**** BEGIN LOGGING AT Sat Jul 21 02:59:58 2012 Jul 21 06:26:17 kerio: ping Jul 21 07:37:31 ~rootsh is rootsh gives you "sudo gainroot", to get root privs but user environment (similar as "su"), and "root", which gives you root privs root env (similar to "su -") Jul 21 07:37:32 ...but rootsh is already something else... Jul 21 07:37:51 ~rootsh Jul 21 07:37:52 it has been said that rootsh is an easy way to get root and it's found here: http://maemo.org/downloads/product/OS2008/rootsh/, or http://maemo.org/downloads/product/Maemo5/rootsh/ Jul 21 07:38:45 ~rootsh is also rootsh gives you "sudo gainroot", to get root privs but user environment (similar as "su"), and "root", which gives you root privs root env (similar to "su -") Jul 21 07:38:45 okay, DocScrutinizer05 Jul 21 07:43:44 jon_y: thumb2 is an instruction set, the cortex A8 could run stuff with that instruction set, but there are some hardware problems that have to be worked around, in the kernel and in the linker Jul 21 07:44:10 if you have the workaround in the kernel, and your linker is new enough, you can run thumb2 binaries with no problem Jul 21 07:45:14 (except that the workaround for it requires doing something that causes a bit of performance loss) Jul 21 07:48:44 and no, it's definitely no solution for crammed rootfs Jul 21 08:09:51 kerio: what kind of loss? Jul 21 08:12:43 jon_y, "and also flushes the branch target cache at every context switch." Jul 21 08:13:26 it doesn't seem like a big performance hit, and the reduced size of thumb binaries weighs it out Jul 21 08:14:57 ok, don't thread on thumb2 Jul 21 08:15:06 got it :) Jul 21 08:19:36 DocScrutinizer05: optify *all* the things! _ò/ Jul 21 08:20:40 kerio: eh? Jul 21 08:21:08 cool, can I have optified Perl too? :) Jul 21 08:21:21 5.16 please Jul 21 08:22:01 well, I had ln -sv /opt/stuff /usr/local now Jul 21 08:22:06 works well enough Jul 21 08:22:50 sometimes, apps are too clever for their own good and this fails Jul 21 08:24:12 python has a ton of bindmounts for that Jul 21 08:26:01 yeah >:-( Jul 21 08:26:12 spamming my mount Jul 21 08:26:13 bindmounts are grrrrrrreat! Jul 21 08:26:59 it's about time to get a smarter ,ount command, that filters output Jul 21 08:27:19 mount | grep -v bind Jul 21 08:27:23 :) Jul 21 08:27:25 kinda Jul 21 08:38:33 is it possible to find out how much warranty is left on my N900? Jul 21 08:38:52 i have lost the proof of purchase :( Jul 21 08:46:36 then it's getting difficult Jul 21 08:46:40 trumee: unless you're prepared for a battle (to get a replacement or more likely demand a refund) the warranty is pretty worthless. Nokia doesn't care about N900 customers. :( That said, I've never needed proof of purchase at a Nokia Care Centre, as long as your phone isn't reported stolen you should be okay. Jul 21 08:47:11 Nokia has a webpage to check based on serial#, but I doubt that is accurate for devices sold late Jul 21 08:47:35 also what zeq said Jul 21 08:50:50 trumee: generally I think there's nokia warranty and there's legal warranty. Both are unrelated. You claim your legal warranty at your shop/seller, based on whatever the laws in your country. Independently Nokia offered a 2(?) year manufacturer guarantee, which is basically void by now I'd say Jul 21 08:52:07 sellers tend to send their customers to manufacturer for claiming warranty. They MUSTN'T Jul 21 08:52:15 btw, anybody know where the sdcard automounter script is located at? Jul 21 08:52:24 yes Jul 21 08:52:29 I'd like to use ext3 card Jul 21 08:52:34 get CSSU Jul 21 08:52:53 how is it related to cssu? Jul 21 08:53:08 iirc atm it wants vfat and only vfat mmc1 Jul 21 08:53:13 well, CSSU has a vastly improved "automounter" Jul 21 08:53:29 that mounts wirtually *all* Jul 21 08:53:48 I have cssu enabled, how do I find out if the cssu automounter is working? Jul 21 08:54:39 ah, I see my alternate easy debian partition Jul 21 08:54:43 zeq: is it expensive to replace the usb port at Nokia Care centre? Jul 21 08:54:45 looks like its already working Jul 21 08:55:00 DocScrutinizer05: thanks so much for working on cssu Jul 21 08:55:18 :-) Jul 21 08:55:24 trumee: that's something they won't do. There are people who will, or you can DIY Jul 21 08:55:36 i am a bit wary of walking into a local phone shop who has never seen an N900 before Jul 21 08:56:12 trumee: get photo printouts of Jul 21 08:56:15 ~usbfix Jul 21 08:56:16 i heard usbfix is http://talk.maemo.org/showthread.php?t=75920 - and **NEVER** use epoxy (unless you want to seal your device for underwater) Jul 21 08:56:43 DocScrutinizer05: i dont have the tools to replace the usb port Jul 21 08:56:45 trumee: it's good to be wary. unless you know they're good, I wouldn't trust a random local phone repairer. Jul 21 08:57:07 my phone only charges using the computer these days. Jul 21 08:57:10 but you have a printer? then take a paper with the photos to your local repair shoppee Jul 21 08:57:40 trumee: that's highly unlikely any USB receptacle problem Jul 21 08:57:58 DocScrutinizer05: i have reflashed it, it is not a software issue Jul 21 08:58:02 DocScrutinizer05: btw, a side effect of all visible partitions is that the tracker will look through ALL of them Jul 21 08:58:25 kind of painful if you have a large easy debian install Jul 21 08:58:25 DocScrutinizer05: o/ Jul 21 08:58:31 jon_y: sounds like a flaw in tracker conf. We might need to address it in CSSU Jul 21 08:58:52 well, iirc there is a built in list of places to track already Jul 21 08:58:59 jon_y: but you can tweak tracker settings easily yourself Jul 21 08:59:06 so I just added "/" to the exclude list and redo the index Jul 21 08:59:16 turns out it works properly Jul 21 08:59:16 kinda Jul 21 08:59:17 jon_y: Install the trackercfg application Jul 21 08:59:29 I will do that, thanks Jul 21 08:59:46 jon_y: It makes changing tracker ignore folders etc nice and easy Jul 21 08:59:54 you already got cssu, iirc it also has a fixed trackerd which actually _obeys_ the blacklists ;-) Jul 21 09:00:02 stock trackerd has a bug there Jul 21 09:00:18 I was wondering why an extra entry showed up in the file browser util after a cssu update :) Jul 21 09:00:32 * jon_y just reflashed the n900 Jul 21 09:00:32 also, contrary to popular belief do not turn on 'low memory mode' or throttling. Jul 21 09:00:59 err, not? Jul 21 09:01:02 Let tracker do it's thing quickly and be done with it. Jul 21 09:01:07 throttling seems fine here Jul 21 09:01:33 though I forbid tracker to index when on battery Jul 21 09:01:36 Throttling DOES work, however it takes longer for tracker to finish. Jul 21 09:02:16 feck tracker Jul 21 09:03:08 I want to install openmode kernel on my n950 Jul 21 09:03:35 is it possible to get imei via xterm? Jul 21 09:03:35 Do I only flash the kernel or do I need to flash the emmc etc as well? Jul 21 09:03:38 a concept designed for >2GHz >1GB RAM desktop PCs, brought to embedded as a fake alternative filesystem Jul 21 09:03:51 trumee: yes Jul 21 09:04:03 trumee: e.g with pnatd Jul 21 09:04:10 trumee: Yes, there is a dbus call. SOmething else as well, cannot remember the name. Jul 21 09:04:20 also, any util to disable the update apt db on internet connection? Jul 21 09:04:37 its annoying to find out you can't use apt right after connecting to wifi Jul 21 09:04:41 and sth like AT+CMGR or whatever the cmd Jul 21 09:05:06 DocScrutinizer05: dbus-send --system --type=method_call --print-reply --dest=com.nokia.phone.SIM /com/nokia/phone/SIM Phone.Sim.get_imsi ? Jul 21 09:05:18 sounds good Jul 21 09:05:30 jon_y: gconftool -s --type int /apps/hildon/update-notifier/check_interval 9999999 Jul 21 09:05:40 cool, thanks Jul 21 09:05:48 DocScrutinizer05: nope this one, dbus-send --system --print-reply --type=method_call --dest=com.nokia.phone.SIM /com/nokia/phone/SIM/security Phone.Sim.Security.get_imei Jul 21 09:06:15 sets check interval to like 30 years or something. Jul 21 09:06:32 vi_: that's insane Jul 21 09:06:40 you don't wanna do that Jul 21 09:07:04 though 9999999 isn't exactly 30 years and mostly fine Jul 21 09:07:56 I added some caveats to the wiki entry in "customizing" Jul 21 09:08:29 I'll run update when I want to :) Jul 21 09:08:43 ERROR: APE algorithm has to be provided to flash all the subimages Jul 21 09:08:43 which boils down to "1 year in seconds is fine, using MAXINT is going to mess up things" Jul 21 09:09:03 DocScrutinizer05: it was hyperbole. It is not really 30 years. Jul 21 09:09:08 btw, updating cssu should still be done in the Application Manager? Jul 21 09:09:15 jon_y: YES! Jul 21 09:09:25 despite what some people say Jul 21 09:09:29 good yes, I'll need to remember that Jul 21 09:09:51 apt-get for everything else Jul 21 09:09:55 CSSU does NOT support, and *deprecates*, fapman Jul 21 09:10:39 the other application manager? Jul 21 09:10:46 Fast Application Manager? Jul 21 09:10:54 using apt-get should be mostly safe on CSSU, as long as you don't do autoremove Jul 21 09:11:11 and better stay away from dist-upgrade (and maybe also upgrade) Jul 21 09:11:32 DocScrutinizer05: can you point me to somthing that will tell me how to flash open mode kernel on my n950? Jul 21 09:11:35 what will go wrong with upgrade and dist-upgrade? Jul 21 09:11:37 jon_y: commonly (formerly) known as fapman Jul 21 09:11:56 ok Jul 21 09:12:03 vi_: a mini-howto on my server Jul 21 09:12:11 vi_: I want one too :| Jul 21 09:12:12 vi_: ask in #harmattan Jul 21 09:12:22 ~openmode Jul 21 09:12:29 meh Jul 21 09:12:48 ~listkeys openmode Jul 21 09:12:49 Factoid search of 'openmode' by key returned no results. Jul 21 09:12:59 ~listvalues openmode Jul 21 09:13:00 Factoid search of 'openmode' by value (1): aegis-no-thanks. Jul 21 09:13:17 ~aegis-no-thanks Jul 21 09:13:17 well, aegis-no-thanks is http://maemo.cloud-7.de/HARM/N9/openmode_kernel_PR1.1/, or http://maemo.cloud-7.de/HARM/N9/1.2/openmode-kernel/ Jul 21 09:14:45 I have the kernel, however I cannot seem to flash the sucker. Jul 21 09:15:01 flasher -k ./zImage-2.6.32.39-dfl61-20113701 --flash-only=kernel -f -R Jul 21 09:15:05 does not work. Jul 21 09:15:17 it says: Jul 21 09:15:45 DocScrutinizer05: meh, tracker behaves decently now that i told it to only index a couple of directories Jul 21 09:16:22 also, the automounter means that i can't easily unmount from h-e-n :( Jul 21 09:16:34 this will stop as soon as you got 20GB of mp3 on your MyDocs ;-P Jul 21 09:16:42 i've got 16 Jul 21 09:16:54 uggh, http://www.nokiawarranty.co.uk/index.html quoted 75 pounds to replace usb port Jul 21 09:16:59 mixed mp3/aac in m4a Jul 21 09:17:34 trumee: that sounds quite reasonable Jul 21 09:17:55 I doubt they'll keep this offer though Jul 21 09:18:05 most likely obsolete Jul 21 09:18:16 if not, please let us all know Jul 21 09:19:00 trumee: but again, if your device is charging from PC, then your USB *IS* ok Jul 21 09:19:16 DocScrutinizer05: why does it fail from the wall charger? Jul 21 09:19:28 your wallcharger defect? Jul 21 09:19:32 DocScrutinizer05: ERROR: APE algorithm has to be provided to flash all the subimages Jul 21 09:19:40 ^DAFUQ this means? Jul 21 09:19:42 DocScrutinizer05: is there a way to forcibly enable all buttons in h-e-n? it would be useful if i'm using an external power source Jul 21 09:19:54 vi_: ask in #harmattan, known problem with newer NOLOs Jul 21 09:19:57 DocScrutinizer05: no it charges my second N900 fine Jul 21 09:20:16 DocScrutinizer05: but they are a flock of dicks. Jul 21 09:20:35 who's "they"? Jul 21 09:21:09 kerio: no Jul 21 09:21:18 kerio: use scripts & xterm Jul 21 09:22:10 damn cant find any website for a shop in london Jul 21 09:22:41 DocScrutinizer05: k :( Jul 21 09:22:48 hi Jul 21 09:23:18 ~hen Jul 21 09:23:18 do ~hostmode and ~factinfo hostmode, this is your maemo council :-/ Jul 21 09:23:26 k Jul 21 09:23:48 infobot: oh you Jul 21 09:23:49 extra, extra, read all about it, infobot is me, http://infobot.svn.sourceforge.net/svnroot/infobot/ // The original infobot is no more, it has been replaced by the latest version of blootbot, accompanied by a rename back to infobot // apt/ibot/infobot/purl is now an infobot run by TimRiker Jul 21 09:25:03 ~tell kerio about hostmode-powered Jul 21 09:25:50 ~hostmode-powered Jul 21 09:25:50 i heard hostmode-powered is http://talk.maemo.org/showthread.php?p=921203#post921203 Jul 21 09:27:04 ~charger Jul 21 09:27:16 DocScrutinizer05: the thing is, it doesn't really work for stuff that requires more than vboost can provide to work Jul 21 09:27:33 kerio: ping Jul 21 09:27:38 pong! Jul 21 09:28:13 You were asking about hildon-welcome before? Jul 21 09:28:17 http://repository.maemo.org/pool/maemo5.0/free/h/hildon-welcome/ looks like the code for it Jul 21 09:28:49 DocScrutinizer05: for those, i just dick around randomly until it works :) Jul 21 09:30:24 kerio: then connect the Y-cable to N900 *after* you've hit "VBoost" Jul 21 09:30:31 although I think you were actually thinking of osso-startup-wizard as the one that runs the first time you power up the phone :) Jul 21 09:31:53 Anyone know who has been doing work on alternate kernels for Maemo5 and/or on improving the WiFi driver? Jul 21 09:40:05 Is there no way to flash this thing without using that bloody navifirm program to steal firmwares from nokia servers. Jul 21 09:40:08 ? Jul 21 09:42:55 jonwil: alternative kernels: mainly Kernel-Power, now pali's domain. WiFi drivers: lxp? Jul 21 09:43:38 vi_: no, but still that's #harmattan topic Jul 21 09:43:52 ok Jul 21 09:44:43 vi_: since otherwise my answer should've been "What? use flasher-3.5, see ~flashing!" Jul 21 09:47:45 DocScrutinizer05: w00t! I have PERMANENTLY invalidated my warrenty! Jul 21 09:48:08 you've been warned, haven't you? Jul 21 09:49:10 DocScrutinizer05: oooh. yes. Jul 21 09:49:44 vi_: still a #harmattan topic, but I think it's reversible nevertheless Jul 21 09:53:31 ~seen lxp Jul 21 09:53:33 lxp is currently on #maemo (19h 57m 53s), last said: 'i am personally using a version of the driver based on a little bit newer wireless-testing with some rewritten patches i made after submitting the patches to the kernel mailinglist, but it doesn't contain the injection patches'. Jul 21 10:01:21 jonwil: You know about lxp right? Jul 21 10:01:33 no Jul 21 10:02:19 They say he only listens to bluegrass music... Jul 21 10:02:37 jonwil: He wrote the injection drivers for the N900. Jul 21 10:02:51 ok Jul 21 10:03:16 jonwil: He was hired to do it by the original backtrack linux on n900 people. Jul 21 10:03:20 They never paid him. Jul 21 10:03:41 So he released his driver to the maemo community. Jul 21 10:05:41 ok Jul 21 10:18:57 ok, basically I am trying to find out if anyone is actually doing any improvements to the N900 WiFi driver (e.g. for CSSU or otherwise) so I can see about contributing some info to that Jul 21 10:27:43 jonwil: the only possibly modified wifi driver I've seen is in kernel power, that packet injection driver Jul 21 10:28:44 technically it's not in kernel-power, it's an out of tree module Jul 21 10:28:58 iirc powersave is the blocker for merging Jul 21 10:31:15 ok, it sounds like no-one is really doing any meaningful work on the driver then Jul 21 10:43:05 jonwil: That about sums it up. Jul 21 10:43:28 in that case I wont bother looking into wl1251-cal Jul 21 10:43:44 jonwil: lxp, while more than happy to point you in the right direction is thoroughly sick of the whole thing. Jul 21 10:44:01 zeq: did you succeed to build the toolchain? Jul 21 10:44:01 I have no kernel skills :P Jul 21 10:44:25 jonwil: want to help with BME stuff? Jul 21 10:44:39 reverse engineering you mean? Jul 21 10:44:45 no, coding Jul 21 10:45:07 I think you already did a wonderful RE job on BME Jul 21 10:45:28 except we still dont have the math to correctly calculate some figures like temprature Jul 21 10:45:34 jonwil: No kernel skills? Are you shitting me? Jul 21 10:45:44 I have never done Linux kernel dev Jul 21 10:45:52 beyond compiling the kernel on my Gentoo box Jul 21 10:46:06 jonwil: I will do it when we have the other stuff (i.e. charging) in place Jul 21 10:46:13 :) Jul 21 10:46:21 what bits of BME is coding needed on? Jul 21 10:46:33 did you check on my repo? Jul 21 10:46:39 link? Jul 21 10:47:03 just a minute Jul 21 10:47:10 ok Jul 21 10:47:35 https://gitorious.org/hald-addon-bme Jul 21 10:47:49 https://gitorious.org/libbmeipc Jul 21 10:47:56 jonwil: ^^^ Jul 21 10:48:15 ok Jul 21 10:48:36 it needs charging support (the new driver in kp51) and temp reading (in libbmeipc) Jul 21 10:48:59 ok, well I cant help with the charging bit Jul 21 10:49:12 or the temprature bit Jul 21 10:49:15 so I cant really help Jul 21 10:50:07 jonwil: you can, it just needs reading from sysfs, the driver handles all the other stuff Jul 21 10:50:17 AFAIK Jul 21 10:50:20 Might go back into the maw that is the connectivity UI layer :) Jul 21 10:50:26 i.e. widgets, control panels etc Jul 21 10:50:44 aah, where is Pali :( Jul 21 10:51:18 DocScrutinizer05: kernel charger driver handles all the stuff needed for charging, ain't? Jul 21 10:51:30 yeah I wanna talk to pali too :P Jul 21 10:51:58 so userland only reads the current status and reports it through hal/dbus? Jul 21 10:52:00 I think I might consider doing further work on my wlan priority stuff Jul 21 10:54:08 freemangordon: basically yes Jul 21 10:54:08 freemangordon: yes. I've built linaro-4.6-2012.7 and linaro-4.7-2012.7. The 4.6 toolchain sucessfully builds qt, 4.7 is still broken, but the gcc bug report now has it narrowed down to a single commit. Jul 21 10:55:32 jonwil: see DocScrutinizer05's answer, so nothing more but reading the sysfs values Jul 21 10:56:08 zeq: a pity Jul 21 10:56:23 http://gcc.gnu.org/bugzilla/show_bug.cgi?id=53135 Jul 21 10:56:24 04Bug 53135: was not found. Jul 21 10:57:02 zeq: I am going to build the latest fennec, what was that switch you wanr me about? something about webrt. Jul 21 10:57:11 hi, all Jul 21 10:57:33 not really interested in BME work, I think I might consider doing further investigations into my wlan priority stuff :) Jul 21 10:57:38 --disable-webrtc Jul 21 10:57:46 I have problem of the portrait mode, it don't work on my N900 Jul 21 10:58:06 jonwil: ok :) Jul 21 10:58:50 jiero: thats because the n900 is a landscape device. Jul 21 10:58:51 zeq: and why is that? Jul 21 10:59:03 and WTF webrtc is? Jul 21 10:59:04 freemangordon: they've enabled webrtc by default, whereas it was disabled previously. Jul 21 10:59:20 ~webrtc Jul 21 10:59:20 it just hasn't been ported yet Jul 21 10:59:51 it's a real-time communications layer. like SIP over the web. Jul 21 10:59:57 jiero: because 1) portrait sucks Jul 21 10:59:59 vi_ I used portrait mode after instatlled CSSU . but.now it is lost, Jul 21 11:00:04 zeq: ok Jul 21 11:00:09 kerio: so it was removed? Jul 21 11:00:12 no, it's still there Jul 21 11:00:18 you have to enable it in transitions.ini Jul 21 11:00:24 ok Jul 21 11:00:33 "cssu features" is a neat gui for stuff like that Jul 21 11:01:44 just wish more people had knowledge of/interest in icd Jul 21 11:02:02 jonwil: mostly people just want icd to go away, i think Jul 21 11:02:05 * freemangordon feels in the same way re BME :) Jul 21 11:02:59 ok, so what exactly are the end-user benefits of replacing BME anyway? Jul 21 11:03:10 jonwil: easier USB hostmode Jul 21 11:03:10 i.e. why do it other than "because we can" or "because its open" Jul 21 11:03:19 ok Jul 21 11:03:32 if you don't rely on BME for battery indicator, you can get one while doing freaky stuff that make USB engineers cry in their sleep Jul 21 11:03:36 correct capacity measurement Jul 21 11:03:38 *a working one Jul 21 11:03:43 freemangordon: "correct", you mean Jul 21 11:04:00 I mean "the real values" Jul 21 11:04:20 so yeah, fuck bme Jul 21 11:05:18 jonwil: bme spams the i2c bus and does not use it properly. Jul 21 11:05:38 jonwil: It is a total kludge of a program. Jul 21 11:06:28 wellll Jul 21 11:06:45 bme IS an abomination Jul 21 11:07:11 but it's not correct to say it spams the i2c-bus and does not use it properly Jul 21 11:08:41 it's rather like i2c device in itself has some compatibiltiy issues by design, when it comes to kernel modules like bq27x00 Jul 21 11:09:00 and bme is using i2c device Jul 21 11:09:20 since there never been stock kernel modzles for bq27200 and bq24150 Jul 21 11:09:25 as for getting rid of icd, it should be possible IF people wanted to put some effort into it. Firstly one would need to start by replacing the WiFi bits using e.g. icd-network-wpa and the info I posted and stuff. Then one would need to replace the other plugins (i.e. GPRS plugin) then after that one can replace ICD2 itself. Biggest issue is dealing with osso-wlan-security :) Jul 21 11:09:48 understanding what that mess of a daemon (eapd) does Jul 21 11:09:56 or rather, what it talks to Jul 21 11:13:32 Replacing ICD2 wouldn't be hard if people cared enough Jul 21 11:13:37 people with the coding skills to help :) Jul 21 11:14:30 I think it would be a hoot to replace browserd with somthing lighter. Jul 21 11:14:44 or just update it Jul 21 11:14:55 there's been plenty of gecko versions Jul 21 11:15:06 freemangordon: You mentioned you have compiled microb-engine for thumb2. Jul 21 11:15:09 biggest issue with replacing gecko is breaking flash, maps and stuff Jul 21 11:15:37 i.e. if you use a newer gecko you need to make sure its still ABI compatible Jul 21 11:15:37 if I apt-get it, will that pull ALL the dependencies or will I still have to copy libc++ or somthing to the appropriate place? Jul 21 11:16:11 kerio: That is not the point. I do not need the whole gecko engine sitting in RAM on the offchance I might want to read my txt messages. Jul 21 11:20:08 vi_: and it is in thumb repo, latest cssu-thumb update includes it Jul 21 11:20:42 vi_: everything is in place Jul 21 11:20:59 i.e. libs, dependencies, etc Jul 21 11:21:27 and later i will push fennec in the repo, but you MUST have latest cssu-thumb installed Jul 21 11:21:40 vi_: ^^^ Jul 21 11:22:03 freemangordon: Ok Jul 21 11:22:08 imma gonna doit. Jul 21 11:22:31 what? I am not native inglish speaker as you may know? Jul 21 11:22:43 I am going to do it. Jul 21 11:23:23 freemangordon: I'm currently building fennec with linaro-4.7, and patched qt-4.7 Jul 21 11:24:34 zeq: patched qt-4.7? Jul 21 11:24:36 did I ask previously about whether we could upgrade to qt-4.8? Jul 21 11:24:54 qt-4.7 needs a few fixes for gcc-4.7 Jul 21 11:25:13 nothing abi breaking Jul 21 11:25:21 zeq: hmm, what fixes? Jul 21 11:25:37 hold on Jul 21 11:27:47 http://www.snewbury.org.uk/maemo/gcc-4.7.diff Jul 21 11:28:26 aah, ok Jul 21 11:28:42 * freemangordon is going to have some lunch Jul 21 11:29:45 When I type in apt-get upgrade I see loads of stuff with -203 on the end. DO I want to upgrade those? Jul 21 11:30:09 like 20 packages. Jul 21 11:30:21 Hi Jul 21 11:30:57 My N900 is caught in infinite restart loop. Jul 21 11:31:04 * DocScrutinizer05 welcomes Kiborg and his bootloop Jul 21 11:31:11 Yaay! Jul 21 11:31:35 Kiborg: 1. DId you install speed/batterypatch? Jul 21 11:31:37 it's actually flashing day Jul 21 11:31:50 Kiborg: 2. Did you make a backup with backupmenu? Jul 21 11:32:10 Kiborg: 3. What was the last thing you did in order to cause this reboot loop. Jul 21 11:32:32 he downloaded huge stuff (some p0rn ;-P ) from youtoube Jul 21 11:32:48 Kiborg: The answer is disable watchdogs, restore backup from backup menu. Jul 21 11:32:50 seems to me it might have crammed / Jul 21 11:33:01 or /var/* Jul 21 11:33:05 or sth like that Jul 21 11:33:14 then re-enable watchdogs. Jul 21 11:33:19 lots of speculation! ;) Jul 21 11:33:29 reflash rootfs, the end. Jul 21 11:33:41 zeq: s/speculation/experiance Jul 21 11:33:44 yes I have speed patch as sugessted by cssu but i did that half a year ago. the last thing I did was download a clip uising CuteTube-QML, that I am using for a year probably. Jul 21 11:33:56 Kiborg: FFS. Jul 21 11:34:00 COMPLETE REFLASH. Jul 21 11:34:00 WUT??? Jul 21 11:34:00 meh, probably something in the FSes corrupted Jul 21 11:34:10 speed patch as suggested by cssu???????? Jul 21 11:34:49 most if not all script-based *-patches are highly discouraged Jul 21 11:34:52 I don't even know friggin speedpatch Jul 21 11:34:59 DocScrutinizer05: lol. Jul 21 11:35:07 DocScrutinizer05: It is great. Jul 21 11:35:36 is that the hoax of this script kiddie Jul 21 11:35:37 ? Jul 21 11:35:52 it changes the cgroup settings Jul 21 11:35:52 (which script kiddie?) Jul 21 11:36:03 As I recall Jul 21 11:36:10 There are 2 scripts. Speedpatch and batterypatch. Jul 21 11:36:24 yeah, both BS par excellance I heard Jul 21 11:36:49 batterypatch is just a cpufreq policy daemon Jul 21 11:36:53 and NO, CSSU DID NOT recommend them! Jul 21 11:38:13 zeq: whatever it is, iirc it can only be some random piece of script picked up and klufged together by somebody with no clue Jul 21 11:38:57 initial stages of that *-patch development were epic. Doing *obviously* nonsensical stuff Jul 21 11:39:13 and arguing about basic knowledge Jul 21 11:39:26 never ever looked at them since then Jul 21 11:39:49 batterypatch is just a set of dbus scripts Jul 21 11:40:17 * NIN101 never understood why you would something like speedpatch in the first place. Everything just works here and isn't sluggish etc. Jul 21 11:40:40 NIN101, placebo effect Jul 21 11:40:47 yeah. Jul 21 11:40:52 yeah, and doing messy shit to sqlite database isn't helping on it Jul 21 11:40:57 of course a N900 is nice and snappy after a reboot, no matter what crap you installed Jul 21 11:41:05 after 5 days, that's a different story Jul 21 11:41:13 not here Jul 21 11:41:17 I reboot every day. Jul 21 11:41:35 DocScrutinizer05, that is the challenge Jul 21 11:42:01 t900:~# uptime Jul 21 11:42:03 13:50:51 up 112 days, 14:46, load average: 0.00, 0.00, 0.00 Jul 21 11:42:54 it gets sluggish as soon as you run into swap fragmentation Jul 21 11:42:58 known effect Jul 21 11:43:02 known workarounds Jul 21 11:43:14 known root cause - well kinda Jul 21 11:44:06 since my systems are not RAM heavy, it rarely ever runs into swapping Jul 21 11:45:04 of course crap like *-patch helps a lot to make it fragment swap even *faster* Jul 21 11:45:37 as do nonsensical wallpapers etc, widgets, whatnot Jul 21 11:46:07 completely true Jul 21 11:53:06 ShadowJK's reswap script would actually help Jul 21 11:53:30 as does an occasional `killall browserd` Jul 21 11:58:09 I have a problem with evolution-data-server. I want to patch it but when I even only recompile and install it (evolution-data-server-addressbook, libebook, libedata-book, libedataserver) then the Calendar starts to randomly crashing (signal 11). Jul 21 11:59:47 Shit. I recompiled another package :D Jul 21 12:05:43 luf: lolwut Jul 21 12:11:54 apt-get install microb-engine Jul 21 12:12:00 cool. Jul 21 12:12:16 Now system boots to hildon then INSTA REBOOTS. Jul 21 12:12:19 FFS Jul 21 12:13:47 So back to the ground. Jul 21 12:14:05 Maybe I recompiled wrong package or not. Jul 21 12:14:09 vi_: did you install the thumb compiled microb-engine without a thumb kernel? Jul 21 12:14:18 *pat pat* Jul 21 12:14:20 Now I recompiled the right one and the same error occured. Jul 21 12:17:44 freemangordon: ping Jul 21 12:19:51 zeq: I have the thumb kernel from cssu-tumb Jul 21 12:19:55 thumb Jul 21 12:20:16 and now I have a non booting device. This blows goatse Jul 21 12:20:26 oh well, just restore your backupmenu backup Jul 21 12:20:28 Backup menu! Jul 21 12:20:33 yay! Jul 21 12:21:04 unfortunatley my /opt partition backup is 1.5GB. It takes a while to restore. Jul 21 12:21:15 ...wat Jul 21 12:23:27 wat what? Jul 21 12:24:31 it's a lot! Jul 21 12:27:03 I have shit symlinked all over the place. Jul 21 12:27:07 frankly it is a mess. Jul 21 12:27:45 so you can't even blame freemangordon Jul 21 12:28:03 kerio: blame him for what? Jul 21 12:28:13 for borking your hildon Jul 21 12:28:25 I was the one who typed in apt-get install Jul 21 12:28:46 also my symlinking crazy shit all over the place has nothing to do with this Jul 21 12:28:51 yeah but it was his package Jul 21 12:28:53 from his repo Jul 21 12:28:54 or something Jul 21 12:29:14 besides, I have just disabled watchdogs and will now fix the problem. Jul 21 12:29:46 his repo that is 'so fucking experimental it will rip your dick clean off so do not even enable it'. Jul 21 12:29:51 I was warned. Jul 21 12:30:07 hahaha Jul 21 12:30:15 works for me :) Jul 21 12:30:22 why do we need watchdogs in the first place? Jul 21 12:31:31 to be fair though, my system is even more experimental (libstdc++ and libgcc_s from gcc4.7) Jul 21 12:32:21 zeq gives no fucks. Jul 21 12:32:30 ne is on the edge of sanity. Jul 21 12:32:38 I just mean, because it works here, doesn't mean it will anywhere else Jul 21 12:33:08 I've got backupmenu ;) Jul 21 12:33:14 backupmenu ^_^ Jul 21 12:33:20 zeq: that is the beauty of maemo. The fact it is soo customisable means there are infinite ways to fuck it up. Jul 21 12:33:40 would swap be better on a really fast µSD? Jul 21 12:33:56 kerio: what is a really fast uSD? Jul 21 12:34:27 µSD Jul 21 12:34:29 microsd Jul 21 12:34:37 (warning: this is obviously a trick question) Jul 21 12:34:51 what is a *really fast* uSD Jul 21 12:35:13 vi_, something with >4MB/s random write? Jul 21 12:35:13 ? Jul 21 12:35:20 Damn hurrian. Jul 21 12:35:21 yeah, something like that Jul 21 12:35:33 sequential write is kinda useless for swap Jul 21 12:35:41 get a sandisk? Jul 21 12:35:48 I was hovering above the 'AWOOOOOOOOGAA! General ignorance detected' Alarm button. Jul 21 12:36:12 With eds I did: tar xzf evolution-data-server_1.4.2.1-20100623+0m5.tar.gz; cd evolution-data-server_1.4.2.1; dpkg-buildpackage Jul 21 12:36:47 kerio: Hurrian is correct. A sandisk 'class 4' will beat the shit out of pretty much all so called 'class 10' uSD cards. Jul 21 12:37:05 Then I install on the N900 .deb packages and shudown/boot it up. After that Calendar occasionaly fall down with signal 11. Jul 21 12:37:13 Any idea what should be wrong? Jul 21 12:38:21 basically the whole 'class' thing is irrelevent. Jul 21 12:38:29 luf, ask cssu team for tips on packaging for fremantle Jul 21 12:38:37 also, you may have left -mthumb on Jul 21 12:38:39 vi_: sadly, even a sandisk mobile ultra (class 4) only achieves at most 2.7MB/s for random writes Jul 21 12:38:46 (512KB ones, mind you) Jul 21 12:39:05 It is like measuring the viscosity of motorbike suspension oil with SAE measurements. Complete nonsense. Jul 21 12:39:06 for 4K ones, you get barely 1MB/s with a normal sandisk class 4 Jul 21 12:40:08 anyway, still slower than the eMMC Jul 21 12:40:18 i think Jul 21 12:40:43 psycho_oreos, you good at openwrt? :P Jul 21 12:42:45 MohammadAG: I have a custom openwrt build, maybe I can help? Jul 21 12:43:16 or maybe it's *too* custom :P Jul 21 12:43:27 zeq, I flashed OpenWRT on my dusty WRT350N Jul 21 12:43:42 I'm planning to install it in my car (:P) with a 3G modem plugged into the USB port Jul 21 12:43:46 ooh i want help too Jul 21 12:43:48 nice Jul 21 12:43:53 i have a wrt610n Jul 21 12:43:55 I just realized I'm actually connected through it... Jul 21 12:44:08 so far i only installed ddwrt on it, but it's kinda bad i think Jul 21 12:44:09 anyway, ifup ppp3g (my interface name) doesn't seem to do anything Jul 21 12:44:30 kerio, ddwrt doesn't work on my router (v2 board), but openwrt seems nice Jul 21 12:44:46 ddwrt is only half opensource Jul 21 12:44:56 then go openwrt :p Jul 21 12:45:10 MohammadAG: anything in the log? Jul 21 12:45:15 i have no idea if it's properly supported Jul 21 12:45:15 logread? Jul 21 12:45:19 yup Jul 21 12:45:35 besides a wrong date Jul 21 12:45:47 http://pastebin.com/raw.php?i=hkB5DA8f Jul 21 12:45:59 anyway, #openwrt maybe? Jul 21 12:46:21 probably the better place :) Jul 21 12:46:57 force-power-key Jul 21 12:47:11 R&D flag, will this remove act-dead mode? Jul 21 12:47:19 afaik yes Jul 21 12:47:39 what's act-dead? Jul 21 12:47:50 MohammadAG: you need to find out what char device the chat script is running on. Make sure it's opening the modem. Jul 21 12:47:50 kerio, charging only mode Jul 21 12:47:53 oic Jul 21 12:48:25 zeq, I configured it to use /dev/ttyS0 Jul 21 12:49:38 I can seem to cat that, ttyS1 returns IO error Jul 21 12:49:55 can you install minicom or something? Jul 21 12:50:10 installing Jul 21 12:50:17 1MB left on / :/ Jul 21 12:50:33 probably better take it over to #openwrt though Jul 21 12:50:37 I'm on there Jul 21 12:53:04 kerio: fixed! Jul 21 12:53:32 also, biltong+chilli sauce==win Jul 21 12:54:15 vi_: what was broken? Jul 21 12:54:29 zeq: libgcc Jul 21 12:54:39 zeq, that place seems dead as ever Jul 21 12:55:02 1. Disable watchdogs. 2. Update libgcc. 3.enable WD. 4. update system. Jul 21 12:55:19 lots of people on there, but it is *really* quiet! Jul 21 12:55:38 vi_: I thought it might be Jul 21 12:55:53 so.. how do I dial with minicom? :P Jul 21 12:56:10 ATD I think Jul 21 12:56:20 browserd, still hogging all the ram. Jul 21 12:56:38 if it uses AT commands Jul 21 12:56:45 ATDT - tone; ATDP - pulse; Jul 21 12:56:50 ah yes Jul 21 12:56:50 it should, it works fine on Ubuntu Jul 21 12:57:05 dusty memory Jul 21 12:57:28 But maybe ATD choose one of them as primary I don't remember. Jul 21 12:58:42 I think it has some initstring? Jul 21 12:59:11 I don't get this, I turned on echo in minicom but I get nothing Jul 21 12:59:27 PULSE DIALLING?? Does the network even still support that? Jul 21 12:59:52 I have no clue for another countries ;) Jul 21 12:59:55 * vi_ recalls a time when you could dial by repeated slamming the handset up/down. Jul 21 13:00:03 Maybe in cenral Africa ... :D Jul 21 13:00:08 MohammadAG: are you sure the device is ttyS0? Jul 21 13:00:13 MohammadAG: hey, you did it wrong then ;-d Jul 21 13:00:21 not ttyACM0 or something? Jul 21 13:00:44 yes Jul 21 13:00:49 but minicom shows /dev/ttyS1 Jul 21 13:01:02 well that won't help :) Jul 21 13:01:35 MohammadAG: configure your minicom properly Jul 21 13:01:49 DocScrutinizer05, any tips? Jul 21 13:02:05 ctrl-a z Jul 21 13:02:06 I recall there being shortcut keys Jul 21 13:02:10 yes that's it Jul 21 13:02:28 damn dusty memory :P Jul 21 13:02:51 I've not used minicom for >10 years! Jul 21 13:02:55 rather ctrl-a o, as a shortcut Jul 21 13:02:58 ctrl-a h is help or menu or something like that. Jul 21 13:03:09 zeq: the same here :D Jul 21 13:03:11 z is help Jul 21 13:04:11 minicom even has a addrbook Jul 21 13:04:17 for numbers to fial Jul 21 13:04:20 dial Jul 21 13:04:34 ok, configured Jul 21 13:04:46 I still don't get how to pass commands to it Jul 21 13:04:53 to what? Jul 21 13:04:56 minicom Jul 21 13:05:07 ctrl-a Jul 21 13:05:20 everything else goes to serial Jul 21 13:05:21 no, I mean AT commands Jul 21 13:05:33 minicom doesn't take AT commands Jul 21 13:05:35 Fucking Maemo SDK. It seems I find the root cause with eds recompilation. Jul 21 13:05:35 MohammadAG: you just type them Jul 21 13:05:40 it's not a modem Jul 21 13:05:42 zeq, and nothing happens? Jul 21 13:05:51 try atz Jul 21 13:06:14 nada Jul 21 13:06:16 you're probably not connected to the modem Jul 21 13:06:24 if you type AT commands, minicom will send them out to serial, and if there's a modem on serial it will (or will not) react on AT command Jul 21 13:06:28 make sure your serial settings are correct Jul 21 13:06:40 flow control etc Jul 21 13:07:19 * MohammadAG tries minicom on iPhone Jul 21 13:07:25 mhm Jul 21 13:07:30 via USB? Jul 21 13:07:45 no, SSH Jul 21 13:07:50 yay Jul 21 13:08:03 MohammadAG: did you get anything in the kernel log when you inserted the device about which serial device it is? Jul 21 13:08:17 zeq, no, dmesg is vague on openwrt Jul 21 13:08:18 Thanks Pali and DocScrutinizer05 for help in finding the problem. Jul 21 13:08:26 ?? Jul 21 13:08:38 USB Serial support registered for GSM modem (1-port) Jul 21 13:08:38 usbcore: registered new interface driver option Jul 21 13:08:38 option: v0.7.2:USB Driver for GSM modems Jul 21 13:08:39 The SDK has enabled thumb as default. Jul 21 13:08:58 yep Jul 21 13:09:00 So it leads to the crashes when I only recompile evolution-data-server. Jul 21 13:09:12 I disabled it and now everything seems ok. Jul 21 13:09:16 I posted a general warning just yesterday iirc Jul 21 13:09:30 Yes. That helped me. Jul 21 13:09:44 MohammadAG: ^^^ Jul 21 13:09:47 :-P Jul 21 13:09:56 I noticed on the wikis :p Jul 21 13:09:58 err, MLs Jul 21 13:09:59 MohammadAG: does cat /proc/devices tell you anything? Jul 21 13:10:01 'your' old thumbie problem Jul 21 13:10:30 it's actually generic Jul 21 13:10:39 http://pastebin.com/raw.php?i=75hcACK7 zeq Jul 21 13:11:04 ttyACM probably Jul 21 13:11:28 zeq, it's not in /dev Jul 21 13:11:29 http://pastebin.com/raw.php?i=NyYhZdsu Jul 21 13:11:33 err wait, you are trying to get a local /dev/tty* via ssh via wlan? Jul 21 13:11:56 * DocScrutinizer05 feel a headache near Jul 21 13:11:57 if you're not running udev you'll have to make the device node Jul 21 13:12:12 DocScrutinizer05, the problem is on my router Jul 21 13:12:22 my router has a USB port which I plugged a USB modem into Jul 21 13:12:28 aaah Jul 21 13:12:32 and it's running OpenWRT, a linux distro for routers Jul 21 13:12:34 and minicom on your router Jul 21 13:12:53 makes sense Jul 21 13:13:13 should be /dev/ttyUSB or ttyACM or sth Jul 21 13:13:42 but only if udev creates that, first you need to check syslog if it gets detected at all Jul 21 13:13:57 device major 166 and 188 respectively Jul 21 13:14:36 mknod mknod /dev/ttyACM0 c 166 0 Jul 21 13:14:43 shouldn't it remove the lines from /proc/devices on device unplug? Jul 21 13:14:45 fsckng pastebin sets mimetype to txt :-/ Jul 21 13:15:00 only *one* mknod, obviously :) Jul 21 13:15:08 browser offers to download or open in editor Jul 21 13:15:25 /proc/devices is what the kernel knows about, not userspace unless you have udev running Jul 21 13:16:12 yep, the kernel driver for usb tty will need a device node in /dev/ Jul 21 13:16:26 you could live happily with static devicenode Jul 21 13:16:59 root@OpenWrt:~# cat /dev/ttyACM0 Jul 21 13:16:59 cat: can't open '/dev/ttyACM0': No such device Jul 21 13:17:25 mknod /dev/ttyACM0 c 166 0 Jul 21 13:17:38 ^^ Jul 21 13:17:57 read that line again Jul 21 13:18:03 No such device, not no such file Jul 21 13:18:05 mknod /dev/ttyUSB0 c 188 0 Jul 21 13:18:14 root@OpenWrt:~# mknod /dev/ttyACM0 c 166 0 Jul 21 13:18:14 root@OpenWrt:~# cat /dev/ttyACM0 Jul 21 13:18:14 cat: can't open '/dev/ttyACM0': No such device Jul 21 13:18:25 same Jul 21 13:18:33 modprobe dafaq Jul 21 13:18:50 will bind to that devicenode Jul 21 13:18:58 MohammadAG: if it's listed in /proc/devices, that should work Jul 21 13:19:07 zeq, it doesn't :P Jul 21 13:19:19 * MohammadAG starts over -> reboot Jul 21 13:19:25 good plan Jul 21 13:19:49 check what kernel module gets loaded when youplug in the modem Jul 21 13:21:22 nope, still doesn't work Jul 21 13:21:46 http://pastebin.com/raw.php?i=TrnRxvFB zeq DocScrutinizer05 Jul 21 13:21:50 I feel I'm missing something Jul 21 13:21:57 MohammadAG: did the ttyACM and ttyUSB devices only show in /proc/devices after insertion? Jul 21 13:22:16 NO I don't want to open it with kwrite!! Jul 21 13:22:44 DocScrutinizer05, just curl it :p Jul 21 13:22:50 http://pastebin.com/raw.php?i=r9G5vb8A on Ubuntu Jul 21 13:22:54 dafaq I will Jul 21 13:23:10 zeq, I'll reboot again to check, it was plugged in from the start Jul 21 13:24:02 MohammadAG: did you listen to me? toldya to check what kernel module gets loaded when you plug in the modem Jul 21 13:24:35 you'll need 2 things for that: syslog/whatever, and plugging in the modem _after_ booting Jul 21 13:24:41 lsmod might help as well Jul 21 13:25:17 zeq, no, they're there from bootup Jul 21 13:25:19 DocScrutinizer05, no new lines in dmesg after plugging in Jul 21 13:25:34 now that's a bad thing Jul 21 13:25:48 lsusb won't show anything either then Jul 21 13:25:53 can you try rmmod and insmod the driver? Jul 21 13:26:04 only if he knows Jul 21 13:26:05 no lsusb Jul 21 13:26:08 the driver Jul 21 13:26:19 true Jul 21 13:26:42 [2012-07-21 15:18:33] modprobe dafaq Jul 21 13:26:44 [2012-07-21 15:18:49] will bind to that devicenode Jul 21 13:26:48 is it possible to stop the keyboard lights from spazzing out when using R&D mode? Jul 21 13:27:04 errr, yes Jul 21 13:27:11 errr, how? Jul 21 13:27:17 there's a sysfs node to switch the hw on/off Jul 21 13:27:21 /etc/pm tab Jul 21 13:27:27 don't ask me the friggin name Jul 21 13:27:29 and there's a config for bootup :p Jul 21 13:27:45 DocScrutinizer05, no modprobe, no dafaq Jul 21 13:27:58 ouch Jul 21 13:28:07 what a shite OS is THAT? Jul 21 13:28:10 is there a way to know the driver? Jul 21 13:28:15 DocScrutinizer05: a very tiny one Jul 21 13:28:16 lsmod? Jul 21 13:28:18 DocScrutinizer05, something that runs on 4MBs of flash Jul 21 13:28:43 well, without modprobe you're fecked Jul 21 13:28:53 MohammadAG: can you check which module is used on ubuntu and see if it's available on your openwrt? Jul 21 13:28:58 unless the fools thought insmod is sufficient replacement Jul 21 13:28:59 insmod works Jul 21 13:29:00 that's what I want to do Jul 21 13:29:07 you just have to deal with the dependencies yourself Jul 21 13:29:11 DocScrutinizer05: that does allow keyboard lights to operate as normal right? Jul 21 13:29:18 vi_, yes Jul 21 13:29:31 zeq, I can't figure out which driver it's using Jul 21 13:29:34 yes Jul 21 13:29:41 is there a way to limit drivers by device ID? Jul 21 13:29:47 Bus 001 Device 008: ID 19d2:0117 ZTE WCDMA Technologies MSM Jul 21 13:29:54 MohammadAG: there should be a /sys entry to help you Jul 21 13:30:12 MohammadAG: what will your desktop PC syslog say when you plug in modem there? Jul 21 13:30:45 zeq: actually that too, but not on a system that's too tiny for modprobe ;-P Jul 21 13:31:02 http://pastebin.com/raw.php?i=1vGjgfcT Jul 21 13:31:09 *sigh* Jul 21 13:31:29 FSCKYOU pastebin Jul 21 13:31:56 I meant in /sys on the ubuntu when the driver is loaded Jul 21 13:32:02 MohammadAG: fool! paste your shite as proper html, not as mimetype txt! Jul 21 13:32:13 DocScrutinizer05, http://pastebin.com/1vGjgfcT Jul 21 13:32:22 your browser should handle raw text Jul 21 13:32:43 ORLY?! Jul 21 13:32:49 yes Jul 21 13:32:50 :p Jul 21 13:33:07 YOUR browser should rething what's html Jul 21 13:33:15 and what's mimetypes Jul 21 13:33:20 what is 'sleep_ind' in pmconfig? Jul 21 13:33:25 that's it Jul 21 13:33:31 if it's 1, change it to 0 Jul 21 13:33:48 save, reboot, or save and change the sysfs value Jul 21 13:34:03 [ 9265.148196] usb 1-1.1: GSM modem (1-port) converter now attached to ttyUSB0 Jul 21 13:34:21 sleep_ind==sleep indicators. Jul 21 13:34:46 I wonder if the hardware has to wake up to actuate the sleep indicators to indicate it was asleep. Jul 21 13:34:47 zeq, I don't know what path to check Jul 21 13:35:31 MohammadAG: silly question, but do you have enough bus power? Jul 21 13:35:34 the usual fuck I guess, with the modem pretending it's a memstick with windows drivers on it Jul 21 13:35:39 zeq, no idea :P Jul 21 13:35:54 the modem lights up and initializes (red -> green) though Jul 21 13:36:07 MohammadAG: that's probably ok then...? Jul 21 13:36:17 yeah, I remember it staying red on the N900 Jul 21 13:36:17 DocScrutinizer05: There is a linux program for toggling the USB mode of the usb stick. Jul 21 13:36:23 yes Jul 21 13:36:29 modeswitch_usb afaik Jul 21 13:36:31 eject even can do, usually Jul 21 13:36:34 What is the motivation for open-source projects to make things hard to build? Jul 21 13:36:49 fasta: It keeps the noobs out. Jul 21 13:37:00 yes, modeswitch_* sounds correct Jul 21 13:37:08 usb_modeswitch Jul 21 13:37:10 oh well, same shit Jul 21 13:37:21 vi_: I'd say it merely shows how much of a n00b the developers are. Jul 21 13:37:59 vi_: would you also say that's why they write pseudo-C instead of correct C? Jul 21 13:38:00 yoh Jul 21 13:38:25 MohammadAG: if it's not installed there probably is a pkg for it Jul 21 13:38:27 though - what's pseudo-c? Jul 21 13:38:33 vi_: and why it requires a special version of gcc to compile, instead of just a whole range of compilers and at the very least the latest? Jul 21 13:38:45 zeq, if what isn't installed? Jul 21 13:38:45 MohammadAG: but you should be able to tell if it has started as a memstick Jul 21 13:38:55 well, I'd rather get dmesg to show stuff first Jul 21 13:39:03 DocScrutinizer05: pseudo-C is code which is not portable, but runs due to pure luck on some architectures and operating systems. Jul 21 13:39:09 Do any of you guys remmeber how to only see photos taken by the camera in the photo-app? I see all kind of small icons and stuff too Jul 21 13:39:18 fasta: :nod: Jul 21 13:39:23 DocScrutinizer05: for example, all media players on Linux fall under that. Jul 21 13:39:25 what does /proc/usb show Jul 21 13:39:25 ? Jul 21 13:39:39 I connected my N9 to the router, dmesg didn't change Jul 21 13:39:40 kgu: In photo app settings there is a choice for 'show only camera' photos. Jul 21 13:39:48 kgu: check tracker config Jul 21 13:39:54 kgu: No, you cannot set it to use that by default. Jul 21 13:40:05 /proc/bus/usb ? Jul 21 13:40:06 DocScrutinizer05: the player I am using now is just so completely wrong, it's not funny, yet it doesn't crash. Jul 21 13:40:21 zeq, empty dir Jul 21 13:40:22 MohammadAG: if it's there Jul 21 13:40:44 cat /proc/filesystems Jul 21 13:40:55 DocScrutinizer05: Qemu is another project for which the same holds. Jul 21 13:40:55 MohammadAG: are you sure the USB itself is supported on that *wrt? Jul 21 13:41:08 DocScrutinizer05, it provides power, it should work Jul 21 13:41:09 DocScrutinizer05: it does support usb Jul 21 13:41:16 DocScrutinizer05: you would think that Redhat employees would know how to write good code, right? Jul 21 13:41:24 http://pastebin.com/wHGaxcSw Jul 21 13:41:28 NO, definitely not Jul 21 13:41:35 fasta: ^^^ Jul 21 13:41:45 DocScrutinizer05: so, I wonder how every single time everything which is done, is done completely wrong. Jul 21 13:41:50 I mean, lennart is a redhat employee afaik ;-P Jul 21 13:42:05 MohammadAG: you'll have to try to install lsusb I think Jul 21 13:42:46 zeq, I'm not sure which package it's in Jul 21 13:42:51 DocScrutinizer05: there are a few actual professional applications which actually do work, but one can likely count them on two hands on the whole planet. Jul 21 13:43:08 DocScrutinizer05: but that's only because I never have seen the code for those likely. Jul 21 13:43:25 Even QT has some really weird ideas. Jul 21 13:43:28 MohammadAG: you could try insmod usbdevfs first Jul 21 13:43:34 some? SOME??? Jul 21 13:43:42 QT does #include Jul 21 13:44:07 That's certainly not following conventions. Jul 21 13:44:07 * DocScrutinizer05 points at qtm-sensors Jul 21 13:44:35 DocScrutinizer05: so, what's the secret behind all this epic failure? Jul 21 13:44:50 42 ;-D Jul 21 13:44:52 DocScrutinizer05: is it just because it is free that they make it crappy on purpose? Jul 21 13:44:59 DocScrutinizer05: or are they just born that stupid? Jul 21 13:45:35 I have seen some commercial software which has null pointer exceptions in all the logs. Jul 21 13:45:40 fasta: what am I missing? what's wrong with #include ? Jul 21 13:45:47 I asked about it and they said 'yeah, that's normal'. Jul 21 13:45:58 root@OpenWrt:~# lsusb Jul 21 13:45:58 Bus 001 Device 001: ID 1d6b:0002 Linux Foundation 2.0 root hub Jul 21 13:45:59 Bus 001 Device 002: ID 0421:0518 Nokia Mobile Phones Jul 21 13:45:59 MohammadAG: actually attaching a memstick first would tell you a lot about your USB Jul 21 13:46:11 \o/ Jul 21 13:46:26 zeq: #include Jul 21 13:46:33 zeq: that's normal. Jul 21 13:46:41 zeq: or C++ Jul 21 13:46:53 zeq: #include Jul 21 13:47:16 well, #include .h seems pretty normal too Jul 21 13:47:18 zeq: there is just no excuse to not include the .h or .hpp or however you want to call it. Jul 21 13:47:39 DocScrutinizer05: yes, that's normal. Jul 21 13:48:11 But even lots of GNU people don't know how to use their own fucking software configuration tool (autoconf). Jul 21 13:48:31 nobody knows how to use friggin autoconf Jul 21 13:48:33 glib has circular dependencies on pkg-config Jul 21 13:48:46 ok, N9 mounted Jul 21 13:48:52 had to install some kernel modules Jul 21 13:48:57 DocScrutinizer05: I learned how to use it correctly, but yes, it was a pain. Jul 21 13:49:15 MohammadAG: as a modem? Jul 21 13:49:22 no, mass storage device Jul 21 13:49:27 :nod: Jul 21 13:49:33 usb 1-1: GSM modem (1-port) converter now attached to ttyUSB0 Jul 21 13:49:33 option 1-1:1.1: GSM modem (1-port) converter detected Jul 21 13:49:33 usb 1-1: GSM modem (1-port) converter now attached to ttyUSB1 Jul 21 13:49:33 option 1-1:1.2: GSM modem (1-port) converter detected Jul 21 13:49:33 usb 1-1: GSM modem (1-port) converter now attached to ttyUSB2 Jul 21 13:49:33 woo Jul 21 13:49:47 :D Jul 21 13:49:49 hey! Jul 21 13:50:13 usual shite exposing three(!) tty Jul 21 13:50:22 DocScrutinizer05: you are somewhat older than me, I think. Don't you have some explanation which explains all the observed failures? Jul 21 13:50:37 not a good one Jul 21 13:50:50 except it seems to me it's related to youth Jul 21 13:50:59 ok, so how do I test it again? :P Jul 21 13:51:03 DocScrutinizer05: youth of the developers or of me? Jul 21 13:51:10 fasta: Closest thing to an explaination would be limited knowledge or ignorance Jul 21 13:51:13 kids learn coding nowadays before they learn reading Jul 21 13:51:28 and bad habits stick Jul 21 13:51:34 most people have their area of focus Jul 21 13:51:38 zeq: yeah, but it's really 99.999% which is ignorant. Jul 21 13:51:53 zeq: it kind of makes me look like an arrogant prick. Jul 21 13:51:59 "screwit, it works!" Jul 21 13:52:10 DocScrutinizer05: "worksforme" Jul 21 13:52:32 [2012-07-20 14:36:32] FIQ: There is the right way, the wrong way and the DocScrutinizer05 way. The DocScrutinizer05 way being like the right way, but even more correct. Jul 21 13:52:35 (yes, it works for you, because you haven't heard of this thing called a "memory model" yet) Jul 21 13:52:48 ? Jul 21 13:52:59 Sequence point? Dude, get real! Jul 21 13:53:13 FIQ: just quoting a matching post Jul 21 13:53:17 freemangordon: did you build latest fennec? Jul 21 13:53:22 -pedantic -Werror? It's working, man! Jul 21 13:53:23 oh Jul 21 13:53:32 why do usb modems feel the need to expose multiple tty ports? Jul 21 13:53:48 It uses 1000 times more CPU than needed? Yeah, who cares? Jul 21 13:53:50 hell, this one also acts as a usb ethernet card Jul 21 13:53:59 kerio: idiocy? Jul 21 13:54:19 kerio: one for AT commands, one for ppp, one for the fun Jul 21 13:54:22 haha Jul 21 13:54:29 this one has only two, i think Jul 21 13:54:33 One has been designed for the NSA. Jul 21 13:54:38 Duh. Jul 21 13:54:40 LOL Jul 21 13:54:45 plus the fake ethernet Jul 21 13:55:08 All the people on Linux who say they have no working wireless should think again. Jul 21 13:55:20 they are for separate command channels I think Jul 21 13:55:24 It's just not working for them. Jul 21 13:55:36 Echelon is working mighty fine. Jul 21 13:55:49 actually it might save you from implementing friggin mux07.xx Jul 21 13:55:56 grrr Jul 21 13:55:59 freemangordon: I managed to build fennec with linaro-4.7 Jul 21 13:56:01 reconnect Jul 21 13:56:07 actually it might save you from implementing friggin mux07.xx Jul 21 13:56:22 I'll put the toolchains up now in the same place as before. Jul 21 13:56:27 I wonder how one would feel working on a global spy system. You would be like Gestapo++. Jul 21 13:56:51 which would otherwise multiplex all the different chanels for cmd(AT), data, unsol response etc over just one tty Jul 21 13:57:07 DocScrutinizer05: what about the fake ethernet? Jul 21 13:57:17 fake ethernet? Jul 21 13:57:37 kerio: isn't that a packet interface? Jul 21 13:57:50 you probably talk about the NIC that gets exposed for direct IP access to GSM stack Jul 21 13:57:51 for direct access Jul 21 13:57:52 this usb card exposes two ttys and an ethernet Jul 21 13:58:15 hm Jul 21 13:58:56 kerio: so you got the magic 3 again Jul 21 13:59:11 the two ttys *kinda* make sense Jul 21 13:59:23 just one is a eth* so you don't need to mount a converter to the tty Jul 21 13:59:42 yeah but how do i use it? Jul 21 13:59:55 basically like any other NIC Jul 21 14:00:03 the official software doesn't use it Jul 21 14:00:05 at least on os x Jul 21 14:01:08 yeah, they probably do 07.xx mux then, and ppp Jul 21 14:01:34 usb modems are a PITA Jul 21 14:01:44 zeq, http://pastebin.com/MXAzVE65 Jul 21 14:02:06 lol Jul 21 14:02:09 thanks MohammadAG Jul 21 14:02:12 MohammadAG: that doesn't look so bad Jul 21 14:02:25 (for 1. line in that pastie matching my post) Jul 21 14:02:43 nearly there :) Jul 21 14:02:55 yeah Jul 21 14:02:57 that's ttyUSB2 Jul 21 14:03:01 0 and 1 didn't work Jul 21 14:03:33 MohammadAG: I suggest you use minicom to interactively connect Jul 21 14:03:49 they tweak the chat script afterwards Jul 21 14:04:21 s/they/then/ Jul 21 14:04:21 zeq meant: then tweak the chat script afterwards Jul 21 14:04:27 ATI Jul 21 14:04:31 Manufacturer: ZTE CORPORATION Jul 21 14:04:33 at least that works Jul 21 14:04:34 MohammadAG: it looks like everything just fine, just you either have wrong APN, or no credit Jul 21 14:04:44 postpaid and APN is right Jul 21 14:04:54 how do I connect with minicom? Jul 21 14:05:02 then you already are in internet mode I'd guess Jul 21 14:05:13 no more need to talk to modem via minicom Jul 21 14:05:29 DocScrutinizer05: he isn't getting CONNECT Jul 21 14:05:31 you'd probably even need "+++" escape sequence to abort data mode Jul 21 14:05:59 ATD*99# << errors Jul 21 14:06:04 yep Jul 21 14:06:21 that's why I say you either got wrong parameters, or no more credit Jul 21 14:06:30 but it works fine on Ubuntu :/ Jul 21 14:06:34 how do I pass a username and pass? Jul 21 14:06:37 the sequence seems all fine Jul 21 14:06:46 until it refuses atd*99*** Jul 21 14:07:12 Jul 21 13:57:36 OpenWrt local2.info chat[8197]: AT+CGDCONT=1,"IP","internet.pelephone.net.il"^M^M Jul 21 14:07:46 DocScrutinizer05, http://pastebin.com/XudJPxQW Jul 21 14:08:36 looks good, though I dunno who's using it Jul 21 14:08:41 Ubuntu Jul 21 14:09:25 well, you need same on your wrt Jul 21 14:09:27 somehow Jul 21 14:09:48 * MohammadAG tries minicom on Ubuntu Jul 21 14:10:19 you don't wanna use minicom, as the system already is two layer higher than that Jul 21 14:10:31 Jul 21 13:57:36 OpenWrt local2.info chat[8197]: send (ATD*99***1#^M) Jul 21 14:10:37 ^^^ is minicom Jul 21 14:10:39 level Jul 21 14:11:03 DocScrutinizer05: I think he's just trying to see if he gets the same response on ubuntu Jul 21 14:11:04 after that all 3 channels are assigned to ppp, and eth0, and whatnot Jul 21 14:11:05 ATD*99***1# Jul 21 14:11:06 CONNECT 7200000 Jul 21 14:11:15 on Ubuntu Jul 21 14:11:33 i installed touchsearch,but can't find icon on home,or add widget menu.i reboot phone but the same Jul 21 14:11:37 MohammadAG: that's using minicom, or the ppp script? Jul 21 14:11:43 minicom on Ubuntu Jul 21 14:12:03 does it do the same on the wrt? Jul 21 14:12:09 no, it returns ERROR Jul 21 14:12:20 serial error? Jul 21 14:12:21 all parameters the same? Jul 21 14:12:37 Jul 21 13:57:36 OpenWrt local2.info chat[8197]: ERROR Jul 21 14:13:50 DocScrutinizer05: just trying to mimimize the difference between the working and non-working cases Jul 21 14:13:51 Replace 'pincode' with the correct pincode of your SIM card. Note that an disabled pincode on the SIM card is problematic, please enable it. If you are connecting to a phone where the pincode has already been entered, there if no need for this. Jul 21 14:13:53 psychologe: check your .desktop files Jul 21 14:14:18 psychologe: /usr/share/apps/*/* sth Jul 21 14:15:45 freemangordon: ping Jul 21 14:15:51 how do I set a pin code? Jul 21 14:16:09 ugh Jul 21 14:16:11 Is there some way to disable the blinking when a new e-mail arrives? Jul 21 14:16:15 DocScrutinizer05, in /usr/share/applications/hildon-home/ i can see touchsearch .desktop Jul 21 14:16:16 MohammadAG: it's the same "phone", I can't see that being the problem tbh Jul 21 14:16:28 fasta: settings-notificatrions Jul 21 14:16:59 root@OpenWrt:~# gcom -d /dev/ttyUSB2 Jul 21 14:16:59 SIM ready Jul 21 14:16:59 Waiting for Registration..(120 sec max) Jul 21 14:16:59 Registered on Home network: "Pelephone",2 Jul 21 14:16:59 Signal Quality: 10,99 Jul 21 14:17:44 that seems to have worked? Jul 21 14:17:46 DocScrutinizer05: you mean the same menu which contains 'received chat'? Jul 21 14:17:47 zeq, I got CONNECT Jul 21 14:17:54 nothing after that though Jul 21 14:17:54 DocScrutinizer05: because it is already disabled there. Jul 21 14:18:38 MohammadAG: it will be waiting for a ppp handshake or packet network access Jul 21 14:19:16 zeq, success? http://pastebin.com/4bjFQQtA Jul 21 14:19:39 looks good :) Jul 21 14:19:46 but it hungup for some reason :p Jul 21 14:19:47 Does anyone use Monav here? Jul 21 14:20:11 MohammadAG: I'd say that's a new problem :) Jul 21 14:20:39 I went through the trouble of getting a PBF file and then converting it with the preprocessor and now the stupid phone says that the fileformat is not compatible. Jul 21 14:21:11 Are there any instructions to get a working mapping application anywhere at all? Jul 21 14:21:37 fasta: settings Jul 21 14:21:48 fasta: notification light Jul 21 14:21:53 owtte Jul 21 14:22:43 the menu point with the 3 colered overlapping dots Jul 21 14:23:29 hm Jul 21 14:23:38 DocScrutinizer05: I switched my system language to English, because otherwise it's just not convenient. Jul 21 14:24:00 I seem to be to stupid to reflash the system- Jul 21 14:24:02 fasta: so what? Jul 21 14:24:12 fasta: do you expect me to do the same? Jul 21 14:24:19 to do you a favour Jul 21 14:24:25 DocScrutinizer05: so, it is rebooting now and then I will likely see what the same things you are seeing. Jul 21 14:24:59 well, you should've picked 'german' then Jul 21 14:25:02 ;-P Jul 21 14:25:31 "Einstellungen" - "Benachrichtigungslicht" Jul 21 14:25:38 (SIC!) Jul 21 14:25:49 DocScrutinizer05: E-mail message received is _not_ enabled. Jul 21 14:25:57 DocScrutinizer05: and I do get it. Jul 21 14:26:12 DocScrutinizer05: for text messages it is however enabled. Jul 21 14:26:57 MohammadAG: that's weird why it hangs up Jul 21 14:27:10 instantly after chap/pap Jul 21 14:27:30 DocScrutinizer05: it's another option I think. Jul 21 14:27:47 DocScrutinizer05: I likely already found it in the actual E-mail application. Jul 21 14:28:07 nfc, I never really used modest Jul 21 14:28:40 You dont use email on the N900? Or another app? Jul 21 14:28:43 if you wanna know for sure, do dbus-monitor|less, /mce Jul 21 14:28:54 no email on mobile Jul 21 14:28:57 DocScrutinizer05: fixed that but no internet conn yet Jul 21 14:29:05 oh I see Jul 21 14:29:20 MohammadAG: routing? DNS? Jul 21 14:29:23 one account with one mail per month Jul 21 14:30:05 Not that much. Snailmail would be fine for you then :) Jul 21 14:30:25 on t900 though I got modest configured, but it doesn't really work as it collides with my other client using pop and modest using imap, which my mailserver doesn't handle Jul 21 14:30:49 no, interface name has to be wan :D Jul 21 14:31:05 zeq, thanks a million :D Jul 21 14:31:11 DocScrutinizer05, you too :D Jul 21 14:31:16 MohammadAG: sure thing, you need to tell it which interface to use Jul 21 14:31:26 POP is so old fasioned Jul 21 14:31:42 MohammadAG: no problem :) Jul 21 14:31:43 http://www.speedtest.net/result/2076236740.png not bad Jul 21 14:32:10 IMAP is so "I'd use a webmailer but I think it's not as leete as this weird protocol" Jul 21 14:32:26 heh Jul 21 14:32:36 IMAP is a programmatic webmail Jul 21 14:33:23 POP3 is the best, don't need constant connection to work Jul 21 14:33:33 exactly Jul 21 14:33:56 I use IMAP :P Jul 21 14:34:03 and backfiring to server is minimal Jul 21 14:34:59 I don't need the server to tell me I can't read my 3 day old mail because it's out for maintenance Jul 21 14:35:33 well, it's my own server, so ... Jul 21 14:35:43 DocScrutinizer05: well, you can do pop3 on modest Jul 21 14:35:51 and I don't need my server make a mail vanish on all clients because I deleted it on N900 Jul 21 14:36:10 I need that, to keep in sync Jul 21 14:36:37 actually that's the thing I need the least Jul 21 14:38:01 I need it the most - people have different needs, and good that both is possible Jul 21 14:38:42 gave me a friggin headache to make my OM account delete the mails from server while keeping it on my mail client - the OM server offers IMAP only :-( Jul 21 14:38:48 How much bandwidth does it eat? Jul 21 14:39:05 I.e. if you leave it on always? Jul 21 14:39:09 Per month. Jul 21 14:39:21 fasta: enough to drain your battery :) Jul 21 14:39:27 IMAP? probably unpredictabble shitload Jul 21 14:39:50 i don't like IMAP so i'll spread FUD about it Jul 21 14:39:54 As long as I'm on flatrate, I dont care Jul 21 14:40:20 * DocScrutinizer05 hopes for battery flatrate Jul 21 14:40:26 IMAP is superior to pop in EVERY way. Jul 21 14:40:38 lol, no Jul 21 14:40:52 lol, yes Jul 21 14:40:54 HTTP is superior to IMAP in EVERY way Jul 21 14:41:13 http!=imap Jul 21 14:41:24 IRC is superior to MSNP in EVERY way Jul 21 14:41:24 imap!=pop Jul 21 14:41:26 DocScrutinizer05: can the "internal" usb port be used for charging, with an inductive mat and something below the cover? Jul 21 14:41:47 hi all, is this still the preferred way to automount ext2 microsd cards? http://wiki.maemo.org/Ext2_on_microSD_card Jul 21 14:41:48 yes Jul 21 14:42:01 dreamer: last i heard, CSSU's automounter mounted *everything* Jul 21 14:42:06 kerio: not without protective circuitry Jul 21 14:42:07 DocScrutinizer05: probably you want disconnected imap, but if'you're fine with downloading the mails, pop and imap are equivalent Jul 21 14:42:16 dreamer: use cssu Jul 21 14:42:18 zeq: which protective circuitry? i'd charge using the n900 itself Jul 21 14:42:19 kerio: it doesn't here. still have to mount -t ext this card Jul 21 14:42:23 StyXman_: exactly Jul 21 14:42:26 jon_y: afaik I have cssu installed .. Jul 21 14:42:38 dreamer: as I found before, cssu mounter auto mounts my ext3 card Jul 21 14:42:40 StyXman_: modulo imap gives you more headache configuring it Jul 21 14:42:45 *automounter Jul 21 14:42:48 I have CSSU Jul 21 14:42:53 and an ext3-formatted SD card Jul 21 14:42:55 I see it in the nokia stock file browser Jul 21 14:43:02 DocScrutinizer05: the server or the client? both should be faily equivalent Jul 21 14:43:06 I still get these "not supported!" errors for the microsdcard thingy Jul 21 14:43:20 kerio: there is no protective circuitry on the internal usb Jul 21 14:43:24 hmm, not sure, you using the stock kernel? Jul 21 14:43:28 oh, that kind of protection Jul 21 14:43:29 DocScrutinizer05: what's the problem with imap4? Jul 21 14:43:29 i see Jul 21 14:43:32 hrmpf :c Jul 21 14:43:33 StyXman_: client side, see my comment about deleting mails from server above Jul 21 14:43:37 jon_y: I use powerkernel Jul 21 14:43:45 and nobody made an inductive-capable BL-5J yet, huh? Jul 21 14:43:48 to be able to use mobilehotspot (or I wouldn't be connected atm :P) Jul 21 14:43:51 hmm, me too, not sure why aren't you mounting it Jul 21 14:44:14 FIQ: check with the nokia file browser app Jul 21 14:44:21 amything new? Jul 21 14:44:21 jon_y: sec Jul 21 14:44:35 nope Jul 21 14:44:50 well, I see memory card there, but opening it just shows "format not supported" Jul 21 14:44:51 kerio: I have an inductive stove :) Jul 21 14:45:01 FIQ: try dmesg|tail ? Jul 21 14:45:18 kerio: I thought about exactly that: BL-5 with inductive charging Jul 21 14:45:21 DocScrutinizer05: most clients support «delete after downloading» Jul 21 14:46:02 StyXman_: yeah, which makes your mail vanish on all other clients :-/ Jul 21 14:46:20 the wonders of imap Jul 21 14:46:29 DocScrutinizer05: it's exactly the same as POP in that regard Jul 21 14:46:30 isn't it Jul 21 14:46:36 no way Jul 21 14:46:41 DocScrutinizer05: cloud computing!! Jul 21 14:46:46 thin clients!! Jul 21 14:46:46 pop download is download Jul 21 14:47:08 no way to remote delete a mail on any remote client Jul 21 14:47:21 man, I get sick of getting told how much cloud integration $new_phone has Jul 21 14:47:25 sure, and if you IMAP download from all your clients, you get the same behavior as POP Jul 21 14:47:40 DocScrutinizer05: no, but once you downloaded and deleted, either with pop or imap, other clients won't see those emails either Jul 21 14:47:42 so, I need to sign up with dropbox to get any external storage :( Jul 21 14:47:55 haha Jul 21 14:47:59 jon_y: google drive! Jul 21 14:48:07 yeah, skydrive too Jul 21 14:48:14 what you can do is have your mail in a dirmail setup and sync it with git Jul 21 14:48:15 so bottom line: fsck imap, pop worked fine for 4 decades Jul 21 14:48:19 jon_y: how can I check if I have the cssu automounter? Jul 21 14:48:34 dreamer: it should be installed as part of cssu Jul 21 14:48:43 after you enabled cssu of course Jul 21 14:49:05 "i don't like imap, so nobody should use it" Jul 21 14:49:37 jon_y: well, I have the 'CSSU Features Configuration' .. Jul 21 14:49:43 damn, too long since I touched this stuff :P Jul 21 14:49:48 ok, some experimentation Jul 21 14:50:02 StyXman_: hahaha what Jul 21 14:50:04 and now I recall what the fs-table is on my sd-card Jul 21 14:50:05 dreamer: try also dmesg|tail? Jul 21 14:50:09 mail is not source code Jul 21 14:50:15 * DocScrutinizer05 kicks kerio, just for fun Jul 21 14:50:27 and git is a bad VCS anyway, so... Jul 21 14:50:32 p1 is formatted as vfat/fat32/whatever you want to call it, contains an uboot image Jul 21 14:50:33 ouch! Jul 21 14:50:41 git is Superior Jul 21 14:50:46 see? he likes it! Jul 21 14:50:52 p2 is formatted as ext4, contains the rest of the content (i.e. an arch installation) Jul 21 14:50:53 * DocScrutinizer05 kicks kerio, just for more fun Jul 21 14:51:00 stop it dammit Jul 21 14:51:09 * kerio IMAPs DocScrutinizer05 Jul 21 14:51:15 you want more? sure thing! Jul 21 14:51:19 jon_y: ^ is my structure, do you know why it causes fails? Jul 21 14:51:29 should be iMaps Jul 21 14:51:32 dunno if this line was sent: (16:45:51) (FIQ) the only sd-related thing I see is "Driver 'sd' needs updating - please use bus_type methods" Jul 21 14:51:34 FIQ: not sure Jul 21 14:51:47 kerio: that *should* solve email sync'ing Jul 21 14:51:53 FIQ: can you mount manually? Jul 21 14:51:56 Doc doesn't like my trolling, Doc must be an asshole Jul 21 14:51:57 yes Jul 21 14:51:58 I just did Jul 21 14:52:02 :c Jul 21 14:52:06 to re-check if what I remembered was true Jul 21 14:52:11 and it worked just fine Jul 21 14:52:23 i love you docky! <3 Jul 21 14:52:26 kerio: not that I pretend to make a lot of sense... Jul 21 14:52:39 StyXman_: people keep using git for everything Jul 21 14:52:53 FIQ: sorry, no idea Jul 21 14:52:56 if you just want to sync a directory, just use rsync or something like that Jul 21 14:53:05 kerio: yes, even home dirs... Jul 21 14:53:10 i.e. I did: mount -t vfat /dev/mmcblk1p1 /mnt && mount -t ext4 /dev/mmcblk1p2 /media/mmc1 Jul 21 14:53:13 kerio: even several GB size binary Jul 21 14:53:19 and they mounted fine Jul 21 14:53:22 Ok, oh well Jul 21 14:53:26 I can live with that error :p Jul 21 14:53:34 kerio: right Jul 21 14:53:48 previously, before the automounter script was around, I needed to mess with the automount script Jul 21 14:53:54 can't remember where it sits Jul 21 14:54:10 iirc just need to remove -t vfat Jul 21 14:54:25 let mount guess the fs Jul 21 14:54:44 the problem is that every time I omit -t vfat, I get "Invalid argument" Jul 21 14:54:56 get cssu! Jul 21 14:55:09 DocScrutinizer: I already have SSU.. Jul 21 14:55:20 s/SS/CSS/ Jul 21 14:55:20 FIQ meant: DocScrutinizer: I already have CSSU.. Jul 21 14:55:37 weird, without -t works for me the last time Jul 21 14:55:49 maybe I remembered wrongly Jul 21 14:57:00 http://wiki.maemo.org/Community_SSU/Changelog ctrl-f ke-recv Jul 21 14:57:26 oh Jul 21 14:57:30 4th jul Jul 21 14:57:32 I rarely update my system Jul 21 14:57:36 maybe that's why :D Jul 21 14:57:53 let's update cssu and see if it makes a difference Jul 21 14:58:04 LOL Jul 21 14:58:13 hm Jul 21 14:58:18 make sure to reboot just in case Jul 21 14:59:29 Your download starts in 5 seconds... Jul 21 14:59:33 Why not in 0 seconds? Jul 21 14:59:45 DocScrutinizer: I never want to touch the "program manager" except if there's a good reason for it, as it's snailspeed is extremly annoying Jul 21 14:59:45 Does it have no mentally prepare the server? Jul 21 15:00:17 *its' Jul 21 15:00:51 FIQ: well, I dunno if you might get away with apt-get upgrade Jul 21 15:00:57 I don't suggest it Jul 21 15:01:09 CSSU tested against HAM *only* Jul 21 15:01:11 DocScrutinizer: last time I did that with a CSSU update, it kinda half-updated or something Jul 21 15:01:27 I always upgrade, except for CSSU Jul 21 15:01:38 *I always apt-get upgrade Jul 21 15:02:09 you'll survive the unbearable delay, get a good cup of tea! ;-) Jul 21 15:02:31 it improves the suspense Jul 21 15:02:37 heh Jul 21 15:03:43 which remembers me.. I was going to install bounce evolution some days ago to have something to do when not having a connection on the phone... it failed after 50% progress on HAM (after taking like 30min) :< Jul 21 15:04:04 and it didn't really tell why except that it... well, failed Jul 21 15:04:26 HAM is large ham and slow Jul 21 15:04:43 no idea why it takes so long to wait between app installs Jul 21 15:07:20 jon_y: some partly nonsesical recheck of _all_ dependencies, plus a rebuild of that friggin heavy graphical scroll-list Jul 21 15:07:37 :( Jul 21 15:07:57 intern wrote it? Jul 21 15:08:06 :nod: Jul 21 15:08:38 if it has to go that route, make it at least take a checkout list or something Jul 21 15:08:53 so I don't have to wait for each install to complete before looking for the next app Jul 21 15:09:24 hey, if somebody really bothered: HAM is FOSS, you could augment it to have checkboxes to mark several packages as "install please", then do it all in one go Jul 21 15:09:27 shopping cart/check list or whatever its called Jul 21 15:09:39 actually there's already such a thing: in backup restore Jul 21 15:10:14 there's also such a thing in fapman, but it doesn't work for ovi stuff for obvious reasons Jul 21 15:10:16 * jonwil wonders if there is genuine interest in replacing icd or if people just like to bitch about how crap it is Jul 21 15:10:44 * DocScrutinizer05 several times thought about creating a generic backup list that just contains _all_ packages Jul 21 15:10:58 oh Jul 21 15:11:01 so you could select/deselect to your liking, then give HAM a go Jul 21 15:11:03 it's done finally Jul 21 15:11:12 just took 15 minutes Jul 21 15:11:15 ... to load the UPDATE LIST Jul 21 15:11:19 now let's update CSSU Jul 21 15:11:20 LOL Jul 21 15:11:33 um Jul 21 15:11:34 maybe you should disable some repos? Jul 21 15:11:44 (null) Install anyway? Jul 21 15:11:47 ^ huh? Jul 21 15:11:55 DocScrutinizer: maaybe Jul 21 15:11:56 :( Jul 21 15:12:12 ssu/mr0/extras-devel/community-testing Jul 21 15:12:26 and HAM is bearable Jul 21 15:12:40 kick extras-devel and you're fine Jul 21 15:12:43 erm Jul 21 15:12:49 "not enough battery" Jul 21 15:12:52 what is the requirement? Jul 21 15:12:55 i have 70% atm Jul 21 15:13:00 o.O Jul 21 15:13:10 use a charger? Jul 21 15:13:18 no Jul 21 15:13:21 broken microUSB Jul 21 15:13:27 the only charger I have is external Jul 21 15:13:29 never heard about requirements re battery in HAM Jul 21 15:13:43 YAy Jul 21 15:13:51 DocScrutinizer05: maemo without the extras repo sucks Jul 21 15:14:04 so let's see now Jul 21 15:14:17 I can't remove this requirement? Jul 21 15:14:20 kerio: shutup fool. I said extras-devel Jul 21 15:14:41 extras-devel is like extras, but stuff isn't ancient Jul 21 15:14:49 software in extras is *not* more stable Jul 21 15:14:54 no, it's also 30 times the size Jul 21 15:14:55 it should be, but it isn't Jul 21 15:15:14 I'm very sure that the battery will not drop before, and also, I've been able to update PRs before with 20% (through xterm) battery Jul 21 15:15:25 and nobody suggested you keep it disabled Jul 21 15:15:58 because if I can't, I see no way of properly upgrading CSSU if HAM is the most stable way Jul 21 15:16:46 FIQ: as mentioned before, I don't know of any such battery check in HAM (maybe that doesn't mean a thing though) Jul 21 15:16:59 DocScrutinizer: but there is one Jul 21 15:17:10 But not with regular updates Jul 21 15:17:14 Just with "system upgrades" Jul 21 15:17:20 you're sure you didn't try to flash a new kernel? Jul 21 15:17:22 and CSSU counts as one Jul 21 15:17:33 :nod: Jul 21 15:17:41 sounds about right Jul 21 15:18:07 anyway 70% should suffice Jul 21 15:18:32 i.e. when updating CSSU, there's a window like with PR updates, "Update for Maemo: blablablabla Create backup copy or continue?" and then a battery check Jul 21 15:18:45 hmmm Jul 21 15:18:49 that's when you're updating a package marked sys, or something Jul 21 15:19:00 and mp-community-fremantle-pr is one of those Jul 21 15:19:01 or whatever it's called Jul 21 15:19:05 never seen that, maybe because usually my battery is 100% when doing updates Jul 21 15:19:21 Does FCamera york with cssu-thumb? Doesn't work for me Jul 21 15:19:24 kerio: that's what FIQ said Jul 21 15:19:33 s/york/work/ Jul 21 15:19:38 qwazix meant: Does FCamera work with cssu-thumb? Doesn't work for me Jul 21 15:19:43 qwazix: quite probably not Jul 21 15:20:08 though freemangordon(?) clainmed it does Jul 21 15:20:26 qwazix: i think it's made to work with the stock kernel and/or the kernel-power from Pali ONLY Jul 21 15:20:34 yep Jul 21 15:20:35 if it works anywhere else, it's just because of a coincidence Jul 21 15:20:38 vi_: pong Jul 21 15:20:40 DocScrutinizer: so should I try to update with apt-get? Jul 21 15:20:44 zeq: ping Jul 21 15:20:56 Hi freemangordon Jul 21 15:21:00 FIQ: you should shut down, charge your battery, make a backup with backupmenu, and then update with HAM Jul 21 15:21:09 freemangordon! ^_^ Jul 21 15:21:13 freemangordon: should FIQ upgrade to CSSU-T latest with apt-get upgrade? Jul 21 15:21:20 zeq: how to make fennec link with --rpath? Jul 21 15:21:47 DocScrutinizer05: lemme see tha backscroll Jul 21 15:22:04 hm, okay are there any fcam drivers package specifically compiled for kernel-cssu that I should try? Jul 21 15:22:10 [UNOFFICIAL OPINION DONT TAKE SERIOUSLY]upgrading with apt-get should work, but reboot as soon as possible once you're done Jul 21 15:22:13 export LDFLAGS="-Wl,-rpath-link,$PWD/dist/bin/:/usr/lib:/lib -Wl,-rpath,/opt/qtm Jul 21 15:22:21 12/lib Jul 21 15:22:29 jon_y: I definitely have cssu installed, at least it's in the settings Jul 21 15:22:35 but ext2 doesn't get mounted Jul 21 15:22:45 zeq: that in debian/rules? Jul 21 15:22:47 dreamer: sorry, I really have no idea Jul 21 15:22:54 try ext3 for kicks? Jul 21 15:22:58 dreamer: which version? Jul 21 15:23:01 that's in mozconfig, just the extra -rpath part Jul 21 15:23:09 FIQ: backscroll too long, would you repeat the essencials? Jul 21 15:23:13 -Wl,-rpath,/opt/qtm12/lib Jul 21 15:23:14 kerio: An interesting thing is that I upgraded to PR1.2 with apt-get w/o problems (I got some strange bugs before reboot however, but that's expected) Jul 21 15:23:40 there's also a patch in our tree which deals with the fennec install directory rpath Jul 21 15:23:53 DocScrutinizer05: how can I see that? Jul 21 15:24:02 DocScrutinizer05: ah Jul 21 15:24:05 settings - about_cssu Jul 21 15:24:09 that gets added by configure since it knows where the install directory is going to end up Jul 21 15:24:12 DocScrutinizer05: 21.2011.38-1Smaemo3 Jul 21 15:24:32 zeq: so, I should puth LDFLAGS = $(LDFLAGS) + -Wl,-rpath,/opt/qtm12/lib Jul 21 15:24:43 in debian/tuls that is Jul 21 15:24:44 dreamer: should work aiui Jul 21 15:24:59 can do, as long as it isn't overridden by mozconfig Jul 21 15:25:28 dreamer: please /join #maemo-ssu and ping freemangordon there Jul 21 15:25:42 freemangordon: cannot update via HAM due to battery requirement (I have 70% atm), i have updated "normal" updates from Nokia through apt-get before just fine with 20%, but: (17:00:40) (FIQ) DocScrutinizer: last time I did that with a CSSU update, it kinda half-updated or something Jul 21 15:26:06 "I did that" as in "I did update CSSU with apt-get" Jul 21 15:26:17 freemangordon: I'm just about to take my dog for a walk, I'll take my N900, but will chat S L O W L Y :P Jul 21 15:26:29 DocScrutinizer05: hokay .. Jul 21 15:26:30 FIQ: so... charge your battery Jul 21 15:26:36 eventually you're going to *have* to Jul 21 15:26:41 FIQ: connect a charger Jul 21 15:26:44 unless it's 20% of a nuclear battery Jul 21 15:26:47 kerio: my external charger only goes to 90% Jul 21 15:26:52 freemangordon: no microUSB Jul 21 15:26:54 freemangordon: usb port is bork Jul 21 15:26:58 FIQ: so? Jul 21 15:27:03 FIQ: does not matter, just connect it and update via HAM Jul 21 15:27:04 RTBGs, last 50 years eaasily Jul 21 15:27:12 aah Jul 21 15:27:28 well, 20% seems too risky to me Jul 21 15:27:34 freemangordon: connect what? Jul 21 15:27:34 um huh? connecting with broken microUSB wont do a thing Jul 21 15:27:40 but.. it's 70% Jul 21 15:28:07 I just told that I've done fine with 20% before (so the battery shouldn't be a problem here, just apt-get vs HAM) Jul 21 15:28:24 if there's a battery check in HAM for system upgrades, it shouldn't trigger at >=70% Jul 21 15:28:29 maybe at 40% Jul 21 15:28:44 maybe 30 Jul 21 15:28:48 not 70 Jul 21 15:28:57 :nod: Jul 21 15:29:00 let's see if my other battery is charged yet Jul 21 15:29:08 what is the message from HAM? Jul 21 15:29:28 "Not enough battery - connect a charger and continue" or something Jul 21 15:29:29 verbatim Jul 21 15:29:44 if possible ;-) Jul 21 15:29:51 I've swedish locale at the moment Jul 21 15:29:57 I guess freemangordonwants to grep for it Jul 21 15:30:00 FIQ: and how many bars you have? Jul 21 15:30:11 on battery metter Jul 21 15:30:36 freemangordon: I can't count them, but it's over half Jul 21 15:30:59 hmm, did you try to reboot? Jul 21 15:31:02 sorry for the lack of info here, but the battery meter doesn't say much Jul 21 15:31:08 yeah, the bars on battery icon are no bars actually ;-P Jul 21 15:31:36 freemangordon: No, but I've come to a new conclusion - just wait at the moment until my other battery is charged fully and try again Jul 21 15:31:41 Do you know of any mail-apps that could replace modest? Jul 21 15:31:42 If it still errors, i'll come back Jul 21 15:32:03 FIQ: I'd guess your previous run with exploding HAM might have messed up things. So a reboot might be fine Jul 21 15:32:19 kgu: microB :P Jul 21 15:32:24 btw, anybody know why my nokia about page says version unknown? Jul 21 15:32:41 its a pr1.3 system Jul 21 15:33:12 kerio: Not really a mail-app.. :) Jul 21 15:35:27 Um, the battery indicator changed its' mind after reboot, so now HAM would definetely complain Jul 21 15:35:35 I've created new tarballs for the linaro-4.6/7 compilers http://www.snewbury.org.uk/maemo Jul 21 15:35:38 now it's more like 40% Jul 21 15:35:42 so I guess I'll just wait Jul 21 15:35:48 until my other battery is charged Jul 21 15:39:54 zeq: how we are supposed to use that? Jul 21 15:40:06 jon_y: you messed a lot with kernels and stuff? Jul 21 15:40:16 hmm Jul 21 15:40:42 speaking of updates, why did nokia release a PR1.3.1? Jul 21 15:40:52 cert issue Jul 21 15:41:00 security update Jul 21 15:41:14 DocScrutinizer05: well, it is flashed with uboot Jul 21 15:41:26 loaded with kp50 Jul 21 15:41:34 ah Jul 21 15:41:36 that might be the reason Jul 21 15:41:58 loading a kernel that is different from flash will give you that? Jul 21 15:42:10 not exactly, but related Jul 21 15:42:39 it's not all clear what this closed lib to gather all the systeminfo does Jul 21 15:43:02 ok, what happened? nobody knows exactly? Jul 21 15:43:28 Nokia would know Jul 21 15:44:02 search tmo for "CAL" and "version unknown" Jul 21 15:44:24 tmo? Jul 21 15:44:38 ~tmo Jul 21 15:44:41 it has been said that tmo is http://en.wikipedia.org/wiki/TMO, or http://talk.maemo.org, or http://de.wikipedia.org/wiki/Terrestrial_Trunked_Radio#TMO. It's *not* T-MO (see ~T-MO) or trolls, morons, oxen. Jul 21 15:44:54 ok Jul 21 15:45:10 ~CAL Jul 21 15:45:11 from memory, cal is a calendar. try $(cal 1752) Jul 21 15:45:24 freemangordon: just unpack into /scratchbox/compilers Jul 21 15:45:29 so CAL calendar? Jul 21 15:45:47 search for "What is CAL?" in tmo Jul 21 15:46:03 this time it *should* have the right kernel headers Jul 21 15:46:03 ok Jul 21 15:46:05 CAL==config Jul 21 15:46:13 cat /proc/mtd Jul 21 15:47:26 iirc that lists some memory layout Jul 21 15:48:20 freemangordon: are you hinting I need to write that howto? Jul 21 15:49:14 I feel more confident doing so without having to mess with broken kernel headers. :) Jul 21 15:51:46 DocScrutinizer05: will the version unknown cause any issues? Jul 21 15:52:02 jon_y: not afaik Jul 21 15:52:28 ok, so its just cosmetic, not too bad Jul 21 15:53:53 zeq1: well, actually no, but it will be a good idea actually to have a howto in wiki Jul 21 16:03:52 ok, let's see how this goes with this charged battery Jul 21 16:15:30 hi. are there any good tutorials how to create maemo apps out there? i want to start creating some stuff for n900 Jul 21 16:17:14 qpx: http://maemo.org/intro/developer_overview/ Jul 21 16:17:21 * jonwil wonders if anyone actually cares about replacing closed-stuff with open-stuff or they just keep complaining about how crap the closed source bits are without being willing to do anything about it... :P Jul 21 16:17:58 Maybe we are just not that skilled :) Jul 21 16:18:05 jonwil: those that complain are not doers, just moaners ;) Jul 21 16:18:38 the question then is, are there any doers out there or am I the only one left who actually wants to do REAL work on this stuff? Jul 21 16:18:53 freemangordon: thanks Jul 21 16:19:02 or just moaning so those who know what to do know that something is to be done :) Jul 21 16:19:34 jonwil: why only you? remember camera-ui , OMP, adeclock, etc,etc Jul 21 16:19:46 hmmm true Jul 21 16:19:51 * jon_y setting up scratchbox soon-ish to screw with n900, but not too soon, busy irl is busy :( Jul 21 16:20:02 not to mention the amount of work Pali did on bq-* stuff Jul 21 16:21:28 qpx: you are welcome Jul 21 16:22:50 jonwil: you were talking about the icd stuff earlier, what does mer use for network management? wouldn't it be easier to port something that works, rather than re-implement the whole thing? Jul 21 16:23:12 yeah I wasnt suggesting writing something new Jul 21 16:23:48 what I was saying is that we need to come up with something that talks to the same dbus interfaces, gconf keys etc that are expected by other parts of the system Jul 21 16:23:53 specifically the bits we cant easily change Jul 21 16:25:10 I think thats why Jul 21 16:25:35 I think 'that's why things like camera-ui, new clock etc have been attractive for devs is that they can be done without breaking stuff Jul 21 16:26:11 https://p.twimg.com/AyVoeE2CIAECLM_.jpg:large Jul 21 16:26:30 out cycling with people, including 3iphones and 2 galaxy s2's Jul 21 16:26:32 jonwil: but your question re BME is applicable here as well, for what purpose you want to reimplament icd? Jul 21 16:26:36 guess which poor N900 was doing all the navigation? Jul 21 16:27:17 Lava_Croft: Which app did you use? Jul 21 16:27:19 well lots of people have said ICD sucks, also by replacing it with something open it can support things like some of the fancy WiFi security bits that the stock stuff doesn't do Jul 21 16:27:23 kgu: mappero Jul 21 16:27:36 i almost exclusively use mappero Jul 21 16:27:46 sometimes marble, but i cant really get around its gui Jul 21 16:28:20 now, stick a control system on it and have the phone steer for you :) Jul 21 16:28:42 mappero better than modrana? Jul 21 16:28:58 i think i saw modrana for 5 seconds Jul 21 16:28:59 none of the map apps ever work for me :( Jul 21 16:29:02 then decided this GUI is not for me Jul 21 16:29:11 ok Jul 21 16:29:16 should i try it? Jul 21 16:29:20 like, give it a chance? Jul 21 16:29:29 mappero isnt really updated anymore, iirc Jul 21 16:29:43 Lava_Croft: does mappero require 3G to operate? Jul 21 16:29:44 maybe, I can't really decide which one I prefer Jul 21 16:29:59 maep is good Jul 21 16:30:06 so yeah replacing ICD is useful for much the same reasons as replacing BME is useful Jul 21 16:30:11 but it doesn't do voice navigation Jul 21 16:30:15 or... navigation, i think Jul 21 16:30:27 hey, it worked! Jul 21 16:30:27 jon_y: depends Jul 21 16:30:35 3g is handy if you didnt pre-download maps Jul 21 16:30:45 and now my SD card mounts properly too :p Jul 21 16:30:47 ah ok, it can predownload Jul 21 16:30:50 yeah Jul 21 16:30:53 not localized though Jul 21 16:30:55 and it does voice nav Jul 21 16:30:56 its just that everyone likes to complain about the suckiness of ICD but no-one except me has shown any interest in actually doing anything about it :P Jul 21 16:31:18 "14,4 gb volume" Jul 21 16:31:19 the last time I tried, modrana doesn't do predownloads Jul 21 16:31:24 aw Jul 21 16:31:37 i think Marble can pre-download Jul 21 16:31:42 no 3G/data plans make it kind of useless Jul 21 16:31:56 well, yeah Jul 21 16:32:05 but no 3g/data plan would make my entire n900 pretty useless Jul 21 16:32:28 nope, I live off wifi for connectivity :) Jul 21 16:32:40 occasionally over USB rndis Jul 21 16:32:46 mostly wifi and 3g here Jul 21 16:32:48 im on the road a lot Jul 21 16:32:52 actually Jul 21 16:32:53 FredrIQ|n900: hey!! :-D Jul 21 16:32:55 90% is 2g Jul 21 16:33:04 i can stream music fine over 2g Jul 21 16:33:10 32kpbs streams:) Jul 21 16:33:14 kbps Jul 21 16:33:25 I keep my phone in 2G most of the time (i.e. unless I am doing heavy data stuff) as it saves battery usage Jul 21 16:33:30 aye Jul 21 16:33:31 same Jul 21 16:33:39 i switch to 3g only when i feel i need it Jul 21 16:33:42 strange that I can get 3G signals in rural areas, but data plans are expensive and idiotic Jul 21 16:33:50 damn packages Jul 21 16:33:59 i get 3g fine throughout entire NL Jul 21 16:34:02 but its a small, flat country Jul 21 16:34:10 urban areas are worse for 3g here, old buildings ec Jul 21 16:34:15 they block out everything Jul 21 16:34:42 yeah, but enough microcells around that signal strength isn't really an issue Jul 21 16:35:17 3g coverage in NL is ridiculous for such a shit tiny country Jul 21 16:35:39 a lot of 'dead' spots Jul 21 16:36:18 I get 3G signal out in rural areas, with wooden huts and stuff around :| Jul 21 16:36:32 no, I don't have a data plan Jul 21 16:36:46 so its left to emergency use only Jul 21 16:36:54 aw Jul 21 16:37:07 if i had no data plan, i would just put a 1650 in my pocket Jul 21 16:37:16 I've rarely *no 3G at all*, but in areas like this (in the middle of nowhere) it can be a bit ridiculous @stability Jul 21 16:37:24 freemangordon: upgraded to latest testing cssu-thumb. It totally rules. Thank you. Jul 21 16:37:51 as seen in my nice d/c's all the time :3 Jul 21 16:38:10 dont trust 3g for constant connectivity Jul 21 16:38:11 ever Jul 21 16:38:16 trust 2G with that:P Jul 21 16:38:23 Lava_Croft: i have dual Jul 21 16:38:29 i dont like dual Jul 21 16:38:37 it just annoys me Jul 21 16:38:44 so it *should* switch if it miss 3g, shouldn't it? Jul 21 16:38:49 yes Jul 21 16:39:00 it does so too, for me Jul 21 16:39:05 which makes my audio stream disconnect Jul 21 16:39:11 idiot packaging gives you expensive data plans with crappy caps, in addition to ridiculously postpaid voice/sms plans Jul 21 16:39:12 I've never got 2g at all here Jul 21 16:39:22 **ridiculously expensive Jul 21 16:39:23 either 3g or nothing, constantly swapping Jul 21 16:39:24 i have about 1gb of data shoving per month Jul 21 16:39:29 unlimited sms Jul 21 16:39:32 and 100 minutes of calling Jul 21 16:39:38 i got a galaxy nexus with it, which i sold right away Jul 21 16:39:43 so i could get N9 Jul 21 16:40:09 why didn't you get another n900? :c Jul 21 16:40:10 you traitor Jul 21 16:40:13 i alreayd have two Jul 21 16:40:19 and n9 is cute Jul 21 16:40:23 DocScrutinizer05: holy shit you weren't lying about extras-devel being huge Jul 21 16:40:25 it is! Jul 21 16:40:30 yup! Jul 21 16:40:37 and theres a lot of shit in it! Jul 21 16:40:39 kerio: was that news to you? :P Jul 21 16:40:55 yes im being called a traitor by someone who is also amazed at the size of extras-devel Jul 21 16:40:58 oh god Jul 21 16:41:00 i guess its time to go do something else Jul 21 16:41:03 o/ Jul 21 16:41:19 DocScrutinizer05: does maemo create a DCIM directory in fat32 partitions? Jul 21 16:41:21 my n9 comes with angry birds and Need For Speed Jul 21 16:41:27 jon_y: mine too Jul 21 16:41:30 and golf lol Jul 21 16:41:35 first things to remove:) Jul 21 16:41:36 lol yeah that too Jul 21 16:41:52 why would you remove angry birds? :o Jul 21 16:41:58 its boring? Jul 21 16:42:06 no interest in it Jul 21 16:42:17 hm golf Jul 21 16:42:22 the golf game is annoying too Jul 21 16:42:32 my previous phone came with some golf app Jul 21 16:42:36 i think the only N9 games that i actually play is Pocket RPG Jul 21 16:42:41 game* Jul 21 16:43:04 now if it came with Nokia 2/3xxx snake game... :) Jul 21 16:43:18 the 3d appearence was quite impressive for such a phone (se p1i), but I didn't play it more than to try Jul 21 16:43:45 jon_y: :P Jul 21 16:44:27 the N9 kind of feels weird to use, if you're dealing with call menus Jul 21 16:44:40 no hard keys make it awkward :( Jul 21 16:45:06 I guess I'll get that too with any andoid/iphone out there Jul 21 16:45:49 how do you guys deal with call menus? repeatedly press 0 until the operator shows up? Jul 21 16:46:49 can anything bad happen if i remove osso-games-* Jul 21 16:46:49 ? Jul 21 16:47:18 well, the trend to have less and less hw-buttons is insane Jul 21 16:47:42 * FIQ|n900 doesn't like that trend Jul 21 16:47:44 I *hate* shooting photos with a touchscreen trigger Jul 21 16:47:44 I for one do NOT want a phone without a hardware keyboard Jul 21 16:48:02 but I fear that I wont have a choice when I am shopping for my next phone Jul 21 16:48:11 jonwil: there are no phones with hw keyboard Jul 21 16:48:19 jonwil: that were released in last year Jul 21 16:48:24 jonwil: I will never get one, unless the number of alternatives hits 0 Jul 21 16:48:50 hm, i thought the droid 3 had a bunch of keys Jul 21 16:48:55 jacekowski: that's evil :< Jul 21 16:49:08 I think the one company that has made the most phones with hwkbd is RIM Jul 21 16:49:16 most BlackBerries seem to have hardware keyboard Jul 21 16:49:23 jacekowski: yep, you're lying Jul 21 16:49:34 no, no you're not D: Jul 21 16:49:41 the droid 3 was released on july 14 2011 Jul 21 16:50:16 hm, the droid 4 has a hw kbd Jul 21 16:50:21 so yeah, jacekowski: you're wrong Jul 21 16:50:48 some manufacturer should realize that they could do some money of hw-kbd, it would be kinda unique this day.. Jul 21 16:51:07 *off Jul 21 16:51:12 I was told that you're supposed to use bluetooth keyboards Jul 21 16:51:21 bluetooth keyboards suck Jul 21 16:51:25 awkward is still awkward Jul 21 16:51:31 you cant use em when you have no surface to put them on Jul 21 16:51:35 my hopes for jolla doing that got kinda crushed Jul 21 16:51:59 DocScrutinizer05: huh? Jul 21 16:52:07 bt keyboards are fine if you have 3 arms and 3 hands Jul 21 16:52:19 * FIQ|n900 also has hopes for that, but.. "crushed"? Jul 21 16:52:26 sometimes, I miss my old nokia series60 candy bar when dealing with call menus Jul 21 16:52:29 smashed, whatever Jul 21 16:52:35 my guess is that with more and more people shopping for phones based on looks rather than usage, hardware keyboards are no longer in vogue Jul 21 16:52:37 nuked Jul 21 16:53:06 jonwil: I guess most people don't deal with call menus Jul 21 16:53:10 DocScrutinizer05: oh yeah i get the definition.. but why? Jul 21 16:53:11 initially tales were jolly will build two devices, one for dumbass users and one for geeks like us Jul 21 16:53:22 yeah Jul 21 16:53:25 call menus are hard enough as it is on my N900 :) Jul 21 16:53:33 agreed Jul 21 16:53:36 then some jolla dude corrected that Jul 21 16:54:28 and said that they'll only make the dumbass users one? Jul 21 16:54:40 DocScrutinizer: has that changed? Jul 21 16:54:55 [2012-07-21 18:53:36] then some jolla dude corrected that Jul 21 16:55:03 to? Jul 21 16:55:14 what kerio said? Jul 21 16:55:16 only one device, with a developer mode Jul 21 16:55:34 I went "WAAAHHAAAHAAAAAAAAAAA! aeggis!" Jul 21 16:55:36 hm... Jul 21 16:55:46 a real developer mode, or pretend like N9? Jul 21 16:55:57 I want a phone with real ethernet jacks :) Jul 21 16:56:03 allegedly a real one, according to stskeeps Jul 21 16:56:11 and I guess that means, N9 style no-hw-kbd? Jul 21 16:56:23 but my hopes for decent hw kbd are mostly gone now Jul 21 16:56:26 hrmpf, the droid 4 has no umts Jul 21 16:56:53 I'm not sure I'd trust the "real" part until I see it :P Jul 21 16:57:14 * DocScrutinizer05 throws a USB->ethernet adapter in jon_y's general direction Jul 21 16:57:15 * FredrIQ|n900 will not express his feelings right now Jul 21 16:57:35 they would just spam this channel Jul 21 16:57:50 :nod: Jul 21 16:57:51 * FredrIQ|n900 afk Jul 21 16:57:52 DocScrutinizer05: no, USB ethernets don't go 1Gbps Jul 21 16:58:07 jon_y: how convenient, the n900 couldn't do anything remotely useful with 1gbps Jul 21 16:58:08 neither does your average SoC Jul 21 16:58:12 heh Jul 21 16:58:25 yeah, not yet Jul 21 16:58:53 file transfer would be at least pretty fast Jul 21 16:58:55 hell, it couldn't even STORE that crap anywhere Jul 21 16:59:05 neither deliver Jul 21 16:59:05 fast enough to saturate the fastest SD cards Jul 21 16:59:49 and not be bogged down by CPU usage on the host Jul 21 17:00:05 well, its not happening in reality Jul 21 17:00:21 * jon_y goes of dreaming again Jul 21 17:01:43 goddamn Jul 21 17:01:49 n900 still makes such nice photographs Jul 21 17:06:12 till they deliver, joll is vapourware. Jul 21 17:06:26 Tech companies have hurt my feelings to many times. Jul 21 17:06:33 I cannot trust anymore. Jul 21 17:06:38 :'( Jul 21 17:06:49 ^too Jul 21 17:07:22 vapourware is a term that is generally not used for stuff like Jolla Jul 21 17:07:38 which is just announced, which sounds rather feasible and which hasnt passed any release dates or whatever, yet Jul 21 17:08:17 expectations on the other hand Jul 21 17:08:17 preorder scam? Jul 21 17:08:18 dont trust those Jul 21 17:10:42 well, announcements had: one device til end of this year. Targeted at both mass market (aka dumbass) and geek users. With some sort of developer mode, which means they will put truszone[TM] to purpose, including the whole broken chain-of-trust concept that is TPM Jul 21 17:11:10 wait, what's broken in the concept of TPM? Jul 21 17:11:11 actually, they never said that a device is promised this year Jul 21 17:11:14 let alone two devices Jul 21 17:11:27 they even emphasized that they make no promises Jul 21 17:11:28 over and over Jul 21 17:11:37 but their *goal* is a device this year Jul 21 17:13:28 ~tell kerio about aegis Jul 21 17:13:48 more like tell kerio about Trusted Computing Jul 21 17:13:55 DocScrutinizer05: that's not the concept of tpm, that's a particular implementation of tpm Jul 21 17:14:03 and i personally dont oppose Aegis at all Jul 21 17:14:04 i'm all for tpm where i'm the one with the key to my device Jul 21 17:14:13 sadly, that doesn't really happen Jul 21 17:14:16 you need something like Aegis in the dumbuser market Jul 21 17:14:35 and to pander to the service providers Jul 21 17:14:46 thats included in 'dumbmarket', but yeah Jul 21 17:14:58 so they can flood you with the latest angry birds promo Jul 21 17:15:31 that's nonsense Jul 21 17:16:08 ~aegis Jul 21 17:16:08 http://www.developer.nokia.com/Community/Wiki/Harmattan:Developer_Library/Developing_for_Harmattan/Harmattan_security/Security_guide , or "The purpose of this framework is: ... to make sure that the platform meets the requirements set by third party software that requires a safe execution environment.", or http://en.wikipedia.org/wiki/Trusted_Computing#Criticism, or http://en.qi-hardware.com/w/images/1/10/ME_382_LockedUpTechnology2.gif Jul 21 17:16:24 "The purpose of this framework is: ... to make sure that the platform meets the requirements set by third party software that requires a safe execution environment." is the key Jul 21 17:16:34 im actually impressed by that whole Aegis thing, that they actually tried to pull it off Jul 21 17:16:49 jolla will need something better protected than aegis tho! :) Jul 21 17:16:50 for *evrything* else there are better more mature less prone to abuse solutions already Jul 21 17:16:55 yep, those dancing angry birds ads on top of the screen Jul 21 17:17:11 safe from you Jul 21 17:17:16 Everybody else always knows a better solution to the problem(s) you solved Jul 21 17:17:18 safe from being deleted Jul 21 17:17:25 but rarely, those people actually create the solution they advertise Jul 21 17:17:36 *mostly*, they just whine about the solution you chose to use Jul 21 17:18:30 chain of trust is a fucked up concept Jul 21 17:18:53 God, computers and mobile phones are full of fucked up concepts Jul 21 17:18:58 since the root mandatorily isn't at user Jul 21 17:19:19 i never ever can understand all the fuss that people can create over stuff like this Jul 21 17:19:36 and im not trying to ridicule you in any way Jul 21 17:19:46 sure Jul 21 17:20:02 i can get the general idea of it, and how its inherently broken and bad Jul 21 17:20:06 but really, why bother Jul 21 17:20:26 are you maybe in the wrong channel? shouldn't you enjoy walled garden over at #harmattan? Jul 21 17:20:39 i love my N900 a whole lot more than my N9 Jul 21 17:20:48 and i can already see the attitude that spring up in so many places like this Jul 21 17:20:52 zeq: i built fennec (cloned from github) but it is very slow compared to previous builds, any clue? Jul 21 17:20:57 and again you are the source of such an attitude Jul 21 17:21:30 and you seem to be the source of fuzzy grunt Jul 21 17:22:24 if my attitude doesn't please you... well, I couldn't bother less, esp since you seem to content with fuzzy BS statements Jul 21 17:22:30 but why make a remark like "go to #harmattan" Jul 21 17:22:32 i mean really Jul 21 17:22:47 apart from it not being a really productive argument Jul 21 17:22:53 you are also trying to actively shit on other folks Jul 21 17:23:03 in this case #harmattan, which are nice people and which is a nice phone with nice software Jul 21 17:23:08 [2012-07-21 19:20:25] are you maybe in the wrong channel? shouldn't you enjoy walled garden over at #harmattan? Jul 21 17:23:09 so what is your problem really Jul 21 17:23:13 please reread! Jul 21 17:23:26 its a broken statement Jul 21 17:23:39 it's no statement at all Jul 21 17:23:42 ETX Jul 21 17:23:54 I find aegis a broken statement - tho I like the idea of a security framework Jul 21 17:23:57 its a statement that disguided as question Jul 21 17:24:14 i dont oppose aegis, but enjoying it is the other side of the spectrum Jul 21 17:24:40 poor animosity Jul 21 17:24:45 then again I've argued over the need to be in the users hand - not under a certificate that walls the developing Jul 21 17:25:05 i would love to place all the power in the user's hands Jul 21 17:25:09 thats how it should be Jul 21 17:25:37 but practice proves over and over that the majority of people just cant and dont want to deal with such freedom Jul 21 17:25:40 you can't be creative with your linux if it's behind a dozen closed doors Jul 21 17:25:54 closed doors are an incentive to be creative, arent they? Jul 21 17:26:19 its not like the closed door on the n9 made it really hard to enter through the window Jul 21 17:26:28 the freedom to develop apps that could kick ass > the ability to make 100 different camera apps Jul 21 17:26:46 just because you can't Jul 21 17:26:58 the majority of people just dont care Jul 21 17:27:02 they just want 'shit to work' Jul 21 17:27:08 i cant help it:< Jul 21 17:27:35 but they CARE for the apps provided (the "ecosystems" people tend to em about) Jul 21 17:27:49 but not as much as you would think Jul 21 17:28:04 and most of the time shit does work on thos walled garden systems Jul 21 17:28:08 uhh Jul 21 17:28:10 no matter how wrong they are in so many ways Jul 21 17:28:17 I updated CSSU recently Jul 21 17:28:19 that's the only pitfall I've been hearing about - with non-tech people Jul 21 17:28:23 everything worked etc Jul 21 17:28:24 they don't have the apps Jul 21 17:28:25 but uh Jul 21 17:28:40 now when checking what I have left to update Jul 21 17:28:46 CSSU still show up Jul 21 17:29:05 why? Jul 21 17:29:21 much like the inability to get a "rare" opportunity to work over VPN Jul 21 17:30:05 i own n900 because i get immensely annoyed by android, ios whatever Jul 21 17:30:06 every fucking other phone gives you this - if we're not talking about S40-series symbian Jul 21 17:30:16 because they limit my freedom to do things the way i want Jul 21 17:30:26 but i rarely ever come across anyone in real life who really cares about such stuff Jul 21 17:30:34 and if i do, its a nerd Jul 21 17:30:57 so why would a company thats trying to be a success in the mobile phone world have to care about user freedom Jul 21 17:31:13 business wise, its just not really that big of an issue if its open or not Jul 21 17:31:34 I'm talking about businesses adopting any "smartphones" Jul 21 17:31:46 emphasis on "" indeed Jul 21 17:32:05 at this point a 10million eur business Jul 21 17:32:39 what's with walled gardens being used a lto :P Jul 21 17:32:44 if one can't do a simple module to enable such communication - it's dropped already Jul 21 17:33:13 hundreds of devices left in the stock Jul 21 17:33:16 ~lto Jul 21 17:33:21 a lot Jul 21 17:33:39 I' Jul 21 17:34:19 d enjoy a device that has the things needed - no the things you could do, but you can't because you're not a nokia developer x) Jul 21 17:35:23 a simple loop/tap device needed + a few dbus permissions -> all is there but no way to use = poor shit Jul 21 17:36:58 jabis: and still that's the whole concept of chain-of-trust Jul 21 17:38:30 DocScrutinizer05 well not exactly - usually the superuser-level is in the hands of user - not an outside authority Jul 21 17:39:22 no way Jul 21 17:39:52 it starts with there's no such role as superuser in that concept. There isn't _any_ roles Jul 21 17:41:18 I'm talking about authority - not privileges Jul 21 17:42:30 thus authority has superceding control over the chain of trust - and authority being in hands of user - not distributor Jul 21 17:43:08 that's not chain of trust Jul 21 17:43:33 that's age old permissions Jul 21 17:45:08 no - chain of trust relies for example some certificates - but authority is the user accepting the certs Jul 21 17:45:23 i will probably never understand why locking devices down is said to protect users :/ Jul 21 17:45:32 what nokia enforces is more a monopoly than a chain of trust Jul 21 17:45:33 DocScrutinizer05: here is an OT question. I want to deply a ftp server in a place with lot of power outages (no backup UPS). a) Is it possible to build a read only linux root. b) Is SSD a better choice than HDD for power failures? Jul 21 17:46:38 Monopoly -> A single CA acting also as RA -> One key to rule them all... -> All signing is done with this single key Jul 21 17:46:44 * trumee would have preferred a router running openwrt. but these things dont have the cpu grunt to do transcoding Jul 21 17:47:39 DocScrutinizer05: the server will be a mythtv server actually Jul 21 17:49:25 chain of trust paradigm is much more hard to grasp Jul 21 17:49:57 trumee: it's definetely possible to build a read only linux root Jul 21 17:50:06 and if the system is read only, it should make no difference Jul 21 17:50:33 kerio: is hdd prone to failure on power outage comapred with ssd? Jul 21 17:50:33 so i'd go ssd Jul 21 17:50:48 ssd's problems are with writing, not reading Jul 21 17:50:54 so... sure, i guess Jul 21 17:50:57 zeq1: dammit, fennec nightly scores 8300 ms vs 6133ms for previous builds, WTF? Jul 21 17:51:16 zeq1: sunspider that is Jul 21 17:51:25 ssd is still a mess - I'd go with hdd Jul 21 17:51:37 perhaps a raptor ^^ Jul 21 17:52:06 jabis: it's going to be a read-only system Jul 21 17:52:19 trumee: hell, go with something even smaller and easily replaceable Jul 21 17:52:20 kerio: the only problem is that i need a mysql server running, and that will need a rw fs :( Jul 21 17:52:21 like a sd Jul 21 17:52:37 so a raptor Jul 21 17:52:53 trumee: how much space do you need? Jul 21 17:53:33 kerio, read-only or not - I still feel hdd more solid than ssd in longevity Jul 21 17:55:19 kerio: few gb i would have guessed. Jul 21 17:55:43 jabis: the machine is going to see about random 5-6 power cycles each day Jul 21 18:15:21 trumee: use a hdd with a journaling fs (ext3/4 for example) Jul 21 18:16:45 trumee: if you wanna do even more: there are PSU with some 2 or 5 seconds backup via capacitors Jul 21 18:17:04 a proper linux should be able to emergency shutdown in that time Jul 21 18:17:15 I.E. do sync Jul 21 18:17:29 DocScrutinizer05: this PSU is interesting, link? Jul 21 18:17:44 no, sorry Jul 21 18:18:06 DocScrutinizer05: i have had corruption with reiserfs before due to power failure Jul 21 18:22:29 zeq: ping Jul 21 18:22:57 reiser isn't a popular choice for concurrency Jul 21 18:23:48 jabis: this was about 10 years ago Jul 21 18:24:26 jabis: i dont have much expectations on journalling fs since then Jul 21 18:24:39 I'm talking about 2012 - reiser isn't a good FS choice for what you're talking about Jul 21 18:25:18 * ShadowJK imagines a ups with that psu is even better Jul 21 18:25:34 but people tend to forget to change the batteries every 2 years Jul 21 18:25:55 ShadowJK: i have using a APC ups and have had lot of problems Jul 21 18:26:21 ShadowJK: these UPS are not built for an environment which sees 5-6 outages each day Jul 21 18:26:35 ShadowJK: the batteries die away very quickly Jul 21 18:26:56 so i would rather hook the system to the mains rather than go with a UPS Jul 21 18:27:48 The "entry level" or "small office" UPSen dont work with active pfc psus very well Jul 21 18:28:33 It's important to never deplete the batteries, so you need to dimension them according to that, and do quick shutdown of computer Jul 21 18:28:57 this is what i have, http://www.apc.com/products/resource/include/techspec_index.cfm?base_sku=SUA1000UXI&total_watts=50 Jul 21 18:29:10 ShadowJK: it is a sine-wave ups for active pfc Jul 21 18:29:59 how long do batteries endure? Jul 21 18:29:59 ShadowJK: buying an active pfc psu was a mistake i commited. should have used a passive psu, i could have used an entry level ups Jul 21 18:30:07 ShadowJK: 2 months Jul 21 18:30:10 wow Jul 21 18:30:13 hot? Jul 21 18:30:18 ShadowJK: exactly! Jul 21 18:30:28 how hot? Jul 21 18:30:39 ShadowJK: 45+ Jul 21 18:30:46 ShadowJK: 45C + Jul 21 18:31:21 ShadowJK: that is the peak temperature though Jul 21 18:31:34 ah that explains it Jul 21 18:32:37 so i am seeking an alternative of a readonly fs and get rid of UPS Jul 21 18:33:00 only problem i have is a damn mysql server i need Jul 21 18:33:36 laptop? Jul 21 18:33:40 ;P Jul 21 18:33:59 though battery there would die in 6 months too Jul 21 18:35:09 ShadowJK: is there no smart mysql which keeps a copy of itself, and uses a previous copy if the current one is corrupt Jul 21 18:35:35 * trumee doesnt care much about the mysql data loss since it is storing only the tv program guide Jul 21 18:35:44 *sql* and smart rarely go together Jul 21 18:37:35 ShadowJK: how about managing db backups using a script? rootfs copies over the db on each bootup Jul 21 18:38:10 ShadowJK: and saves db every hour Jul 21 18:38:40 also, is it possible to have the partition/usb drive with db to have no cache Jul 21 18:39:01 that way everything will be in sync to minimise loss Jul 21 18:43:24 trumee: actually a laptop is a smart alternative Jul 21 18:44:51 trumee: you just should make sure the battery doesn't get too warm Jul 21 18:46:03 DocScrutinizer05: the server hosts a few tv cards. should i use laptop just for the rw fs and server for tv cards and readonly fs Jul 21 18:47:19 hmmm Jul 21 18:49:53 i'm in seattle Jul 21 18:50:05 connecting.. to get to the rough and tough mean streets of pasco washington lol Jul 21 18:51:44 blah Jul 21 19:06:28 trumee: if your UPS batteries get too warm, consider placing them outside the UPS case Jul 21 19:08:21 freemangordon: pong Jul 21 19:08:30 freemangordon: I was out having dinner :) Jul 21 19:09:35 freemangordon: I'll rebase again and push if it helps Jul 21 19:09:59 trumee: however I think most important thing is to shut down PC and UPS quickly Jul 21 19:10:48 I don't get it. I tried to flash the system to fix the infinity rebooting problem. The problem is that the Flasher program just starts and closes instantly. Jul 21 19:11:07 ~flashing Jul 21 19:11:08 i heard maemo-flashing is http://wiki.maemo.org/Updating_the_tablet_firmware Jul 21 19:11:25 Kiborg: make sure you are running a proper version of flasher Jul 21 19:11:27 Kiborg: the flasher program doesn't "open" and "close" Jul 21 19:11:44 it's a commandline program Jul 21 19:11:44 you also have to run it as administrator or root from the prompt Jul 21 19:11:48 or commandline Jul 21 19:12:04 DocScrutinizer05: what ever happened to the windows program that did it all fo ryou? :) Jul 21 19:12:13 like.. next next next flash done Jul 21 19:12:23 dafaq, I thought that been a hoax? Jul 21 19:12:27 you can't even find that naymore Jul 21 19:12:41 DocScrutinizer05: huh it was a nokia app Jul 21 19:12:57 the one that came with the N900 and N810... it downloaded the update and flashed it Jul 21 19:13:02 well, a windows app. So NFC here Jul 21 19:13:07 it opens in cmd doesn't it? Jul 21 19:13:11 LOL Jul 21 19:13:22 Kiborg: flasher has to be ran manually Jul 21 19:13:30 so yes... you open cmd Jul 21 19:13:39 then run it Jul 21 19:14:38 http://wiki.maemo.org/Updating_the_tablet_firmware#Windows 6. Jul 21 19:15:12 ugh Jul 21 19:15:18 stuck waiting for my airplane Jul 21 19:15:35 boarding time is 1:05pm pst Jul 21 19:15:49 50 more minutes before i can boad Jul 21 19:18:59 For EU I take global one? Jul 21 19:19:21 i think nowadays you pick global for everything Jul 21 19:19:32 ok Jul 21 19:20:29 * DocScrutinizer05 cackles comparing length of windows related instructions vs linux related, on http://wiki.maemo.org/Updating_the_tablet_firmware Jul 21 19:23:18 dang it MUST be flashing weekend or somesuch Jul 21 19:23:26 hehh Jul 21 19:23:31 i still have to flash my n810 Jul 21 19:23:44 i don't even know why Jul 21 19:24:03 i was going to toss it in a car to do something with it i suppose Jul 21 19:24:12 with an bt obd2 reader Jul 21 19:24:19 an/a Jul 21 19:24:30 obd2? Jul 21 19:25:14 :P Jul 21 19:25:40 o.O Jul 21 19:25:47 ~wiki obd2 Jul 21 19:25:51 At http://en.wikipedia.org/wiki/Obd2 (URL), Wikipedia explains: "{{Cleanup|date=July 2009}} {{Refimprove|date=June 2009}} 'On-Board Diagnostics', or 'OBD', in an automotive context, is a generic term referring to a vehicle's self-diagnostic and reporting capability. OBD systems give the vehicle owner or a repair technician access to state of health information for various vehicle sub-systems. The amount of diagnostic information available via OBD has ... Jul 21 19:25:59 aaaah Jul 21 19:26:41 cyeah Jul 21 19:26:44 maemo4 still has carman Jul 21 19:27:04 which looks up the codes for you if a test engine light comes on.. also shows you your fuel mixture on the fly Jul 21 19:27:13 and alerts you if you are running too rich or lean Jul 21 19:27:26 zeq: yeah, rebase it. I was thinking to but fennec build in cssu-thumb repo, but it is unusably slow now :( Jul 21 19:27:45 nice Jul 21 19:28:09 obd2 does have quite a bit of information Jul 21 19:28:22 I ponder to use my N810 as a door bell plate, with camera and voice Jul 21 19:28:29 i wanted to use the n810 on my 94 caprice but it is an older obd... some weird hybrid Jul 21 19:28:37 lol... the cam sucks tho :) Jul 21 19:28:48 plus... the cam doesn't support h264 Jul 21 19:28:57 you'd have to write something that supported lame h263 or something Jul 21 19:29:01 just fine enough for my ugly unsolicited visitors ;-P Jul 21 19:29:23 well.. it is more the supported codec than anything Jul 21 19:29:38 which is why the lame ass cam wouldn't support gtalk video chatting :-/ Jul 21 19:29:45 I don't care about codec as long as other end can decode it Jul 21 19:29:54 lol Jul 21 19:29:58 if it even can Jul 21 19:30:04 :) Jul 21 19:30:25 even thought about using mirror on my other N810 and sshfs mounting the /dev on the door plate Jul 21 19:30:29 the cam was the n810s biggest blunder Jul 21 19:30:44 they just took the n800 and shoved it into a case with a keyboard and slapped a gps on it in favor for a radio receiver Jul 21 19:31:09 and called it "new" ... i think that's where apple got the iphone upgrade path idea from Jul 21 19:31:19 hehehe Jul 21 19:31:47 what type of freq does the gps system use? Jul 21 19:31:54 hmm Jul 21 19:31:55 ? Jul 21 19:32:02 i'm wondering if they just reprogrammed the fm radio receiver Jul 21 19:32:08 noooo Jul 21 19:32:10 lol Jul 21 19:32:14 to receive gps signals instead of fm :) Jul 21 19:32:17 MADE IN CHINA! Jul 21 19:32:19 some 1.6GHz Jul 21 19:32:24 oh Jul 21 19:32:29 guess not then :) Jul 21 19:33:02 i wonder how much power those sats use Jul 21 19:33:05 to send the signal Jul 21 19:33:21 i mean.. they probably receive sunlight all the time.. but still Jul 21 19:33:36 I'd really wonder what ntfs/whatever bandwidth you'd need to access raw /dev/video Jul 21 19:34:17 on the n810 Jul 21 19:34:17 ? Jul 21 19:34:21 yup Jul 21 19:34:30 i doubt much Jul 21 19:34:40 well.. not much by today's standards Jul 21 19:34:54 (power) guestimate: 100W Jul 21 19:34:55 freemangordon: pushed Jul 21 19:36:15 GPS IIR 1,136 kW total Jul 21 19:38:13 1,100 kilowatts?? Jul 21 19:38:37 isn't tha tlike... a MW? Jul 21 19:38:38 http://en.wikipedia.org/wiki/GPS#Satellite_frequencies Jul 21 19:38:46 that's german Jul 21 19:38:53 kW? Jul 21 19:38:59 so 1.1 kW Jul 21 19:39:00 kW = Mhz? Jul 21 19:39:13 or power output? Jul 21 19:39:23 total power Jul 21 19:39:39 because a megawatt is a LOT Jul 21 19:39:44 for a radio Jul 21 19:39:49 just 1.1kW, not 1100kW Jul 21 19:40:01 oh Jul 21 19:40:11 . is far different than , Jul 21 19:40:16 in germany they use , instead of . ? Jul 21 19:40:19 yep Jul 21 19:40:28 how does that not confuse people? Jul 21 19:40:40 Europe is used to it ;-P Jul 21 19:40:49 fair enough Jul 21 19:40:52 used to the metric system too ;) Jul 21 19:41:03 we're over here on the imperial system still... we just hate europe Jul 21 19:41:07 lets be honest Jul 21 19:47:05 http://de.wikipedia.org/wiki/GPS#GPS_IIR Jul 21 19:47:10 Only 3 countries in the world still use the imperial system. Jul 21 19:47:15 Liberia. Jul 21 19:47:19 Burhma. Jul 21 19:47:24 And USA. Jul 21 19:47:28 Also even if you had a software radio, you would probably not have the CPU to decode GPS Jul 21 19:47:30 I think that says a lot. Jul 21 19:47:43 aren't those all US type places? Jul 21 19:47:53 i know Liberia was created to send negroes back to africa Jul 21 19:48:10 not too sure about Burhma .. but i thought it didn't exist anymore Jul 21 19:48:45 freemangordon: my latest fennec build with gcc47 seems ok. still really slow to load pages though! Not sure if MOZ_ACCELERATED=1 is working or not... Jul 21 19:49:02 Honestly, the imperial measurments system should have been BURIED decades ago. Jul 21 19:49:17 I do not measure things in bushels or rods thank you. Jul 21 19:49:33 vi_: LOL.. yeah.. honestly.. i agree ;) Jul 21 19:49:44 but try converting hillbillies and ignorant people to a new system Jul 21 19:49:47 it would be chaos Jul 21 19:49:47 zeq: MOZ_ACCELERATED=1 means enable HW acceleration, but you don't really need it Jul 21 19:50:03 there are entries in about:config for that purpose Jul 21 19:50:07 zeq: ^^^ Jul 21 19:50:13 vi_: the one that confuses me the most.. is the "dozen" Jul 21 19:50:19 up to now it's had the effect of black flickering you described before Jul 21 19:50:23 wtf is up with the number 12 that it gets its own special name? Jul 21 19:50:44 now no errors, maybe the env var is ignored now? Jul 21 19:50:46 too bad the world didn't start off working with an 8 base system Jul 21 19:50:57 imagine how easily peopl ewould understand binary numbers nowadays if it were heh Jul 21 19:51:02 dozen is the only one that survives, there were many other names too Jul 21 19:51:19 ShadowJK: for 12? Jul 21 19:51:21 zeq: NFC, justpulled master, lemme build and check what's going on Jul 21 19:51:43 we use the 12" foot heh Jul 21 19:51:54 zeq: could that regression of speed be related to some jscrib zlib changes in Makefile? Jul 21 19:51:59 because i guess 12" was a common foot size in the colonial days? Jul 21 19:52:04 not for 12, but for all sorts of measurements Jul 21 19:52:04 *jscript Jul 21 19:52:37 i don't understand why the colonists weren't using the metric system anyways Jul 21 19:52:47 didn't the metric system come out before the US was reated? Jul 21 19:52:51 created Jul 21 19:53:02 http://en.wikipedia.org/wiki/Dozen Jul 21 19:53:09 freemangordon: I may not be losing noticible performance because I compiled with -ftree-vectorize, -ffast-math and -fgraphite? Jul 21 19:53:26 it seems fast except for the page loading time Jul 21 19:54:21 freemangordon: I can't see how it could have anything to do with zlib Jul 21 19:54:49 zeq: run sunspider please. wothout OC and tell me the result Jul 21 19:55:03 ShadowJK: wow.. lots of history behind the dozen Jul 21 19:55:11 ah well. my flight is about to start boarding in about 10 minutes.. ttyl Jul 21 19:55:17 freemangordon: you might want to try enabling elfhack. It doesn't work with the qemu-arm-sb that comes with Gentoo scratchbox, I don't know if the same applies for you. Jul 21 19:55:30 freemangordon: ok Jul 21 19:55:37 elfhack? Jul 21 19:55:42 ~elfhack Jul 21 19:56:14 https://wiki.mozilla.org/Elfhack Jul 21 20:00:27 zeq: how to enable that? Jul 21 20:01:59 freemangordon: it's disabled explictily in mozconfig Jul 21 20:02:02 default is enabled Jul 21 20:02:07 (I think) Jul 21 20:03:48 ok, lemme check it Jul 21 20:05:27 zeq: yes, ac_add_options --disable-elf-hack Jul 21 20:05:34 zeq: did you add that? Jul 21 20:05:36 hillbillies and rednecks, I get it now. We use metric, base10 because we have 10 fingers, toes etc. Inbred hillbillies use base12 because of the extra fingers. Jul 21 20:05:49 It fails in my qemu Jul 21 20:06:00 ok, i'll try it here Jul 21 20:06:35 < Macer> didn't the metric system come out before the US was reated? Jul 21 20:06:45 freemangordon: assuming my battery lasts sunspider is underway Jul 21 20:06:49 there are bars in Edinburgh that are older than USA. Jul 21 20:08:36 freemangordon: I just accidently locked my N900! Never done that before, don't know the default code...?!?! Jul 21 20:09:01 12345 or something Jul 21 20:09:20 :D Jul 21 20:09:26 merlin1991: thanks! :) Jul 21 20:09:32 unlocked? Jul 21 20:10:39 yup, although also rebooted Jul 21 20:10:49 /home/maemo/workspace/thumb/tmp/fennek/fennec-qt-maemo5/obj-build/build/unix/elfhack/elfhack -b -f test.so Jul 21 20:10:50 test.so: terminate called after throwing an instance of 'std::runtime_error' Jul 21 20:11:12 elfhack don't work here too Jul 21 20:16:56 zeq: but it seems to be a linker problem, not qemu Jul 21 20:17:06 could be wrog of course Jul 21 20:17:42 I thought that first of all, but last time I tried it I got an illegal instruction from qemu Jul 21 20:17:52 still could be binutils though Jul 21 20:18:21 Macer: there's no SI equivalent for one hilbilly - it's inconvertible Jul 21 20:18:43 maybe we should use 2.22.5x.x? Jul 21 20:19:15 hmm, most probably we should use older binutils than newer Jul 21 20:19:23 2.22 is what I'n using on my laptop Jul 21 20:19:46 anyway, I will leave that for now, lets see how is the performance with latest nightly Jul 21 20:19:47 2.22.52.0.4.20120604 on my server LOL Jul 21 20:20:25 sunspider is running at the moment Jul 21 20:22:42 zeq: well, we'll compare compilers too that way :D Jul 21 20:23:11 not too impressive Jul 21 20:23:31 ? Jul 21 20:23:32 @600Mhz 8565.7 Jul 21 20:23:50 hmm, it is even worse :( Jul 21 20:23:53 WTF? Jul 21 20:24:03 no pint to compile it here Jul 21 20:24:03 feels quite quick though Jul 21 20:24:04 :/ Jul 21 20:24:10 *point Jul 21 20:24:42 obviously some JIT engine "upgrade" Jul 21 20:25:09 Compiler probably wouldn't make much difference to jscript Jul 21 20:25:33 afterall it's a code generator itself Jul 21 20:26:08 it's not going to be the bottle-neck Jul 21 20:26:41 zeq: but WTF, almost 2 times slower. damn Jul 21 20:27:42 zeq: what was the addres of mozzilla chat server? Jul 21 20:27:49 *address Jul 21 20:27:55 irc.mozilla.org Jul 21 20:29:27 #mobile? Jul 21 20:29:34 yup Jul 21 20:30:19 i'll try to ping romaxa for some help on that issue Jul 21 20:30:39 ok Jul 21 20:31:51 freemangordon: browserd is now using around half the RAM it did before. Jul 21 20:32:07 vi_: :) Jul 21 20:32:15 vi_: well, maybe 10-20% less :) Jul 21 20:33:06 freemangordon: well for me browserd used to sit around 10-13% Jul 21 20:33:18 Now it sits around 6% Jul 21 20:33:25 which is great. Jul 21 20:33:39 So great in fact I am dedicating this next beer to you. Jul 21 20:33:45 * vi_ pssssht Jul 21 20:34:33 freemangordon: I've gotta go. Might be back later... Jul 21 20:34:40 zeq: ok Jul 21 21:15:53 freemangordon: just learnt of "about:support" Jul 21 21:16:27 and...? Jul 21 21:16:36 anything helpful here? Jul 21 21:16:47 *there Jul 21 21:17:38 usefully it shows crash info (supposedly) and it confirmed for me I'm not using hw accel Jul 21 21:19:29 in microB? Jul 21 21:19:32 no kidding Jul 21 21:19:45 kerio: fennec Jul 21 21:19:50 even worse! Jul 21 21:20:12 kerio: surely hw accel would be a good thing? :) Jul 21 21:20:26 yeah, but i don't expect that to work on a linux desktop Jul 21 21:20:35 let alone the n900 Jul 21 21:21:06 romaxa has had it working (so he says) with ~60fps page rendering Jul 21 21:21:53 sweet Jul 21 21:22:22 that was a while ago and the patches need reworking I believe Jul 21 21:22:43 zeq: I was thinking... isn't it better to use some older and stable version. Jul 21 21:22:50 i.e 14 or 15 Jul 21 21:23:11 ]16 gets worse with every nightly Jul 21 21:23:55 freemangordon: yes, and no. we're more likely going to get help from romaxa getting support mainlined with mozilla-central... but I think it might be worth taking a dual approach Jul 21 21:24:19 I think we can alwasy backort whatever romaxa do Jul 21 21:24:25 *backport Jul 21 21:25:08 freemangordon: I have to say, I'm a bit surprised about the amount of churn, things applied then reverted hours later... Jul 21 21:25:48 zeq: do you remember when fennec was fast? I mean around what date, so we can do bisect easy Jul 21 21:26:16 I suggest you look in the irc log for your previous sunspider results Jul 21 21:27:25 hmm, I think it was around 9 or 10 of July, lemme check Jul 21 21:28:20 freemangordon: I don't remember the plugin-container taking up more memory than the main process when we started Jul 21 21:30:25 freemangordon: we need romaxa to pass on his hw accel patchset and work from whatever the latest release that applies to Jul 21 21:31:28 ls Jul 21 21:35:21 zeq: http://mg.pov.lt/maemo-irclog/%23maemo.2012-07-04.log.html#t2012-07-04T15:36:16 Jul 21 21:36:03 freemangordon: that was @600Mhz? Jul 21 21:36:11 i think so, not sure Jul 21 21:37:14 fwiw they seem to have fixed that https google problem now Jul 21 21:37:25 but it does not matter if it is 600 or 805 Jul 21 21:37:29 and page-loading is faster than before Jul 21 21:37:45 hmm, i'll need to backport the fix to microb :D Jul 21 21:38:16 freemangordon: but only your thumb-compiled one! Jul 21 21:38:21 vendor lockdown, baby! Jul 21 21:38:34 freemangordon: how similar is the mozilla codebase to microb-engine? Jul 21 21:40:20 apart from the lower sunspider score, my current fennec build isn't too bad. most serious bug I've found is when attempting to use hw accel crashes it, I have to nuke ~/.mozilla/fennec to get it to start-up again without crashing! Jul 21 21:42:46 freemangordon: are you going to rebase on a commit from that date and test? Jul 21 21:42:56 kerio: it works in non-thumb microb afaik, at least that is what was reported, didn't tried it myself Jul 21 21:43:15 feel free to push it to a new branch on github Jul 21 21:43:22 zeq: I am still wondering what to do, but first I will wait for current build to finish Jul 21 21:43:32 may as well keep tracking mozilla-central on master Jul 21 21:45:24 zeq: WTF? virtual keyboard does not pop up when address bar is clicked Jul 21 21:45:36 that was working, for sure Jul 21 21:45:46 let me check Jul 21 21:46:01 worksforme Jul 21 21:46:03 aah, my fault Jul 21 21:46:24 it's my second n900 with "use virtual keyboard" disabled Jul 21 21:46:33 ah Jul 21 21:46:52 we still need to rip out all that keyboard handling code Jul 21 21:47:10 yep Jul 21 21:48:15 going to let the dog in the garden brb Jul 21 21:48:20 zeq: debian/fennec//opt/fennec-17.0a1 Jul 21 21:48:23 ??? Jul 21 21:48:40 this is fennec 17, LOL Jul 21 21:48:50 they don't hang around! :P Jul 21 21:57:15 what the heck jollamobile quite since almost 6 hours Jul 21 21:57:57 actually *almost* since 24 hours Jul 21 22:00:06 Weekend secret meeting. Jul 21 22:00:26 :nod: probably even close to truth Jul 21 22:00:43 they have private channels Jul 21 22:01:09 but the missing noise from eventual lurkers is quite scary Jul 21 22:02:07 what, is there a channel? Jul 21 22:02:14 esp since yesteray morning the TV news opened with "Nokia made another 1.1 E9 EUR minus in last quartal" Jul 21 22:02:17 #jollamobile ? Jul 21 22:02:21 oh Jul 21 22:02:39 DocScrutinizer, amusingly analysts took that as positive news, nokia stock went up like 18% or something crazy Jul 21 22:02:49 Heh. Jul 21 22:02:51 yeah ROTFL Jul 21 22:03:32 There's some sliver of truth in that, the costs of downsizing are pretty extreme, but one-time in nature Jul 21 22:03:35 cought 18 bullets but still alive. I put all my money on the bloody guy on the floor Jul 21 22:03:58 Like, Jolla itself allegedly is mostly funded by Nokia at this point? Jul 21 22:04:16 I doubt that Jul 21 22:04:28 For sure they have received seed capital Jul 21 22:04:30 Contractual commitments to me ego? Jul 21 22:04:34 though, you never know Jul 21 22:04:53 Or seed capital from redundancies? Jul 21 22:05:23 I guess not public. Jul 21 22:05:29 According to .fi newspaper reporting, Jolla cam out of the "incubator program" that Nokia had to setup, to support former employees setting up new companies Jul 21 22:05:39 Ah. Jul 21 22:05:42 what are a few dozen mio if you can pay 1.1 bio per 3 months from your coffe stash Jul 21 22:06:35 but honestly I guess there are other VC entities involved Jul 21 22:07:12 I wonder how much money do you need to gave foxconn or someone make 50,000 devices Jul 21 22:07:13 and that "incubator program" was about as much as "you get fired in 6 month, but you can take some time off right now" afaik Jul 21 22:07:59 ShadowJK: easy: 50,000 * sales/3 Jul 21 22:08:11 DocScrutinizer05: funniest thing with the positve reaction to the financial report, was of the smartphones shipped, most were NOT WP7 Jul 21 22:08:28 yeah, that too Jul 21 22:08:59 zeq: probably the reason for the +18% Jul 21 22:09:30 WP7 sales did double, and they said "sales" this time, not "shipped" Jul 21 22:09:50 from 1000 to 2000? Jul 21 22:09:53 ;-P Jul 21 22:09:54 And they said they were late to WP7 party, but have front row seats at WP8 party Jul 21 22:09:55 sales to AT&T warehouses? Jul 21 22:10:41 nokia should've realized that the wp-train was a bad idea Jul 21 22:10:49 oh... duh, it's elop Jul 21 22:10:52 nvm Jul 21 22:11:33 DocScrutinizer05: fixed jollamobile for you ;) Jul 21 22:13:43 zeq: https google still gives certificate error, probably you have tls disable in about:config Jul 21 22:14:02 nope Jul 21 22:14:04 interesting Jul 21 22:14:17 freemangordon: must be fixed by my toolchain...? Jul 21 22:14:23 weird Jul 21 22:14:25 yep Jul 21 22:14:50 seems toolchain related, time to switch to yours :) Jul 21 22:15:55 freemangordon: note though you'll need to rebuild qt4 with the gcc47 patch to build against it, including fennec Jul 21 22:16:14 why i should rebuild qt? Jul 21 22:16:27 and qt4 won't itself build with 4.7 currently, not a huge problem though Jul 21 22:17:09 zeq: for what eason i need to rebuild qt with gcc 4.7 Jul 21 22:17:32 there's a header with a bug Jul 21 22:17:42 you could just change the header I guess Jul 21 22:17:55 to me that's even worse ;) Jul 21 22:17:59 hmm Jul 21 22:18:36 and what version is yours 4.6? 4.6.3? Jul 21 22:18:51 which? Jul 21 22:19:07 I built fennec with 4.7 Jul 21 22:19:12 I know Jul 21 22:19:23 but you said you built 4.6.x too Jul 21 22:19:29 the toolchain i mean Jul 21 22:19:33 I have build linaro-4.6-2012.7 and linaro-4.7-2012.7 Jul 21 22:19:54 s/build/built/ Jul 21 22:19:54 zeq meant: I have built linaro-4.6-2012.7 and linaro-4.7-2012.7 Jul 21 22:19:59 does fennec builds with 4.6 without your patches? Jul 21 22:20:08 should do Jul 21 22:20:09 *build Jul 21 22:20:12 haven't tried Jul 21 22:20:28 well, maybe i should use that then Jul 21 22:20:36 was more interested in testing 4.7 :) Jul 21 22:20:53 could you try to build fennec with your 4.6 to see of google https error is there? Jul 21 22:20:59 zeq: I know Jul 21 22:21:09 ok, but it will take a while... Jul 21 22:21:19 zeq: no hurry Jul 21 22:22:09 I really kicked things off over at #jollamobile Jul 21 22:22:19 * zeq trolls well :P Jul 21 22:23:10 zeq: but we deffinitely should should stick to some version around 4-10 july, the current has HUGE memory consumtion and is slow as hell Jul 21 22:23:41 sonspider: 8920 ms Jul 21 22:23:46 *sunspider Jul 21 22:23:52 @600 Jul 21 22:25:02 zeq: seems gcc 4.7 saves about 300ms Jul 21 22:28:07 ok Jul 21 22:28:37 not bad really, didn't expect much difference from the JIT Jul 21 22:29:39 Where I expected to see most improvement was page layout Jul 21 22:30:28 why? Jul 21 22:32:28 very visible for a start ;) Jul 21 22:33:10 it's also native code rather than JS (right?) Jul 21 22:33:35 yeah, sounds sane Jul 21 22:34:31 anyway, it is late for today, but I will try to do git bisect tomorrow Jul 21 22:36:05 zeq: it is strange there is nor release for 16a1 is not released: http://releases.mozilla.org/pub/mozilla.org/firefox/releases/ Jul 21 22:36:21 I told you it is late here :D:D:D Jul 21 22:36:22 I'll push my gcc47 build fixes for fennec, then get started on the linaro-4.6 fennec build Jul 21 22:36:51 zeq: it is strange that 16a1 is not released: http://releases.mozilla.org/pub/mozilla.org/firefox/releases/ Jul 21 22:36:58 looking Jul 21 22:38:00 don't see 15 there either? Jul 21 22:38:35 there is 15b1 Jul 21 22:38:41 ah ok Jul 21 22:39:03 aurora maybe? Jul 21 22:39:03 but seems latest not in alpha or beta is 13 Jul 21 22:39:20 http://www.mozilla.org/en-US/firefox/all-aurora.html Jul 21 22:39:37 there is is :) ^ Jul 21 22:39:41 it is* Jul 21 22:40:34 but "Download Firefox Aurora in your language to experience the newest features and innovations in an unstable environment even before they go to Beta" Jul 21 22:41:53 14.01 is the latest release according to http://www.mozilla.org/en-US/firefox/new/ Jul 21 22:42:25 and it is released on 17 jul Jul 21 22:42:35 i think that is a good candidate Jul 21 22:42:49 b7 was performing pretty well Jul 21 22:43:09 is there a tag in the git repo? Jul 21 22:43:25 hmm, lemme try to checkout Jul 21 22:43:58 freemangordon: I'm keeping the gf awake, going to have to sign-off... Jul 21 22:44:51 night all! :) Jul 21 22:44:55 * zeq waves Jul 22 01:43:33 hi **** ENDING LOGGING AT Sun Jul 22 02:59:59 2012