**** BEGIN LOGGING AT Thu Feb 14 03:00:00 2013 Feb 14 09:39:44 Q: Is it possible to force n900 to keep showing pictures gotten from different IMs even if the account is disabled? I mean ...i'm adding my gtalk just to get some pictures for few friends. But if I disable the account , the pictures are not shown anymore. So my Q is , is there a way to force showing pics of disabled accounts? Feb 14 10:05:13 how do revert back to stock camera app? Feb 14 10:06:02 iluminator105: what do you mean? Feb 14 10:06:47 install by hand again, without the other repo active?! Feb 14 10:10:37 i am a commandline guy i dont wanna do thats why i asked Feb 14 11:06:30 is there a way to use instant messaging enabled/disabled account profiles, without contact lists get messed up? Feb 14 11:52:42 amospalla, why do you need it? account can be enabled and always offline :) and account contacts dont get messed up. Feb 14 11:55:24 raandoom: for example, Mary has a jabber account, that I have inside his contact info, when I enable/disable accounts, his jabber account gets outside of his contact entry Feb 14 11:56:31 I like to disable accounts sometimes, but I don't because I have to merge contacts again Feb 14 11:57:20 somebody told me here that it works that way, and I can not do anything, but wanted to ask again, it's strange to me it can not work in a more friendly way Feb 14 11:57:53 wow. i didnt know about that Feb 14 11:59:34 i usually dont merge contacts Feb 14 12:00:02 I like to have all info about "Mary" on a single contact Feb 14 12:12:14 merging contacts is the shit Feb 14 12:13:49 DocScrutinizer05: ping Feb 14 12:15:57 kerio: pong Feb 14 12:16:27 DocScrutinizer05: <3 Feb 14 12:17:05 feeling slightly better? or still monumentally pissed off? Feb 14 12:26:27 wrong question Feb 14 12:26:31 indeed Feb 14 12:30:17 MT took initiative to drive our migration adventure and I'm helping as much as I can. I'm just ignoring my feeling of pissed-off for now, until the issue again gets topic #1. Actually right now our problems are not with approving volunteers and Rob's/HiFo's funny ideas how to manage a tech staff Feb 14 12:36:50 seems HiFo is better at pissing of community than Nokia? ;) Feb 14 12:38:22 a lot better Feb 14 13:00:02 meh, it seems about the same, it's just our expectation that's different Feb 14 13:14:22 did you saw this? http://www.freedesktop.org/wiki/Software/systemd/PredictableNetworkInterfaceNames Feb 14 13:14:37 * teotwaki waves at Jaffa Feb 14 13:14:46 'lo Feb 14 13:15:02 teotwaki: Thought about jumping into your Twitter conversation yesterday, but decided better of it Feb 14 13:15:20 Jaffa: yeah, got a few DMs afterwards. Feb 14 13:15:26 so new udev will rename eth0 to some enp5s0? Feb 14 13:16:18 teotwaki: FWIW, I'm with you. He's often worried about follower numbers or people downvoting things on Planet Feb 14 13:19:41 Pali: hahaha systemd hahaha Feb 14 13:19:53 part of systemd is udev Feb 14 13:20:06 and as I understand this is implemented in udev Feb 14 13:20:15 I want working udev!! Feb 14 13:20:42 enp2s0 or enx78e7d1ea46da as interface name is nonsense Feb 14 13:20:43 yeah, the udev takeover is some major bullshit Feb 14 13:20:55 eth0 is normal name Feb 14 13:21:05 also wlanX Feb 14 13:22:57 ok, somebody should create posters "Wanted: Lennart dead or alive" Feb 14 13:24:42 s/or alive// Feb 14 13:29:15 +10000 Feb 14 13:29:56 is there already a website lennart-no-thanks.net ? Feb 14 13:30:38 see prefix names: http://cgit.freedesktop.org/systemd/systemd/tree/src/udev/udev-builtin-net_id.c#n30 Feb 14 13:30:41 or maybe stop-lennart.org Feb 14 13:30:50 en* = ethernet Feb 14 13:30:52 WTF? Feb 14 13:31:09 we had always eth* for ethernet Feb 14 13:32:12 now this enought Feb 14 13:32:36 SPC TERMINAL DATION INOUT ALPHIC DIM(,) TFU MAX FORWARD CONTROL (ALL); Feb 14 13:33:28 Why somebody accepted this patch wich breaking name conventions for network interfaces??? Feb 14 13:33:53 Finally kernel drivers started using eth* for ethernet and wlan* for wifi cards Feb 14 13:34:36 and now some idiot got nice idea how to revert kernel patches in userspace... Feb 14 13:34:48 and also break everything... Feb 14 13:35:06 that's Lennart's preferred game Feb 14 13:35:34 if they wanted *pemanent* numbers for devices, why not to use scheme wlan and eth?? Feb 14 13:35:51 Now I really do not understand this step Feb 14 13:36:19 I guess each time he watches whole linux world stumpling and falling for a major fuckup he introduced, he gets a day-long orgasm Feb 14 13:37:29 next few weeks he has fever with delusions of thousands of devels patching their code to adopt his braindead new patch Feb 14 13:38:22 "no matter if the love or hate you. as long as everybody speaks about you, you're famous" Feb 14 13:38:29 they* Feb 14 13:40:12 ~poettering Feb 14 13:40:12 'sth is poettering' means it acts invasive, possessive, destructive, and generally in an egocentric exacerbating negative way. ``this cancer is extremely poettering'' Feb 14 13:42:19 does anyone know whats the n900 timer granularity, please? is it as bad as 50msec as my first test implies? Feb 14 13:42:32 which timer? Feb 14 13:42:55 all timers rely on kernel timer granularity, don't they? Feb 14 13:43:52 sure Feb 14 13:44:16 I need to measure time intervals in my c++ code as precisely as possible Feb 14 13:44:58 I tried QTime::elapsed and it implies a resolution of 50ms. Feb 14 13:45:04 "implies" how? Feb 14 13:45:35 it returns ms, but all results I get are a multiple of 50 Feb 14 13:48:06 what does gettimeofday() do? Feb 14 13:48:36 kerio: didn't try and have my sdk at home. so i'll give it a try tonight. Feb 14 13:49:13 I'll let you know ;) anyway, thanks for the hint Feb 14 13:57:22 topro: timer precision is based on the platform Feb 14 14:02:40 topro: you may be able to use the PMU to get some more precise timers, but it's going to be very platform-specific (cortex v8 in this case) Feb 14 14:02:56 topro: http://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0344k/ch14s10s01.html Feb 14 14:12:17 generally "jiffies" is the buzzword Feb 14 14:14:44 teotwaki: hey, ETM, my pet topic :-) Feb 14 14:15:10 how do high-resolution clocks deal with relativistic effects? Feb 14 14:15:17 teotwaki: don't you think ETM would make for a really great extension to what gdb can do? Feb 14 14:16:05 teotwaki: AIUI ETB is available on platform as well, not only via the jtag interface Feb 14 14:17:57 kerio: absolutely identical to the way my butt deals with them Feb 14 14:18:27 but... your butt isn't used to keep the time! Feb 14 14:18:47 ...is it? Feb 14 14:20:15 by the way Feb 14 14:20:37 a guy on my team, friend I've known for decades, a small god when it comes to debugging and hunting down bugs Feb 14 14:20:56 his 4-letter name is JTAG Feb 14 14:21:05 haha Feb 14 14:21:06 nice Feb 14 14:21:11 (first initial then 3 first letters of last name) Feb 14 14:21:20 his email is actually jtag@... Feb 14 14:23:38 DocScrutinizer05, do you have any news about extras-devel? when will be working again? Feb 14 14:24:05 topro: > >The accuracy of various system calls that set timeouts, (e.g., select(2), sigtimedwait(2)) and measure CPU time (e.g., getrusage(2)) is limited by the resolution of the software clock, a clock maintained by the kernel which measures time in jiffies. The size of a jiffy is determined by the value of the kernel constant HZ.<< ( man 7 time ) Feb 14 14:24:10 Pali: ask merlin1991 Feb 14 14:24:31 autobuilder for extras-devel still not working Feb 14 14:24:38 kerio, why merlin? Feb 14 14:24:52 he's the repo dude now Feb 14 14:25:03 merlin1991: ping Feb 14 14:25:24 Pali: possibly never, since HiFo is too concerned any volunteer touching it might be a rogue hacker that messes up our infra Feb 14 14:25:51 oh yeah, there's that too Feb 14 14:26:12 so this is end of maemo, fremantle and n900? Feb 14 14:26:18 Pali: also HiFo does not trust in my abilities to properly handle permissions (or they have never heard about permissions) Feb 14 14:28:48 I just wonder wtf been the purpose of HiFo appointing me for maemo administration manager then, when they want to doublecheck and approve every single chmod or useradd I do Feb 14 14:29:48 DocScrutinizer05, these correspondence emails, are they going between you and Rob or you and the whole board? Feb 14 14:30:31 one been from him to me, with no visible CC. The others are between board and council Feb 14 14:33:30 Pali: (end of...) depends on very next actions and/or statements from HiFo. Possibly yes Feb 14 14:35:45 Pali: NB while 2 out of 3 of HiFo opted for "let them do their job, we can review and revoke everything any time later on", one (guess who) insists in HiFo staying in control of everything related to daily operational business, while same time they say HiFo can't decide anything any faster than ~2 weeks delay minimum Feb 14 14:36:08 DocScrutinizer05: so... can't they outvote him? Feb 14 14:36:26 kerio: NFC what's the deal with HiFo Feb 14 14:36:44 kerio: you need to find evidence of misconduct to do so; although IMO his actions are enough of gross misconduct Feb 14 14:36:59 once I thought I know what they are supposed to do and how they might do that. Now I really don't know anymore about what ideas they got Feb 14 14:37:17 also they can't outvote him; Council decide on misconduct demeanour's Feb 14 14:37:29 ...really Feb 14 14:37:37 DocScrutinizer05, and some info from nokia about maemo swadmin gpg key? Feb 14 14:37:43 Pali: nothing, so far Feb 14 14:37:59 that i mean to kick him out; outvote the issue is they inadvertently end up following him for whatever it is Feb 14 14:40:36 Pali: btw, i was looking for you a while ago Feb 14 14:40:40 my Conversations have a scrollbar Feb 14 14:40:55 the kind that you have to drag around Feb 14 14:41:05 kerio, I had same Feb 14 14:41:05 and trying to scroll in the middle selects text instead Feb 14 14:41:10 i know, that's why i wanted to ask you Feb 14 14:41:24 i already tried deleting my .mozilla Feb 14 14:41:26 didn't fix it Feb 14 14:41:29 * DocScrutinizer05 feels like in school, asking "may I visit the toilet" and answer is "send in filled form 27B/2 for that" Feb 14 14:42:01 DocScrutinizer05: obviously, piss in a corner Feb 14 14:42:03 kerio, check permission in .mozilla Feb 14 14:42:12 also owner/group Feb 14 14:42:35 I deted that folder more times and killed browser/rtcom processes Feb 14 14:42:45 and this fixed it Feb 14 14:46:50 kerio: either that, or just ignore the answer and leave room without further comment [yet to be decided upon whether to come back after toilet or just go home and have a morepleasant day] Feb 14 14:49:59 Pali: the thing is, i deleted it when the system was off Feb 14 14:50:04 so it should've been completely rebuilt afterwards Feb 14 14:53:39 hm, i think i botched the optification Feb 14 14:55:31 DocScrutinizer05: concerning kernel constant HZ: afaik this is true for kernels not supporting high resolution timers, or is hr_timers something x86 specific? Feb 14 14:56:07 sorry, that's too specific for my knowledge about the topic Feb 14 14:56:08 iirc hr timers came with ~2.6.25, right? Feb 14 14:56:59 Pali: no luck :( Feb 14 14:57:56 kerio: (optification) toldya ;) Feb 14 14:58:05 DocScrutinizer05: i also undid the optification Feb 14 14:58:07 and the problem stayed Feb 14 14:58:17 so i don't know if it's the optification or my weird configuration Feb 14 14:58:25 ...i don't receive messages that often Feb 14 14:58:45 kerio: some maybe useless guessoid: I seem to recall this stuff is related to themes Feb 14 15:01:16 what was the non-systemd udev project again?= Feb 14 15:02:37 ~seen Aranel Feb 14 15:02:49 aranel is currently on #n9 (15h 33m 55s) #harmattan (15h 33m 55s) #meego (15h 33m 55s), last said: 'thedead1440: oh okay, It was weird to see all kinds of errors when running a apt-get purge'. Feb 14 15:03:14 kerio, ask Aranel, he had same problem Feb 14 15:03:54 merlin1991, udev is part of systemd... Feb 14 15:04:29 Pali: there is a non systemd udev poject Feb 14 15:05:30 merlin1991, I heard about it, but devs who forked it told that this non systemd udev is not for normal usage... Feb 14 15:05:45 DocScrutinizer05: Please down leave - hope you will return to the class, so you can enjoy all the fun from your classmates - no matter how stupid and incompetent the supervisors are :) Feb 14 15:06:05 yeah, down leave! Feb 14 15:06:06 wait, what Feb 14 15:06:28 ~s/down/dont/ Feb 14 15:08:31 Pali: doesn't say that in their readme: https://github.com/gentoo/eudev Feb 14 15:09:18 merlin1991, devs who forked udev told that on fosdem Feb 14 15:09:38 ah okay Feb 14 15:09:56 they need to get convinced for the better Feb 14 15:10:03 well, they said e stands for experimental Feb 14 15:10:40 or could stand Feb 14 15:10:50 so far they mostly just re-plumbed the buildsystem and tweaked few things here and there Feb 14 15:11:30 but the goal is of course to have something in case systemd devs go borg on init systems Feb 14 15:11:38 "in case"? Feb 14 15:12:26 no comment :] Feb 14 15:12:40 I'm happily using mdev and openrc atm Feb 14 15:12:54 fsck systemd Feb 14 15:13:00 ~poettering Feb 14 15:13:01 'sth is poettering' means it acts invasive, possessive, destructive, and generally in an egocentric exacerbating negative way. ``this cancer is extremely poettering'' Feb 14 15:13:28 hm never heard of openrc Feb 14 15:15:17 anyone using x11vnc on the n900? Feb 14 15:15:54 meh, i should reflash Feb 14 15:15:55 s/dead or alive/dead: reward 10000EUR, alive: reward 1000EUR"/ Feb 14 15:16:05 starting over is always fun Feb 14 15:16:14 "dead or alive, but preferrably dead" Feb 14 15:17:13 s/dead or alive/dead: reward 10000EUR, alive: reward 1000EUR (only when conveniently packaged)"/ Feb 14 15:17:34 DocScrutinizer05: do you have a BM backup of a freshly-flashed system? Feb 14 15:18:14 DocScrutinizer05: systemd is very good compared to PA Feb 14 15:18:21 kerio, try to install new theme Feb 14 15:18:37 switch to that new theme and reboot Feb 14 15:19:18 kerio: sorry, probably not Feb 14 15:19:41 hm, it's probably even better to grab the .fiasco images, unpack, nandsim, and grab the files Feb 14 15:20:48 kerio: i might deliver any files you ask for Feb 14 15:21:23 nah, i'll do it myself after i boot my linux partition Feb 14 15:21:25 no nandsim on os x Feb 14 15:21:28 and no ubifs Feb 14 15:21:30 and no ext Feb 14 15:21:34 hehe Feb 14 15:22:02 DocScrutinizer05: can you check if /usr/share/themes is a symlink on your system(s)? Feb 14 15:22:11 sure Feb 14 15:24:17 lrwxrwxrwx 1 root root 21 2010-06-23 06:03 /usr/share/themes -> /opt/usr/share/themes Feb 14 15:24:24 ok, so it wasn't mine :) Feb 14 15:25:04 incredible how much customization hides in /etc etc Feb 14 15:25:29 ssh iron900: ssh: Could not resolve hostname iron900: Name or service not known Feb 14 15:26:01 lrn2.ssh/config Feb 14 15:26:05 moving to a new system *always* is weeks of annoyance Feb 14 15:26:21 that's why you put ssh fun into .ssh/config Feb 14 15:26:28 so you keep at least those aliases ;) Feb 14 15:26:32 Pali: still no dice Feb 14 15:26:40 but the theme thing *is* a bit weird Feb 14 15:26:42 kerio: i'm not going to assign IP addr to hostnames inside my ssh config Feb 14 15:26:43 some things don't change properly Feb 14 15:27:09 kerio: since that's clearly a job for /etc7hosts Feb 14 15:27:09 also what the fuck, hildon-theme-beta is ugly as fuck Feb 14 15:29:31 kerio, try to spawn browser[d] and rtcom-* with run-standalone.sh Feb 14 15:30:48 grrr, no fan RpM on linux sensors for this board Feb 14 15:32:37 Manufacturer: "Intel Corporation" Product: "DH77EB" Version: "AAG39073-304" Feb 14 15:34:53 kerio, I updated battery status plugin, bme replacement packages and kernel-power v52 Feb 14 15:35:06 yay Feb 14 15:35:13 there was bug in rx51_battery driver which reported temperature in bad units Feb 14 15:35:22 so I changed it also in bme replacement packages Feb 14 15:35:34 hm, is that the only fix? Feb 14 15:35:57 ~pali Feb 14 15:35:57 well, pali is http://atrey.karlin.mff.cuni.cz/~pali/ Feb 14 15:36:11 kerio, next patch is for bq2415x driver Feb 14 15:36:25 I renamed none mode to off Feb 14 15:36:39 is "none" also still accepted? Feb 14 15:36:45 and added new none mode which enable charger with current limit 100mA Feb 14 15:36:50 oh Feb 14 15:36:51 neat Feb 14 15:37:14 Pali: why not "low" or something like that? Feb 14 15:37:29 breaking backwards-compatibility, even for bleeding-edge things, always makes me sad :( Feb 14 15:37:33 usb spec say that 100mA is max limit before usb enumeration Feb 14 15:37:47 yeah, but why not call it "low"? Feb 14 15:37:50 so 100mA is safe Feb 14 15:38:09 because none is already used by some scripts/people Feb 14 15:38:25 ...that's precisely the wrong reason to call it none Feb 14 15:39:06 unknown/none/na charger Feb 14 15:40:21 kerio, make sure you update kernel-power and bme replacement packages at one time without rebooting Feb 14 15:40:37 instead you can get bootloop because of "hot" device Feb 14 15:40:38 yeah, i figured :) Feb 14 15:41:28 now I pushed also update for status area battery plugin to cssu devel Feb 14 15:42:47 Pali: what's the directory for kernel-power on your site? Feb 14 15:43:19 it is not in directory Feb 14 15:43:23 :( Feb 14 15:44:28 http://talk.maemo.org/showpost.php?p=1270141&postcount=168 Feb 14 15:45:03 wait, that's september 21 Feb 14 15:46:11 links are same Feb 14 15:49:25 Pali: have you also gave a look at my patch for u-boot-tools? Feb 14 15:49:41 kerio, not yet... Feb 14 15:50:17 merlin1991, when will be next cssu meeting? Feb 14 15:53:58 Pali: it boots! yay! Feb 14 15:56:31 holy shit, 3 seconds for 2 reboots Feb 14 15:57:31 what? Feb 14 16:11:03 Pali: the hwclock Feb 14 16:11:11 it drifts like hell Feb 14 16:18:20 kerio, I forgot to update hald-addon-bme, now I uploaded new deb package Feb 14 16:19:44 D: Feb 14 16:23:57 kerio: is that face at or about me? Feb 14 16:24:11 MentalistTraceur: no, it's just that i have to reboot Feb 14 16:24:25 [notice] mail back to normal operation Feb 14 16:24:33 hi MentalistTraceur Feb 14 16:24:37 DocScrutinizer05: but is it a board-approved operation? Feb 14 16:25:05 kerio: obviously not, so i gonna have to delete all mails that would get sent to board Feb 14 16:25:18 ;-P Feb 14 16:27:32 kerio: (hwclock drift) fix your adjfile! Feb 14 16:27:38 DocScrutinizer05: oh you Feb 14 16:28:41 I honestly never seen RTC drifting much. I seen hwclock --adjfile messing stuff up fubar several times Feb 14 16:29:06 DocScrutinizer05: the hwclock has 1-sec granularity Feb 14 16:29:29 i guess that you'll lose half a second, on average, for each write/shutdown/boot/read Feb 14 16:31:14 oh FFS Feb 14 16:31:33 * DocScrutinizer05 wishes he hadn't fixed mail VM Feb 14 16:32:18 :-( Feb 14 17:09:41 kerio, I updated kernel-power deb packages Feb 14 17:10:00 I forgot to include one patch for bq2415x Feb 14 17:10:32 without it sysfs node reported bad detected charger Feb 14 18:02:11 Pali: blah Feb 14 18:02:17 one more reboot Feb 14 18:02:25 son, i am disappoint Feb 14 18:02:45 Pali: i want ksplice for kernel-power! Feb 14 18:21:23 udev - that seams to be work of some alien! I do not understand this lanuage at all. I think, if that shit would happen, i am going to put all my interfaces on bridge and build brtables and arptables to get back both sensul name and normal NIC behavour Feb 14 18:21:51 how expensive is a hitman? Feb 14 18:22:03 we can pool some bitcoins Feb 14 18:26:42 cut his hands, so he will not be able to write any new line of code Feb 14 18:29:29 Pali: there's speech-to-text now Feb 14 18:29:40 i believe that any solution has to be more... radical Feb 14 18:30:52 i'm afraid redhat would pull a clone out of their fridge if anything ever happened to poettering Feb 14 18:30:56 hahaha Feb 14 18:31:09 DocScrutinizer05: systemd's bus factor is 1 Feb 14 18:31:56 which in itself disqualifies systemd for any serious system Feb 14 18:32:33 since eventually he *will* get hit by a bus, and then? watch systemd bitrotting? Feb 14 18:32:42 DocScrutinizer05: gleefully Feb 14 18:32:55 DocScrutinizer05: btw, can you drive a bus? Feb 14 19:02:45 also, should i reflash, at last? Feb 14 19:29:44 kerio: I can drive everything from a micro rc-car (28mm total length) to bus, tank, even helicopter Feb 14 19:30:45 You don't drive a helicopter. Feb 14 19:32:41 So, Doc, how big a... mhm... helicopter have you driven so far? Feb 14 19:50:13 how big is this Bell thing? Feb 14 19:51:26 (it only been 3 minutes, on a ride I hired where some other pilot did lessons and they wanted to earn a hundred bucks for taking me with them) Feb 14 19:52:09 the other ones been much smaller, much much much smaller ;-) Feb 14 19:52:46 (or even virtual) Feb 14 20:01:43 GeneralAntilles: a good friend of me stated I don't drive a car either - he supposed I operate a mobile Otto engine Feb 14 20:03:30 DocScrutinizer05: i bet that a helicopter-shaped dent in lennart poettering would do wonders for systemd Feb 14 20:04:59 I'd bet on that. Less though on the details of that wonder. Might be rather unexpected Feb 14 20:11:15 I wonder if red bull went through / completed their planned modifications to their Bo105 that would enable it to hover upside down Feb 14 20:11:34 seriously, what the hell's wrong with my browserd Feb 14 20:13:12 ShadowJK, terrifying Feb 14 20:24:01 browserd[2232]: New e:0x60530, g:0x2f65e8, ctx:(nil) Feb 14 20:24:08 what does this mean? Feb 14 20:28:42 kerio: thismeans it got a flue, and wants a hooker Feb 14 20:30:26 E: Couldn't find package microb-engine-hooker Feb 14 20:30:44 aaaaaw Feb 14 20:31:23 isn't browserd open? Feb 14 20:31:31 not rtcom-messaging-ui though Feb 14 20:31:43 Hello :) I have a problem after battery calibration. Polarcell ~1550 mAh. I don't have any low battery alert notification or sound. Feb 14 20:31:53 this clearly states the process is browserd Feb 14 20:32:03 DocScrutinizer05: smartarse Feb 14 20:32:28 spoofy: i doubt it was caused by the calibration itself Feb 14 20:32:47 anyway, bme won't really care about your calibration Feb 14 20:32:52 kerio: so I should recalibrate it again? Feb 14 20:33:11 why would you calibrate first instance? Feb 14 20:33:11 bme doesn't care about calibration Feb 14 20:33:48 unless you're using bq27k-detail or bq27200.sh you don't even notice calibration Feb 14 20:33:57 or the bme replacement! Feb 14 20:34:07 except that it tends to use bogus data Feb 14 20:34:07 he didn't even mention that Feb 14 20:34:09 Pali: >:C Feb 14 20:34:22 Hmm.. I used a sq27200.sh script Feb 14 20:34:32 holy shit the microb-engine tarball is 58mb, compressed Feb 14 20:34:34 bq* Feb 14 20:34:47 then you should be able to tell if your calibration been successful or not Feb 14 20:34:49 spoofy: yeah but the battery meter doesn't use that Feb 14 20:35:20 is CI=0? what's cycles-since-last-learning? Feb 14 20:35:25 spoofy, the "calibration" only makes a difference for the values reported by bq27200.sh and similar programs. It makes no difference for the default built-in battery meter or low battery alarms... Feb 14 20:36:03 Damn it :/ Feb 14 20:36:33 spoofy: I suggest `stop bme; sleep 5; start bme; sleep 5; stop mce; sleep 5; start mce` Feb 14 20:37:09 DocScrutinizer05: hmmm.. ok I'll try Feb 14 20:37:25 probably your calibration deactivated bat-low warnings and your 2forgot2 to re-enable it Feb 14 20:37:38 "forgot"* Feb 14 20:37:42 DocScrutinizer05: You think It will help? Feb 14 20:37:48 sure Feb 14 20:37:48 or he just didn't start bme again Feb 14 20:37:52 I used polarcell since mounth or more Feb 14 20:37:56 although... nah Feb 14 20:38:00 kerio: that's basically what I meant Feb 14 20:39:46 also the new "battery applet" from dev repo is not working right Feb 14 20:40:02 It shows 1280 mAh but I have 155x mAh Feb 14 20:40:44 spoofy: the battery applet from cssu-devel prefers to display utterly bogus data instead of nothing at all Feb 14 20:41:10 and it prefers to display utterly bogus data instead of data that's reported as perfectly accurate, and *is* perfectly accurate, but it "might not be" Feb 14 20:41:26 is there even a way for the n900 to measure actual capacity of the battery? That "capacity resistor" definitely does not do justice to degraded cells... Feb 14 20:41:37 ecc3g: the capacity resistor is the utterly bogus data Feb 14 20:41:45 bq27k can be calibrated, and it's fairly accurate Feb 14 20:42:06 ~bq27k Feb 14 20:42:11 meh Feb 14 20:42:19 ~listkeys bq27 Feb 14 20:42:21 Factoid search of 'bq27' by key (2): bq27k-detail ;; bq27200. Feb 14 20:42:28 Last mesured Discharge: 1557 mAh Feb 14 20:42:37 ~bq27k-detail Feb 14 20:42:37 extra, extra, read all about it, bq27k-detail is http://maemo.cloud-7.de/maemo5/usr/local/sbin/bq27k-detail2 Feb 14 20:42:48 ~bq27200 Feb 14 20:42:49 [bq27k-detail] http://maemo.cloud-7.de/maemo5/usr/local/sbin/bq27k-detail2 Feb 14 20:43:50 also: Feb 14 20:43:55 ~batteryfaq Feb 14 20:43:56 i heard batteryfaq is http://wiki.openmoko.org/wiki/Battery_Questions_and_Answers Feb 14 20:44:18 (which has built-in bq27000) Feb 14 20:45:38 spoofy: also you should use my script later on in that thread, rather than manual method as suggested by Estel Feb 14 20:46:00 ...for bq27200 calibration Feb 14 20:46:58 or simply look there: http://maemo.cloud-7.de/maemo5/patches_n_tools/calibrate-bq27k.sh Feb 14 20:47:18 1557 sounds fine for polarcell anyway Feb 14 20:48:05 yep Feb 14 20:48:11 w8.. I just want to have: a) battery icon working - showing actual +/- battery state, b) Alert notifications before completly discharge, c) Real battery captivity Feb 14 20:48:25 and 1280 for design capacity sunds abut right as well Feb 14 20:48:29 What should I use to have all of these "features" :D Feb 14 20:48:47 what kind'a script, calibratin or something :D Feb 14 20:49:09 I don't think there's such thing 8yet) Feb 14 20:49:16 (yet) Feb 14 20:49:31 Doc, Why should I use your script? Feb 14 20:49:40 Is it better than estels bqxxxx? Feb 14 20:50:08 DocScrutinizer05; ffs, his issue isn't calibration Feb 14 20:50:08 because it takes care about stopping and starting bme at the right moments, and works unattended Feb 14 20:50:42 ShadowJK: his issue are pink ponies, for all I can tell Feb 14 20:51:29 or, right now, a misconception about calibration paired with a flawed calibration procedure resulting in bme not working Feb 14 20:51:34 Ok.. so when I'm calibrating the battery (learning cycle - stop bme, charge&discharge > bme learned new captivity > charge again) I should stop bme few times again? Feb 14 20:51:59 All those random scripts, including all of mine, wont do anything for him. Feb 14 20:52:06 indeed Feb 14 20:52:26 spoofy; no, bme never learns anything, ever. Feb 14 20:52:28 except save him from messing up stuff during manual calibration Feb 14 20:53:09 [2013-02-14 21:48:11] w8.. I just want to have: a) battery icon working - showing actual +/- battery state, b) Alert notifications before completly discharge, c) Real battery captivity Feb 14 20:53:19 ENOTAVAIL Feb 14 20:53:27 DocScrutinizer05: coffcoffbme replacementcoffcoff Feb 14 20:54:20 The bme replacement might be able to do it eventually, but I don't know anything about it, except that kerio seems to curse its bugs on a regular basis :) Feb 14 20:54:32 oh ok.. nvm I'm totally stupid today ;] Feb 14 20:54:42 ShadowJK: it works as coded Feb 14 20:54:46 hell, it works as intended Feb 14 20:54:56 the bugs are in a particular detail of the intention Feb 14 20:55:12 So the bug is in the spec then? :P Feb 14 20:55:23 kinda, yeah Feb 14 20:57:09 Discovered such a bug at $work yesterday.. Our intern had wired up a electrical cabinet, I was attaching datalines and sensorlines and happened to noticed miswiring that would've smoked everything if it was energized. Wasn't the intern's fault though, the bureau that did the schematics messed up Feb 14 20:57:28 bugfree implementation, buggy spec :D Feb 14 20:57:39 Why do we even bother with this 'battery capacity' abstraction (obv. for end-users, but meh)? Is there a battery applet that shows something more concrete, like voltage being pulled at a given time? Feb 14 20:58:50 (Disclaimer, I have very little understanding of batteries, so I could be way off here) Feb 14 20:59:12 MentalistTraceur; "pulling voltage" is meaningless gibberish :P Feb 14 20:59:23 K. Feb 14 20:59:28 voltage pushes, current is pulled, power = volts * amps Feb 14 20:59:31 MentalistTraceur: current being pulled, maybe Feb 14 20:59:39 that would be neat Feb 14 21:00:02 and energy = power * time, capacity = current * time Feb 14 21:00:16 wait Feb 14 21:00:28 isn't capacity charge over voltage? Feb 14 21:00:29