writing on it Feb 03 18:39:01 in my custom cover :P Feb 03 18:39:04 including a "> 2000µAh" Feb 03 18:39:20 kerio, scud is quite long living Feb 03 18:39:44 mine got still ~2700 about all those years of usage Feb 03 18:39:50 well, 2600 currently Feb 03 18:39:52 to be fair Feb 03 18:39:59 during heavier usage Feb 03 18:40:29 freemangordon, out of curiosity - is there anything that can be done about glibc errors polluting syslog? Feb 03 18:40:37 it makes sniffing for errors a little harder Feb 03 18:40:44 iirc you can configure syslog to only report ERROR or above Feb 03 18:40:52 those glib things are INFO Feb 03 18:41:02 "> 2000µAh"? haha, I am willing to believe them Feb 03 18:41:07 well, true, but won't I miss smth important? Feb 03 18:41:10 DocScrutinizer05: and it was! :D Feb 03 18:41:21 I wonder what they had in mind Feb 03 18:41:25 it was around 1400000µAh Feb 03 18:41:45 well, I suppose it was about something inside protective PCB Feb 03 18:41:50 not a cell itself :P Feb 03 18:42:03 kerio, how you've emasured it to be 1400? Feb 03 18:42:12 constant discharge rate? Feb 03 18:42:46 constant at ~200mA Feb 03 18:42:56 idling it would be a lot more i suppose Feb 03 18:43:04 so rest assured, you've got as good battery as vi's Feb 03 18:43:13 he measured it on device via average discharge Feb 03 18:43:15 during normal usage Feb 03 18:43:22 his relative capacity may be as high Feb 03 18:43:30 but at you conditions he wuld get the same Feb 03 18:43:48 I got rid of skype, we'll see how much I would get of bl-5j Feb 03 18:43:56 my 3100 mAh (when new) battery was made using 2 cells reporting ~1430 during constant discharge Feb 03 18:44:09 google for "silver needle in the skype" pdf Feb 03 18:44:16 and you won't even return to skype Feb 03 18:44:18 strange - they suddenly don't like charging outside of fridge Feb 03 18:44:25 getting smaller and smaller Feb 03 18:44:30 especially, that their fiasco during vita release just reinforce silver needle theory Feb 03 18:44:36 SpeedEvil: i forgot again, how am i supposed to figure out if the battery is healthy? i've got VOLT and CUR on both idle and load Feb 03 18:44:48 well, once I explained you why you're doing it wrong with that fridge Feb 03 18:45:03 why it's hoax placebo, and harmfull one (to cells) too Feb 03 18:45:08 have you missed it? :P Feb 03 18:45:27 kerio: in short - apply a rapid load, and measure the voltage drop Feb 03 18:45:35 much drop means high ESR Feb 03 18:45:41 Estel_: as long as it is above 0C it should be fine Feb 03 18:45:46 kerio, we need to convince Pali about adding edv1 and vdq to things exported by bq27x00_battery Feb 03 18:45:54 actually I was getting more and more charge this way Feb 03 18:45:56 Estel_: well, they're exported Feb 03 18:45:59 but badly Feb 03 18:46:04 kerio, where? Feb 03 18:46:17 WizardNumberNext, you was getting more *relative* charge Feb 03 18:46:23 aka wasted power during charging Feb 03 18:46:26 to previous charge\ Feb 03 18:46:33 Silver Needle, wasn't that the BH paper by two french dudes? Feb 03 18:46:34 which never ever turn into relative discharge capacity oincrease Feb 03 18:46:43 I think I met one of 'em while I was consulting for EADS. Feb 03 18:46:51 Estel_: $(( ( $( grep 0x0a= registers | cut -d= -f2 ) & 4 ) / 4 )) Feb 03 18:46:53 teotwaki, no idea who did it, but they were right Feb 03 18:46:55 that's VDQ Feb 03 18:47:04 some time later skype segfaulted on vista dropping some debug data Feb 03 18:47:14 EDV1 is $(( ( $( grep 0x0a= registers | cut -d= -f2 ) & 2 ) / 2 )) Feb 03 18:47:17 that allowed to get info about skype logging motherboard serial Feb 03 18:47:22 it's not a valid discharge unless it'w over 7c Feb 03 18:47:24 it's Feb 03 18:47:33 SpeedEvil: yeah but which values are good? Feb 03 18:47:34 they explained, that it was to "easily catch people stealing on their paid services" Feb 03 18:47:35 over 0.7C? Feb 03 18:47:37 i don't know if the voltage drop is a lot Feb 03 18:47:42 no, 7 Feb 03 18:47:59 kerio, does it work when bq is loaded? doesnt it need i2c? Feb 03 18:48:04 how 7C? this would be 1320*7 Feb 03 18:48:05 why isn't it properly exported? Feb 03 18:48:11 ~1320*7 Feb 03 18:48:12 9240 Feb 03 18:48:15 to /sys/class/power_supply/bq27x00 Feb 03 18:48:19 7 celcius? Feb 03 18:48:21 Estel_: /sys/class/power_supply/bq27200-0/registers :) Feb 03 18:48:23 Estel_: I don't much care for tracking, there's a bazillion reasons for that. I do, however, care about adding anti-debugging features to your binaries :) Feb 03 18:48:24 kerio: search ohm.sh on the wiki Feb 03 18:48:29 ShadowJK: yes Feb 03 18:48:30 oh, sorry misunderstood Feb 03 18:48:35 teotwaki, exactly Feb 03 18:48:53 I just mean that they indeed have thing to avoid determining what skype is really doing Feb 03 18:48:55 most of my discharges are in room temperature Feb 03 18:48:55 on ones pc Feb 03 18:49:13 ShadowJK: the bq* chip will not learn a new capacity if the temp is under 7, when evdf is hit Feb 03 18:49:15 SpeedEvil: can't find shit Feb 03 18:49:21 also, *edv1 Feb 03 18:49:24 WizardNumberNext: charging of LiIon at low temperatures definitely NOT recommended Feb 03 18:49:26 kerio: sorry, I'm going to sleep Feb 03 18:49:30 well, even today, some fool (foolisness? as it was female) tried to force me into installing skype Feb 03 18:49:33 which is going to fall, as I lowered voltage on my CPUs in desktop and server, so temperature in room would fgall a bit Feb 03 18:49:34 evdf isn't used for learning at all, edv1 is Feb 03 18:50:22 teotwaki__: you noticed my request to temporarily stop syncing to repo.m.o? Feb 03 18:50:31 DocScrutinizer05: I think I am going to finish this experiment anyway, as they charge shorter (maybe it better efficiency), so might be getting less charge Feb 03 18:50:33 DocScrutinizer05: not at all Feb 03 18:50:40 42V drop with a 137mA drop Feb 03 18:50:53 DocScrutinizer05; a local-ish EV guy claims it's fine with the right cells, they just go high impedance briefly while warming up and then temp and impedance stabilizes to something sensible.. Feb 03 18:50:59 with battery at 66% Feb 03 18:51:01 DocScrutinizer05: I'll shut off the sync of everything. Feb 03 18:51:06 kerio, what you propose as sh line of script to get vdq value and edv1 value? Feb 03 18:51:19 Estel_: as in the flags? Feb 03 18:51:24 i told you Feb 03 18:51:25 I would update bnf to be able of cooperating with palis-yours battery applet Feb 03 18:51:36 $edv= Feb 03 18:51:42 no, edv= Feb 03 18:51:49 and then call $edv Feb 03 18:51:50 DocScrutinizer05: done Feb 03 18:51:52 kerio; fair Feb 03 18:51:53 for example Feb 03 18:52:18 I would like to add some if's to bnf, so it would use bq27x00 and i2c when bq27x00 is absent Feb 03 18:52:23 ShadowJK: well, the datasheets I've seen so far are like: "discharge allowed temp: -20°C .. +60°C, charge 0°C .. 50°C" Feb 03 18:52:31 teotwaki__: thanks Feb 03 18:52:33 but bq27x00 doesnt allow me to get edv1 and vdq holy way Feb 03 18:52:42 Estel_: http://pastebin.mozilla.org/2111699 Feb 03 18:52:49 rmo needs to have a rsync service for mirrors, and then we need to add mirrors to repository.maemo.org A records for mirror-owner selectavle amount of time :D Feb 03 18:52:57 ShadowJK: we just need a CDN Feb 03 18:53:10 kerio, same for "calibration needed" flag Feb 03 18:53:11 so my charges are out of it, as I have less then 0C in frigdge Feb 03 18:53:17 kerio; meh Feb 03 18:53:26 ShadowJK: cloudflare is fscking **free** Feb 03 18:53:29 kerio, thanks Feb 03 18:53:31 ShadowJK: that's a problem since we don't own maemo.org DNS yet Feb 03 18:53:37 are you ware of calibration needed flag? Feb 03 18:53:42 if you have less than 0C in fridge your milk will freeze Feb 03 18:53:45 Estel_: notice that those two functions return the opposite numerical value Feb 03 18:53:46 ShadowJK: the way we do it on LMDE, is simply to have a package that lists all available mirrors, with a script which attempts to estimate the best mirror for one person, and simply update the source list file based on the user's choice. Feb 03 18:53:53 why bq27x00 can't export all values re charging chip obtainable via i2c? Feb 03 18:53:57 I do use milk Feb 03 18:53:58 as it blocks i2c Feb 03 18:54:03 teotwaki__: why not do something like http.debian.net? Feb 03 18:54:04 it should provide replacement Feb 03 18:54:10 Pali, ^ Feb 03 18:54:14 Estel_: they're exported, ffs Feb 03 18:54:16 kerio: yes, but you have to use them as a nameserver. Feb 03 18:54:21 kerio: infrastructure, mainly Feb 03 18:54:22 just not in a practical way Feb 03 18:54:31 kerio, I mean practical way Feb 03 18:54:33 warfare: checked that CNAME thing? Feb 03 18:54:35 why not? Feb 03 18:54:38 Estel_: i2cget isn't practical at all Feb 03 18:54:47 and checking a numerical value in a registers file is about the same Feb 03 18:54:49 no, no, you got me wrong Feb 03 18:54:54 sure it is Feb 03 18:54:58 it would be nice, yes Feb 03 18:55:03 kerio: LMDE is a very small distro, I haven't looked at http.d.n, but when we set up the system, this seemed easiest without requiring a central server to be up 24/7. Feb 03 18:55:05 but if bq27x00 exports many values practical way Feb 03 18:55:11 kerio: yes, but thats not in the free plan. Feb 03 18:55:11 why not those useful ones too? Feb 03 18:55:39 warfare: well i, for one, i welcome our new dns overlords Feb 03 18:55:50 s/i welc/welc/ Feb 03 18:55:51 kerio meant: warfare: well i, for one, welcome our new dns overlords Feb 03 18:55:53 kerio, how would you export calibration needed flag to? Feb 03 18:56:03 kerio: The domain is still with nokia. Feb 03 18:56:05 Estel_: by doing the same numerical manipulations Feb 03 18:56:17 go figure out what's the appropriate register and the appropriate bit for it Feb 03 18:56:21 I don't get syntax, how do you know that, where it is documented Feb 03 18:56:36 rebooting, brb Feb 03 18:56:38 i dunno, i just checked bq27200.sh Feb 03 18:56:39 Estel_, you can access i2c bus when bq27x00_battery is loaded Feb 03 18:56:43 use I2C_SLAVE_FORCE Feb 03 18:56:52 Pali: oh hell naw Feb 03 18:57:01 (charging at low temp) just saying. Generally if something is specified as "allowed range: A .. B" then your sweet spot is neither A nor B but (A+B)/2 Feb 03 18:57:28 that's a general pattern Feb 03 18:57:34 DocScrutinizer05: usually because at A-little and at B+little, you get actual damage :) Feb 03 18:58:20 Estel, kerio, see: https://lkml.org/lkml/2013/1/19/162 and https://lkml.org/lkml/2013/1/20/20 Feb 03 18:58:23 Pali: it would be really useful for scripts and such to get the flags in a more simple way - i'm not sure if there's something similar in the kernel api Feb 03 18:59:07 kerio, debugfs regmap interface Feb 03 18:59:23 but I'm not going to touch bq27x00_battery kernel driver Feb 03 18:59:34 Pali: not even for "cosmetic" changes? :( Feb 03 18:59:40 if you need something, write patch and upstream it Feb 03 18:59:50 kerio, I mean nothing new Feb 03 18:59:58 maaaaaaaaan, having to parse "registers" sucks :( Feb 03 19:00:13 there's another pattern: "allowed range: < B", you pick a spot as much below and away from B as possible. See spec for R_short of USB ID: <200R. You gonna use short, not a resistor of any size Feb 03 19:00:15 If there will be some bugs, I can look at it Feb 03 19:00:22 but bq27x00_battery driver working fine Feb 03 19:00:25 (now) Feb 03 19:00:36 DocScrutinizer05: "You can still access it if you use the I2C_SLAVE_FORCE flag." is bad, bad, bad advice, right? Feb 03 19:00:56 huh? Feb 03 19:01:06 https://lkml.org/lkml/2013/1/20/20 Feb 03 19:01:27 isn't i2c the thing where if you force things, you get magic smoke? Feb 03 19:01:33 *might gt Feb 03 19:01:36 **get Feb 03 19:01:50 kerio, you are reading properties Feb 03 19:02:01 not writing to registers Feb 03 19:02:34 kerio, if you want to export that properties (CI, ...), write regmap interface to kernel driver Feb 03 19:02:47 kerio: yes Feb 03 19:03:07 kerio, if your patch will be accepted to upstream, I can include it into kernel-power :-) Feb 03 19:03:14 kerio: nfc what I2C_SLAVE_FORCE does, but you mustn't do concurrent access to I2C bus Feb 03 19:04:21 #define I2C_SLAVE_FORCE 0x0706 /* Use this slave address, even if it is already in use by a driver! */ Feb 03 19:05:09 still doesn't say if they do any locking to ensure no two processes concurrently mess with I2C bus Feb 03 19:05:20 think of I2C like a printer Feb 03 19:05:32 you don't want two apps printing same time Feb 03 19:05:54 DocScrutinizer05: /nick badanalogyguy Feb 03 19:06:14 you need to finish one print job before allowing any other job to print Feb 03 19:06:25 you mean like a semaphore? Feb 03 19:06:42 mutex is enought Feb 03 19:07:41 Feb 03 19:07:43 Feb 03 19:07:43 Last time I taught C/C++, I had a student ask "Excuse me, but what's a whole phore?" "I said semaphore, not semi-phore." Feb 03 19:07:44 Feb 03 19:07:46 Feb 03 19:07:47 Feb 03 19:07:49 Feb 03 19:07:50 Feb 03 19:07:52 Feb 03 19:07:53 Feb 03 19:08:11 that's a bit harsh doc Feb 03 19:08:29 * RST38h yawns Feb 03 19:08:34 /cs quiet #maemo user Feb 03 19:08:56 weird mispaste Feb 03 19:08:58 only newlines Feb 03 19:09:20 I think I've put plate on spce or something similar Feb 03 19:09:34 sorry, got phone Feb 03 19:09:50 what happened with backscroll? Feb 03 19:09:57 never seen error like that Feb 03 19:09:57 I would think kernel would at least prevent same bus to be used twice in same time Feb 03 19:10:13 WizardNumberNext: at least apologize. Feb 03 19:10:16 Me neither, but as you can see plates can do miracles Feb 03 19:10:24 I did Feb 03 19:10:38 directly below, when I realized what happened Feb 03 19:10:42 nha Feb 03 19:10:46 your client buffered the output Feb 03 19:10:51 you were flooding one line every second Feb 03 19:10:55 it never ran to the end Feb 03 19:11:42 anyway: I apologize for my weird behaviour - have no idea what exactly happened Feb 03 19:11:59 no problem, we're always happy to kick someone. Feb 03 19:12:16 ouc Feb 03 19:12:18 =D Feb 03 19:12:28 that hurts ;) Feb 03 19:12:42 at least this time, we can use the excuse "For the greater good" Feb 03 19:12:50 * teotwaki__ winks at Estel_ Feb 03 19:12:54 you actually can Feb 03 19:13:08 but anyway I got ban striaght away Feb 03 19:13:12 Pali, kerio, DocScrutinizer05, summing it up, it is safe for me to access i2c for READ using slave_i2c, or is it as bad as infamous kernel patch? Feb 03 19:13:21 ChanServ was fast enough to react Feb 03 19:13:24 I banned ya (only a speaking ban) Feb 03 19:13:30 teotwaki, cheers ;) Feb 03 19:13:56 I got speaking ban and few seconds after completely disconnected Feb 03 19:14:25 this ban was usefull anyway - who know what would happen, if this would continue Feb 03 19:14:27 because two people at once decided to take care of you Feb 03 19:14:27 that was doc kicking you Feb 03 19:15:15 whatever - good he did that, I don't want to know how many empty lines it would send Feb 03 19:15:16 the kick was probably more useful, it cleared your client's sending cache instantly. The mute would have had you send stuff for quite a bit of time, depending on the paste Feb 03 19:15:21 Pali, kerio, DocScrutinizer05, as said, I would like to be sure that I'm not incorporating anything stupid into bnf. Many people use it, and blue smoke isn't fun, even if it's theoretical risk Feb 03 19:16:04 so what about this slave i2c, is it safe for read, or very, very bad as said infamous patch eons ago? Feb 03 19:16:25 if we are speaking about cigarettes - it is ok for me, if it is about n900 - it is definitely NOT Feb 03 19:16:44 kerio, write a patch and upstream it - you will get timeless glory as kernel hacker :P Feb 03 19:16:47 Estel_: I dunno what's the deal with slave_i2c, but there's no such thing like READ on I2C Feb 03 19:16:49 Estel_, I do not know, better ask on LKML Feb 03 19:17:05 Estel_: if it is true you are able to use same bus twice in same time, then nothing is safe Feb 03 19:17:09 cheeesus, I think I'll stay with registrars Feb 03 19:17:34 ask on LKML and/or i2c kernel maintainer what current linux code doing Feb 03 19:17:38 if there would be write going on, and you would try to read, you can actually write your command, instead of getting anything from device Feb 03 19:17:58 kerio, wtf si that pastebin you've send me Feb 03 19:18:08 my calibration script Feb 03 19:18:10 Pali, I'll Feb 03 19:18:13 kerio, nice Feb 03 19:18:19 ah, I see Feb 03 19:18:27 notice especially the vdq and edv1 functions Feb 03 19:18:32 OK, so now I only need code for "calibration needed" flag Feb 03 19:18:35 yea Feb 03 19:18:36 (and notice that their output is reversed) Feb 03 19:18:41 I think that nobody here not know how parallel i2c reading via kernel API working Feb 03 19:19:09 so Feb 03 19:19:11 vdq=return $(( 1 - ( ( $( grep 0x0a= registers | cut -d= -f2 ) & 4 ) / 4 ) )) Feb 03 19:19:19 will give me vdq as 0 or 1? Feb 03 19:19:27 when invoking $vdq Feb 03 19:19:32 Pali: I never have read that code (if I would understand it first), so I do not Feb 03 19:19:33 or is it more complicated? Feb 03 19:19:57 kerio, why it's reversed, btw? Feb 03 19:20:14 Estel_, kernel i2c mailinglist: linux-i2c@vger.kernel.org Feb 03 19:20:23 ~snack Pali Feb 03 19:20:24 * infobot throws Pali a doggy treat Feb 03 19:20:39 ~thanks Pali Feb 03 19:20:39 de rien, Estel_ Feb 03 19:20:55 ~french Feb 03 19:20:56 Vous pouvez obtenir de l'aide sur Debian sur le canal #debian-fr - For help in french, please join #debian-fr Feb 03 19:20:57 (from $ ./scripts/get_maintainer.pl -f drivers/i2c ) Feb 03 19:21:19 ~slovak Feb 03 19:21:34 ok lets try: Feb 03 19:21:38 ~polish Feb 03 19:21:38 haha :D Feb 03 19:21:39 somebody said polish was something used to make things shiny, or some language with MUCH too high a consonant-to-vowel ratio. /j #debian.pl, or a really good sausage Feb 03 19:22:05 last one is true! Feb 03 19:22:22 english one is uneatable Feb 03 19:22:27 ~english Feb 03 19:22:28 i heard english is "In the words of the great Samuel L. Jackson; English, motherfuc*er. DO YOU SPEAK IT?!" - It is also more geographically spread than any other language Feb 03 19:22:35 read: poisonous Feb 03 19:22:47 ~german Feb 03 19:22:48 german is, like, #debian.de - deutschsprachige Debian-Hilfe gibt es in #debian.de Feb 03 19:22:57 ~hail Feb 03 19:22:58 i heard hail is chimp Feb 03 19:23:36 ~teotwaki Feb 03 19:23:42 :( Feb 03 19:24:03 ~Estel_ Feb 03 19:24:03 it has been said that estel_ is Estel Feb 03 19:24:11 wut? Feb 03 19:24:17 ~Estel Feb 03 19:24:31 I do not know, but it worked opposed to ~teotwaki Feb 03 19:25:14 ~kindergarden Feb 03 19:26:38 kerio Feb 03 19:26:41 your code returns Feb 03 19:26:56 grep: registers: No such file or directory Feb 03 19:27:08 -sh: arithmetic syntax error Feb 03 19:27:14 it's due to: Feb 03 19:27:31 ( $( grep 0x0a= registers | cut -d= -f2 ) Feb 03 19:27:33 probably Feb 03 19:27:53 are you using kp52? Feb 03 19:28:05 sure Feb 03 19:28:22 well, is there a "registers" file? Feb 03 19:28:32 ash/bash compliance? Feb 03 19:28:35 yes Feb 03 19:28:47 grepping it causes xterm to hangs, though Feb 03 19:29:10 emr Feb 03 19:29:19 whele this file is supposed to be? Feb 03 19:29:41 my script cds to /sys/class/power_supply/bq27200-0/ Feb 03 19:30:17 silly me Feb 03 19:30:42 doesnt have it loaded Feb 03 19:30:44 :P Feb 03 19:32:11 -sh: 0: not found Feb 03 19:32:14 wut the?... Feb 03 20:01:05 ~kill kerio Feb 03 20:01:06 * infobot shoots a inverse graviton gun at kerio Feb 03 20:01:15 blast you, it was all about your inversion of value Feb 03 20:01:30 I spend last 30 minutes debugging why it doesn't want to work Feb 03 20:01:37 and it was one ) too much :P Feb 03 20:01:44 finally, vdq is: Feb 03 20:01:59 $(( ( $( grep 0x0a= registers | cut -d= -f2 ) & 4 ) / 4 )) Feb 03 20:02:09 minus $ Feb 03 20:06:39 Estel_: it's clearly stated Feb 03 20:06:49 vdq() returns success if vdq is set, failure otherwise Feb 03 20:06:55 success is zero, failure is nonzero, in shellscripting Feb 03 20:07:59 kerio, I have better solution for checking version of bme/hald-addon-bme Feb 03 20:08:15 adding new property to hald-addon-bme which tell us that we using new replacement Feb 03 20:08:26 fucking... yes Feb 03 20:08:45 although, don't call it "bme replacement" Feb 03 20:08:47 so will know at runtime if we are using replacemnt version of hald-addon-bme or not Feb 03 20:08:58 kerio, version? Feb 03 20:09:04 well, not really Feb 03 20:09:07 what does it signify? Feb 03 20:09:13 give me some key name and value Feb 03 20:09:28 that even without the binary running, it still updates its values Feb 03 20:09:50 meh, it's just a cosmetic thing anyway Feb 03 20:12:05 kerio, what about key "maemo.bme.version" with value "1.0" ? Feb 03 20:12:25 in hal device bme? Feb 03 20:12:52 meh, whatever you want Feb 03 20:12:59 perhaps bme-replacement is the most accurate, actually Feb 03 20:13:27 we can increase that version if we add something in future Feb 03 20:13:50 kerio, run: $ hal-device bme Feb 03 20:14:05 to see all key/values in hald-addon-bme Feb 03 20:15:18 maemo.version is probably right, i suppose Feb 03 20:15:20 or maemo.bme.version Feb 03 20:16:39 whats the commandline for the camera Feb 03 20:18:22 Pali, does time to empty and time to end is calculated differently in bq27x00 than chip reports (via i2c, for example)? Feb 03 20:18:47 there is average and "now", I wonder which one is more in line with old bq27200.sh script. ShadowJK? Feb 03 20:19:07 Estel_, bq27x00_battery kernel driver only export what chip reports Feb 03 20:19:14 I see Feb 03 20:19:30 Estel_; dunno what you mean? Feb 03 20:19:33 Estel_, look at source code Feb 03 20:19:47 Rate and TTF/TTE are as reported by chip Feb 03 20:20:27 Estel_: https://garage.maemo.org/plugins/ggit/browse.php/?p=kernel-power;a=tree;f=kernel-power-2.6.28/debian/patches;hb=HEAD Feb 03 20:20:32 to bring up the camera Feb 03 20:20:35 ShadowJK, bq27200.sh reports time to empty and time to full as you've said, but pali also extracted "time to empty average" and "time to empty now" Feb 03 20:20:45 Pali, thanks Feb 03 20:20:59 Estel_, bq27200.sh reporting more times Feb 03 20:21:04 ShadowJK, I wonder which one bq27200.sh uses, average or now? Feb 03 20:21:11 well the chip doesn't have a time to empty average. Feb 03 20:21:11 right Feb 03 20:21:26 #define BQ27x00_REG_TTE 0x16 Feb 03 20:21:26 so how bq27x00_battery calculates average? Feb 03 20:21:31 #define BQ27x00_REG_TTF 0x18 Feb 03 20:21:36 #define BQ27x00_REG_TTECP 0x26 Feb 03 20:21:42 ough Feb 03 20:21:56 these 3 ^^ values are used in driver Feb 03 20:22:01 * Estel_ nods Feb 03 20:22:37 register BQ27x00_REG_TTECP is used for time_to_empty_avg sysnode Feb 03 20:22:59 I see Feb 03 20:23:12 well, the one I've used is 0x16 Feb 03 20:23:44 no reason why not update bnf to report both of them Feb 03 20:27:11 oh, so like same as TTE more or less, but adjusted for constant power rather than constant current discharge? Feb 03 20:28:31 So it should be a 5.12 sec average too Feb 03 20:29:22 It's a miracle, but RMO seems to be up. :-D Feb 03 20:29:33 been up for a while Feb 03 20:29:38 been slow for a while Feb 03 20:29:50 ye, been up for many hours. Feb 03 20:31:36 what the command to bring up the camera in commandline Feb 03 20:34:10 i am out thanks guys catch you later will get to it later, kerio got ssh working thanks later Feb 03 20:44:14 Pali: what about using bq27k data if it's available, over everything else? Feb 03 20:44:28 where? Feb 03 20:44:35 battery applet Feb 03 20:45:29 kerio, if you have idea for "design cpacity", create wiki page and write it there Feb 03 20:45:35 i already told you my idea Feb 03 20:45:37 I already told you this Feb 03 20:45:51 calibrated bq27200 trumps **everything else** Feb 03 20:45:53 irc is not good for this discussion Feb 03 20:46:32 Pali: what's the difference between battery.reporting.last_full and battery.charge_level.last_full? Feb 03 20:47:45 Pali, afaik most kernel changes were done aftr irc discussions Feb 03 20:47:54 in hald addon bme: battery.charge_level.last_full = 8*battery_info->power_supply_charge_full/battery_info->power_supply_charge_design Feb 03 20:47:56 why userland battery applet must be different? Feb 03 20:48:28 it's obvious that some random value reported via resistor in battery's pcb is worse than anything from bq27x00, even when uncalibrated Feb 03 20:48:42 you could as good hardcode anything instead of using rx51_battery data Feb 03 20:48:47 Estel_, need to write all positive & negative aspects for solutions Feb 03 20:48:52 it would be as close to reality as that resistor Feb 03 20:48:56 same as wiki page for kernel in cssu Feb 03 20:49:16 doesn't we know al pros and cons already? I'm all for discutting things, but there is no other solution Feb 03 20:49:23 no different approach to consider Feb 03 20:49:39 sure, thats why we dont have kernel in cssu and wont have in years :P Feb 03 20:49:46 Pali: anyway, i doubt we'll ever going to get anywhere with this - is there some facility to set configurations in the battery applet? what should we use? making it configurable (with some hidden gconf keys maybe?) would be better Feb 03 20:49:46 joking Feb 03 20:50:07 pali, all resistors in batteries are for around 1300mAh Feb 03 20:50:18 if bq27x00 isn't calibrated, just display "unknown" there Feb 03 20:50:28 or "need calibration" Feb 03 20:50:30 whatever Feb 03 20:50:40 Estel_, I do not remember what everybody said on IRC, and I really do not want to search IRC logs for that Feb 03 20:50:44 and show percents as bme is doing it, via voltage Feb 03 20:50:46 so reason why to create wiki page Feb 03 20:50:49 I see Feb 03 20:50:52 have all info at one place Feb 03 20:50:54 well, it's good reason Feb 03 20:50:56 Estel_: bme voltage is fucked up, though Feb 03 20:51:16 I'll try to find some spare time for that wiki - actually did it already but wiki worked at 17b/s at the time Feb 03 20:51:26 kerio, better than design capacity :P Feb 03 20:51:35 now we have working maemo wiki, so we can use it Feb 03 20:51:37 remember, it's failsafe to bq not calibrated Feb 03 20:51:38 yea Feb 03 20:51:38 is it? it assumes a specific voltage profile Feb 03 20:51:43 doesn't it Feb 03 20:51:43 well, I agree Feb 03 20:52:15 kerio, in worst case user have what he have with stock bme, in any other cvase (bq calibrated) user have precise value Feb 03 20:52:41 can't think of better solution Feb 03 20:53:20 so what we need is: which icon show in battery applet (we have 8 icons), next how to calculate percentage (if will be shown in applet), next what to show for design capacity (if something) Feb 03 20:53:25 provide solution on wiki Feb 03 20:54:04 we can modify it and then choose final solution which will be (re)implemented in hald-addon-bme and battery applet Feb 03 20:54:16 kerio, Estel_ ^^^ Feb 03 20:57:01 good idea, Pali Feb 03 20:58:53 sources which can be used for applet: hald-addon-bme, hal bq27x00_battery device (created automatically), hal rx51_batttery device (created automatically) Feb 03 20:59:49 sources which can be used in hald-addon-bme: anything (because running as root), but preffered is to read only values from /sys/class/power_supply/ Feb 03 21:00:13 current implementation of applet is on: https://gitorious.org/community-ssu/status-area-applet-battery Feb 03 21:00:42 current implementation of hald-addon-bme is on: https://gitorious.org/rx51-bme-replacement/hald-addon-bme Feb 03 21:02:12 kerio, when you will have your idea (on wiki), write me Feb 03 21:05:52 Hi, Can anyone help me with vnc viewer? Feb 03 21:06:21 we can't know until you give us a bit more details Feb 03 21:07:02 Well, I can connect fine when I remove my password on my vps Feb 03 21:07:13 But when there's a password vnc viewer always says invalid password Feb 03 21:07:19 Even if it's the right password Feb 03 21:13:15 the vnc viewer where? Feb 03 21:14:32 on nokia n900 Feb 03 21:15:00 which viewer Feb 03 21:15:15 tried using abcdef as password? Feb 03 21:15:26 to filter out symbols problems? Feb 03 21:15:39 windows server won't let me use anything simple Feb 03 21:15:46 confirmed, that from other viewer, in same situation, it works? Feb 03 21:16:02 erm? Feb 03 21:16:08 I've only tried the VNC Viewer on n900 Feb 03 21:16:14 so use tight vnc or anything else configurable Feb 03 21:16:29 well, I recommend presenceVNC for vnc and rdesktop for rtp Feb 03 21:16:32 anyway Feb 03 21:16:32 iirc there's two vnc viewers Feb 03 21:16:48 I'll try presencevnc Feb 03 21:16:48 which vnc server you use on desktop Feb 03 21:17:02 tightvnc is on my desktop Feb 03 21:17:18 I would say it's problem with server not viewer - I recommended presencevnc as it's more adapted to N900 Feb 03 21:17:28 so why it won't let you any simple password? Feb 03 21:17:40 last time I've checked tight wasn't picky Feb 03 21:18:19 my usual way for connecting to vnc is setting server to accept local connections only, and ssh'in from device, creating local port forwarding Feb 03 21:18:37 so on device i connect to 127.0.0.1: Feb 03 21:18:47 this way everything works ok, and i can screw vnc password Feb 03 21:18:57 as ssh private key auth takes care of it Feb 03 21:19:28 not that I doesn't have vnc password set anyway - and it works. as a security paranoia level, and just for the fun of it Feb 03 21:19:38 so it should work for you too.... hm. Feb 03 21:19:51 but you used term vps Feb 03 21:19:58 was it meant to be vnc or you really mean vps? Feb 03 21:20:31 and if yes, what vps have to do with it? Feb 03 21:25:56 i was connecting to my windows vps Feb 03 21:26:10 it's working now, for some reason when i used the on screen keyboard it accepted my password Feb 03 21:26:26 but when i type it from the keyboard it says invalid password Feb 03 21:47:33 framework_nulled, you mean on-screen keyboard on N900? Feb 03 21:47:52 vnc viewer seems to not ccept some symbols from hardware key mappings, then Feb 03 21:48:04 it's probably because it's qt Feb 03 21:48:09 and qt fails HARD at keyboard input Feb 03 21:48:21 true, sadly Feb 03 21:48:40 yeah estel_ Feb 03 21:48:49 when i use the on-screen keyboard i can login fine Feb 03 21:48:57 when i don't use it i get an invalid password error Feb 03 21:49:21 can you test in Notes if what you think you're inputting is actually being input? Feb 03 21:49:49 notes isn't qt, is it? Feb 03 21:50:02 he should try it at some other field in vnc viewer Feb 03 21:50:04 maybe he has caps lock enabled :P Feb 03 21:50:17 very funny :P Feb 03 21:50:23 i don't Feb 03 21:50:26 i turned that off Feb 03 21:50:32 and yeah it shows up fine in the notes Feb 03 21:50:46 kerio, how to get imin from registers? again, where have you found explanation where some value is hardcoded in registers? Feb 03 21:50:53 or is it raw sniff from i2c Feb 03 21:51:02 and you use same calculation as for i2c values? Feb 03 21:52:09 Estel_: read the bq27000.sh source Feb 03 21:52:21 or the bq27k_detail.sh source Feb 03 21:57:34 kerio, so it's the same as i2c Feb 03 21:57:41 i assume so Feb 03 21:57:48 and trust me, I've been through bq27200.sh sources many times :P Feb 03 21:57:48 well, it's not sniffing Feb 03 21:57:50 it's polling Feb 03 21:57:57 yep, it was coloqualism Feb 03 21:58:09 as in sniffing other, more physical things Feb 03 21:58:17 if you get the pun. Or not. Feb 03 21:58:49 btw define difference between sniffing and polling? ;) Feb 03 22:00:04 sniffing is passive, polling is active Feb 03 22:00:16 well, true Feb 03 22:00:42 you actually talk to i2c, not just read data Feb 03 22:00:49 like with registers file Feb 03 22:00:56 traps, traps everywhere! Feb 03 22:01:21 excuse my specifinc sense of humour, while scripting new version on bme, I'm in "coding" madness ;) Feb 03 22:01:26 scripting sadness Feb 03 22:01:27 etc Feb 03 22:06:44 freemangordon: Welcome back man. What about you health? Is it better now? Feb 03 22:30:40 Pali, sorry for interrupting you, but I got small dillemda related to your newest implementation of bq27x00_battery as in newest kp's Feb 03 22:31:01 as you remember, there were earlier version of that module in olders KP's, that had wrong calculations Feb 03 22:31:10 there is still a chance that someone may be using it, though Feb 03 22:31:36 now, I want to check, in my script, that bq27x00_battery is *loaded*= but *only* the "fixed" one Feb 03 22:31:55 irrelevant for bme replacement which working only with >= kp52 Feb 03 22:32:01 now, how to distinguish it from older one on older kernels? Feb 03 22:32:09 well, my script works with or without bme replacement Feb 03 22:32:15 use uname for kernel version Feb 03 22:32:24 it uses bq27x00_battery exported data, and if not, i2cget Feb 03 22:32:27 $ uname -r Feb 03 22:32:32 well, kernel present != module loaded Feb 03 22:32:55 check if exists dir /sys/class/... Feb 03 22:32:56 of course I though about that, but I don't want to force usage of this module on BNF users, even if the6're kp users Feb 03 22:33:08 sure, but it exist for older kenrnel too Feb 03 22:33:18 and old module with wrong values Feb 03 22:33:24 check kernel version Feb 03 22:33:30 $ uname -r Feb 03 22:33:37 at the same time... hm, ok Feb 03 22:34:00 thanks Feb 03 22:34:38 btw on which version of KP you've fixed bq27x00_battery, Pali? Feb 03 22:34:42 since which version? Feb 03 22:35:17 I do not know, I fixed different problems in different versions Feb 03 22:35:40 and now I see that in upstream kernel and kernel-power temperature of bq27x00_battery is reported in bad units Feb 03 22:36:22 kernel api say that it should be reported in 1/100°C but bq27x00_battery report it in 1/10°C Feb 03 22:36:41 0_o deja-vu Feb 03 22:36:42 rx51_battery reporting temperature in 1/100°C Feb 03 22:36:59 I do not know if changing it in kernel-power is good idea Feb 03 22:37:00 do you remember ammount of false bug reports, when you've changed it? :P Feb 03 22:37:16 but I will sent patch to upstream kernel Feb 03 22:37:19 sure Feb 03 22:37:27 if they accept it, then bring it to kp Feb 03 22:37:50 when will this hash sums stuff be fixed for extra-devel ? Feb 03 22:37:54 I cannot seem to install some packages from extra-devel because of it Feb 03 22:38:08 Estel_, see: https://lkml.org/lkml/2013/2/3/274 Feb 03 22:38:38 does the CSSU kernel have hardware support for aes ? Feb 03 22:38:50 http://maemo-community-mailing-lists.2589537.n2.nabble.com/N900-AES-and-SHA1-MD5-hw-acceleration-drivers-td5495027.html Feb 03 22:38:55 I cam across this Feb 03 22:38:59 yes Feb 03 22:38:59 LinuxCode, do you mean omap AES support? Feb 03 22:39:05 yes Feb 03 22:39:09 no Feb 03 22:39:12 no? Feb 03 22:39:16 yes Feb 03 22:39:17 kernel power had aes support Feb 03 22:39:21 no kernel support Feb 03 22:39:21 it needs some special NOLO Feb 03 22:39:32 it needs special *x-loader* Feb 03 22:39:32 not that special Feb 03 22:39:51 it's only one mshield register that is set differently Feb 03 22:39:51 jacekowski: hmm? Feb 03 22:40:12 well, truecrypt would benefit like hell from hardware aes support, I suppose Feb 03 22:40:17 will I start a war, if I ask if it will be in eventually ? Feb 03 22:40:25 Estel_, so would dm-crypt Feb 03 22:40:32 * Estel_ nods Feb 03 22:40:43 Estel_: IIRC according to benchmarks HW is slower than SW Feb 03 22:40:46 I asked nokia developer and he told me that special x-loader has included special binary blob in secure mode which doing something whith mshield Feb 03 22:40:46 ever since I heard reports that the UK police will now just read people's mobiles out Feb 03 22:40:57 I am feeling distinctly uneasy Feb 03 22:40:59 something which cannot be configured from kernel or nolo Feb 03 22:41:00 never though we have smth like it. Well, even without hw aes, twofish and serpent works quite fast, still Feb 03 22:41:15 freemangordon, was afraid about smth like that Feb 03 22:41:19 jacekowski: we don't have code in kernel running in secure mode Feb 03 22:41:33 freemangordon: it's not about secure mode Feb 03 22:41:34 jacekowski, do you have some more info about it? Feb 03 22:41:48 freemangordon: it's about enabling the hardware, and that is controlled by mshield stuff Feb 03 22:41:52 freemangordon, so any reason to use hw aes "acceler"? Feb 03 22:42:14 LinuxCode, about police - ha ha, I would like to see one reading N900's encrypted content Feb 03 22:42:23 freemangordon: according to benchmark HW AES was faster than SW Feb 03 22:42:29 just be sure to have lock code set to 5 minutes Feb 03 22:42:43 >> You need special boot loader, because it has secure code to enable it.. You cannot do it from non-secure side. << Feb 03 22:42:49 freemangordon: as in, someone tried it in this channel Feb 03 22:43:00 >> Nolo has binary blob for that. << Feb 03 22:43:18 jacekowski: ok, my memories are vague. However, we can't change anything but AUX reg from the kernel. And we don;t have the keys to sign PPA application Feb 03 22:43:41 >> In order to use AES hw it is necessary to enable it during boot from Secure side. It is done by special code which is compiled as part of bootloader. << Feb 03 22:43:46 so I guess we wont see it then Feb 03 22:43:50 lets slice clod brute force on keys during every kp and cssu installation Feb 03 22:43:53 ;) Feb 03 22:44:06 this is what nokia developer told me Feb 03 22:44:25 Pali: could there be some call to the secure monitor? Feb 03 22:44:25 LinuxCode, don't be sad, though - IDK why, but on N900, serpent seems fastest during benchmarks Feb 03 22:44:28 no idea why Feb 03 22:44:38 as we have more support for twofish and aes, theoretically Feb 03 22:44:46 but I have that *special* flasher, x-loader, nolo and source code of kernel drivers Feb 03 22:44:59 freemangordon: we don't have to Feb 03 22:45:07 Pali: the problem is that it is only for one HW revision iirc Feb 03 22:45:16 freemangordon, I asked that nokia developer and he told me that it is not possible Feb 03 22:45:17 freemangordon: once bootloader enables it, you can use it fine from unsecure kernel Feb 03 22:45:38 jacekowski: sure, but we don;t have that bootloader Feb 03 22:45:40 freemangordon, all 21xx and 22xx have same xloader and nolo Feb 03 22:45:43 freemangordon: we do Feb 03 22:45:49 hmm, ok Feb 03 22:45:55 so that is not for one but for more hw revision Feb 03 22:46:22 there are only two revisions in practical existence, and another 2 virtually non-existent now Feb 03 22:46:24 someone linked it like 2 years ago in this channel Feb 03 22:46:27 since the signature is same for all hw rev Feb 03 22:46:31 only reason why I mentioned AES hw support is, that surely there will be power savings Feb 03 22:46:34 jacekowski, do you know if it is possible to enable it from non secure side? Feb 03 22:47:00 Pali: I doubt that will be possible Feb 03 22:47:24 from limited amount of information i could find about mshield, yes it is possible Feb 03 22:47:28 if mshield is configured to allow it Feb 03 22:47:40 so that is no in our case Feb 03 22:47:47 DocScrutinizer05, does the extra-devel repo need updating or something, I keep getting hash sums mismatches, which then means I cant install stuff Feb 03 22:47:50 jacekowski, do you have some information about mshield? Feb 03 22:47:56 Pali: very limited Feb 03 22:48:06 LinuxCode: the key is expired Feb 03 22:48:06 merlin1991, mentioned something about rebuilding/re-checkign that ?!?! Feb 03 22:48:07 I was not able to find anything Feb 03 22:48:13 Pali: mostly marketing stuff Feb 03 22:48:20 Pali: no real technical info Feb 03 22:48:22 DocScrutinizer05, elaborate please Feb 03 22:48:23 LinuxCode: ask pali about it, he knows all the details Feb 03 22:48:32 gpg key used for signing ? Feb 03 22:48:46 jacekowski, ok Feb 03 22:48:50 err, or maybe I'm completely wrong Feb 03 22:49:01 LinuxCode, GPG key on downloads.maemo.nokia.com was expired Feb 03 22:49:07 ahh k Feb 03 22:49:11 so SSU and OVI repositories not working Feb 03 22:49:12 I guess the extras(-devel) is autobuilder related Feb 03 22:49:16 LinuxCode, either use apt directly, or fapmanp or red pill mode Feb 03 22:49:18 so how can I disable gpg key checks for extra-devel ? Feb 03 22:49:24 well I did apt Feb 03 22:49:30 it bitches there too Feb 03 22:49:36 if 3th option, then disable those superiority of signed content in options Feb 03 22:49:37 merlin1991 and thedead1440 should know what's the problem with hashsum Feb 03 22:49:45 extras-devel is on maemo.org Feb 03 22:49:51 and GPG key is without expiration date Feb 03 22:49:59 hmmm Feb 03 22:50:09 LinuxCode, maybe you're using some mirror? Feb 03 22:50:11 I tried to install cryptfs Feb 03 22:50:21 Estel_, let me double check Feb 03 22:50:31 LinuxCode, why the hell encrypt home? Feb 03 22:50:38 no extras-devel Feb 03 22:50:43 r.m.o Feb 03 22:50:53 Pali, jacekowski: I guess we can disassemble xloader with and without that mshield stuff and check what needs to be dome Feb 03 22:51:01 You're keeping that confidential non-police-eyes data on opt, or what? Feb 03 22:51:14 Estel_, ? Feb 03 22:51:15 truecrypt partition seems more reasonable re performance and power savings Feb 03 22:51:30 well, I wonder why do you want to encrypt whole $HOME Feb 03 22:51:34 I havent done anything yet, I just want to install a few things from extra-devel Feb 03 22:51:37 freemangordon, that nokia developer wrote me that he included into x-loader/nolo some binary blob without source code Feb 03 22:51:39 without getting the error Feb 03 22:51:50 there is GUI'fied (without sacrificing CL) latest TrueCrypt in repos Feb 03 22:52:09 just repartition your emmc or sd card and encrypt whole partition via truecrypt Feb 03 22:52:12 or two of them Feb 03 22:52:36 one for less-secure things, and then config maemo to keep addressbook, mails, passwords, settings, photos and videos there... Feb 03 22:52:41 that is one option Feb 03 22:52:47 and auto-mount it at boot time, just provising password and keys Feb 03 22:52:49 LinuxCode: Estel_ is in a rage, he highlighted you on accident Feb 03 22:52:58 k Feb 03 22:53:02 wtf? Feb 03 22:53:05 no, I don't. Feb 03 22:53:12 so, how do I fix this hash sums issue for extras-devel ? Feb 03 22:53:23 that would really help me out ;-} Feb 03 22:53:30 actually, we're talking about encyption, and you interlooped, DocScrutinizer05 ;) Feb 03 22:53:35 also stop me from hammering the repo twice Feb 03 22:54:00 LinuxCode, no idea about repo key, but as for your needs about encrypting, it seems to me that truecrypt is more feasible Feb 03 22:54:25 according to http://omappedia.org/wiki/Bootloader_Project there is linux tool generate_MLO which is MShield signing tool Feb 03 22:54:37 such partition mounted at boot time ensures that all your private things are saved there. If you have lock code, no one can ever bypass it without rebooting device... Feb 03 22:54:44 which always unmount said partition Feb 03 22:54:49 Pali: sure, you just need the keys :D Feb 03 22:54:55 Estel_, I know Feb 03 22:54:59 = no one, thief or police, will access your photos, addressbook, mails etc Feb 03 22:55:04 Estel_, Im just exploring options atm Feb 03 22:55:08 I see Feb 03 22:55:23 well, just tiving advice, as I was through all this before Feb 03 22:55:30 checking every encryption option ;) Feb 03 22:55:30 I must admit though, I kind of like the patch I saw, where one has to enter the /home key at boot Feb 03 22:55:48 sure, but it hurts performance and battery life for whole device Feb 03 22:55:51 because that would confuse anyone who is not a linux user Feb 03 22:55:52 confidential data or not Feb 03 22:56:10 Estel_, well, why would truevrypt have better performance ? Feb 03 22:56:58 because, when you encrypt whole /dev/mmcblk0p2 , you also do so for many programs Feb 03 22:57:04 that don't need it's data encrypted Feb 03 22:57:10 ohh I wouldnt do that Feb 03 22:57:19 every second in your device runtime = accessing some encrypted thing Feb 03 22:57:22 I already resized the partitions Feb 03 22:57:22 well Feb 03 22:57:28 /home/user/.config Feb 03 22:57:29 ;) Feb 03 22:57:37 freemangordon, btw, we have in maemo kernel mshield RNG driver Feb 03 22:57:37 because I previously hit the 2GB max limit Feb 03 22:57:42 so why to encrypt home at all? Feb 03 22:57:44 which is not in upstream kernel Feb 03 22:58:01 RNG = random number generator Feb 03 22:58:07 just prepare separate encrypted partition and throw there everything that need to be encrypted, symlinking it Feb 03 22:58:20 so, for example, mailbox settings or addressbook Feb 03 22:58:20 Estel_, I will do that with dm-crypt Feb 03 22:58:23 hardware AES is still pretty slow Feb 03 22:58:25 I think Feb 03 22:58:29 LinuxCode: we have an expired signing key on Nokia SSU, we have hashsum errors on maemo (probably introduced by autobuilder, ask merlin1991 and thedead1440 and brkn about it) Feb 03 22:58:31 Pali: http://beagleboard.googlecode.com/files/signGP.c, this is for BB GP aiui Feb 03 22:58:32 well, why not Feb 03 22:58:40 if i remember numbers correctly it's somewhere aroudn 15M/s Feb 03 22:58:43 for hardware Feb 03 22:58:47 DocScrutinizer05, k Feb 03 22:58:52 LinuxCode: and we have a repo.m.o that's slow like molasses Feb 03 22:59:03 LinuxCode, using truecrypt, I have some nice gui window loaded just after hildon-desktop, that ask for password and/or keys Feb 03 22:59:11 Pali: scratch that, it does nothing Feb 03 22:59:13 DocScrutinizer05, yeah, but I am wondering if partially people are double hammering the server , because of it Feb 03 22:59:26 HAM downloads twice the same package, then asks if ya want to give up Feb 03 22:59:30 if not provided, addressbook is empty as in fresh device, no mail accounts, etc Feb 03 22:59:37 yet everything *seems* to work Feb 03 22:59:51 LinuxCode: the mirrors mostly have the hashsum problem as well, afaik. One has not as it didn't pick up on the faulty original rmo (aiui) Feb 03 22:59:56 Estel_, I saw a set up like that yes, just over-mount /home/opt or whatever Feb 03 22:59:57 "ask for password and/or keys" Feb 03 23:00:05 So, where do you get the keys from? A USB fob? Feb 03 23:00:10 DocScrutinizer05, I see Feb 03 23:00:17 teotwaki, keys can be files too Feb 03 23:00:17 freemangordon, I know what signGP.c doing Feb 03 23:00:26 you have whole N900 of file candidates :) Feb 03 23:00:35 mixing it with passwords is quite nice Feb 03 23:00:45 Estel_: what's the point of having the keys on the device, if it's just to query for the password? Feb 03 23:00:52 it may be something from mydocs, or from opt, pr from anywhere Feb 03 23:01:19 Either way, you can't revoke the certificate remotely, you can't do anything that would be a benefit when using keys. Feb 03 23:01:20 because possible passwords * all files on device * all combination of those files is more than just password? Feb 03 23:01:35 merlin1991, ping Feb 03 23:01:38 you're just adding 1 bit of entropy Feb 03 23:01:39 certificate? I'm not sure if we're talking about same thing Feb 03 23:01:47 no, certainly not Feb 03 23:02:02 teotwaki__: same screwed thinking that makes devels store pop3 passwords in obfuscated form in a config file Feb 03 23:02:20 I'm talking about truecrypt encrypted partition, which uses password and first 1MB of data from every key file (no limit of ammount of key files) Feb 03 23:02:22 to encrypt Feb 03 23:02:25 and then decrypt Feb 03 23:02:48 oh fucking hell, stop drinking the truecrypt koolaid. Feb 03 23:02:57 not to mention random pool which is irrelevant, btw Feb 03 23:02:59 hm? Feb 03 23:03:12 technical arguments against it or it doesn't happened :P Feb 03 23:03:29 if you store the keyfiles on the same device, there is no added security. Feb 03 23:03:36 There is an added security bias, but no added security. Feb 03 23:03:52 seriously? so how anyone is supposed to know which files to use even with known password? Feb 03 23:03:54 teotwaki__: futile Feb 03 23:04:04 well, I'm asking about technical answer Feb 03 23:04:13 if my thinking is wrong, I'll gladly confirm it Feb 03 23:04:19 maybe by using file `find /` ? Feb 03 23:04:24 and? Feb 03 23:04:26 Because if truecrypt can use the files, it means it has to validate those files as valid input, which means it's either just a hash (which is a terrible way to decrypt files), or the files can be found through find -exec. Feb 03 23:04:36 wrong. Feb 03 23:04:40 right. Feb 03 23:04:41 it doesnt validate them Feb 03 23:04:48 it uses everything you throw at it Feb 03 23:04:50 So it's just a hash of the first 1MB. Feb 03 23:04:57 and it doewsn't event know if partition is really encrypted Feb 03 23:05:06 until you feed it with correct password and keyfiles Feb 03 23:05:08 which is about as useful as a plaintext string written down in /etc. Feb 03 23:05:14 nope. Feb 03 23:05:24 yes Feb 03 23:05:47 first of all, truecrypt have no idea if partition is encrypted or just random Feb 03 23:05:57 second, it have no idea about files you're throwing at it Feb 03 23:06:13 the shellscript to "crack" this "encryption level" is about one line, and takes about 20 min runtime on N900 Feb 03 23:06:22 hash or not, one have to physically test mixing every keyfile with password, during bruteforcing Feb 03 23:06:34 DocScrutinizer05, suuure, elaborate more Feb 03 23:06:35 Estel_: but you agree the device is stolen Feb 03 23:06:40 Estel_: and the files are on the device Feb 03 23:06:41 teotwaki, sure Feb 03 23:06:43 yep. Feb 03 23:06:46 so bruteforce time has little impact Feb 03 23:06:53 yet all the info needed is still on the device Feb 03 23:06:59 it's just a matter of time, correct? Feb 03 23:07:07 it have, when it's more than lifetime of sun ;) Feb 03 23:07:08 And you can eliminate any file under 1MB Feb 03 23:07:13 nope Feb 03 23:07:19 files underr 1MB can be used too Feb 03 23:07:27 it's just that it parses up to 1MB of file Feb 03 23:07:31 if there are 10000 files to test, this is equivalent to 2 more chars in your password Feb 03 23:07:35 roughly Feb 03 23:07:40 and treat is as part of random data input due encryption, methinks Feb 03 23:07:51 DocScrutinizer05, wrong as usual Feb 03 23:07:53 Estel_: it's just pseudo entropy. Feb 03 23:08:01 you don't know if you need 2 files, one, or 10 Feb 03 23:08:05 as *your* usual? Feb 03 23:08:08 + anu combination Feb 03 23:08:10 then I agree Feb 03 23:08:14 s/anu/any/ Feb 03 23:08:15 Estel_ meant: + any combination Feb 03 23:08:28 you don't even know *if* there is key file at all Feb 03 23:08:34 Estel_: and do you really think that people will click on 10 or 20 files, in the right order, everytime they restart their phone? Feb 03 23:08:47 insulting as usual, and clueless as estel usually is Feb 03 23:08:47 Oh no, that's right, they'll do it once, forget the combination, and then come cry here on how to reset it. Feb 03 23:08:52 teotwaki, I restart it ~once per month, so yes Feb 03 23:08:58 :P Feb 03 23:09:10 since I uploaded truecrypt to devel, no one ever did that Feb 03 23:09:21 wth Feb 03 23:09:24 truecrypt for maemo? Feb 03 23:09:28 and it have some big pool of users, even if 1/10of downloader use it :P Feb 03 23:09:32 how exactly does that work Feb 03 23:09:35 there is a password and username for ddownloads.maemo.nokia.com lol ?!?! Feb 03 23:09:35 FIQ, since ages? Feb 03 23:09:43 or was Feb 03 23:09:50 or maybe, they will not forget it, but because their phone is always on 24/7, for 8 months at once (ZOMG uptime of 183 days!), you just need to press the unlock button, and have access to the... that's right, the decrypted partition. Feb 03 23:09:53 FIQ, gimme a second Feb 03 23:10:05 teotwaki, nope. Feb 03 23:10:11 if phone is stolen or seized Feb 03 23:10:22 one need to reboot it at least once to get pass lock code Feb 03 23:10:27 either reflash or easy crack Feb 03 23:10:30 either reflash or easy crack Feb 03 23:10:32 Estel_, why dont you give FIQ a how to Feb 03 23:10:33 Estel_: you do realise I've deployed security systems for banks and governments, right? Feb 03 23:10:35 anyway, part get unmounted Feb 03 23:10:51 there are a few Feb 03 23:10:52 a how to for truecrypt? I don't feel I need it Feb 03 23:11:02 FIQ, k Feb 03 23:11:04 Just curious how it worked for maemo Feb 03 23:11:09 FIQ, I'll link you to thread about truecrypt on maemo Feb 03 23:11:10 teotwaki__: no he doesn't realize Feb 03 23:11:12 just put the files on a usb fob, use the host mode to retrieve the files during boot, and don't keep the fob and the phone together. Done. Feb 03 23:11:12 just a minute Feb 03 23:11:14 teotwaki__: futile Feb 03 23:11:16 Estel_: sure Feb 03 23:11:24 teotwaki, sure, that's and option Feb 03 23:11:41 but, frankly, it all depends on what you're securing Feb 03 23:11:57 I haven't said that it is feasible to encrypt boot-time mounted partition wit keys Feb 03 23:11:58 Estel_: it's the only option, yours doesn't even pass FIPS Feb 03 23:12:10 99% of times, password is enough Feb 03 23:12:23 and that 1% of the time Feb 03 23:12:25 for more touchy data, use encrypted partition that is mounted only when you need it Feb 03 23:12:35 or hidden partition inside outer one Feb 03 23:12:42 they'll use a $2 pair of scissors and make you unlock the focking phone. Feb 03 23:12:50 s/foc/fuc/ Feb 03 23:12:50 teotwaki__ meant: they'll use a $2 pair of scissors and make you unlock the fucking phone. Feb 03 23:13:01 well, that's outside of scope and can be used against any security anywhere Feb 03 23:13:05 on usb stick too Feb 03 23:13:21 teotwaki__: now you remind me of a certain xkcd Feb 03 23:13:26 yea Feb 03 23:13:29 it's popular one Feb 03 23:13:37 FIQ: I had that xkcd happen to me, or rather, a project I worked on. Feb 03 23:13:51 teotwaki, frankly, my setup is low-security partition mount at boot time for photos, addresbook etc Feb 03 23:14:04 http://xkcd.com/538/ found it Feb 03 23:14:06 so random thief/finder won't access my photo Feb 03 23:14:16 or addressbook or configs Feb 03 23:14:18 We deployed a Chip&Pin system for a bank in the UK. The bank gave its customers a small device they could slide their card into, type their PIN, and get either an OTP to login to the website, or a CR to authorise transactions. Feb 03 23:14:18 or whatever Feb 03 23:14:31 A couple months after it was deployed Feb 03 23:14:55 a couple of french dudes got tortured inside their apartment, after two days of torture, both of them died. Feb 03 23:15:04 + I use 2nd partition *not*mounted at boot time, that I keep empty just in case i ever need to store something secure Feb 03 23:15:15 for this purpose, I have script to encrypt swap too Feb 03 23:15:20 The thieves used the bank's device to validate whether the kids were telling them the right PIN. Feb 03 23:15:26 of course tracker etc keeps away from both partitions Feb 03 23:15:26 * DocScrutinizer05 thought for random finder/thief we got device lock code Feb 03 23:15:39 despite what you're trying to say, I see this layout both practical and feasible ;) Feb 03 23:15:52 teotwaki__, lol, in the UK ? Feb 03 23:15:56 DocScrutinizer05, bullshit, as easy reflash will reveal mydocs Feb 03 23:15:56 I have one fo those Feb 03 23:15:56 LinuxCode: aye Feb 03 23:15:57 with photos Feb 03 23:15:59 etc Feb 03 23:16:08 you would just give people the pin then call the bank and police after Feb 03 23:16:21 who cares about the money Feb 03 23:16:22 LinuxCode: not if you're dead when they leave. Feb 03 23:16:37 teotwaki, but it isn't any argument against security Feb 03 23:16:37 was that in the news ? Feb 03 23:16:43 yeah, big story Feb 03 23:16:44 hang on Feb 03 23:16:47 it doesn't have anything to do for passwords Feb 03 23:16:52 Estel_: sure? Feb 03 23:17:12 teotwaki__, I hope this wont be a story in the Sun or Daily Mail Feb 03 23:17:13 and you still haven't told me why storing keys on device is no better than key-less setup, security wise Feb 03 23:17:23 against bruteforce crack Feb 03 23:17:24 * DocScrutinizer05 thinks Estel_'s info is obsolete since PR1.3 Feb 03 23:17:32 and I don't mean bruteforce as in scizzors :P Feb 03 23:17:42 DocScrutinizer05, hm? Feb 03 23:17:45 LinuxCode: it was in 2008 Feb 03 23:17:59 I'm sure that reflashing allows me to boot device without lock code question Feb 03 23:18:03 even if lock code is set Feb 03 23:18:12 you probably could reflash a PR1.2 image though Feb 03 23:18:13 teotwaki__, k gfound it Feb 03 23:18:14 it isn't changed, but not asked Feb 03 23:18:18 I must have missed that Feb 03 23:18:22 unless you enable it again Feb 03 23:18:31 http://en.wikipedia.org/wiki/New_Cross_double_murder Feb 03 23:18:42 teotwaki__, it was several hours not days Feb 03 23:18:45 isn't there a master lock code? Feb 03 23:18:49 LinuxCode: my bad Feb 03 23:18:52 but yeh, then you just give them your pin Feb 03 23:18:52 in case lock code is forgotten Feb 03 23:19:01 FIQ: nope Feb 03 23:19:07 which basically renders lock code useless if you know the algoritm Feb 03 23:19:08 call the police call bank, you will get the mney back Feb 03 23:19:09 oh okay Feb 03 23:19:14 FIQ, during device reflash Feb 03 23:19:19 asking for lock code is removed Feb 03 23:19:25 as you werent reckless... Feb 03 23:19:28 so you can boot device and access mydocs etc freely Feb 03 23:19:31 anyway Feb 03 23:19:32 on pr1.3 too Feb 03 23:19:52 then what's the point? Feb 03 23:20:05 FIQ: and iirc in PR1.3 you need to flash eMMC too, to reset / disable the lockcode Feb 03 23:20:07 that being said Feb 03 23:20:07 also, after that, you may use either john to crack lock code in 3 seconds, or qwerty's file to reset it Feb 03 23:20:17 DocScrutinizer05, wrong Feb 03 23:20:20 did it zillion of times Feb 03 23:20:27 nothing is impossible, you can just make stuff harder for evil people Feb 03 23:20:29 no need to flash emmc ever, to achieve anything Feb 03 23:20:35 LinuxCode: actually, the reason why banks are pushing for Chip&Pin is because it moves all liability to the customer. Feb 03 23:20:38 but not impossible Feb 03 23:20:45 no need to flash emmc ever, to achieve anything Feb 03 23:20:47 that's how I see on things anyway Feb 03 23:20:52 teotwaki__, of course, but that has already been proven as ineffective Feb 03 23:20:55 DocScrutinizer05: ok Feb 03 23:20:58 by security researchers Feb 03 23:20:59 that's why I was just to said that I don't held bank and goverment security in high regard Feb 03 23:21:03 LinuxCode: look at the booklet/contract you got when you got your chip&pin card, it says that if you disclose your PIN, you are not legible for reimbursement. Feb 03 23:21:05 when you told me about working on it Feb 03 23:21:10 just didn't wanted to be inpolite Feb 03 23:21:12 in Africa they use biometrics Feb 03 23:21:18 bwhahahahaha Feb 03 23:21:20 biometrics Feb 03 23:21:27 better than chip and pin Feb 03 23:21:33 no, definitely not. Feb 03 23:21:37 eye vein scan Feb 03 23:21:37 +1 Feb 03 23:21:45 biometrics is a hollywood gimmick. Feb 03 23:21:47 too much s-fi Feb 03 23:21:48 or something Feb 03 23:21:48 it isn't better Feb 03 23:22:01 better than chip and pin Feb 03 23:22:01 drug the user down, force him to authencate Feb 03 23:22:04 you don't use public information to authenticate a person Feb 03 23:22:12 problem solved Feb 03 23:22:18 it reminds me or idiotic programs for laptops Feb 03 23:22:23 *authenticate Feb 03 23:22:23 FIQ: not even that, fingerprints can be recovered at the mark's building, car, office, bus. Feb 03 23:22:27 that use face image Feb 03 23:22:29 to login Feb 03 23:22:32 or users photo Feb 03 23:22:33 :P Feb 03 23:22:34 teotwaki__: that too :P Feb 03 23:22:51 well, best security is multi-layered one Feb 03 23:22:53 though it looks cool! Feb 03 23:22:55 face image or retinal scan can be obtained with $2000 worth of equipment from 200m distance. Feb 03 23:23:07 if attacker doesn't know number of layers required, it's even bette Feb 03 23:23:08 r Feb 03 23:23:22 Chip&Pin is infinitely better than biometrics Feb 03 23:23:23 (biometric auth that is) Feb 03 23:23:35 it has its faults, but still Feb 03 23:23:45 thats why i still think that keyfiles the way truecrypt uses them is still better than nothing, and it certainly complicates things Feb 03 23:23:59 teotwaki, now interesting thing Feb 03 23:24:02 Estel_: just drop it. Feb 03 23:24:14 truecrypt allow you to have set of password and keyfiles for partition revealing data Feb 03 23:24:20 and other set for *same partition* Feb 03 23:24:24 revealing other data Feb 03 23:24:32 without limit of number of such layers Feb 03 23:24:45 so just steganography Feb 03 23:24:46 no technical possibility to determine if there are more layers inside Feb 03 23:24:50 nope Feb 03 23:24:56 you may reveal 10layers Feb 03 23:25:00 due to exortion Feb 03 23:25:01 complicates things, yes, a lot! ... for the user Feb 03 23:25:11 sounds like it would allow multiple partitions on one place which could have other uses Feb 03 23:25:11 and no one will be ever sure if there is no 11 one Feb 03 23:25:25 it is steganography... you hide indistinguishable information in meaningful information Feb 03 23:25:28 no way to determine if empty space is just encrypted empty space or some data in another layer Feb 03 23:25:36 not exactly Feb 03 23:25:43 yes, exactly. Feb 03 23:25:50 not even truexcrypt knows if partition contain it Feb 03 23:26:08 nope, as in steganography it is still obtainable Feb 03 23:26:15 ... Feb 03 23:26:18 that multilayer stuff to create plausible deniability causes even more torture Feb 03 23:26:29 if you forget how many layers you have or password/keyfiles for it, no fuckin way to even determine it was even there Feb 03 23:26:31 since they never will stop until you die Feb 03 23:26:33 aka plausible deniability Feb 03 23:26:43 "I heard the definition of steganography 10 years ago on hackersnews, so now I know that all steganography is based on hiding text in pictures" Feb 03 23:26:52 DocScrutinizer05, at this point you will not know yourself how many layers you had Feb 03 23:27:00 so they still won't get what they want Feb 03 23:27:08 as you're dead anyway, no loss for you Feb 03 23:27:24 there's no secret big enough to die for Feb 03 23:27:25 teotwaki, please, read what plausible deniability is Feb 03 23:27:36 fuck off you twat Feb 03 23:27:38 "you're dead, no loss for you" Feb 03 23:27:44 sounds positive eh Feb 03 23:27:48 or not Feb 03 23:28:14 *Pk Feb 03 23:28:36 I try to have an educated discussion with you, Estel_, and this is how you treat people? By patronising them on things they probably worked on during their masters? Feb 03 23:29:02 teotwaki, pathetic way to lose discussion by use chanop "quiet" :/ I'm dissapointed in you Feb 03 23:29:11 that's an interesting cloak Feb 03 23:29:24 bye, will return when you won't be drunk... Feb 03 23:29:29 drunk/ Feb 03 23:29:32 right. Feb 03 23:30:01 because anyone who doesn't agree with you is a drunk. Feb 03 23:30:15 teotwaki__: futile, toldya Feb 03 23:30:24 indeed Feb 03 23:30:42 "No, but you have to understand that abusing your operator powers by silencing people who disagree with you is wrong." Feb 03 23:30:54 estel doesn't sound like a contributor but that might just be me Feb 03 23:31:08 well, if you don't want to discuss ig with me, just stop. silencing or kickung via chanop is pathetic kindergarden Feb 03 23:31:16 FIQ: I actually wondered where's that cloak from Feb 03 23:31:27 Estel_: and patronising people and acting like a dick isn't? Feb 03 23:31:28 DocScrutinizer05: from council Feb 03 23:31:36 according to an old meeting Feb 03 23:31:44 old council members got that cloak Feb 03 23:31:46 btw, plausible deniability is something different from steganography, and you could educate yourself, before attacking others. Some concepts shared, some different. Feb 03 23:31:49 FIQ: nope, council cloak is council Feb 03 23:31:52 FIQ: i'd rather he has the contributor cloak than the council one which he abused for ages. Feb 03 23:31:53 when they left Feb 03 23:32:04 teotwaki, patronizing. said guy who stopped civilized discussion by logging to chanop? Feb 03 23:32:07 pathetic Feb 03 23:32:09 ooh Feb 03 23:32:10 DocScrutinizer05: yes but not when they left Feb 03 23:32:18 I see Feb 03 23:32:21 I'm out, go and try "educated discussion" in real life Feb 03 23:32:34 Estel_: when did I say plausible deniability was the same as steganography, exactly? Feb 03 23:32:36 by jumping with punches and silencing others that doesn't share your point of view. Feb 03 23:32:48 I silenced you because you were a dick. Feb 03 23:32:52 we'll see where it leads you Feb 03 23:32:53 sure Feb 03 23:33:03 so enjoy yuor extended e-penis now, bye. Feb 03 23:33:15 bye Estel_ Feb 03 23:33:44 Estel_: my e-penis has been big and hard for ages, no need for you to increase its size, but thanks for lowering yourself to viagra. Feb 03 23:34:21 That feels better :) Feb 03 23:34:52 sorry for flaming and trolling, by the way. I'm done. Feb 03 23:36:00 main problem are guys who *know* they're right, know it so certainly that they never even smell when somebody maybe has yet a better idea about that particular topic Feb 03 23:36:29 DocScrutinizer05, I was like that when I was 16 Feb 03 23:36:41 often leads to patronizing Feb 03 23:36:43 with my first proper g/f, she put me right, quickly Feb 03 23:36:49 hehe Feb 03 23:37:01 then you feel dumb, and stop doing that Feb 03 23:37:07 i bet she does, i bet she does Feb 03 23:37:10 because you realise you were a twat Feb 03 23:37:12 nudge nudge saynomore Feb 03 23:37:18 You had your first proper girlfriend at age 16? Feb 03 23:37:23 kerio, I am 34 now Feb 03 23:37:26 Dude, hand in your geek card. Feb 03 23:37:50 hah Feb 03 23:37:55 teotwaki__, lol, when I say proper, I mean longer than 1 year Feb 03 23:38:00 LinuxCode: what if everybody teling you that you're twat but you just don't listen? Feb 03 23:38:13 DocScrutinizer05, then it makes you look stupid Feb 03 23:38:17 You had a relationship that lasted over a year? Feb 03 23:38:18 very stupid indeed Feb 03 23:38:21 LinuxCode: we heard the first time, hand in your geek card Feb 03 23:38:47 stop lying! We all know humans who date for over a year don't know how to operate IRC, just Facebook. Feb 03 23:38:59 and your n900 priviledges are revoked, go buy an iphone Feb 03 23:39:13 iphone...... Feb 03 23:39:15 hahah Feb 03 23:39:15 Or an S3, if you're feeling adventurous. Feb 03 23:39:17 teotwaki__: and whatsapp! Feb 03 23:39:24 * LinuxCode throws up a little Feb 03 23:39:36 S3 isn't that bad Feb 03 23:39:51 teotwaki__: isn't the nexus4 about the same, but cheaper? Feb 03 23:39:52 kerio: actually, he's 34, so he probably has a decent job now... Feb 03 23:40:00 kerio: yeah, but you can't get one xD Feb 03 23:40:01 so..... now you all made fun of me.... how do I tell apt not to bitch about this expired key crap Feb 03 23:40:18 apt won't bitch about the expired key Feb 03 23:40:35 it bitches about hash sums, and mentions the expired key Feb 03 23:40:43 you'll sometimes get a "install these packages without verification?" Feb 03 23:40:52 hash sum mismatch is a critical error Feb 03 23:41:41 kerio, ok, so is this something local I can fix Feb 03 23:41:49 'this glass is empty, how do i drink water from it anyway?' Feb 03 23:41:51 it is only with extras-devel Feb 03 23:41:58 LinuxCode: use a mirror Feb 03 23:42:01 best comment on slashdot this week, a guy trying to explain Steve Jobs is dead: http://apple.slashdot.org/comments.pl?sid=3429511&cid=42774203 Feb 03 23:42:08 for instance, extras-devel-light Feb 03 23:42:17 "Steve is in the iCloud now. Someone tried turning him off and then back on again. In an appeal to hipsters, he's gone underground to sell more macs. He is permanently 404. There is no way he could have had anything to say about this. There is no app for that. Get it?" Feb 03 23:42:30 deb http://extras-devel.merlin1991.at/ fremantle free non-free Feb 03 23:42:45 k Feb 03 23:42:47 thank you Feb 03 23:43:07 it's a really nice mirror, supports pdiffs, only lists the latest versions, and doesn't afraid of anything Feb 03 23:43:26 grammar error Feb 03 23:43:27 so is this a direct copy of extra-devel, but just works ? Feb 03 23:43:36 * DocScrutinizer05 just noticed that you probably can detect such an individual (twat, partonizing, overly self esteem type) by *none* of their statements about factoids they think to know about ever gets qualified by them adding link a sounrce of info, or by a "probably", "iirc" or similar indication that they might be wrong on this one Feb 03 23:43:50 pdiff=? Feb 03 23:44:12 FIQ: incremental changes in Packages Feb 03 23:44:30 hmmmm Feb 03 23:44:36 so you don't end up downloading the whole 9mb on each update Feb 03 23:44:46 yeah I understand, but... Feb 03 23:44:59 DocScrutinizer05: don't know if that one is for me as well, but I've learned a long time ago, especially when talking about bugs or deliveries, to speak in the conditional tense, and say "we" instead of "I". Feb 03 23:45:05 doesn't that mean it has to keep a diff of every single version? Feb 03 23:45:16 FIQ: the repository does Feb 03 23:45:21 yeah Feb 03 23:45:21 but no Feb 03 23:45:22 repo Feb 03 23:45:33 when the CRM dude asks an update: "We have not been able to reproduce this problem, however, it would appear to be related to..." Feb 03 23:45:37 just from a version to the next Feb 03 23:46:02 and if I have a really old version? Feb 03 23:46:14 or for deliveries, "As planned, we should be able to provide you with..." Feb 03 23:46:31 FIQ: then apt will download the whole file Feb 03 23:46:39 makes sense Feb 03 23:46:45 and from then on, partials :) Feb 03 23:46:58 or simply when they're being idiots, "Being relatively swamped with work, I shall attempt to provide you with an answer within the new 24 hours." Feb 03 23:47:00 so it basically use pdiff if the version I had in question was archived Feb 03 23:47:14 and a diff can be made of it Feb 03 23:47:17 teotwaki__: wasn't specifically for you Feb 03 23:47:28 which can be explained, later on, as "I only promised to try and get back to you within the next 3 business days." Feb 03 23:47:47 24 hours, 3 business days, getit? :P Feb 03 23:48:05 s/new/next/ Feb 03 23:48:07 FIQ: much easier, it stores the diffs Feb 03 23:48:08 FIQ, despite you don't see me as contributor, I've promised you a link to truecrypt on Maemo: Feb 03 23:48:09 http://talk.maemo.org/showthread.php?t=81435 Feb 03 23:48:17 repos are static, in nature Feb 03 23:48:29 DocScrutinizer05, what was syntax for factoids only applicable to #maemo? Feb 03 23:48:34 whenever the maintainer changes something, files are created/added Feb 03 23:48:47 ty Feb 03 23:49:01 I wouldn't want every infobot user tog et redirected to tmo thread when asking about truecrypt, ya know Feb 03 23:49:19 well Feb 03 23:49:23 this should answer your question Feb 03 23:49:24 ~french Feb 03 23:49:25 Vous pouvez obtenir de l'aide sur Debian sur le canal #debian-fr - For help in french, please join #debian-fr Feb 03 23:49:26 <#channel> factoid-key Feb 03 23:49:49 ohhh so you're given a contributor cloak when you release a devel package? interesting Feb 03 23:50:02 FIQ: no, you're given a cloak when you ask for it Feb 03 23:50:13 hehe, yes Feb 03 23:50:16 basically Feb 03 23:50:25 teotwaki__: I see Feb 03 23:50:30 FIQ: you want one? Feb 03 23:50:47 I guess you have to have some kind of reason though Feb 03 23:50:49 * kerio felt special :( Feb 03 23:51:17 ~#maemo truecrypt is http://talk.maemo.org/showthread.php?t=81435 Feb 03 23:51:18 okay, Estel_ Feb 03 23:51:18 teotwaki__: well I don't think I've done significant contributions to the community besides minor help in tmo Feb 03 23:51:23 ~truecrypt Feb 03 23:51:24 methinks truecrypt is http://talk.maemo.org/showthread.php?t=81435 Feb 03 23:51:47 well, seriously, I never asked for contributor cloak Feb 03 23:51:54 it isn't provided when you release a package Feb 03 23:52:22 I have no idea who gave it to me and when, and I don't care either - still, it's nice to know that someone respect my small and humble contributions Feb 03 23:52:33 speaking of contributions Feb 03 23:53:03 should try and get an overview of maemo stuff that could benefit from being localized Feb 03 23:53:49 "maemo stuff" aka things provided by CSSU Feb 03 23:54:09 freemangordon, that special nolo is completly different as nolo in PR.1.3 FW Feb 03 23:54:22 and untranslated things in core maemo (mostly terminal applications) Feb 03 23:54:28 I have diff between two objdump Feb 03 23:54:30 and teotwaki, frankly, I shouldn't ever bother talking to you, but... You know, I've fought for freedom of my country at not-so-long-ago times, when you were put in prison for it. I'm really not one, who will get "terrorized" by some random, pathetic dude, talking about dicks and just happening to have chanop here. I thought you were partner for educated discussion, when you pinged me about my discussion with someone else, re encryption. Feb 03 23:54:30 I was wrong, it seems, as you turned out to be regular dickhead Feb 03 23:54:30 teotwaki__: I want one if it's that easy Feb 03 23:54:38 in real world, we call those people bandits Feb 03 23:55:21 I went to prison? That's news to me. Feb 03 23:55:23 here, call yourself whatever you want, and still pretend that you're better due to your master degree or whatever, even if yo7're just talking BS and ending discussion in disgracing way of logging to chanop operator status Feb 03 23:55:36 well, you may re read it without being drunk. Feb 03 23:55:39 or not Feb 03 23:55:49 * kerio braces Feb 03 23:56:03 * kerio unbraces Feb 03 23:56:07 in fact, it's quite end of topic for me Feb 03 23:56:17 then shut up NOW Feb 03 23:56:38 * Estel_ shrugs Feb 03 23:57:05 Estel_: that's nice. Have a great evening. Feb 03 23:57:28 hmm Feb 03 23:57:37 DocScrutinizer05: do you remember who the Group Contact is for Maemo? Feb 03 23:57:38 same to you :) everyone loves happy endings. Feb 03 23:57:39 DocScrutinizer05: just wondering, were red pill settings window available in non-cssu maemo? Feb 03 23:57:47 X-Fade: Feb 03 23:58:07 FIQ: if you enable it Feb 03 23:58:11 ~red-pill Feb 03 23:58:12 rumour has it, redpill is http://wiki.maemo.org/Red_Pill_mode Feb 03 23:58:16 well, apparently he is the person who assigns cloaks. Feb 03 23:58:17 FIQ: in the latest couple of versions you can only enable it by editing the config file Feb 03 23:58:27 But I thought X-Fade completely stopped services for #maemo? Feb 03 23:58:41 teotwaki__: yes, and last such session been on a request by me Feb 03 23:58:44 kerio: yea but I wondered if settings was added by cssu Feb 03 23:58:45 DocScrutinizer05, when will be autobuilder working? Feb 03 23:58:52 as it's not localized at all Feb 03 23:59:00 when X-Fade fixed it, Pali Feb 03 23:59:08 unlike *everything else* in non-cssu maemo UI Feb 03 23:59:29 FIQ: it wasn't added by cssu afaik Feb 03 23:59:36 ok Feb 03 23:59:37 it's just... hidden Feb 03 23:59:43 yeah Feb 03 23:59:52 that might explain it not being localized Feb 04 00:01:00 We should make a list of people who need a cloak Feb 04 00:01:05 and send it to X-Fade by email. Feb 04 00:01:13 s/need/want/ Feb 04 00:01:53 but seriously, I went to prison? Fucking A! Feb 04 00:02:01 teotwaki__: we/I did that just ~8 weeks ago Feb 04 00:02:14 DocScrutinizer05: well, then you forgot FIQ! Feb 04 00:02:59 teotwaki__: I also had trouble to parse that statement with prison. I think that "you were" was an "I could" Feb 04 00:03:30 (FIQ) he didn't ask for a cloak Feb 04 00:03:54 He's being humble. Feb 04 00:03:56 or he did ask but didn't fulfill the requirements to grant it Feb 04 00:04:05 Requirements? Feb 04 00:04:11 there are Feb 04 00:04:21 There's requirements on a cosmetic icon? Feb 04 00:04:23 defined on wiki somewhere Feb 04 00:04:30 [01:03:57] or he did ask but didn't fulfill the requirements to grant it Feb 04 00:04:34 I did? :o Feb 04 00:04:45 I don't know Feb 04 00:04:57 no wonder Nokia went down hill, so much bureaucracy even the FOSS Communities they inspired were riddled with it. Feb 04 00:04:57 I said *or* Feb 04 00:05:31 oh right Feb 04 00:05:39 teotwaki__: I think it's just a recommendation Feb 04 00:06:08 to give council a means to say "no" Feb 04 00:06:23 wtf is this channel about? Feb 04 00:06:40 wtf is /topic meant for? Feb 04 00:06:41 this channel is about the operating system maemo Feb 04 00:06:47 which I thought was obvious Feb 04 00:07:05 as it's written in topic, kinda :P Feb 04 00:07:48 i think that has been mispelled Feb 04 00:08:25 linuxman44: did you come here from #freenode? Feb 04 00:08:33 no Feb 04 00:09:21 incredible, >>Registered : Jan 26 03:44:22 2013 (1 week, 1 day, 20:23:43 ago)<< and already a cloak Feb 04 00:09:33 who? Feb 04 00:09:42 aah sorry Feb 04 00:09:48 User reg. : Feb 29 01:40:39 2012 (48 weeks, 4 days, 22:27:26 ago) Feb 04 00:10:53 the topic should be changed Feb 04 00:11:02 Why? Feb 04 00:11:02 suggestions please! Feb 04 00:11:19 'MEMO' Feb 04 00:11:22 not maeamo Feb 04 00:11:24 maemo* Feb 04 00:11:30 bullshit Feb 04 00:11:36 not sure if troll or just really stupid Feb 04 00:11:44 are you are troll or just not informed? Feb 04 00:11:48 he goes to harmattan and ask the same Feb 04 00:11:52 s/you are/you a/ Feb 04 00:11:52 indeed, or was that meant to be funny? Feb 04 00:11:53 FIQ meant: are you a troll or just not informed? Feb 04 00:11:58 I would say troll Feb 04 00:12:32 that ident rings a bell Feb 04 00:12:42 i'd say bot Feb 04 00:12:44 I think we banned him in Fedora before Feb 04 00:12:53 a few months back Feb 04 00:12:55 you can always check :P Feb 04 00:13:12 "are you a troll or just not informed?" "or both?" Feb 04 00:13:39 Blargh, they don't make them as they used to Feb 04 00:14:06 [20:36] im ddosing fedora website Feb 04 00:14:06 Good trawls made you sweat and shed tears before you realised you'd been had. Feb 04 00:14:10 yep Feb 04 00:14:12 the troll Feb 04 00:14:51 Pali, where one can find all most up-to-date bme replacement bits you've made? Feb 04 00:14:54 he cam in and claimed one of our contributors had died Feb 04 00:14:58 came* Feb 04 00:15:00 as per "user installablen Feb 04 00:15:04 not source code Feb 04 00:15:20 now we just get germans who scream "AD HOMINEM AD HOMINEM" and "DRUNK DRUNK", and trolls that sound like poorly written jokes, a bit as if Joe Pasquale came 'round. Feb 04 00:15:25 LinuxCode: then the contributor proceed to join? :P Feb 04 00:15:33 FIQ, eventually yes Feb 04 00:15:41 heh Feb 04 00:15:44 gg troll Feb 04 00:15:46 I had a freenode admin chase him around Feb 04 00:15:51 all day Feb 04 00:15:57 banning him every chan he joined Feb 04 00:16:01 lol Feb 04 00:16:03 ;-p Feb 04 00:16:04 I remember one guy on #boost Feb 04 00:16:10 who would PM people at random Feb 04 00:16:25 send them links to chicks' facebook pages and pictures Feb 04 00:16:47 and ask if they thought her hot, then claim he went to school with, that they were sluts, and they should convert to $religion. Feb 04 00:17:20 Estel_, on: https://gitorious.org/rx51-bme-replacement and http://atrey.karlin.mff.cuni.cz/~pali/rx51-bme-replacement/ Feb 04 00:17:27 now that's an interesting way of trolling Feb 04 00:17:32 DocScrutinizer05, I don't belive you're so bad with english as you oretend to be. Sentence "I did smth in times, where you went in prison for doing that" is quite easy to parse ;) Feb 04 00:17:40 or my english started to suck? Feb 04 00:17:46 it always sucked. Feb 04 00:17:51 s/where/when/ Feb 04 00:17:58 well, thats true too! Feb 04 00:18:23 He thought it was a lot less funny when I rang up the IT department of his university (not that I ever did, but the troll got trolled) Feb 04 00:18:24 anyway, if you still talk about it - it was just about "when one were sent to prison for doing so" Feb 04 00:18:40 "was sent" Feb 04 00:18:54 were == plural. Feb 04 00:19:07 well, not quite. Feb 04 00:19:11 right. My bad Feb 04 00:19:16 or not Feb 04 00:19:18 anyway Feb 04 00:19:26 i think you got the message pretty well Feb 04 00:19:31 if not, well... Feb 04 00:19:36 Pali, thanks Feb 04 00:19:52 I didn't understand a word, which is why I replied "that's nice". Feb 04 00:20:18 http://www.youtube.com/watch?v=ZUQ1LXCA6t0 Feb 04 00:21:02 no problem. In simpler words, I mean that some random funny guy with ego problems won't impress me by going berseker and logging into chanop, abusing virtual "strenght" to silence one, when discussions goes badly, as I've meet and fought such nadit behavior in real life Feb 04 00:21:10 which make you really, really pathetic in my eyes Feb 04 00:21:24 I hope that message went correct way, this time Feb 04 00:21:27 cool, cool-cool-cool. Feb 04 00:21:42 s/nadit/bandit/ Feb 04 00:21:47 http://www.youtube.com/watch?v=8IIrf_JSuQk Feb 04 00:22:13 I'm really to tired for clowns, though. Feel free to kick me from channel, anyone Feb 04 00:23:11 http://www.youtube.com/watch?v=d9sTAJOOnCo Feb 04 00:23:26 Estel_: I told you to shut up Feb 04 00:23:29 Who's having fun? Are you guys having fun? Feb 04 00:24:19 Estel_: that's 2nd warning on a new case now Feb 04 00:24:49 moo, GeneralAntilles Feb 04 00:24:50 the threat about instant ban as of a few days ago stays unaffected Feb 04 00:24:54 GeneralAntilles: yaaaay Feb 04 00:24:59 GeneralAntilles: sup buddy? Feb 04 00:25:08 GeneralAntilles: can I just say something? Feb 04 00:25:08 nice to see you after so much time :) Feb 04 00:25:09 I'm watching the Superbowl. :| Feb 04 00:25:14 No. Shush. Feb 04 00:25:18 GeneralAntilles: your state produces a big pile of shit tv shows. Feb 04 00:25:26 My state? Feb 04 00:25:30 Florida. Feb 04 00:25:33 I don't think my state produces any TV shows Feb 04 00:25:37 that I know of, anyway. Feb 04 00:25:49 I mean, a lot of Cops is probably filmed down here Feb 04 00:26:00 but that just because every felon and escaped convict in the south runs to Florida. Feb 04 00:26:04 You can't die of exposure here. Feb 04 00:26:54 Python Hunters, Billy the Exterminator (actually, that may be Louisiana, but same hick kind of show) Feb 04 00:27:12 Haha Feb 04 00:27:17 Pali, you have writtern all of those parts? I feel ashamed of not being able to write that proposition wiki page today Feb 04 00:27:24 I mean, we even get that shit on French TV, seriously. Feb 04 00:27:29 your productivity amazes me Feb 04 00:27:34 Python Hunters sounds really stupid. Feb 04 00:28:24 (parts as in all those things that communicate with bits relying on bme, that *many* have said to be impossible to re-create, feasibly) Feb 04 00:28:37 Well, basically, it's just a bunch of dudes driving boats in the everglades, smashing through bushes and yelling "SNAKE!" "Is it a python?" "No." Feb 04 00:28:59 Are they state/federal officers? Feb 04 00:29:10 And the only pythons they find either have a bullet in their head, or have been run over. Feb 04 00:29:11 Estel_, kernel drivers are written by me, hald-addon-bme was by freemangordon, libbmeipc was taken meego (which is open) and patched for our kernel drivers Feb 04 00:29:23 GeneralAntilles: nha, just some random joes as far as I can tell Feb 04 00:29:42 Sounds excruciating. Why are you watching it? :P Feb 04 00:29:44 still, it's outstanding work. May I ask about caveats, as I underastand it, it isn't finished and don't mimic everything yet? Feb 04 00:29:57 what doesn't work currently, as compared to closed bme? Feb 04 00:30:04 GeneralAntilles: no superball. Feb 04 00:30:14 bowl* Feb 04 00:30:33 http://www.alsq-project.com/barf/bucket/pictures/now%20stop%20asking.jpg Feb 04 00:30:49 BTW, I wonder about feasibility of shutting device down if voltage is < 3000mV for 15 seconds straight, instead of 2800 mv, like now (in replacement) Feb 04 00:31:02 or, even better, making treeshold user-configurable Feb 04 00:31:02 Do they call it a superbowl, because that's the size of bucket people in front of the telly use to store booze, popcorn or other things they might consume during ad-riddled games? Feb 04 00:31:19 GeneralAntilles: you bastard, you so pre-empted me. Feb 04 00:31:31 Pali, basically, any thing like "if more than x seconds" is feasible? Feb 04 00:31:40 or would increase battery usage due to polling? Feb 04 00:31:41 GeneralAntilles: hi mister CO Feb 04 00:31:48 :-) Feb 04 00:31:51 Post-season college football games are called bowl games. Feb 04 00:31:54 It's derived from that. Feb 04 00:32:02 I mean, it must be some straight-now value, or we may use more complicated method? Feb 04 00:32:06 DocScrutinizer05, howdy. Feb 04 00:32:09 I don't feel any stronger. Feb 04 00:32:14 lol Feb 04 00:32:32 I want to know why they call it football when most of the time you never even use your feet to interact with the ball... Feb 04 00:32:37 GeneralAntilles: so it's post-season? Feb 04 00:32:47 GeneralAntilles: and only super colleges can participate? Feb 04 00:32:55 teotwaki__, I'll go with that, sure. Feb 04 00:33:17 GeneralAntilles: you're so regretting saying a single line in the channel, right now, aren't ya? Feb 04 00:33:42 jonwil, because game where you use foot is called soccer there? :P Feb 04 00:33:46 patents, patents! Feb 04 00:34:09 teotwaki__, I'm very bored. :P Feb 04 00:34:14 Alcohol isn't an option. Feb 04 00:34:15 GeneralAntilles: who's playing? Feb 04 00:34:21 Alcohol is always an option. Feb 04 00:34:23 49er's and Ravens Feb 04 00:34:30 btw, today brazillian judge denied apple right to call their products "iPhone" in brazil Feb 04 00:34:37 Unless you're considering refreshments at an AA meeting. Then alcohol isn't an option. Feb 04 00:34:50 due to someone registering iPhone name in 2000 in Brazil Feb 04 00:34:51 http://privatepaste.com/14dc6d91dc <-linuxman troll in /query (not really funny) Feb 04 00:34:59 with apple trying to do the same 6 years later Feb 04 00:35:50 DocScrutinizer05, top tip, leave ban in place for 4 months Feb 04 00:35:51 revenge on/of patent trolls? New SW episode by disney Feb 04 00:36:06 teotwaki__, no alcohol for me. :[ Feb 04 00:36:07 I doubt any troll uses static IP Feb 04 00:36:11 not to mention proxies Feb 04 00:36:24 we didn't ban his IP. Feb 04 00:36:36 he uses same registered nick? 0_o Feb 04 00:36:42 Estel_, he does Feb 04 00:36:47 well ident Feb 04 00:36:56 he's not the sharpest cookie Feb 04 00:36:58 well, people never stop to amaze you. Feb 04 00:37:00 but always linuxman linuxman* Feb 04 00:37:06 he's an idiot Feb 04 00:37:10 in other words Feb 04 00:37:22 even to the trolls Feb 04 00:37:22 poor sob, so many other things to do in life... Feb 04 00:37:23 A bit like some german guy creating puppet accounts on tmo Feb 04 00:37:26 what else would you expect from a troll Feb 04 00:37:48 well, as for tmo trolls, anyone noticed guy called "Piotr"? seems like retard trolling on purpose Feb 04 00:37:50 LinuxCode: well at least some try to ban evade Feb 04 00:38:00 as that's typical troll acting Feb 04 00:38:03 btw it's "Peter" in polish language Feb 04 00:38:08 FIQ, for this imbecile, that is even too much effort Feb 04 00:38:15 lol Feb 04 00:38:20 no loss I guess :P Feb 04 00:38:23 GeneralAntilles: whatchabeenupto? Feb 04 00:38:33 I guess, if you go into freenode asking for a cloak every 5 mins Feb 04 00:38:41 GeneralAntilles: do you have a gun? Feb 04 00:38:45 a permanent ban, will come quicker than you could cough Feb 04 00:39:28 GeneralAntilles: there's some company in downtown miami causing us a lot of grief, mind shooting them? It's just a bunch of lowlife french criminals trying to live the american dream, nothing too bad. Feb 04 00:40:35 teotwaki__, I would be careful with suck requests lol, with all those "thought crimes" aka conspiracy to commit... , in the US Feb 04 00:40:40 such* Feb 04 00:40:53 LinuxCode: I'm not in the US, only GeneralAntilles can get in trouble ;) Feb 04 00:40:57 BTW I mentioned that guy on TMO due to strange fact - registered in 2010, most posts in last few days Feb 04 00:41:05 looks like someone's sub-account Feb 04 00:41:09 teotwaki__, ohh dont you worry, they extradite people for less Feb 04 00:41:34 teotwaki__, like the UK student, for running a torrent site here, which is not a crime here Feb 04 00:41:41 Yeah, I'm not too worried, I've got three passports. Just that will take the french administrations two years to figure out who to arrest. Feb 04 00:41:57 hehe Feb 04 00:42:10 kerio, thank you for that mirror Feb 04 00:42:20 it doesnt have the hash sums issue, and is quite fast too Feb 04 00:42:24 much appreciated Feb 04 00:42:25 you can always rape a young girl too, french goverment will surely give you sanctuary rights Feb 04 00:42:36 ask Polansky Feb 04 00:42:41 he will confirm Feb 04 00:44:14 she was willing Feb 04 00:45:12 I know, just pulling his leg Feb 04 00:45:45 in fact this whole thing with polansky was quite pathetic from us side, but, anyway Feb 04 00:46:18 power complex couldn't stand anyone running from iron hand of justice, even if he actually was right Feb 04 00:46:47 teotwaki__, why do you ask? Feb 04 00:46:55 Ah Feb 04 00:46:56 lol Feb 04 00:47:12 jacekowski, still, I'm not sure if pedophile is much better than rapist Feb 04 00:48:02 So we can conclude it's both bad ;) Feb 04 00:51:42 Now, women pedophiles are becoming more common, here in the netherlands :o Feb 04 00:57:12 ~20580/60 Feb 04 00:57:12 343 Feb 04 00:57:49 ~19980/60 Feb 04 00:57:49 333 Feb 04 00:58:05 Pali, which units bq27x00 reports time to empty in? Feb 04 00:58:42 reading from i2c and from /sys/class/power_supply/bq27x00_battery/time_to_empty_avg I get different reults Feb 04 00:58:47 if it's in seconds Feb 04 00:58:53 and not different by small factor Feb 04 00:59:07 but like 333 minutes vs 400 minutes Feb 04 01:00:21 if it's in minuts*100, difference is even greater Feb 04 01:00:25 minutes* Feb 04 01:02:40 Estel_, time values are in seconds Feb 04 01:02:52 check if time%60 == 0 Feb 04 01:03:08 because bq chip reporting time values in minutes Feb 04 01:04:20 mhm, that's why i calculated in in public here via /60 Feb 04 01:04:27 as per specification you've linked me to Feb 04 01:04:44 I'm just curious why it produces so different values... lemme check smth Feb 04 01:05:38 ~24060/60 Feb 04 01:05:39 401 Feb 04 01:06:11 well, maybe it's measurement moment difference, after all... Feb 04 01:29:40 ~mobilehotspot Feb 04 01:34:56 now I found that NOLO can send string error messages via usb to flasher Feb 04 01:35:12 harmattan flasher using it, but fremantle not Feb 04 01:35:23 maybe I will implement it in 0xFFFF Feb 04 01:35:43 n900 NOLO sending that error strings Feb 04 01:48:57 Pali, would be nice Feb 04 01:49:13 damo22, you want to search repos for qtmobilehotspot Feb 04 01:49:22 as mobilehotspot is earlier, buggy implementation Feb 04 01:49:42 qtmobilehotspot is based on mobilehotspot and carried concept to usable state Feb 04 01:54:36 i cant find qt versoin in stable repo Feb 04 02:28:24 damo22, it may be in devel Feb 04 02:28:32 many stable things are in devel, sadly Feb 04 02:28:51 erm, who remember how mili-thing/1000 is called? Feb 04 02:29:00 nano? Feb 04 02:29:25 * Estel_ is writing comments for almost-finished new version of bnf Feb 04 02:31:17 micro, it seems Feb 04 02:31:21 Off the top of my head, micro. Feb 04 02:31:28 Oh, beat me to it. Feb 04 02:32:06 oh, hello MentalistTraceur Feb 04 02:32:32 *Waves* Feb 04 02:32:40 wouldn't ever expect to meet you here outside CM, even less at thi hour Feb 04 02:32:49 s/thi/this/ Feb 04 02:32:50 Estel_ meant: wouldn't ever expect to meet you here outside CM, even less at this hour Feb 04 02:33:17 How is it in the "big politic"? ;) Feb 04 02:34:44 Yeah, I said I'd try to get on IRC more often. Feb 04 02:36:07 It's like the only place you end up with live updates if shit hits the fan, plus the most active council member and all the volunteers who do infra maintenance are largely on here. Feb 04 02:36:14 maybe you just need a irc bouncer? or not, as it would mean reading through 1000 lines daily, filtering out important discussions out of noise Feb 04 02:37:15 irc is great for latest and greatest, but also, extremely noisy Feb 04 02:37:32 and when I'm here it's even more noisy than usual :P Feb 04 02:38:30 As for "this hour", it's only 21:40 here. Feb 04 02:38:40 Usually I stay up way later. Feb 04 02:38:53 Granted, today I am actually expetionally in need of sleep. Feb 04 02:39:08 so have a nice night :) Feb 04 02:39:43 I'm not getting off yet - going afk for about 30 minutes though. Feb 04 02:39:43 btw in case you need a irc bouncer, you always have a free slot on mine ;) Feb 04 02:40:51 Thanks for the offer, I'll take you up on it if I end up feeling the need for one. **** ENDING LOGGING AT Mon Feb 04 02:59:59 2013