**** BEGIN LOGGING AT Mon Nov 14 02:59:57 2011
Nov 14 03:01:51 <DocScrutinizer> you must be dreaming
Nov 14 03:04:34 <DocScrutinizer> btw no worries about Nokia ever disclosing bme src, they don't even dare to give specs on how to *charge* the battery
Nov 14 03:05:20 <DocScrutinizer> as if they would be (more) liable of any damage done when they tell hackers what to _avoid_
Nov 14 03:06:41 * jonwil doesnt understand why anyone cares about BME code
Nov 14 03:06:52 <jonwil> I have no wish whatsoever to mess with BME
Nov 14 03:07:10 <DocScrutinizer> I asked several times "what is the NTC sensor reading for bat temp where we should shut down for safety reasons, to avoid battery cell overtemp?" - A: "we mustn't tell as we don't want to take responsibility"
Nov 14 03:08:03 <DocScrutinizer> I have no wish to look at this abomination called bme src
Nov 14 03:08:35 <jonwil> None of the bits of code I want seem to be particularly risky in terms of security or safety or legal issues
Nov 14 03:08:54 <DocScrutinizer> there's definitely *nothing* in it I'd want to know about
Nov 14 03:10:58 <DocScrutinizer> ShadowJK lives happily now since > 1 year with letting bq24150 charger chip just do its job, rather than trying to teach it how to do it better like bme does
Nov 14 03:20:14 <DocScrutinizer> btw please help me out, what's been the major argument for meego? wasn't it sth like "we'll do it THE RIGHT WAY[TM] this time, no closed source anymore"? Well, fine, now we have a mostly published hildon and a closed source swipe, and bme is as closed as it ever was X-P
Nov 14 03:21:44 <jonwil> Harmattan is not MeeGo
Nov 14 03:21:59 <jonwil> MeeGo is as open as Nokia can legally get it
Nov 14 03:22:03 <DocScrutinizer> I know, but then WTF did they do all the time?
Nov 14 03:23:13 <DocScrutinizer> where from are all those closed bits of swipe/MTF?
Nov 14 03:24:00 <jonwil> no idea
Nov 14 03:24:01 <jonwil> :)
Nov 14 03:26:21 <DocScrutinizer> what a major brainf*ck
Nov 14 03:27:19 * DocScrutinizer notices he gets his grumpy 15 minutes and waves a cyal8r
Nov 14 03:47:05 * jonwil cant understand why people seem to think that having the MCE source code is of any benefit
Nov 14 03:47:53 <jonwil> like the guy on TMO who said "I could probably be very happy with pr 1.1 BME source"
Nov 14 04:19:45 <SpeedEvil> It'd be useful for a proper replacement.
Nov 14 05:17:58 <LaoLang_cool> hello
Nov 14 05:18:15 <LaoLang_cool> looking for a picture viewer, any recommended?
Nov 14 05:18:55 <robbiethe1st> Uh, stock one?
Nov 14 05:19:24 <slonopotamus> robbiethe1st: that woud be too simple
Nov 14 05:19:38 <LaoLang_cool> robbiethe1st: stock one?
Nov 14 05:19:44 <LaoLang_cool> ...
Nov 14 05:19:47 <LaoLang_cool> oh
Nov 14 06:07:31 <slonopotamus> do we have source for retutime from initfs?
Nov 14 06:19:44 <jonwil> Do you mean for N900 or N8xx?
Nov 14 06:20:01 <slonopotamus> N900 has initfs?
Nov 14 06:23:06 <jonwil> I cant find any indications that retutime on the N810 is open source
Nov 14 06:26:29 <slonopotamus> yep. sad.
Nov 14 06:27:12 * slonopotamus tries to figure out what to do with clock on his n800 with 3.1 kernel
Nov 14 06:50:36 <pabs3> jonwil: what does retutime do?
Nov 14 06:50:41 <jonwil> no idea
Nov 14 06:50:51 <jonwil> google seems to suggest that its clock related
Nov 14 06:50:58 <jonwil> i.e. it helps set the clock properly
Nov 14 06:52:01 <pabs3> seems to be a proprietary userspace util for doing what the kernel RTC drivers do I guess
Nov 14 06:53:24 <jonwil> yeah something for doing things to avoid needing to document certain things about the clock logic
Nov 14 06:53:28 <jonwil> I am guessing
Nov 14 06:55:00 <pabs3> this lwn comment seems to suggest there is kernel code to replace that: http://lwn.net/Articles/230227/
Nov 14 06:57:29 <pabs3> and it seems like there is qemu code to emulate them here: http://www.weilnetz.de/qemu-doc/ARM-System-emulator.html
Nov 14 07:12:55 <LaoLang_cool> Hello, how to auto-solve depend issue with dpkg -i  ?
Nov 14 07:13:18 <LaoLang_cool> Is there an option to let dpkg auto install the depend?
Nov 14 07:16:01 <pabs3> LaoLang_cool: apt-get -f install
Nov 14 07:16:20 <pabs3> LaoLang_cool: dpkg doesn't have any dependency resolving, thats in apt
Nov 14 07:18:21 <LaoLang_cool> pabs3: thank you, that's it
Nov 14 08:27:43 <cityLights> morning
Nov 14 08:51:53 <LaoLang_cool> Hello, morning there. How to invoke a app to view a pic file in xterm? How to run a app to listen to a music file in xterm?
Nov 14 09:08:38 <Sicelo> phone-control is easiest package to do both things
Nov 14 09:09:09 <Sicelo> if u want 'hard' way, check out Phone Control page in wiki
Nov 14 09:10:28 <woldrich> LaoLang_cool, 2) mplayer
Nov 14 09:10:55 <LaoLang_cool> woldrich: seems mplayer is buggy on n900, vol setting will make it no responce
Nov 14 09:11:04 <woldrich> works fine for me
Nov 14 09:11:29 <LaoLang_cool> woldrich: press 0 and 9 to adjust vol is fine for you?
Nov 14 09:11:32 <LaoLang_cool> hmm..
Nov 14 09:14:27 <woldrich> LaoLang_cool, yes, with -ao pulse
Nov 14 09:14:47 <LaoLang_cool> woldrich: oh, maybe it's the problem, I will try, thank you
Nov 14 09:32:08 <mihu> Hi. The "Maps" application is not starting up any more. When I run "nokia-maps" from X terminal, I see a couply of messages, but then it gets stuck after "Plugin::Setup()", then "device language: GER" and then "set def coord ...". After that it just hangs. Of course I cannot remember fiddling with anything that might cause this. Any ideas?
Nov 14 09:36:59 <LaoLang_cool> woldrich: thank you, works fine!
Nov 14 09:37:04 <LaoLang_cool> mplayer -ao pulse
Nov 14 09:51:27 <achipa> X-Fade: heya, got another weird one - i pushed libqtm-12+git*fremantle5 but it kinda didn't appear in the user interface
Nov 14 09:52:04 <achipa> also libqtm-12 seems to have lost the promotion block so it's borked in general as it propagated to testing
Nov 14 09:52:09 <jonwil> I have never been able to run nokia-maps from xterminal, only from control panel
Nov 14 09:52:20 <jonwil> I mean app view
Nov 14 09:52:24 <jonwil> not control panel
Nov 14 09:58:34 <MaceSyn> :) ...
Nov 14 09:58:35 <MaceSyn> omnia> ipkg list gcc
Nov 14 09:58:36 <MaceSyn> gcc - 4.2.3-1 - The GNU Compiler Collection.
Nov 14 09:58:48 <MaceSyn> ok.. a native working gcc on a synology disk station but not on an n900?
Nov 14 10:00:31 <mihu> jonwil: If I do that, then "maps" will never start up either. It will keep showing that scroll bar forever. 
Nov 14 10:07:51 <ruskie> MaceSyn, so add the sdk repo and install it
Nov 14 10:08:20 <crashanddie> MaceSyn, there's gcc on the n900
Nov 14 10:09:56 <MaceSyn> crashanddie: a whole dev environment on the n900?
Nov 14 10:10:06 <crashanddie> sure
Nov 14 10:10:10 <MaceSyn> using maemo.. so you don't have to cross compile in sb?
Nov 14 10:10:14 <MaceSyn> :)
Nov 14 10:10:19 <MaceSyn> wow really? does it work?
Nov 14 10:10:24 <crashanddie> of course
Nov 14 10:10:29 <crashanddie> that's how liqbase was developed
Nov 14 10:10:43 <crashanddie> entirely on-device.
Nov 14 10:10:57 <MaceSyn> nice
Nov 14 10:11:02 <crashanddie> it's linux, innit?
Nov 14 10:11:05 <MaceSyn> what about a functional dpkg?
Nov 14 10:11:21 <crashanddie> I thought dpkg was on there?
Nov 14 10:11:32 <MaceSyn> i'm asking ;)
Nov 14 10:12:28 <MaceSyn> that's kind of neat.. so you're saying i can install dpkg and gcc and get some src and build ? :)
Nov 14 10:12:43 <MaceSyn> i will try that out
Nov 14 10:12:45 <ruskie> dpkg is already on there iirc
Nov 14 10:13:08 <MaceSyn> i remember there being problems building and packaging on device
Nov 14 10:13:48 <crashanddie> MaceSyn, compiling C++ is very long.
Nov 14 10:13:54 <crashanddie> MaceSyn, C code is kind of OK.
Nov 14 10:30:31 <jonwil> its possible to compile on device but it uses up large amounts of space to do so
Nov 14 10:30:35 <jonwil> because of all the libs you need
Nov 14 10:30:41 <jonwil> and all the headers
Nov 14 10:30:43 <jonwil> and tools
Nov 14 10:31:15 <jonwil> my include folder alone is about 71 megs
Nov 14 11:43:36 <cityLights> DocScrutinizer!
Nov 14 11:43:42 <cityLights> waht up dud?
Nov 14 11:46:15 <cityLights> remeber me asking about dnsmasq yesterday
Nov 14 11:46:31 <cityLights> ?
Nov 14 11:52:59 <Jaffa> X-Fade: ping - planet.maemo.org doesn't seem to have updated with http://mwkn.net/issues.rss
Nov 14 11:53:43 <X-Fade> Jaffa: The maintenance node is down, notified the isp.
Nov 14 12:02:51 * RST38h yawns
Nov 14 12:06:51 <vdv> i'm trying to build (from qtcreator) an app for a maemo5 target, but since dependency libs (hunspell, vorbis) are not present in madde, build fails. how can fix that?
Nov 14 12:14:33 <jonwil> Who was it that was working on alternative charging options for the N900 (i.e. using the charging chip instead of using BME)?
Nov 14 12:15:05 <SpeedEvil> ShadowJK: ?
Nov 14 12:15:40 <ShadowJK> yes
Nov 14 13:00:55 * jonwil is surprised at the effort Nokia have gone with the notification system. They have an open source daemon (hildon-sv-notification-daemon) that talks to a closed source shared library (libhildon-plugins-notify-sv.so) which then talks to a closed source daemon (nsv-decoder-service)
Nov 14 13:13:27 <jonwil> aha, nsv-decoder-service exists only to decode compressed audio into wav files
Nov 14 13:13:38 <jonwil> and isn't used if the wav file already exists
Nov 14 13:18:52 <pabs3> jonwil: I guess it uses some hardware decoding thing
Nov 14 13:19:03 <jonwil> well its making gstreamer calls
Nov 14 13:19:09 <jonwil> so yeah it must be doing something like that
Nov 14 13:19:19 <jonwil> what surprises me is that Nokia felt the need to keep this particular item closed
Nov 14 13:19:31 <jonwil> given that I dont see it doing anything much other than handling notifications
Nov 14 13:19:35 <jonwil> like calls and SMS
Nov 14 13:19:42 <jonwil> to e.g. play ringtones
Nov 14 13:19:45 <jonwil> or vibrate
Nov 14 13:21:31 <jonwil> although the more I look at it, the more it looks clonable
Nov 14 13:21:36 <jonwil> which would be nice
Nov 14 13:34:36 <LaoLang_cool> hi, any recommended timer for n900?
Nov 14 13:34:53 <LaoLang_cool> timer, alarm
Nov 14 13:39:33 <HaXeri> the app called "Clock"
Nov 14 13:40:20 <LaoLang_cool> HaXeri: the built-in clock? Too limit, how to count down/up?
Nov 14 13:41:04 <LaoLang_cool> and I can't see the second, it's only precise up to minute
Nov 14 13:56:00 <mihu> Hi. Does anybody know which clock rates are supported by the SD card controller of the N900? UHS-I is certainly not supported, but HS mode? What is the maximum clock?
Nov 14 13:56:30 * ShadowJK uses tickstill for counting down/up
Nov 14 14:01:10 <SpeedEvil> mihu: 25MHz is used.
Nov 14 14:01:20 <SpeedEvil> I'm unsure of the hardware maximums as I haven't been arsed.
Nov 14 14:01:45 <SpeedEvil> The hardware can easily do 12 meg a second read and write from SD.
Nov 14 14:02:38 <mihu> SpeedEvil: Thanks.
Nov 14 14:03:13 <mihu> Anybody ever tried to hook up a SDIO card (for example a GPS received) to the N900 SD card slot? 
Nov 14 14:03:26 <SpeedEvil> The hardware supports it.
Nov 14 14:04:30 <mihu> SpeedEvil: Yes, I think so. One problem is that most cards are SD form factor, so you need a SD to micro SD adapter. Another thing that worries me is that the kernel is so old.
Nov 14 14:06:50 <SpeedEvil> And yes.
Nov 14 14:07:05 <LaoLang_cool> No timer good enough on n900? :(
Nov 14 14:09:44 <NIN101> i like "stopish"
Nov 14 14:15:00 <LaoLang_cool> NIN101: thanks, it's simple and work :)
Nov 14 14:22:48 <neal> SpeedEvil: What card / workload did you get 12 mb/s on?
Nov 14 14:26:17 <SpeedEvil> old kingston 1G, and dd
Nov 14 14:26:35 <neal> reading or writing?
Nov 14 14:26:42 <SpeedEvil> both
Nov 14 14:26:47 <neal> hmm
Nov 14 14:27:17 <SpeedEvil> The actual 'real life' speed will vary, but this is mostly down to the card, not the n900 hardware.
Nov 14 14:27:48 <neal> I still don't understand why I get 6mb/s with my n900 and 12mb/s with my thinkpad
Nov 14 14:27:58 <SpeedEvil> Good question.
Nov 14 14:28:03 <SpeedEvil> H?ow are you testing it?
Nov 14 14:28:10 <neal> I used dd
Nov 14 14:28:17 <neal> I also used bonnie
Nov 14 14:28:40 <SpeedEvil> I'd ispect syslog, and see what it says about speeds.
Nov 14 14:29:14 <SpeedEvil> Also - top - check some other proces isn't busy at the same time
Nov 14 14:29:21 * DocScrutinizer moos at SpeedEvil
Nov 14 14:29:23 <neal> the system was idle
Nov 14 14:29:51 <ShadowJK> iostat is nice, the parameter to optimize for is avgrq-sz
Nov 14 14:30:10 <ShadowJK> (iostat is in sysstat)
Nov 14 14:31:48 * DocScrutinizer has a deja-vu
Nov 14 14:32:06 * DocScrutinizer ponders to find the copy of this convo in chanlog
Nov 14 14:32:38 <neal> we started this conversation a day or two ago, yes
Nov 14 14:33:44 <mgedmin> get an n9 -- no sd card slot, no problem
Nov 14 14:33:55 <neal> have one
Nov 14 14:33:58 <neal> I don't like it :)
Nov 14 14:34:23 <neal> I don't like the lack of hardware keyboard, the capacitive touch screen, swipe, aegis
Nov 14 14:34:36 <neal> those are my main complaints
Nov 14 14:34:53 <ShadowJK> I was going to test the screen in a store, but before I had managed to convince it to boot I was drawing too much attention from the storekeepers :)
Nov 14 14:35:25 <DocScrutinizer51> haha
Nov 14 14:36:32 <Macer> running on Linux  armv5tel
Nov 14 14:36:35 <Macer> nice
Nov 14 14:36:49 <ShadowJK> N8x0?
Nov 14 14:36:52 <jacekowski> thing is, once you have more than 16-32G of internal flash
Nov 14 14:36:53 <Macer> the synology web ui is eyeos heh
Nov 14 14:37:02 <Macer> no. synology diskstation
Nov 14 14:37:15 <Macer> i am wondering if built bins will work on the n8x0 tho
Nov 14 14:37:22 <jacekowski> then you can live without sd card
Nov 14 14:37:26 <jacekowski> Macer: it's not eyeos
Nov 14 14:37:39 <jacekowski> Macer: similiar principle but it's nowhere close to eyeos
Nov 14 14:37:41 <Macer> jacekowski: no? looks incredibly similar
Nov 14 14:37:50 <jacekowski> no
Nov 14 14:37:53 <jacekowski> eyeos is in php
Nov 14 14:37:57 <Macer> as in worse or better or completely different?
Nov 14 14:37:58 <jacekowski> their stuff is in perl
Nov 14 14:38:03 <jacekowski> completly different
Nov 14 14:38:08 <jacekowski> it's mostly management software
Nov 14 14:38:08 <Macer> heh
Nov 14 14:38:10 <jacekowski> rather than OS
Nov 14 14:38:13 <ShadowJK> The benefit of SD card is that when the device fails to boot for mysteric reasons and troubleshooting is prevented by aegis&friends, you can just take out the SD card and get your data. :)
Nov 14 14:38:23 <Macer> i dont consider php an "os" :)
Nov 14 14:38:34 <Macer> but their ui is nice
Nov 14 14:38:39 <Macer> pretty awesome tbh
Nov 14 14:38:43 <jacekowski> do you have diskstation?
Nov 14 14:38:47 <Macer> yes
Nov 14 14:38:51 <Macer> just got it friday
Nov 14 14:38:51 <jacekowski> which one/
Nov 14 14:38:52 <DocScrutinizer51> ShadowJK: exactly
Nov 14 14:38:56 <Macer> ds411j
Nov 14 14:39:00 <jacekowski> hmm
Nov 14 14:39:06 <jacekowski> have you got HDDs for it as well
Nov 14 14:39:07 <ShadowJK> (Nokia Care has a trained army of monkeys that extract the internal emmc chips and hten hit it with a hammer). data^Wsilicon recovery
Nov 14 14:39:14 <Macer> ram is in short supply on it
Nov 14 14:39:21 <Macer> but 1 hd went 40MB/s
Nov 14 14:39:25 <Macer> no
Nov 14 14:39:31 <Macer> getting some this week
Nov 14 14:39:36 <Macer> 4x3TB
Nov 14 14:39:36 <jacekowski> 128M is plenty for that sort of device
Nov 14 14:39:51 <jacekowski> i had WD hdd 
Nov 14 14:39:51 <Macer> but flooding in taiwan is hurting me emotionally to buy them
Nov 14 14:39:53 <jacekowski> network hdd
Nov 14 14:40:02 <jacekowski> Macer: i bought mine like month before that flood
Nov 14 14:40:05 <Macer> they are 30%+ now
Nov 14 14:40:11 <ShadowJK> I had a NSLU2, 332M ram.. seemed to be quite sufficient
Nov 14 14:40:14 <Macer> 250 is the cheapest
Nov 14 14:40:24 <DocScrutinizer51> ShadowJK: for some reason romBL first tries USB, then uSD(!), then finally NAND
Nov 14 14:40:45 <jacekowski> Macer: buy just 2 
Nov 14 14:40:45 <Macer> i couldhave gotten them for 150 a while back :(
Nov 14 14:40:53 <jacekowski> Macer: you can always add them later
Nov 14 14:40:58 <Macer> can i add to raid 5?
Nov 14 14:41:02 <jacekowski> yes
Nov 14 14:41:16 <jacekowski> Macer: and 40M/s is bit slow for that device
Nov 14 14:41:21 <jacekowski> i managed to get a lot more out of it
Nov 14 14:41:26 <Macer> it was a pos hd :)
Nov 14 14:41:35 <Macer> and only 1 80G i dug out
Nov 14 14:41:40 <Macer> to test
Nov 14 14:41:58 <Macer> i am sure a raid config will go faster.. does it use md?
Nov 14 14:42:06 <jacekowski> yes
Nov 14 14:42:07 <jacekowski> and lvm
Nov 14 14:42:10 <jacekowski> 1048576000 bytes (1.0 GB) copied, 9.67831 s, 108 MB/s
Nov 14 14:42:19 <Macer> md is kind of lame isnt it?
Nov 14 14:42:23 <jacekowski> why?
Nov 14 14:42:29 <Macer> doesnt it have cpu overhead?
Nov 14 14:42:35 <jacekowski> it does
Nov 14 14:42:37 <Macer> and ram?
Nov 14 14:42:40 <jacekowski> no
Nov 14 14:42:43 <jacekowski> just cpu
Nov 14 14:42:48 <Macer> how bad?
Nov 14 14:42:50 <jacekowski> and well, that cpu doesn't have anything else to do
Nov 14 14:43:00 <Macer> i wanted to run services ;)
Nov 14 14:43:06 <Macer> it has a ton of them
Nov 14 14:43:16 <jacekowski> Macer: 1048576000 bytes (1.0 GB) copied, 8.75929 s, 120 MB/s
Nov 14 14:43:19 <jacekowski> that's raid5
Nov 14 14:43:21 <jacekowski> 120MB/s
Nov 14 14:43:26 <Macer> nice!
Nov 14 14:43:34 <Macer> full gbit huh?
Nov 14 14:43:40 <jacekowski> that's locally
Nov 14 14:43:45 <jacekowski> on device 
Nov 14 14:43:48 <Macer> oh
Nov 14 14:43:50 <jacekowski> DiskStation> dd if=/dev/md3 of=/dev/null bs=1M count=1000
Nov 14 14:43:53 <DocScrutinizer51> hoiw does cpu overhead matter as long as it's faster than the net bw of hdd?
Nov 14 14:43:54 <Macer> kinda slow :)
Nov 14 14:43:57 <Macer> lol
Nov 14 14:44:08 <jacekowski> Macer: well, HDD only delivers 100M/s
Nov 14 14:44:16 <Macer> i mean it is ok for a $300 device
Nov 14 14:44:38 <jacekowski> it delivers more than i've ever managed to get out of 3k dell md3000u
Nov 14 14:44:40 <Macer> http://tech.rancorous.net 
Nov 14 14:44:47 <Macer> i have old and new pics there
Nov 14 14:44:52 <jacekowski> in terms of sequential reads
Nov 14 14:44:58 <Macer> my old server is a 10 hard drive beast
Nov 14 14:45:26 <jacekowski> Macer: you really want to get normal bash on it
Nov 14 14:45:28 <Macer> i figured they would have at least thrown in a small processor to help the raid
Nov 14 14:45:29 <jacekowski> Macer: and stuff like that
Nov 14 14:45:33 <Macer> i tried!
Nov 14 14:45:43 <Macer> ssh wont connect using /opt/bin/bash
Nov 14 14:45:50 <jacekowski> ?
Nov 14 14:45:54 <jacekowski> works for me
Nov 14 14:45:56 <Macer> don't know whats up with that
Nov 14 14:46:05 <Macer> not me for some reason
Nov 14 14:46:12 <jacekowski> have you installed everything correctly?
Nov 14 14:46:35 <Macer> what was there to install?
Nov 14 14:46:35 <Macer> ipkg install bash :)
Nov 14 14:46:52 <jacekowski> so you've got ipkg working?
Nov 14 14:47:00 <Macer> every time i try to connect as a user with bash in /etc/passwd it denies the login
Nov 14 14:47:18 <jacekowski> /etc/shells
Nov 14 14:47:21 <DocScrutinizer> prolly bash missing in /etc/shells
Nov 14 14:47:22 <jacekowski> is bash in there
Nov 14 14:47:27 <Macer> hm
Nov 14 14:47:58 <DocScrutinizer> bash missing there is what saves most people messing with that from a bootloop on N900
Nov 14 14:48:40 <DocScrutinizer> at least it used to
Nov 14 14:48:44 <Macer>  /opt/bin/bash wasnt but /bin/bash was
Nov 14 14:48:51 <Macer> and i cp bash to /bin
Nov 14 14:48:59 <DocScrutinizer> back in times of pre-PR1.2, never checked since
Nov 14 14:49:00 <Macer> but let me try /opt/bin/bash
Nov 14 14:49:05 <Macer> and see what happens
Nov 14 14:49:35 <DocScrutinizer> WUT?
Nov 14 14:50:08 <DocScrutinizer> /opt/bin/bash is a magnitude more gasheaded idea than /usr/bin/bash
Nov 14 14:51:04 <DocScrutinizer> shells clearly belong into /bin, nowhere else. Esp if you consider to use one of those as login shell
Nov 14 14:52:30 <Macer> it is because ipkg pkgs are "optional" :)
Nov 14 14:52:32 <DocScrutinizer> and as your PATH usually isn't supposed to have /opt/bin at all, you will frequently want to get symlinks from /bin to /opt/bin for those pkgs that actually may go there
Nov 14 14:53:17 <DocScrutinizer> means calling anything via FQN /opt/bin/* is usually odd
Nov 14 14:54:12 <DocScrutinizer> unless it's another component of same package that acquires knowledge about where the package lives on a dynamic way
Nov 14 14:54:19 <Macer> mace@omnia's password: 
Nov 14 14:54:19 <Macer> Permission denied, please try again.
Nov 14 14:54:30 <Macer> see? :)
Nov 14 14:56:04 <DocScrutinizer> your login shell is invalid?
Nov 14 14:56:14 <DocScrutinizer> e.g not in /etc/shells ?
Nov 14 14:56:30 <Macer> i checked. it is
Nov 14 14:56:40 <Macer> i just put it there
Nov 14 14:57:28 <Macer> so im lost ;)
Nov 14 14:57:40 <Macer> is there some ssh setting that would prevent this?
Nov 14 14:59:36 <jacekowski> Macer: just add bash to your .profile
Nov 14 15:00:03 <DocScrutinizer> haha, nice
Nov 14 15:00:03 <Macer> lol
Nov 14 15:00:15 <r00t|n900> exec bash
Nov 14 15:00:18 <Macer> typical maemo hackery there jacekowski :)
Nov 14 15:00:27 <DocScrutinizer> so bash executes .profile and starts bash, which executes .profile and...
Nov 14 15:00:36 <Macer> i would like to find the underlying issue :)
Nov 14 15:00:49 <r00t|n900> typical unix hackery
Nov 14 15:01:32 <Macer> ash executes profile.. the bash... then... is more like it
Nov 14 15:01:34 <DocScrutinizer> I can see this will work excellent
Nov 14 15:01:39 <Macer> it uses ash by default
Nov 14 15:01:48 <Macer> i should just replace the ash bin hahaha
Nov 14 15:02:00 <Macer> that would be even more ghetto
Nov 14 15:02:32 <DocScrutinizer> making bin/sh a link to bash is not that much ghetto
Nov 14 15:03:11 <DocScrutinizer> bash is prepared to get installed/used that way
Nov 14 15:04:03 <neal> Here are my results: reading/writing from an ext3 partition on the microsd card: 15.7/5.5 mb/s; from internal flash: 7.5/20.3.
Nov 14 15:04:03 <DocScrutinizer> of course you should make damn sure you got a sane conservative bash installation then, not some shit in /oipt
Nov 14 15:06:44 <DocScrutinizer> halley:~ # ll /bin/sh
Nov 14 15:06:45 <DocScrutinizer> lrwxrwxrwx 1 root root 4 Dec 16  2009 /bin/sh -> bash
Nov 14 15:07:06 <DocScrutinizer> halley:~ # which sh
Nov 14 15:07:07 <DocScrutinizer> /usr/bin/sh
Nov 14 15:07:09 <DocScrutinizer> halley:~ # ll `which sh`
Nov 14 15:07:11 <DocScrutinizer> lrwxrwxrwx 1 root root 9 Dec 16  2009 /usr/bin/sh -> /bin/bash
Nov 14 15:08:57 <DocScrutinizer> halley:~ # whichgrep sh
Nov 14 15:08:59 <DocScrutinizer> /usr/bin/sh
Nov 14 15:09:00 <DocScrutinizer> /bin/sh
Nov 14 15:10:24 <DocScrutinizer> halley:~ # cat `which whichgrep`
Nov 14 15:10:25 <DocScrutinizer> #!/bin/sh
Nov 14 15:10:27 <DocScrutinizer> find `echo $PATH|tr ":" " "` -name "$1"
Nov 14 15:32:50 <DocScrutinizer> jacekowski: Macer: http://zsh.sourceforge.net/FAQ/zshfaq01.html#l8 ;-D
Nov 14 15:44:37 <Macer> yeah im lost
Nov 14 16:36:00 <Shapeshifter> ???
Nov 14 16:53:43 <neal> [50242.337280] EXT3-fs error (device mmcblk0p1): ext3_journal_start_sb: Detected aborted journal
Nov 14 16:53:51 <neal> lovely
Nov 14 16:54:51 <DocScrutinizer> ugh
Nov 14 16:55:06 <neal> that's the memory card
Nov 14 16:55:19 <neal> but I'm guessing the problem is the n900's software
Nov 14 16:59:38 <neal> that was a wasted 50 euros :/
Nov 14 17:02:09 <DocScrutinizer> uSD with IO errors / bad blocks ?
Nov 14 17:02:36 <neal> I don't think it is the usd card.
Nov 14 17:02:41 <neal> I think it is the software
Nov 14 17:02:52 <DocScrutinizer> which software?
Nov 14 17:02:59 <neal> e.g., having to use the fixed omap_hsmmc module
Nov 14 17:04:23 <DocScrutinizer> ooh, you refer to your tests as of above
Nov 14 17:05:02 <neal> yes, I was just doing another test and it didn't work because the file system was suddenly read-only
Nov 14 17:36:07 <Macer> weird
Nov 14 17:36:14 <Macer> root can ssh in and bash works
Nov 14 17:36:21 <Macer> a user gets permission denied
Nov 14 17:36:31 <Macer> bash is in /etc/shells
Nov 14 17:36:51 <Macer> is there some weird sshd setting which would deny a user a shell?
Nov 14 17:44:07 <Sicelo> maybe user has no pass
Nov 14 17:44:47 <Macer> the user does
Nov 14 17:44:59 <Macer> does pam affect using a shell?
Nov 14 17:46:52 <Macer> hm.. doesnt seem so
Nov 14 17:48:33 <Macer> Sicelo: you may be right.. maybe for some reason or another sshd isn't passing a pw over to pam?
Nov 14 17:49:11 <mgedmin> Macer, locking a user account with passwd -l forbids ssh logins too
Nov 14 17:49:28 <mgedmin> in later Maemo releases (and Harmattan) the user account is locked by default
Nov 14 17:49:46 <mgedmin> get root, vi /etc/passwd, replace the user:! with user:* (or the other way around, I don't remember), then you can ssh in
Nov 14 17:50:21 <mgedmin> or passwd -d user, (IIRC)
Nov 14 17:53:26 <merlin1991> yep passwd -d works for me
Nov 14 18:36:26 <Macer> jacekowski: i figured it out
Nov 14 18:36:34 <Macer> the synology sshd sucks
Nov 14 18:36:40 <Macer> had to install openssh
Nov 14 18:37:04 <Macer> but i messed it up and locked root out... they have differend sshd_config files
Nov 14 18:59:35 <badcloud> is there a package manager which can select multiple packages (to install/uninstall) via gui?
Nov 14 19:02:03 <HaXeri> why gui
Nov 14 19:03:30 <NIN101> FAM. 
Nov 14 19:03:42 <badcloud> HaXeri: doesn't have to be
Nov 14 19:03:45 <badcloud> so... apt-get
Nov 14 19:03:53 <badcloud> thanks
Nov 14 19:06:51 <befr0d> FAM is awesome
Nov 14 19:06:57 <badcloud> noted :)
Nov 14 19:10:41 <merlin1991> FAM comes from deeper layers of hell
Nov 14 19:12:56 <NIN101> so we are bound to use the superfast HAM otherwise we risk fscking the system. Great. 
Nov 14 19:17:09 <merlin1991> exactly :D
Nov 14 19:17:12 <NIN101> :-)
Nov 14 19:17:25 <merlin1991> one could do patches for ham though
Nov 14 19:17:39 <merlin1991> I guess a lot of hams processing can be cached at some point
Nov 14 19:24:28 <NIN101> I think it's ok to install stuff like GSM/3G switcher over FAM, but I would never use apt, FAM etc. to update CSSU, install kernels etc. If one wants to be on the safe side, he should use HAM and nothing else.  Using something different is the users own problem and choice.  
Nov 14 19:26:39 <Sicelo> but ham wasn't able to downgrade kernel for me. apt was only option, and went well
Nov 14 19:34:52 <merlin1991> NIN101 the problem with fapman is that it runs apt-get autoremove
Nov 14 19:35:00 <merlin1991> wich is known to be b0rked on the n9
Nov 14 19:35:03 <merlin1991> err n900
Nov 14 19:35:23 <merlin1991> apt ist just fine when you know when to restart the device afterwards
Nov 14 19:39:49 <NIN101> once an "apt-get upgrade" _killed_ my system :-). every program segfaulted.  
Nov 14 19:40:11 <Sicelo> ;)
Nov 14 19:41:26 <Sicelo> btw, since apt-get upgrade does the actual upgrading, what is cli command to check which packages are upgradeable, without actually doing so? (or is it apt-get -s upgrade?)
Nov 14 19:42:42 <neal> -s does a dry run
Nov 14 19:48:56 <merlin1991> Sicelo: I use -s to check
Nov 14 19:49:01 <merlin1991> dunno if there's a better way
Nov 14 19:56:44 <Sicelo> i'm rtfm now to see
Nov 14 20:00:25 <Sicelo> apt-get -u upgrade, according to manual.. but, hmm
Nov 14 20:09:56 <neal> the woodchuck packages need one more (positive) vote to enter extras.  Can someone please review them?  Thanks.  http://maemo.org/packages/package_instance/view/fremantle_extras-testing_free_armel/murmeltier/0.4~20111102-7/ (and s/murmeltier/apt-woodchuck/)
Nov 14 20:31:22 <DocScrutinizer51> ~aegis
Nov 14 20:31:23 <infobot> 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
Nov 14 20:36:17 <DocScrutinizer51> unrelated ^
Nov 14 20:38:33 <Shapeshifter> DocScrutinizer: if I were to connect two batteries to my n900, what would I have to do? is it possible to just connect them using Y cables to each pin or is it more conplicated?
Nov 14 20:38:57 <Shapeshifter> DocScrutinizer: or to get to the point: I'm pondering if it is possible to hot-swap batteries
Nov 14 20:39:02 <Shapeshifter> in theory
Nov 14 20:39:44 <neal> Shapeshifter: there are a couple of posts about both on talk.maemo.org
Nov 14 20:40:35 <Shapeshifter> neal: ah yes, found some
Nov 14 20:40:52 <SpeedEvil> Connecting two batteries of different charge levels is a very bad dea
Nov 14 20:41:06 <SpeedEvil> It may cause very large currents to flow, and internal fuses to blow.
Nov 14 20:41:16 <ShadowJK> + and - connected together, third pin only from one battery. Batteries must be same state of charge when you connect them together (this means you can't connect batteries together for purpose of hotswapping)
Nov 14 20:41:35 <ShadowJK> and by "+ and - connected together" I mean + from bat1 to + on bat2, etc
Nov 14 20:42:10 <Shapeshifter> mh, too bad
Nov 14 20:42:44 <Shapeshifter> ha I didn't expect hot swapping while actually connected to a wall charger to be such a big deal
Nov 14 20:42:57 <Shapeshifter> I assumed *that* would simply work
Nov 14 20:43:58 <ShadowJK> It's not able to consume sufficient amount of power through that path :)
Nov 14 20:44:19 <ShadowJK> Or not able to regulate it fast enough
Nov 14 20:44:44 <ShadowJK> so it only works if no gsm/3g activity goes on when the battery is removed
Nov 14 20:45:04 <ShadowJK> (and by default the device shuts down if it notices battery removed)
Nov 14 20:45:51 <Shapeshifter> that's a shame.
Nov 14 20:46:57 <ShadowJK> I do occasionally do hotswap though
Nov 14 20:47:17 <ShadowJK> If I come home from work and have to leave again for a longer time, I might hotswap in a charged battery
Nov 14 20:48:05 <Shapeshifter> ShadowJK: from cars I know that there are things called 'isolators' which make it possible to use 2 batteries in one car with one battery not draining the other one (i.e. no cross-battery energy flow). any idea if something like that exists in miniature?
Nov 14 20:48:17 <SpeedEvil> In principle, yes.
Nov 14 20:48:52 <SpeedEvil> In practice, it's more difficult with low voltages.
Nov 14 20:48:58 <SpeedEvil> And nobody makes them
Nov 14 20:49:02 <Shapeshifter> mh.
Nov 14 20:51:16 <ShadowJK> actually..
Nov 14 20:52:10 <SpeedEvil> And yes, there are dieal diode chips which would work
Nov 14 20:53:44 <ShadowJK> I was just going to paste link to one :O
Nov 14 20:56:52 <ShadowJK> ideal diode search on ebay gives 1 result (but no the right thing), so nobody makes ready circuits :)
Nov 14 21:10:28 <rly> I tried tuxracer just for fun. I managed to exit in an unconventional way. Is there also a conventional way to exit a full-screen command line application?
Nov 14 21:17:33 <ShadowJK> ctrl-backspace? 
Nov 14 21:17:36 <ShadowJK> camera door? :)
Nov 14 21:17:38 <badcloud> anyone here installed 'remind' on their n900?
Nov 14 21:18:36 <badcloud> it's my calendar of choice but I ****ed up my system trying to install it via dpkg
Nov 14 21:21:00 <SpeedEvil> camera button, then exit camera app drops you to desktop
Nov 14 22:01:02 <rly> SpeedEvil: pressing the 'turn off' button also does that.
Nov 14 22:09:24 <Tenzen> Does anyone know if i'm able to unlock a nokia N900 by flashing it?
Nov 14 22:10:24 <MohammadAG> If the SIM's locked it's probably stored in CAL, so flashing it won't change anything
Nov 14 22:10:43 <MohammadAG> (I didn't know operators sold it locked, I knew they could, just didn't know there was one doing it)
Nov 14 22:11:49 <SpeedEvil> I doubt it's in CAL.
Nov 14 22:11:54 <SpeedEvil> I'd suspect it's in the modem
Nov 14 22:12:54 <Tenzen> I only suspect it is locked. I'm not certain eyt
Nov 14 22:13:00 <Tenzen> yet*
Nov 14 22:13:42 <Tenzen> But i live in New Zealand (where there are only 3 main cell network providers), so i wouldn't be surprised if they do lock it.
Nov 14 22:13:49 <V13> ehlo... Any hints on how to detect current desktop orientation as it was added in latest CSU? I have a desktop widget that needs to change behavior when desktop switches to portrait mode.
Nov 14 22:15:28 <Tenzen> if it is locked in the CAL, is there any easy way to unlock it? ie not through a dodgy website
Nov 14 22:17:28 <Sicelo> Tenzen: what exactly is locked? 
Nov 14 22:23:28 <Tenzen> I haven't bought it yet, and they haven't replied to my questions about the item yet either. I'm just afraid that if i buy it and the sim is locked to the Vodafone NZ network, i won't be able to use it on the 2degrees network i use.
Nov 14 22:24:22 <merlin1991> ~seen jonwil
Nov 14 22:24:29 <infobot> jonwil <~jonwil@27-33-137-199.static.tpgi.com.au> was last seen on IRC in channel #maemo, 9h 2m 49s ago, saying: 'which would be nice'.
Nov 14 22:25:21 <luke-jr> SIM is locked? O.o
Nov 14 22:40:11 <DocScrutinizer> SpeedEvil: yes, it's definitely in cmt
Nov 14 22:42:21 <DocScrutinizer> Tenzen: no, it's not possible to unlock SIMlocked N900
Nov 14 22:42:52 <DocScrutinizer> except of course with proper unlock code
Nov 14 22:43:32 <DocScrutinizer> this code is device/IMEI specific though, and only the provider who has set the lock can give you the unlock code
Nov 14 22:57:27 <badcloud> device off, two-three yello led blinks when charger plugged in, then just a flash of the green (then shuts off)
Nov 14 22:57:36 <badcloud> s/yello/yellow
Nov 14 22:57:42 <badcloud> is this normal?
Nov 14 22:58:00 <ShadowJK> how long are the blinks?
Nov 14 22:58:04 <ShadowJK> and how long in between them?
Nov 14 22:58:46 <ShadowJK> and do you get them right away when you plug in charger, or do they appear some time later?
Nov 14 23:00:00 <badcloud> ShadowJK: takes time before they start blinking
Nov 14 23:00:15 <badcloud> and they blink like regular charge
Nov 14 23:00:38 <badcloud> then one green, screen off except for dot animation
Nov 14 23:00:44 <badcloud> then screen goes blank
Nov 14 23:01:23 <badcloud> I checked if battery is fully charged, it's 88%
Nov 14 23:01:44 <badcloud> couldn't find a relevant bug reprt
Nov 14 23:01:48 <badcloud> report even
Nov 14 23:02:13 <ShadowJK> got anything exotic installed, like power kernel?
Nov 14 23:02:21 <ShadowJK> uboot, multiboot
Nov 14 23:02:24 <badcloud> nope
Nov 14 23:02:27 <ShadowJK> rescue console
Nov 14 23:02:31 <ShadowJK> or whatever it was called
Nov 14 23:02:50 <ShadowJK> weird
Nov 14 23:03:06 <badcloud> only thing somewhat advanced was a clocking util I think
Nov 14 23:03:09 <badcloud> but didn't use it
Nov 14 23:03:32 <ShadowJK> So it's in act_dead mode probably, and crashes when charge reaches full
Nov 14 23:03:47 <badcloud> ShadowJK: sounds about right
Nov 14 23:04:08 <badcloud> when I unplug, screen is faintly on
Nov 14 23:04:11 <badcloud> then dimms
Nov 14 23:04:13 <badcloud> dims
Nov 14 23:05:00 <DocScrutinizer> screen going black after knightrider dots animation isn't unusual, just it should come out of it. Unrelated to charging/battery
Nov 14 23:05:05 <ShadowJK> I use act_dead so rarely, most of the time if battery ran empty I start the device as soon as there's enough power to do so
Nov 14 23:05:28 <ShadowJK> so I'm not that familiar with how it's supposed to behave in that "pretend we're dead and just charging" mode :)
Nov 14 23:06:08 <badcloud> ShadowJK: use act_dead? wasn't something I actively did
Nov 14 23:06:51 <ShadowJK> it's what the "pretend device is 'off' and only 'charging'" mode is called
Nov 14 23:07:05 <ShadowJK> or "device off and charging" mode
Nov 14 23:07:18 <badcloud> so, in order for me to get a full charge, phone has to be on?
Nov 14 23:07:55 <ShadowJK> no
Nov 14 23:08:28 <badcloud> battery was at 88% on boot (after unplugging charger)
Nov 14 23:08:41 <badcloud> doesn't seem full
Nov 14 23:09:34 <DocScrutinizer> unrelated
Nov 14 23:09:35 <ShadowJK> All 3 battery meters are not synchronized with the charger, and the one used by the UI is fully reset on boot, and it makes a guess when started
Nov 14 23:10:32 <ShadowJK> Also, the "charge full" event is unrelated to both charger status and battery meter status :-)
Nov 14 23:10:47 <badcloud> I'm all confused now
Nov 14 23:11:02 <badcloud> shouldn't the act_dead be issued at 100% bat cap?
Nov 14 23:11:06 <ShadowJK> In practice most of the time for me the green light turns on half an hour or an hour after battery can be considered "full"
Nov 14 23:12:05 <ShadowJK> no act_dead mode is when device is booted in order to display the charging led and turn led from yellow to green, but prevented from showing the normal user interface
Nov 14 23:13:03 <merlin1991> in the harmattan docs there's eve docs on where to place deamon start scripts to have them running in act_dead :D
Nov 14 23:13:59 <ShadowJK> Does aegis actually let you have apps running when in "Off, charging only" state on N9? :)
Nov 14 23:14:49 <badcloud> couldn't find anything that would change charging habits
Nov 14 23:14:54 <badcloud> in terms of apps installed
Nov 14 23:15:33 <ShadowJK> Anything that changes the kernel, or modifies the boot process
Nov 14 23:15:48 <merlin1991> dunno ShadowJK at least the documentation tells you where to place configuration for deamons to start in act_dead
Nov 14 23:16:11 <ShadowJK> "clocking app" might have pulled in power-kernel, if you're talking about overclocking or something..
Nov 14 23:16:22 * ShadowJK forgets what power-kernel's real name is
Nov 14 23:16:28 <badcloud> turns out I didn't install anything like that
Nov 14 23:17:29 * badcloud contemplates if it's worth reflashing over something like this
Nov 14 23:17:55 <badcloud> just reflashed over some dpkg screw-over
Nov 14 23:18:13 <ShadowJK> Well I wouldn't
Nov 14 23:18:26 <ShadowJK> but that's because I can't remember the last time my device would've been turned off :P
Nov 14 23:18:34 <badcloud> good point
Nov 14 23:19:26 <ShadowJK> my spare N900 with uboot does some extra reboots on "recover from flat battery"
Nov 14 23:22:27 <badcloud> nutbar
Nov 14 23:25:32 <DocScrutinizer> listen!! do you? IT's UNREALTED TO CHARGING
Nov 14 23:26:33 <badcloud> DocScrutinizer: then what does it pertain to
Nov 14 23:26:52 <DocScrutinizer> sth in rootfs messing up your boot
Nov 14 23:27:14 <badcloud> DocScrutinizer: is there some file I should be comparing with stock?
Nov 14 23:27:28 <DocScrutinizer> how would you do that?
Nov 14 23:27:49 <badcloud> dl stock rootfs onto host pc, extract
Nov 14 23:28:05 <ShadowJK> "extract" part is kinda nontrivial :-)
Nov 14 23:28:06 <DocScrutinizer> AIUI you got no access to device whatsoever
Nov 14 23:28:20 <ShadowJK> DocScrutinizer, nah his device boots fine
Nov 14 23:28:43 <DocScrutinizer> err so what's the problem then ? o.O
Nov 14 23:28:58 <ShadowJK> it reboots on full-battery when in "off" mode and on charger :-)
Nov 14 23:29:06 <badcloud> ShadowJK: not reboot
Nov 14 23:29:14 <badcloud> "charges" for a few secs
Nov 14 23:29:17 <badcloud> then shuts off
Nov 14 23:29:29 <badcloud> I don't see a solid green
Nov 14 23:29:33 <badcloud> just a blink
Nov 14 23:29:40 <DocScrutinizer> that's normal
Nov 14 23:29:44 <DocScrutinizer> during boot
Nov 14 23:30:20 <badcloud> shouldn't I see a solid green led?
Nov 14 23:30:27 <DocScrutinizer> eventually yes
Nov 14 23:30:54 <badcloud> ok
Nov 14 23:31:58 <DocScrutinizer> though no matter if you plug in the powered-down device to charge it, or you power up device via power button, you always *boot* the system, just the boot looks slightly different. So in the end it's still the booting that got messed up on your device
Nov 14 23:31:58 * ShadowJK sleeps
Nov 14 23:32:24 <badcloud> well, after Nokia usb screen, 2-3 yellows, one green, knightrider, faint blue/black screen, device is OFF off, no leds, nothing going on the screen
Nov 14 23:32:32 <DocScrutinizer> badcloud: you got powerkernel or hotspot?
Nov 14 23:32:38 <badcloud> DocScrutinizer: nope
Nov 14 23:33:42 <DocScrutinizer> when device is in that OFF off state, while still hooked to USB wall charger, what happens when you press the power button?
Nov 14 23:33:56 <badcloud> short or long
Nov 14 23:33:58 <DocScrutinizer> will the white LED ramp up in brightness?
Nov 14 23:34:24 <DocScrutinizer> badcloud: well, does sth happen when you press short?
Nov 14 23:34:28 <badcloud> yes when long press
Nov 14 23:35:04 <badcloud> long=same as before+normal boot
Nov 14 23:35:10 <badcloud> gonna try short now
Nov 14 23:36:44 <DocScrutinizer> what's happening when you lock screen/kbd with slider switch when device is full booted up to GUI and connected to charger? what does the LED show then?
Nov 14 23:36:53 <badcloud> short=white for the duration of press
Nov 14 23:37:00 <DocScrutinizer> :nod:
Nov 14 23:37:59 <badcloud> booting
Nov 14 23:38:32 <DocScrutinizer> first blind shot to the blue: mce.ini messed up, or bme confused
Nov 14 23:39:12 <badcloud> ^ no led
Nov 14 23:39:21 <DocScrutinizer> do you think your device charges even when it doesn't show the yellow/green LED?
Nov 14 23:39:26 <badcloud> but I think the setting for led charging is off
Nov 14 23:39:34 <DocScrutinizer> LOL
Nov 14 23:39:52 <badcloud> DocScrutinizer: I figured that doesn't effect the charge mode when off
Nov 14 23:39:58 <badcloud> only when in user gui
Nov 14 23:40:06 <DocScrutinizer> I dunno, but obviously it does
Nov 14 23:40:36 <DocScrutinizer> replace /etc/mce/mce.ini by original
Nov 14 23:41:17 <badcloud> DocScrutinizer: turned setting on
Nov 14 23:41:21 <DocScrutinizer> and - you know it - you owe me a beer for kidding me that much ;-D
Nov 14 23:41:23 <badcloud> locked kbd, yellow
Nov 14 23:41:48 <badcloud> done, draft or bottle? ;)
Nov 14 23:41:58 <DocScrutinizer> bottle please
Nov 14 23:42:19 <badcloud> shutting off and trying "off" charging mode
Nov 14 23:43:06 <badcloud> leds work
Nov 14 23:43:38 <badcloud> jeebus that was a dumb, nervewrecking, dumb dumb dumb ride
Nov 14 23:44:18 <DocScrutinizer> so make it 2 beer
Nov 14 23:44:26 <badcloud> 6-pack
Nov 14 23:45:03 <badcloud> so, the notification light setting *is* supposed to affect "off" charging mode?
Nov 14 23:45:28 <badcloud> ^ I'll owe you a chaser for the answer to that one
Nov 14 23:45:43 <DocScrutinizer> whatever it is "supposed to", it obviously does
Nov 14 23:46:16 <DocScrutinizer> I guess there's no clear definition on what it is supposed to do regarding that
Nov 14 23:46:21 <badcloud> +ok
Nov 14 23:46:27 <badcloud> s/+ok/ok
Nov 14 23:46:42 <DocScrutinizer> unless you mean what it "usually does on other devices" - then no clue
Nov 14 23:46:53 <badcloud> that's what I meant
Nov 14 23:47:00 <DocScrutinizer> I never disabled charging indication
Nov 14 23:47:25 <badcloud> well there you go
Nov 14 23:47:34 <badcloud> thanks a beer, I mean billion
Nov 14 23:47:36 <DocScrutinizer> I'm messing with my LED patterns on a way more evil level ;-D
Nov 14 23:47:58 <badcloud> oh?
Nov 14 23:48:36 <DocScrutinizer> direct edit in mce.ini, and the comment there is rather fuzzy about the 5 possible states for a pattern 
Nov 14 23:49:14 <DocScrutinizer> if we take that for a definition what it's "supposed to do" then I have to say probably not even the original author was clear on it
Nov 14 23:49:25 <badcloud> haha
Nov 14 23:50:05 <DocScrutinizer> quote>>>
Nov 14 23:50:08 <DocScrutinizer> # ScreenOn - 0 only show pattern when the display is off
Nov 14 23:50:10 <DocScrutinizer> #            1 show pattern even when the display is on
Nov 14 23:50:11 <DocScrutinizer> #            2 only show pattern when the display is off, including acting dead
Nov 14 23:50:13 <DocScrutinizer> #            3 show pattern even when the display is on, including acting dead
Nov 14 23:50:14 <DocScrutinizer> #            4 only show pattern if the display is off, or if in acting dead
Nov 14 23:50:16 <DocScrutinizer> #            5 always show pattern, even if LED disabled
Nov 14 23:50:58 <DocScrutinizer> so if you explain that to me so it makes sense, I owe *you* a beer
Nov 14 23:52:59 <SpeedEvil> the 'even' patterns, you mean?
Nov 14 23:53:05 <DocScrutinizer> yep
Nov 14 23:53:16 <DocScrutinizer> 4 and 5 sound like on crack
Nov 14 23:53:42 <SpeedEvil> They're different
Nov 14 23:54:01 <SpeedEvil> 5 will show in act_dead, display off, display on
Nov 14 23:54:29 <DocScrutinizer> >>only show pattern when the display is off, including acting dead<< vs  >>only show pattern if the display is off, or if in acting dead<<
Nov 14 23:54:34 <DocScrutinizer> my head bursts
Nov 14 23:55:36 <SpeedEvil> Silly.
Nov 14 23:56:13 <badcloud> unrelated, what would be the best way to install remind without breaking apt?
Nov 15 00:07:18 <badcloud> drinks for everyone on the house if you help me with that one
Nov 15 00:13:39 <badcloud> night
Nov 15 00:48:25 <jonwil> hmmm, https://bugs.maemo.org/show_bug.cgi?id=12388#c3 is interesting
Nov 15 00:48:26 <povbot> 04Bug 12388: Missing source code of certificate control panel applet
Nov 15 00:51:46 <merlin1991> yes
Nov 15 00:51:53 <merlin1991> and as said we'd need the 0.1.4 source
Nov 15 00:54:15 <jonwil> I mean in the context of the Nokia guy saying that the code probably shouldnt have been released in the first place
Nov 15 00:55:00 <merlin1991> it does have gpl on it though
Nov 15 00:55:10 <merlin1991> also according to the bug  report
Nov 15 00:55:22 <jonwil> trye
Nov 15 00:55:27 <jonwil> true, that code has GPL
Nov 15 00:55:58 <merlin1991> so technically we should have the full code anyway when the package comes with a gpl copyright
Nov 15 00:56:15 <jonwil> if the code is 100% Nokia, they own it and can do what they like with it
Nov 15 00:57:09 <merlin1991> sure, but they can't ship a binary with gpl attached without giving out source
Nov 15 00:57:29 <SpeedEvil> Sure they can.
Nov 15 00:57:38 <SpeedEvil> The GPL notice means nothing legally.
Nov 15 00:57:47 <SpeedEvil> It's an indication that the code may be GPL licenced.
Nov 15 00:58:00 <merlin1991> bah
Nov 15 00:58:06 <SpeedEvil> I can ship a GPL notice with a windows DVD, it doesn't mean it's GPL.
Nov 15 00:58:17 <merlin1991> well if they keep it closed they should not attach the notice :D
Nov 15 00:58:20 <merlin1991> sure you're right
Nov 15 00:58:26 <jonwil> The 0.1.4 version in PR1.3 has a "nokia proprietary" license file and a comment in the changelog of "Also replaced the faulty  debian/copyright file with proper license info."
Nov 15 00:58:39 <merlin1991> interesting
Nov 15 00:58:40 <jonwil> indicating that "closed" is the intended license for that package
Nov 15 00:59:01 <luke-jr> jonwil: too late.
Nov 15 00:59:12 <luke-jr> I don't think it would fly in court.
Nov 15 00:59:16 <merlin1991> oh it would
Nov 15 00:59:27 <jonwil> I think legally we can use the 0.0.9 code we do have
Nov 15 00:59:30 <luke-jr> they distributed it under GPL
Nov 15 00:59:31 <merlin1991> yes
Nov 15 00:59:35 <jonwil> and I dont think Nokia can or would stop us
Nov 15 00:59:35 <luke-jr> they can't change it later
Nov 15 00:59:37 <DocScrutinizer> how would you attach a GPL notice to a binary?
Nov 15 00:59:48 <merlin1991> DocScrutinizer: it comes with the deb
Nov 15 00:59:52 <jonwil> but they have no obligations to release code for 0.1.4 or any other later version
Nov 15 00:59:53 <luke-jr> DocScrutinizer: easily. it effectively means nobody else can redistribute :/
Nov 15 00:59:55 <DocScrutinizer> it does?
Nov 15 00:59:57 <merlin1991> the copyright file in the deb claims a gpl
Nov 15 01:00:30 <DocScrutinizer> then they must release the source ;-D
Nov 15 01:00:32 <luke-jr> DocScrutinizer: well, you can only distribute with code, and if you don't have the code, not at all
Nov 15 01:00:40 <merlin1991> as jonwil said, if anything they'd have to disclose the source of the deb that came with gpl notice
Nov 15 01:00:41 <luke-jr> DocScrutinizer: copyright holder is not obliged to do anything
Nov 15 01:01:01 <luke-jr> but it DOES mean we're free to reverse engineer it
Nov 15 01:01:01 <jonwil> Nokia is the copyright holder and is not obligated to release anything at all
Nov 15 01:01:22 <jonwil> me, I intend to use the 0.0.9 code we do have and the 0.1.4 binaries and see if I can update it to be functionally equivalent to to the 0.1.4 version
Nov 15 01:03:35 <SpeedEvil> The key is - does the person releasing it under GPL have the legal authority do do so.
Nov 15 01:03:54 <SpeedEvil> If they do - then even if they do it by mistake - they can't reverse that act - if someone takes a copy.
Nov 15 01:04:07 <SpeedEvil> If they don't - the act of including the GPL notice is a no-op.
Nov 15 01:08:34 <DocScrutinizer> also some version released under GPL doesn't oblige the (c)-holder to do same with next revision
Nov 15 01:11:00 <DocScrutinizer> after it's the original author in the end who has claims on what's been done with the source, and only that original author can enforce GPL on others using his code 
Nov 15 01:11:11 <DocScrutinizer> after all*
Nov 15 01:11:20 <SpeedEvil> Where in some cases, original author may not be a person.
Nov 15 01:11:41 <DocScrutinizer> we, the end users, have no rights at all, except those the original author grant to us
Nov 15 01:12:51 <DocScrutinizer> and we can't legally enforce a 2nd stage offender to comply to the GPL, only original author who holds (c) can do that
Nov 15 01:13:03 <DocScrutinizer> usually
Nov 15 01:13:29 <jonwil> great, gitorius is being annoying
Nov 15 01:13:31 <DocScrutinizer> though this might depend on particular state's jurisdiction and legislative
Nov 15 01:19:32 <jonwil> great, http://gitorious.org/omap3camera just doesnt want to let me get any code
Nov 15 01:22:43 <jonwil> seems like the git repo just vanished off the face of the earth somehow
Nov 15 01:25:59 <jonwil> although from the looks of it, everything for the N900 camera is open source anyway except for camera-ui (which has an open clone in CSSU)
Nov 15 01:30:23 <luke-jr> SpeedEvil: it at least moves the liability, I would think
Nov 15 01:30:48 <SpeedEvil> luke-jr: Indeed.
Nov 15 01:30:54 <luke-jr> if you use code with the reasonable belief it's available under the GPL, and someone distributed it from Nokia that way without authorization, I can't see you being held liable for it
Nov 15 01:30:57 <SpeedEvil> Sure.
Nov 15 01:32:31 <jonwil> hmmm no, camera isnt open, there is libomap3cam and omap3camd, both of which appear to be closed source.
Nov 15 01:32:40 <jonwil> Lets see who actually uses that lib
Nov 15 01:32:47 <jonwil> i.e. what links to it
Nov 15 01:41:14 <jonwil> hmmm, actually, I wonder just what bits are in the non-working git repo at http://gitorious.org/omap3camera
Nov 15 01:58:56 <jonwil> hmmm further analysis suggests its just kernel code
Nov 15 01:59:15 <jonwil> meaning that omap3camd and libomap3cam are definatly closed
Nov 15 02:00:35 <jonwil> actually wait, http://www.gitorious.org/omap3camera/mainline/commits/6252d87 implies it might not be closed
Nov 15 02:01:27 <jonwil> or not, thats kernel code
**** ENDING LOGGING AT Tue Nov 15 02:59:57 2011