**** BEGIN LOGGING AT Mon Jul 13 02:59:59 2015 Jul 13 05:37:52 * Maxdamantus wonders who decided what font point sizes should be available in the terminal application. Jul 13 05:38:52 12 is slightly too small to read acomfortable arm's length away, 16 is way too big. Jul 13 05:42:55 $ gconftool-2 -s /apps/osso/xterm/font_size -t int 14 Jul 13 06:22:11 what's the difference between gconftool and gconftool-2 ? Jul 13 06:47:35 one is for g-1 and other for g-2 ? Jul 13 11:08:36 ~repos Jul 13 11:08:36 it has been said that repos is deb http://mirrors.muarf.org/maemo/apt-mirror/mirror/downloads.maemo.nokia.com/fremantle/ssu/apps/ ./;; deb http://mirrors.muarf.org/maemo/apt-mirror/mirror/downloads.maemo.nokia.com/fremantle/ssu/mr0/ ./ Jul 13 12:47:35 IroN900:~# ll `which gconftool` Jul 13 12:47:36 lrwxrwxrwx 1 root root 27 2014-12-06 12:41 /usr/bin/gconftool -> /etc/alternatives/gconftool Jul 13 12:47:38 IroN900:~# ll /etc/alternatives/gconftool Jul 13 12:47:39 lrwxrwxrwx 1 root root 20 2014-12-06 12:41 /etc/alternatives/gconftool -> /usr/bin/gconftool-2 Jul 13 12:48:40 difference: number of dereferences Jul 13 12:52:32 when in 58 years gnome/gtk/gconf-THREE will roll out, your scripts will still work when they use "gconftool" - unless of course when they once more break the complete UI/syntax Jul 13 13:27:00 does anyone knowwhere do I find ^-character on n900? Jul 13 13:38:06 here, have some ^^^^^^^^^^^^ Jul 13 13:45:34 :) Jul 13 13:46:41 my keyboard and fn+sym don't seem to have it so is there any other way to get it other than copypaste? Jul 13 13:48:13 add it to your keyboard layout Jul 13 13:51:28 veikko: I have it there http://i.imgur.com/lmio0WB.png seems like it's layout dependent Jul 13 13:56:54 how do i wake up a n900 from the yellow hearbeat "charging" state? Jul 13 14:00:44 ^ is on vkbd, open that with blue+Sym Jul 13 14:05:38 it's a deadkey, either hit it twice or do ^+space Jul 13 14:06:05 L29Ah: unlock the screen if it's on Jul 13 14:06:08 or turn it on if it's not on Jul 13 14:06:46 lol wow Jul 13 14:06:54 this time it reacted to the power button Jul 13 14:07:01 :* Jul 13 14:09:15 it failed to calibrate the battery again ;[ Jul 13 14:09:25 stuck at 2038 mAh Jul 13 14:16:14 http://maemo.cloud-7.de/maemo5/patches_n_tools/calibrate-bq27k.sh Jul 13 14:16:46 from 2038 you will need ~4 calibration runs Jul 13 14:29:10 /usr/sbin/i2cset -y 2 0x6b 0x01 0xcc; #just in case... Jul 13 14:29:10 /usr/sbin/i2cset -y -m 0x80 2 0x6b 0x00 0x80; # timer reset Jul 13 14:29:15 i wonder what does this do Jul 13 14:31:07 anyway thanks Jul 13 14:42:41 I can elaborate on it, if you really wanna know Jul 13 14:43:22 "timer reset" is rather self explanatory, it tickles the watchdog timer Jul 13 14:44:07 "just in case" is the disabling of charging which shouldn't be needed to get repeated in the loop, it's done before loop starts Jul 13 14:45:06 both are I2C commands sent to the bq24150 charger chip Jul 13 14:46:09 the bq24150 has a watchdog timer that resets the chip when it expires after 15 to 32 seconds Jul 13 14:46:31 why is bme stopped? Jul 13 14:46:45 * L29Ah tries to get how is it different than w/o the script Jul 13 14:46:50 because bme would configure the bq24150 to do charging Jul 13 14:47:13 ah ok Jul 13 14:48:12 so we stop bme, tell the chip to NOT charge and tickle chip's watchdog until the bq27200 gas gauge finished calibration Jul 13 14:48:42 once calibration accomplished, we restart bme and thus charging Jul 13 14:50:57 i wonder why did it fail to get my battery calibrated before Jul 13 14:51:18 i charged it until i got the valid discharge flag enabled Jul 13 14:51:21 and left it to die Jul 13 14:51:31 it died in some weeks but nothing changed Jul 13 14:51:43 possibly bme shut down system before calibration accomplished Jul 13 14:52:51 anyway the comments and stdout echo's in that script were meant to be sufficiently expanatory. I wonder how to improve that Jul 13 14:53:06 they're fine Jul 13 14:53:25 you can improve that putting a link to the datasheet in a comment :3 Jul 13 14:53:35 k :-) Jul 13 15:07:35 reload please Jul 13 15:12:50 L29Ah: the calibration is doing a max 1/6 adjustment per cycle iirc Jul 13 15:13:08 ~2038 *5/6 Jul 13 15:13:08 1698.333333333333 Jul 13 15:13:21 ~1698 * 5/6 Jul 13 15:13:21 1415 Jul 13 15:13:25 ~2038 * 7/6 Jul 13 15:13:25 2377.666666666667 Jul 13 15:13:31 i thought you meant this ;3 Jul 13 15:13:49 yes, it is limited both direction probably Jul 13 15:14:11 for details check datasheet Jul 13 15:14:31 good explanation .. but i still don't quite get the whole bq24150 & bq27200 thing ... why do we seemingly have two different chips for (i guess) the same thing? the same question is also valid for the related kernel modules ... some have blacklisted bq27x00_battery, some havent .. :( Jul 13 15:14:33 possibly it even was 1/8 Jul 13 15:15:04 Sicelo: bq27200 is meter, bq24150 is charger Jul 13 15:16:00 and the bq27200 kernel module conflicts with BME since it opens a chan to chip exclusively Jul 13 15:17:07 when bme is already started when the bq27200 module gets loaded, it prolly still works. But when you (re)start BME after module been loaded, BME segfaults or craps out in some way Jul 13 15:18:07 ok. makes sense. since i started using the replacement bme, i can no longer use any of the i2c scripts (yours and bq27200.sh) ... and bnf doesn't really give me all the info i like Jul 13 15:18:43 in KP <49(?) it was even worse, somebody hacked the complete I2C kernel driver so it ignores all locks, possibly resulting in hazardous collisions on I2C bus Jul 13 15:20:19 failed to find interesting "6" entries in the both of the datasheets Jul 13 15:20:27 k Jul 13 15:20:35 it should be pretty easy to relace the I2C calls in whatever scropt by cta /sys/foo/bar/bq27200/(capacity|state|whatever) Jul 13 15:21:09 L29Ah: it's a "15% change max" or similar wording Jul 13 15:21:21 in bq27200 DS Jul 13 15:25:14 >> Each new SI value is computed as follows: SINEW = (15/16)*SIOLD + (1/16)*AI<< for example Jul 13 15:29:12 here you are: >> the LMD value is restricted to a maximum LMD learn-down during any single learning discharge of LMD/8.<< Jul 13 15:30:29 so I been wrong 2 times: 1. it's only for learn-DOWN (not up, 7/6) 2. it's a fraction (LMD/8), not a percentage Jul 13 15:33:05 ohmy, lemme reppst that gibberish Jul 13 15:33:32 it should be pretty easy to replace the I2C calls in whatever script by `cat /sys/foo/bar/bq27200/(capacity|state|whatever)` Jul 13 15:34:06 (when you use the bq27k kernel module) Jul 13 15:34:39 pali's bme-replacement has a built-in mode to calibrate, iirc Jul 13 15:35:24 just discharge battery until bme-replacement send event (to mce/dsme) to turn phone off Jul 13 15:39:59 Pali: so connecting device to charger and enabling VBoost should do the trick? Jul 13 15:40:52 I gather it would stay in hostmode/vboost until shutdown, and immediately reboots due to charger attached Jul 13 15:41:09 there is no auto reboot Jul 13 15:41:14 o.O Jul 13 15:41:35 just shutdown when bq detect empty battery Jul 13 15:41:35 you can't stop N900 from booting when VBUS 5V from charger applied Jul 13 15:42:03 > EDVF=3000 Jul 13 15:42:14 i wonder if the device can run on 3V Jul 13 15:43:16 it should Jul 13 15:44:16 also note that for calibration cycle you need EDV1, not EDVF Jul 13 15:46:39 ~2038 *7/8 *7/8 *7/8 *7/8 Jul 13 15:46:39 1194.63818359375 Jul 13 15:46:51 four cycles, toldya Jul 13 15:47:13 ~2038 *7/8 *7/8 *7/8 Jul 13 15:47:13 1365.30078125 Jul 13 15:47:25 hmmm, with a very good battery 3 Jul 13 15:48:57 i need to reach 5A*h :/ Jul 13 15:49:26 https://dump.bitcheese.net/files/avijeju/n900-cutie.jpeg Jul 13 15:49:35 aaah Jul 13 15:49:49 nice ;-D Jul 13 15:50:04 yeah, learning-UP allegedly works in one cycle Jul 13 15:51:03 but I wonder where's the upper limit for bq27200 LMD Jul 13 15:52:59 ~0xffff * 1 Jul 13 15:53:27 ~65535 /20 Jul 13 15:53:27 3276.75 Jul 13 15:53:45 bad luck, 3277mAh Jul 13 15:54:22 oops sorry Jul 13 15:54:52 ~65535 *3.57 /20 Jul 13 15:54:52 11697.997499999999 Jul 13 15:55:43 LMD unit: 3.57Vh ; Divide by Rs in milliohms to convert μV to mA or μVh to mAh. Jul 13 15:56:30 LMD unit: 3.57µVh ; Divide by Rs in milliohms to convert μV to mA or μVh to mAh. Jul 13 16:00:08 so even when it might be signed, it's still 32768 * 3.57 / 20 -> ~5Ah Jul 13 16:03:14 (whether system can run at 3V) TPS65950 datasheet >>2.7 V is the minimum threshold for the battery at which the device will turn OFF<< Jul 13 16:03:57 http://www.google.de/search?q=tps65950+system+shutdown+voltage+threshold Jul 13 16:04:41 :* Jul 13 16:06:26 this however does NOT apply for modem which will bail aut at dunno 3.3V or sth already Jul 13 16:06:33 out* Jul 13 16:08:15 so at end of discharge, expect to see a No-SIM icon or similar indication/note Jul 13 16:37:56 The earphone piece on the n900 is spring contacted and not soldered, right? Jul 13 16:38:54 I hope it's just reactivating/cleaning the contacts to fix the earphone on my n900... Jul 13 16:39:59 most likely its broke Jul 13 16:41:27 I bought my n900 for 50 euros. Did I pay too much of it? It's flawless because it hasn't been used longer than approximately a year. Jul 13 16:41:53 ecc3g: yep Jul 13 16:42:24 veikko, depends, but more important, is it of use to you? Jul 13 16:42:42 it's a fun toy Jul 13 16:42:46 phone part works in any phone Jul 13 16:43:01 but to appreciate n900 you have to be linux freak (a bit) Jul 13 16:43:13 I just installed debian chroot on it Jul 13 16:43:40 do you like to listen to music? Jul 13 16:44:01 propaganda mode enabled Jul 13 16:44:18 propaganda is one way Jul 13 16:44:23 i gather feedback too Jul 13 16:45:04 i bought mine for ~$40, still dunno if it made sense as i'm yet to switch to it from my dying android junk Jul 13 16:45:33 so it just wastes my time on perfectionism :] Jul 13 16:45:42 so the earphone tends to break? how so? Jul 13 16:45:44 hmm Jul 13 16:45:53 KotCzarny: I use it for running emacs, firefox, necroplayer, xterminal, fm radio Jul 13 16:46:02 where can I get another n900 for $50 USD... Jul 13 16:46:14 I should change firefox for something lighter, maybe dillo? Jul 13 16:46:19 veikko: would you like to try the best alternative player? Jul 13 16:46:25 yes sir Jul 13 16:46:25 * ecc3g just bought an android phone in case the n900 cant be resurrected cheaply :( Jul 13 16:47:14 ecc3g: https://www.avito.ru/rossiya/telefony/nokia?q=nokia+n900&s=1 Jul 13 16:47:24 this android phone sucks. I hate android... *sigh* lesser of two evils however... Jul 13 16:47:25 /58 to get usd Jul 13 16:47:34 KotCzarny: what it is? Jul 13 16:49:00 veikko: oscp Jul 13 16:49:13 how it is better Jul 13 16:49:23 http://talk.maemo.org/showthread.php?t=94590 Jul 13 16:49:27 all features here Jul 13 16:49:38 is it stable? Jul 13 16:50:07 hmm, quite a bit, some bugs are still lurking, but overall stable and i use it daily Jul 13 16:52:25 hmm... i should look for a cheap defective one that has some other problem... =\ Jul 13 16:52:49 parameter equalizer would be perfect Jul 13 16:53:06 veikko: if you find me nice and fast algo i can add it Jul 13 16:54:00 not sure I want to import one though...hmm... Jul 13 16:54:22 Should see if I can get the one my coworker has but doesn't want any more...would be nice if I could get it for a song... Jul 13 16:54:41 grab some adroid phone and swap? Jul 13 16:55:27 ecc3g: why don't you buy the earpiece on ebay? Jul 13 16:56:05 KotCzarny: I use TAP Equalizer (LADSPA) in linux. could you use its code? Jul 13 16:56:23 hmm, if i can rip it out, ill look into it Jul 13 17:03:13 (( The earphone piece)) yep Jul 13 17:06:35 ecc3g: it's a known "issue" that the spring contacts of earphone jack might need cleaning to make proper contact to PCB pads Jul 13 17:07:59 using unvarnished carton to wipe off any oxids and dirt usually helps a lot Jul 13 17:11:51 what maemo used for now?? Jul 13 17:12:04 so old, surprised to see people still in here... Jul 13 17:12:19 my n810 has only functioned as an alarm clock for like the last 4 years,, lol Jul 13 17:12:20 i use my n900 continuously since 2009 Jul 13 17:12:34 esaym153: install oscp on it, it will have new life Jul 13 17:14:08 esaym153: n900 users there mostly Jul 13 17:14:20 but, n8x0 is still alive too Jul 13 17:14:36 * esaym153 googles oscp Jul 13 17:14:48 google oscp player Jul 13 17:25:28 alas no, minimum the coworker would trade is an apple. Jul 13 17:25:45 (if it was a trade...) Jul 13 17:26:21 ecc3g: those iphones 3/4 are quite cheap nowadays Jul 13 17:26:32 i think he already has a 5 Jul 13 17:27:55 maybe he wants to have another for kids Jul 13 17:28:02 has no kids Jul 13 17:28:25 maybe he wants to have kids Jul 13 17:28:26 so hence hopefully he'll give up the n900 for almost nothing... Jul 13 17:29:48 ((equalizer)) already exists Jul 13 17:31:52 http://my-maemo.com/software/applications.php?fldAuto=1475&faq=34 Jul 13 17:35:53 doc, does it work for any pulseaudio stream or only mafw ones? Jul 13 17:42:40 it's mafw aiui Jul 13 17:49:01 so mafw/gst only Jul 13 17:53:02 no, just mafw (excluding gst) Jul 13 17:53:40 it replaces the stock mafw gst backend ("renderer") and adds eq support Jul 13 17:54:55 veikko: dillo is great, but there is no JS support Jul 13 17:55:37 and css support isnt "complete" either Jul 13 17:56:41 i need an "optimized for dillo" button Jul 13 17:58:39 <3 Jul 13 17:59:26 http://www.dillo.org/css_compat/index.html Jul 13 17:59:32 and ... that's just for css1 iirc Jul 13 18:45:43 bencoh: love the design. too bad that the modern web sucks Jul 13 18:49:44 should probably avoid the sites anyway that dillo can't render with the old phone Jul 13 18:50:12 hmm ... Jul 13 18:50:42 that means dropping support for mostly everything useful "on the go" Jul 13 18:50:59 but yeah, modern web sucks Jul 13 18:58:36 modern web = "I want your clients to be a VM so I can run what I want on your computer" ... Jul 13 18:59:06 dont think we'll ever see "here's the data, you can show it however you want" Jul 13 18:59:28 anymore, that is Jul 13 19:00:11 well, lots of web services provide access to theirs api Jul 13 19:00:25 but you're not running it on your computer Jul 13 19:00:34 running what? Jul 13 19:00:41 code that access their API Jul 13 19:00:59 i listen music on vk.com and searching google for youtube videos, all without a browser Jul 13 19:01:12 it's the second server that runs that code, then sends you the stuff to run on your computer Jul 13 19:01:26 what Jul 13 19:01:58 that's the "ideal world" these "content providers" want Jul 13 19:02:04 what Jul 13 19:02:09 so you cant pirate their media Jul 13 19:03:11 use the media how they want it to be used - consume only - never shared... That will cost you. Jul 13 19:03:34 what? Jul 13 19:04:02 i saw a slogan some time ago: 'own nothing, have everything' Jul 13 19:04:18 (it was for cable or something, but idea is striking) Jul 13 19:04:26 ... if you pay for it Jul 13 19:04:35 sure, that's the fine print Jul 13 19:04:42 i just don't see the problem you're talking about Jul 13 19:04:47 devil's in the details Jul 13 19:05:02 l29ah: closed firmware blobs is just part of it Jul 13 19:05:08 paid only once for an intangible stuff in my whole lifetime Jul 13 19:05:19 shared it in torrents just after this Jul 13 19:05:39 http://l29ah.blasux.ru/music/fav/ is up for ~6 years, no problems Jul 13 19:05:51 you live in a country where intelectual property isnt of such value for the sharks yet Jul 13 19:06:09 heh... yeah .ru and .cn ... Jul 13 19:06:21 in .pl it was the same 5-20 years ago Jul 13 19:06:32 anyway i won't lose anything worthy Jul 13 19:07:08 me, i'm in the world of lockdown. Fortunately I care not for most of the media out there today. Jul 13 19:07:11 er, exceph sci-hub and libgen Jul 13 19:07:45 is tor banned in .pl? Jul 13 19:07:51 not yet Jul 13 19:08:11 and piracy still thrives Jul 13 19:08:22 tor = circumvention, has nothing to do with the ideals of the content creators/providers Jul 13 19:08:26 but there are movements from the sharks Jul 13 19:08:58 ie. they start raiding people who pirate things for money Jul 13 19:09:08 best you can hope for is reject the media. Jul 13 19:09:21 don't want the stuff they sell. Jul 13 19:09:46 many do want fastfood for the brain Jul 13 19:10:00 yes that's very unfortunate Jul 13 19:55:00 Am I doing something wrong here but I just cannot figure out how to change setting values in MicroB when at about:config. I see the lines I would like to change but theres the problem: how do I modify these values? When I press for long on the value I would like to change nothing opens - or is there some trick that I'm not aware of? Jul 13 19:56:17 try pressing enter? Jul 13 19:57:19 What I'm trying to do is just to enhance the security of the browser by disabling insecure SSL/TLS methods. I also tried modifying via user.js file, rebooted n900 and checked that those values had updated to prefs.js but still the device is using those insecure methods. Jul 13 19:58:46 KotCzarny: Thank you. I am an idiot for not trying out something so bloody obvious :D Jul 13 20:02:18 nah, its just that about:config is kind of hack in microb Jul 13 20:02:50 you also need to be aware that changes you make don't "stick" .. close the UI and they are gone Jul 13 20:03:27 So editing via the user.js is really the only way to modify the settings? Jul 13 20:03:27 duh? Jul 13 20:03:45 isn't there a "save" button somewhere? Jul 13 20:04:04 hmm Jul 13 20:04:16 i think setting images load to disable was a sticky there Jul 13 20:04:27 i might be wrong tho Jul 13 20:04:37 i mean in about:config Jul 13 20:04:42 Could it be that it saves those changes directly to prefs.js which gets "reset" when the daemon restarts? Jul 13 20:04:51 exactly Jul 13 20:05:35 pffft should never try the easy way to do anything :D Jul 13 20:06:08 http://talk.maemo.org/showpost.php?p=1237403&postcount=1 ... Jul 13 20:08:02 I'm just curious but does anyone happen to know the reason for why it resets those settings everytime it restarts? Jul 13 20:08:50 dang, I thought there was a "save" button at the bottom Jul 13 20:11:44 anyway I'm pretty sure my edits were sticky Jul 13 20:16:05 prolly it's similar to xchat, which doesn't write back settings until you quit it regularly Jul 13 20:17:59 so either a normal closing of browser instance or a killall browserd might do Jul 13 20:18:27 (no killall -9 !) Jul 13 20:19:39 maybe -USR ? Jul 13 20:33:32 just accept that maemo is unsafe Jul 13 20:33:41 we only have until december, anyway Jul 13 20:34:54 Yes it is to quite some extent. N900 is my go to "linux" toy with which I experiment when bored. Jul 13 20:35:27 Of course there would be those Raspberry Pis lying around but.. I like this one better :D Jul 13 21:47:17 kerio: what's in December? Jul 13 21:51:07 Luke-Jr: openssl 0.9.8 EOL Jul 13 22:04:14 and what happen after EOL? Jul 13 22:04:28 all bugs disappear? :D Jul 13 22:04:50 lol Jul 13 22:05:07 then we have to upgrade to the infamous 1.0.1 with all its regular exploits <.< Jul 13 22:07:45 after End Of Life, bugs die Jul 13 22:09:02 I mean, it's not like they become butterflies Jul 13 23:00:34 hmm... my n900 is eating more power when idle than usual... :( Jul 13 23:00:58 hmm... Jul 13 23:01:22 i wonder if this new microsd card is leaking more power than the old one... Jul 13 23:24:59 zzztop Jul 13 23:27:48 it's hardly the uSD card, rather indexer/thumbnailer going busy indexing the card's content. Particularly when something is "fishy" with the content from indexing point of view Jul 13 23:31:52 card is empty except for swap... Reboot fixed it for now as far as I can tell, but it likely will show up again... Jul 13 23:32:21 not sure what I installed lately, because I didn't... Jul 14 00:45:13 I'd consider stopping to blame trackerd indexer when you come up with a zzztop output showing it's some other process Jul 14 00:46:05 let it sample over a 1800s Jul 14 00:48:54 theoretically it could be your swap size: when you created several (dozen?) GB of swap, the management tables of swapper process eat all RAM, driving system into a swap hell Jul 14 00:51:12 this is a theory, I dunno if it stands any close evaluation **** ENDING LOGGING AT Tue Jul 14 02:59:58 2015