**** BEGIN LOGGING AT Sun Jun 19 02:59:57 2011 Jun 19 03:11:56 while I'm at it: http://maemo.cloud-7.de/maemo5/usr/sbin/ seems to have a copy I accidentally left here public, for those already screwed by CSSU wrt to default camera-ui Jun 19 03:13:59 cool thanks Jun 19 03:24:15 I wonder if CSSU maintainers pondered there *might* be processes calling maemo-launcher directly via execve() or whatever, and just setting argv[0]="canera-ui", which will inevitably fail when original camera-ui.launch got nuked Jun 19 03:25:24 well, it will fail anyway for canera-ui, but should launch original maemo cam with "camera-ui" ;-P Jun 19 03:25:47 but doesn't when executable got nuked Jun 19 03:54:36 http://www.meegoexperts.com/2011/06/evidence-rumours-meego%e2%80%99s-death-greatly-exaggerated/ Jun 19 03:54:40 Love the tagline. Jun 19 04:09:04 It's pining for the Fjoords. Jun 19 04:10:23 DocScrutinizer, tegra uses a cortex a9 with soft mmu i think Jun 19 04:41:01 Wacky. Jun 19 04:41:08 On inspecting ppp logs. Jun 19 04:41:22 Jun 19 03:06:34 pad pppd[19800]: sent [LCP ConfAck id=0x6 ] Jun 19 04:41:39 * SpeedEvil wonders why my ISP objects to chars 30 and 31 Jun 19 04:42:24 (t-mobile, UK) Jun 19 04:43:18 ver: so you plan to compile maemo for a different hw platform? won't fly for all I know Jun 19 04:44:13 SpeedEvil: lost me Jun 19 04:44:31 asyncmap is 'please don't send these control chars' Jun 19 04:44:39 in a ppp auth sequence Jun 19 04:44:48 aah Jun 19 04:45:27 fail to decode char(30) / (31) there Jun 19 04:45:38 0xa0000 Jun 19 04:45:47 oh wait, that's 5 digits, weird Jun 19 04:45:59 It's a bitmap of chars, including 0 Jun 19 04:46:10 o.O Jun 19 04:47:20 in my book 0x0000 == b[31:0]0 Jun 19 04:48:05 umm Jun 19 04:48:07 yeah Jun 19 04:48:46 I don't understand what the leading 'a' is for. Jun 19 04:48:55 * DocScrutinizer neither Jun 19 04:49:28 tbh I never even heard of asyncmap before Jun 19 04:49:39 It's ooold. Jun 19 04:49:47 I was bored enough to be reading ppp logs. Jun 19 04:50:47 And hacking on a proof of concept non-working demo for the web compressor Jun 19 04:51:25 aah web compressor Jun 19 04:51:47 I#m just wrapping my head around 802.15.4 Jun 19 04:52:21 It'd be nice if zigbee had taken off, and gotten interoperable switches actually cheap Jun 19 04:52:43 Saw the /. on the nanonote Jun 19 04:52:48 yeah Jun 19 04:52:49 had no modpoints though Jun 19 04:53:10 zigbee is >> 802.15.4 though Jun 19 04:53:14 yes Jun 19 04:53:18 lots more stack. Jun 19 04:53:27 wpan only specifies phy and mac layer Jun 19 04:53:36 And zigbee is annotingly nonfree Jun 19 04:53:41 yep Jun 19 04:53:59 while wpan is as free as it gets Jun 19 04:55:13 the wole purpose of wpan been a *simple* protocol, as wifi and BT were considered too complex :-P Jun 19 04:55:56 wpan otoh should survive >1 year on passive standby - one of its design goals Jun 19 04:56:04 The raw air-format of wifi isn't too bad. Jun 19 04:56:06 from bat that is Jun 19 04:56:19 It's just that there are so damn many layers. Jun 19 04:56:39 yeah, as are in zigbee probably Jun 19 04:56:47 Indeed. Jun 19 04:57:36 clustered trees in wpan are just compex enough and exceed the mere 802.15.4 specs, thus creating need for crap like zigbee Jun 19 04:58:08 I'd prefer the LISS approach Jun 19 04:58:13 KISS even Jun 19 06:40:18 tzafrir: hi? Jun 19 06:47:20 tzafrir_laptop: so I i didnt hear from you Jun 19 07:47:40 hm does anyone know how to make links in terminal open with firefox? Jun 19 07:50:14 i use browser switchboard, but when i click links, nothing happens Jun 19 07:50:39 if i run 'browser' in a terminal, firefox opens Jun 19 07:51:05 known problem or just me? Jun 19 10:44:31 Greece is socialist?? Jun 19 10:50:26 any of you guys played with qimsys from extras-devel? Jun 19 10:54:46 @psycho_oreos: not yet, but it looks a lot better than the n900's software keyboard Jun 19 10:56:32 i must say i was a bit saddened that the cssu stuff hasnt given me gapless playback, BUT... Jun 19 10:56:41 ...awesome work on improving the terminal! Jun 19 10:56:54 @HRH_H_Crab: blame MAFW Jun 19 10:56:57 i've not had to use "reset" once! Jun 19 10:57:07 it doesnt matter how many screens within screens im in, Jun 19 10:57:13 keys are doing what they are supposed to! Jun 19 10:57:20 patches welcome for gapless playback :p Jun 19 10:57:37 which reminds me, i need to ask MAG for Home and End buttons for XTerm Jun 19 10:57:48 Hurrian, the latest qimsys (version 2.0) prevents one from using the default onscreen keyboard Jun 19 10:58:13 can't use imswitch Jun 19 10:58:14 ? Jun 19 10:58:45 psycho_oreos, permanently? Jun 19 10:59:10 MohammadAG, I'm guessing at this point in time, permanently yes Jun 19 10:59:43 psycho_oreos, and uninstalling it won't bring that back? Jun 19 11:00:12 MohammadAG: im not trying to nag. i totally understand that there must be some horrific reason why it isnt trivial to implement it. Jun 19 11:00:16 I've removed qimsys-* and still unable to get onscreen keyboard back. My other n900 without having installed the latest qimsys-* but has literally the exact same setup as the one having issues shows no virtual keyboard losses. Jun 19 11:00:38 MohammadAG, nope, not even downgrading, removing and even purging it Jun 19 11:00:53 interesting, /me installs it Jun 19 11:00:54 psycho_oreos, tried running imswitch? Jun 19 11:01:31 psycho_oreos, is it a big improvement over the stock keyboard? Jun 19 11:01:41 Hurrian, no clue, you mean hildom/mscim switcher thing? did that and restarted osso-xterm twice, didn't work Jun 19 11:01:58 MohammadAG, I couldn't get qimsys to work lol Jun 19 11:02:24 not only did I lose virtual keyboard, I couldn't type Japanese at the time Jun 19 11:02:50 @HRH_H_Crab , i'm guessing MAFW lets go of the pulse pipe Jun 19 11:02:58 i'm yet to look at the code Jun 19 11:03:14 hi MohammadAG Jun 19 11:03:15 wait, so it's only japanese? Jun 19 11:03:26 hey cloudyLights Jun 19 11:03:54 anyone experienced in gstreamer and C can add gapless playback Jun 19 11:04:03 no, it doesn't input anything when one enables qimsys (2.0) Jun 19 11:04:06 the python app I worte is not packaged well Jun 19 11:04:11 well, whatever the case, massive thanks to whoever sorted out the terminal! Jun 19 11:04:13 you rock! Jun 19 11:04:15 will try to ask in 5hr here Jun 19 11:04:30 i use my n900 for shelling in to my machines at home a great deal Jun 19 11:04:38 also wanted to say thanks for CSSU Jun 19 11:04:43 and i frequently use nested screens Jun 19 11:04:47 though I did notice there was one extra input box when I was trying to find music to play with mediaplayer Jun 19 11:04:57 apps like mutt were often pretty impossible to use due to keyboard wierdness Jun 19 11:05:01 * Shapeshifter hasn't used the onscreen keyboard even once. doesn't see the point >.> Jun 19 11:05:08 but that now seems to be a thing of the past. Jun 19 11:05:36 best things about CSSU: blurless, zoom-on-press, tactile, camera-ui2, osso-xterm Jun 19 11:06:18 Shapeshifter, I don't use the standard one alot of the time, but its the extra syms that are also affected. This is particularly annoying if you're working with shell and you can use things like: `, |, { and } for instance Jun 19 11:06:32 psycho_oreos, no, I meant is qimsys only for JP input? Jun 19 11:06:47 MohammadAG, it looks like its supposed to be :) Jun 19 11:07:05 * MohammadAG uninstalls Jun 19 11:07:54 mscim allows a broader range of input, though at the time I've only tried Japanese Jun 19 11:09:56 psycho_oreos: I've bound all of those symbols to some combination. I replaced a character I never use (ç on my keyboard) with |, fn+left is {, fn+right is } fn+shift+left is [ fn+shift+right is ], <, > are also there Jun 19 11:11:28 Shapeshifter, nice that's one other way of doing things :) Jun 19 11:11:56 MohammadAG: I see that you fixed the osso pdf app Jun 19 11:12:21 MohammadAG: is the source for calender app also open? Jun 19 11:12:29 can we add features there? Jun 19 11:13:26 cloudyLights, no, I only merged patches for it Jun 19 11:13:35 the calendar app is closed, engine's open Jun 19 11:14:08 what do you think will be the future for getting a better calender? Jun 19 11:14:11 gpe? Jun 19 11:15:53 HI Jun 19 11:16:12 Have enabled sshd on my maemo Jun 19 11:16:25 Disabled root login (for security) Jun 19 11:16:25 zzbottom: did you change root password? Jun 19 11:16:32 Yep Jun 19 11:16:39 But that is not the issue Jun 19 11:16:57 I have created a user, that I want to use for ssh Jun 19 11:17:02 and the to su to root. Jun 19 11:17:04 So... Jun 19 11:17:12 I need to add this to the sudoers. Jun 19 11:17:23 Let me just check to be sure... Jun 19 11:18:02 W.M.O.: Sudo relies on /etc/sudoers, a configuration file that details who can run what commands. /etc/sudoers is traditionally modified by using the 'visudo' command. DO NOT DO THIS ON MAEMO. In maemo sudoers is regenerated periodically from config snippets in /etc/sudoers.d/, using visudo can cause the slow death of your system as your edit gets pushed further down /etc/sudoers. Jun 19 11:18:13 I use visudo and I add ALL=(ALL) ALL Jun 19 11:18:31 >>>>>/etc/sudoers is traditionally modified by using the 'visudo' command. DO NOT DO THIS ON MAEMO. Jun 19 11:18:39 ok Jun 19 11:18:40 ... Jun 19 11:18:41 hmm Jun 19 11:18:52 then how to sudo? Jun 19 11:18:54 zzbottom, the "root" cmd doesnt work for you? Jun 19 11:19:15 Don't think so through ssh. Jun 19 11:19:29 It does from localhost... Jun 19 11:19:31 ssh is just u being on a remote machine Jun 19 11:19:40 shouldnt be any diff Jun 19 11:19:41 but let me try remote Jun 19 11:20:11 is not in the sudoers file. This incident will be reported. Jun 19 11:20:29 is the output when I "root" from the user I have created. Jun 19 11:20:30 what the crap, sudoers gives each program an entry Jun 19 11:20:55 nokia sure did choose to do everything ass-backwards Jun 19 11:21:13 nokia == muppets Jun 19 11:21:14 not cool Jun 19 11:21:38 ok gotta go back to coding Jun 19 11:21:40 laterz Jun 19 11:22:10 On the device, "root" gives me root access. (from the standard "user" login), default. Jun 19 11:22:43 But when ssh'ing and login on to the device using a newly created user, I get: Jun 19 11:22:52 is not in the sudoers file. This incident will be reported. Jun 19 11:23:08 Any workaround for this when sudoers is not possibel? Jun 19 11:25:19 usermod -G wheel username Jun 19 11:25:22 add a script in /etc/sudoers.d Jun 19 11:26:03 I know the wheel from BSD - this will also work on maemo? Jun 19 11:26:34 trying Jun 19 11:26:35 ... Jun 19 11:27:23 it works if group wheel is OK'd in sudoers Jun 19 11:28:09 funny... now I can't gain root on the devise itself... probably because I disabled permitrootlogin in sshd_config Jun 19 11:28:12 )-: Jun 19 11:28:37 it sucks Jun 19 11:29:33 now, how can I edit sshd_config... only through R&D? Jun 19 11:29:38 or any tricks? Jun 19 11:31:21 meego => mount ubifs Jun 19 11:33:05 Not running meego yet Jun 19 11:33:29 zzbottom: http://wiki.maemo.org/Root_access Jun 19 11:34:05 rootsh is installed Jun 19 11:34:19 maybe installing it again will fix sshd_config. Jun 19 11:34:30 sshd_config doesn't prevent root access on device Jun 19 11:35:28 Hurrian: HRH_H_Crab: ( i'm guessing MAFW lets go of the pulse pipe) exactly. you need to set up a new pipe for each file to playback, and there's no proper way to sync them Jun 19 11:35:58 im no coder but that sounds like the horrific reason i alluded to earlier. Jun 19 11:36:12 Mohammad: It shouldn't, but is has been breaking my local gainroot... Jun 19 11:36:26 changing sshd_config what the only thing I did. Jun 19 11:38:06 docscrutinizer: and i guess putting two files together in the background and streaming it into the pulse pipe would eat memory? Jun 19 11:39:04 something like cat Jun 19 11:40:02 zzbottom: refer to http://wiki.maemo.org/User:Joerg_rw/tools for sudoers Jun 19 11:41:19 MAFW doesn't handle pulse pipes, only gstreamer ones Jun 19 11:41:30 it constructs a new pipe for each file, which is kinda dumb Jun 19 11:41:31 Strange - I does actually gain root on the console even though I get the reply " is not in the sudoers file. This incident will be reported." Jun 19 11:41:48 when typing "root" in a console on the device Jun 19 11:41:51 hm.... Jun 19 11:42:35 Hurrian: who cares, but what will you do at end of 2nd file? ;-P Jun 19 11:43:48 zzbottom: take care when messing with etc/sudoers. Real chance to go for a reflash Jun 19 11:43:59 docscrutinizer: keep piling up files :P Jun 19 11:44:19 and if gst is to blame, i think someone may have to do some backporting Jun 19 11:44:56 yeah, that's what you'd like to do, feed audio framework with seamless data. Alas this friggin mafw doesn't work like that Jun 19 11:45:53 basically, in my book a mediaplayer has to do the audio streaming under close control itself, not offload to a stupid framework Jun 19 11:46:31 as soon as you got seamless, you run into next obstacle: crossfade Jun 19 11:47:00 mafw is crap basically Jun 19 11:47:20 it's nice for simple tasks, but completely inappropriate for a decent mediaplayer Jun 19 11:47:37 too bad a lot of maemo apparently relies on it Jun 19 11:48:03 no problem, you don't HAVE TO use it Jun 19 11:48:55 you can open a simple audio device and do all the streaming locally, instead of offloading the whole task to mafw Jun 19 11:49:17 and I bet canola etc are doing exactly that Jun 19 11:49:23 ah Jun 19 11:49:32 Doc: think I will leave sudoers alone... don't really want to flash the device... Jun 19 11:50:23 zzbottom: tweaking sudoers.d/* with an own new file, or even editing files there, is all way safer than messing with etc/sudoers Jun 19 11:51:06 * MohammadAG wouldn't use Phonon over mafw Jun 19 11:51:08 just saying Jun 19 11:51:31 just don't forget to keep one root shell open for emergency recovery :-) then run update-sudoers and try if all your root and whatnot commands work as supposed Jun 19 11:51:49 phonon, BLAERGH! Jun 19 11:52:05 even more abstraction Jun 19 11:52:11 MAG: as it should damn well be, maemo is GTK based, you'd like to keep the middleware centered around GNOME Jun 19 11:52:38 i cringe every time a package pulls QT dependencies and fills rootfs Jun 19 11:52:51 wut, gnome???? another BLAERGH! Jun 19 11:53:02 sshd_config changed back... Jun 19 11:53:03 Doc: GNOME 2 was great Jun 19 11:53:04 ~nuke g* Jun 19 11:53:05 * infobot prepares 100 missle silos, and targets them at g* ... B☢☢M! Jun 19 11:53:11 now I can SSH as root. Jun 19 11:53:29 even though this is not what I want, I will stick with this... Jun 19 11:53:40 zzbottom: btw ssh and sudoers are unrelated Jun 19 11:53:40 leaving it more unsecure... Jun 19 11:53:41 )-: Jun 19 11:53:47 Hurrian, sorry, but I prefer Qt UIs over Gtk ones Jun 19 11:53:54 Well, Nokia... nice try. Jun 19 11:54:02 at least when writing them, Gtk is way too much effort to work with Jun 19 11:54:17 MAG: yes, Qt gives more bling without relying on hacky image engines Jun 19 11:54:31 zzbottom: huh? Jun 19 11:54:37 Hurrian, that would be QML Jun 19 11:54:43 and.... now I do not get " is not in the sudoers file. This incident will be reported." when typing "root" on the device Jun 19 11:55:06 who of you told it would be easy to provide QT3 in fremantle?! :D Jun 19 11:55:35 thopiekar_, probably me :p Jun 19 11:55:41 though it'd be unthemed etc Jun 19 11:55:48 I mean Nice try nokia to try to join the smartphone marked with Maemo/meego. Jun 19 11:56:30 I was trying to build it but it always fails.. today I found out that Qt4 provides Qt3 support .. Jun 19 11:56:38 but disabled on maemo5 Jun 19 11:56:40 zzbottom: why did you create yet another user that's not "user"? I fail to get the point of that anyway Jun 19 11:57:19 deffo cause problems when apps hardcode for the path /home/user Jun 19 11:57:20 thopiekar_, oh right, I forgot about that, it's just a compile flag Jun 19 11:57:26 --no-qt3-support Jun 19 11:57:29 Would like to use my own login - which is not a default one Jun 19 11:57:36 yep, MohammadAG Jun 19 11:58:12 I hate apps that hardcode paths Jun 19 11:58:15 zzbottom, they didnt even try to topple the iPhones and Androids Jun 19 11:58:32 and furthermore to enhance security. Jun 19 11:58:37 it's not that hard to use gettenv("$HOME") Jun 19 11:58:42 I plan to make a package now that would be able to be installed besides the original qt4 package and which will add qt3 support Jun 19 11:58:42 everything default is bad. Jun 19 11:59:02 thopiekar_, where would that install? Jun 19 11:59:09 zzbottom: there's no enhanced security in a new user that has sole purpose to do "rot" Jun 19 11:59:11 poor marketing, very low hq QC standards for a Nokia phone, standard Nokia UI patterns (INTERNET CONNECTIONS, ANYONE?) Jun 19 11:59:20 "root" Jun 19 11:59:38 MohammadAG: $XDG_CONFIG_HOME >:( Jun 19 12:00:09 standard what? Jun 19 12:00:11 I would need it on my N900 to have MythTV on it.. Jun 19 12:00:11 Doc: ssh'ing is more secure if you can't login with default logins. Jun 19 12:00:14 look at http://wiki.maemo.org/User:Joerg_rw/tools, fix sudo to ask for proper root password whenever somebody tries to "root". That's a bit of security Jun 19 12:00:34 if you can login through root, you can just bruteforce it Jun 19 12:00:47 use proper ssh-keys to log in to user account - that's good for security Jun 19 12:00:48 it is way hardware if you also have to guess the login name Jun 19 12:00:52 the -maemo5 confflag disables qt3-support -.-' what does the -maemo5 at all? Jun 19 12:01:23 thopiekar_, a lot Jun 19 12:01:26 sorry ... it is way HARDER if you also have to guess the login name Jun 19 12:02:16 * thopiekar_ will try to prepare the configure file to force qt3 Jun 19 12:02:23 or wouldn't it be better to have something like fail2ban for that matter? Jun 19 12:02:34 zzbottom: no, login in to an account that has no password-auth allowed is just identically hard no matter what's the account name: impossible without the ssh RSA-key Jun 19 12:02:39 thopiekar_, just remove --no-qt3-support from debian/rules Jun 19 12:02:54 Doc.. but I am not that worried about login in from the device itself... I want to enhance the security from remote with SSH Jun 19 12:02:56 done that but the -maemo5 flag also disables it.. Jun 19 12:02:59 MohammadAG: ^ Jun 19 12:03:39 login from remote is at *least* as secure as login from device itself, quite obviously Jun 19 12:04:48 so if you can't get root from user without proper root password, then same applies to a remote login to account user, after somebody has found a way to steal your RSA-key to even log in to sshd on account "user" Jun 19 12:05:10 but on the device I gain root by just typing "root" when rootsh is installed (as in my case). Yes, that is a big security flaw.. but funny enough, I can live with it as I am always carrying my phone. Jun 19 12:05:51 Through remote SSH, I want to disable all default logins, create my own login and be able to su from here. Jun 19 12:05:55 and if you would have bothered to actually look at what I gave URL to you... Jun 19 12:06:08 you would have seen how to fix this issue Jun 19 12:06:48 yep, I look at your URL, but is only fix the local issue, right? Jun 19 12:07:16 the notion that somebody would attempt a bruteforce login from the network is simply unconceivable to me unless pretty obvious passwords ar chosen Jun 19 12:07:17 it fixes "root" no matter how you issue the command Jun 19 12:07:24 Myu concern is the remote SSH issue, where I want to disable default logins and login with my own login as well as being able to su Jun 19 12:07:47 zzbottom: why's that an "issue"? Jun 19 12:08:25 you are far more likely to have your password keylogged than the ssh daemon bruteforced Jun 19 12:08:33 Choom: hunter2 Jun 19 12:09:03 Choom: I also fail to get it Jun 19 12:09:11 The issue is that you can't add a newly created user to the sudoers as I understand (at least without getting into troubles)... that is an issue Jun 19 12:09:24 nevertheless there's definitely no problem in setting up this weird stuff Jun 19 12:09:27 you can't??? Jun 19 12:09:29 since when? Jun 19 12:09:40 just need to do it properly Jun 19 12:09:46 exactly Jun 19 12:09:59 ruskie: hey. did you ever do something new to your urxvt package? I'm still using yours from like a year ago. Jun 19 12:10:02 Ok... then... how to do it proberly? Jun 19 12:10:11 Shapeshifter, nope Jun 19 12:10:12 though I really miss the purpose of the whole nonsense Jun 19 12:10:30 zzbottom, see /etc/sudoers.d and /etc/sudoers Jun 19 12:10:44 it's explained how you need to add new entries to sudeoers Jun 19 12:10:49 ruskie: aww... but... perl support *sob* Jun 19 12:10:49 nah, see man sudo and man sudoers Jun 19 12:11:19 Ok, looking back... Jun 19 12:11:59 DocScrutinizer, iirc sudoers file explains that it is generated Jun 19 12:12:09 yes Jun 19 12:12:18 that's what update-sudoers is for Jun 19 12:12:40 as we got an obsolete sudo that doesn't know how to include sudoers.d/* Jun 19 12:12:45 the rest is just use the right directives in a sudoers.d/file Jun 19 12:13:02 yep Jun 19 12:13:23 and look at http://wiki.maemo.org/User:Joerg_rw/tools for an example how to meddle with it Jun 19 12:13:32 Mh. Can someone summarize for me what (tremendous) work is needed to run another distro on the n900? Jun 19 12:13:40 Think I am beginning to get the idea of sudoers.d Jun 19 12:15:01 the directives in etc/sudoers and etc/sudoers.d/* are identical in syntax and semantics Jun 19 12:15:23 user isn't supposed to mess with /etc/sudoers Jun 19 12:15:56 as it gets generated when running update-sudoers script Jun 19 12:18:58 ok... Jun 19 12:23:07 in sudoers.d, I got a file named "01sudo". Can I simply just edit this to reflect "user all = passwd: /usr/sbin/gainroot" Jun 19 12:23:27 in order to get prompted for pw on the device when rooting? Jun 19 12:23:38 (Now I am mixing things, I know) Jun 19 12:33:43 Thanks DocScrutinizer - been reading your guideline and putting them together. It does seem that I am able to do what I want to accomplish... Jun 19 12:34:35 (the weird stuff) (-; Jun 19 12:35:01 yw Jun 19 12:35:18 bye Jun 19 12:42:40 can anyone see a problem with the following dbus calls? http://pastebin.com/ayAKZVHa (N900) Jun 19 12:45:17 error is included in the pastebin. Jun 19 13:08:24 running it as root Jun 19 13:08:29 ? Jun 19 15:34:07 hey guys :) Jun 19 15:34:46 if i stick to glib and pure c functions am i good to go for maemo? Jun 19 15:35:44 yes - uif you want a command line app Jun 19 15:35:59 well i will add the gui layer after words Jun 19 15:36:03 afterwards Jun 19 15:36:07 sorry brain went elsewhere Jun 19 15:36:32 gf said something to me Jun 19 15:36:39 :) Jun 19 15:36:51 QML is C++ I think - though I've not doe it Jun 19 15:37:02 yah i want to stick with gtk Jun 19 15:37:18 for the desktop and maemo Jun 19 15:37:24 gtk works Jun 19 15:39:24 :) Jun 19 15:39:42 using plain c is a real throwback to days gone by Jun 19 16:47:23 * SpeedEvil finally starts coding a benchmarker for the webproxy idea using old data. Jun 19 17:05:24 does anyone know how to make links in terminal open with firefox? Jun 19 17:05:27 i use browser switchboard, but when i click links, nothing happens Jun 19 17:05:34 if i run 'browser' in a terminal, firefox opens Jun 19 17:05:37 known problem or just me? Jun 19 17:35:07 QML is C++ I think - though I've not doe it Jun 19 17:35:22 no, C++ is compiled, QML is interpreted Jun 19 17:35:32 The syntax is close to JS Jun 19 17:58:55 Jaffa, I don't buy it. Jun 19 17:59:00 I'm planning on getting the accelerometer readings from the N900 and send it to an opengl application to plot stuff.But the readings seems weird.Moving from position to position,there's a huge change in the values. Jun 19 17:59:03 W Jun 19 17:59:13 will I be able to use it?or will accuracy be a problem? Jun 19 17:59:59 worked fine for me. I used the example on maemo wiki Jun 19 18:00:06 reading the values from /proc Jun 19 18:00:22 petteri: I did the same thing Jun 19 18:00:40 but a small change gives me completely different values Jun 19 18:01:30 fish_sticks: maybe your n900 is on some strange state, you might want to reboot and try again Jun 19 18:01:46 did that,no difference Jun 19 18:03:15 petteri: all I'm asking is will that be accurate enough to plot something on opengl? Jun 19 18:03:39 or will I have say a circle being plotted randomly every second Jun 19 18:03:50 definitely accurate enough Jun 19 18:04:16 fish_sticks: What are you trying to do. Jun 19 18:04:24 thanks and its better I do it via usb right? Jun 19 18:04:39 SpeedEvil: trying to plot something using the accelerometer co-ordinates Jun 19 18:04:45 The accellerometer almost can't be used for integratingchanges in position, it's not possible generally. Jun 19 18:05:10 integrating changes?didnt get you Jun 19 18:05:29 What are you trying to do with the data you get drom the accellerometer. Jun 19 18:05:48 this one I used: http://wiki.maemo.org/Accelerometers#Smoothed_C_interface and got working values Jun 19 18:05:56 For working out which direction 'down' is - it works. Jun 19 18:05:58 sending that data via a socket Jun 19 18:06:02 For much more than that - it's complex. Jun 19 18:06:23 petteri: thanks Jun 19 18:06:31 SpeedEvil: you mean to tell me I wont be able to use it? Jun 19 18:06:58 What are you trying to do with the data - not how are you reading it. Jun 19 18:07:16 I'll take that an draw say a circle using opengl Jun 19 18:07:27 how? Jun 19 18:07:29 Why? Jun 19 18:07:43 planning on implementing say a squash game Jun 19 18:07:51 so that the bat moves as I move the N900 Jun 19 18:07:54 something like that Jun 19 18:08:02 reading my accelerometer several times, I get these: -36 -18 -936, -18 -36 -936, -36 -36 -936, -18 -36 -936, -36 -18 -954 Jun 19 18:08:23 I don't know where my phone is right now, but the values are pretty stable to me Jun 19 18:08:46 Choom: its not stablity that I want,but fluidity Jun 19 18:08:52 It is poor to very poor at determining position. This is because you need to integrate the already noisy accellerations, and try to work out where the normally dominant 'down vector' is. Jun 19 18:08:55 continuous precise variations Jun 19 18:09:10 SpeedEvil: okay Jun 19 18:09:14 Note also that all those readings are divisible by 18 Jun 19 18:09:25 SpeedEvil: didnt get you Jun 19 18:09:39 fish_sticks: Do you understand vectors? Jun 19 18:09:48 SpeedEvil: kindof Jun 19 18:10:24 The accellerometer outputs a fairly inaccurate vector, which measures the accelleration it feels, and its current direction. Jun 19 18:10:40 This is however dependant on the orientation of the phone. Jun 19 18:10:54 SpeedEvil: you mean to tell me it will not tell me the current position? Jun 19 18:11:09 An accellerometer can fundamentally never tell you a position. Jun 19 18:11:17 It measures accelleration. Jun 19 18:11:19 it doesn't tell you a position, it tells you acceleration Jun 19 18:11:22 hmm Jun 19 18:11:32 will a gyro help? Jun 19 18:11:43 You need to integrate accelleration to get to velocity, then integrate velocity to get to position. Jun 19 18:11:53 you need to infer position from the acceleration vector, and ignore the acceleration of gravity Jun 19 18:11:56 And a gyro removes some of the error terms, but not all. Jun 19 18:12:03 got it Jun 19 18:12:20 A gyro makes it easier to work out where 'down' really is - and to subtract this vector from your measured accelleration. Jun 19 18:12:35 But you're still faced with the noise of the accellerometer. Jun 19 18:12:40 hmm Jun 19 18:13:00 so I'll have to take the values,integrate it to get the position huh? Jun 19 18:13:08 but then again,noise is a problem Jun 19 18:13:23 The granularity is 18 milligee. This is an accelleration of 9.8mm/s^2. After 10 seconds, the velocity is uncertain by 9.8cm/s. After 100 seconds - 1m/s. Jun 19 18:13:49 hmm Jun 19 18:14:03 Jaffa, given the WeAreMeeGo revive and whatnot. Jun 19 18:14:13 9.8m/s^2 Jun 19 18:14:19 not mm Jun 19 18:14:25 err - no Jun 19 18:14:32 SpeedEvil: I know its out of scope,but is there another way to do it? Jun 19 18:14:34 I screwed up comprehensively. Jun 19 18:14:44 I meant 9.8mm/s, but left off the factor of 18 Jun 19 18:14:51 So the above is out by a factor of 18 Jun 19 18:14:55 k Jun 19 18:15:17 fish_sticks: you can use the accellerometer basically as a 'joystick' Jun 19 18:15:23 fish_sticks: this works fairly well. Jun 19 18:15:51 hmm k Jun 19 18:16:11 Imagine that there is a stick poking out of the screen, always vertical. Jun 19 18:16:46 so just x and y? Jun 19 18:17:36 X,Y,Z can be mapped onto two angles of the 'stick' - if you ignore the magnitude of the accelleration. Jun 19 18:17:47 k Jun 19 18:18:59 You may or may not find http://wiki.openmoko.org/wiki/Technical:Accelerometer_Fundamentals - which I wrote earlier - of use. Jun 19 18:19:19 will check it out Jun 19 18:27:35 GAN900: Indeed, I'm just trying to manage expectations. Too much TMO reading last week ;-) Jun 19 18:28:01 GAN900: Sounds like something which might be Section=front tonight, but no idea what the best thing(s) to link to is/are Jun 19 18:31:51 Hehe Jun 19 18:32:01 Yeah, I can rumor mongor Jun 19 18:36:21 Jaffa, still ... a thread with nearly 6k posts ... Jun 19 18:36:29 who said that Maemo is dead? ;-) Jun 19 18:36:51 this thread probably has more posts than whole of MeeGo forums :-D Jun 19 18:37:02 It's got a device Jun 19 18:37:04 Go figure Jun 19 18:37:26 haha Jun 19 18:38:01 * DocScrutinizer ponders of any metric which could compare tmo to meego forum Jun 19 18:38:22 would probably need that legendary crap filter Jun 19 18:39:45 GAN900, :-) Jun 19 18:40:32 Does anyone happen to know what gzips maximum compression 'window' is with -9? Jun 19 18:40:50 ROFL @ the blog countdown: http://www.nokiablog.ch/ Jun 19 18:40:51 err, I did, but forgot :-S Jun 19 18:42:11 SpeedEvil: I think I#ve seen it once, in a reasoning about gzip's ram fotprint and speed Jun 19 18:43:40 K. Trying stupid metrics at the moment. Jun 19 18:44:19 'What is the increase in filesize of all of the text/* files from this website, in chronological order that you have ever downloaded gzipped, if you add the latest file' Jun 19 18:44:23 For example Jun 19 18:45:53 that's not only about window size but also about dict Jun 19 18:46:06 I know. Jun 19 18:46:18 and if the dict gets reused on new files Jun 19 18:46:30 That's what I'm trying to avoid. Jun 19 18:46:59 err, avoid? Jun 19 18:47:09 reusing the 'dictionary' Jun 19 18:47:09 the first gzip, then tar Jun 19 18:47:19 then* Jun 19 18:47:44 I want to see how to best apply compression with a dictionary comprising the whole of the last 256M or whatever files you've downloaded. Jun 19 18:47:59 otherwise gzip will use one dict for whole tar'ed set of files Jun 19 18:49:03 gzip always works on one file only, and uses dict for the whole file Jun 19 18:49:41 I'm not sure that's true - I seem to recall the dictionary can get at least partially purged - though I've forgotten the details. Jun 19 18:50:03 sure, it adapts Jun 19 18:50:19 which makes sense Jun 19 18:50:24 usually Jun 19 18:51:06 anyway NB gzip seems pipe safe Jun 19 18:51:22 so it can't do optimum Jun 19 18:51:40 for ptimum you'd need a two-pass compression Jun 19 18:51:49 Indeed. Jun 19 18:52:37 and that's the reason why the dict must change during compression Jun 19 18:53:03 and due to ram limitations dict can't grow infinitely Jun 19 18:53:36 FIrst pass test results. 12M uncompressed text in 325 files. gzipping them individually 2.5M. gzipping them in chronological order - 2.4M. bzip2 - 1.3M Jun 19 18:54:56 what's "chronological order"? you concatenated them? Jun 19 18:55:08 Concatenated in the order they were downloaded. Jun 19 18:55:18 :nod: Jun 19 18:55:21 amazing Jun 19 18:55:32 which factor? Jun 19 18:55:40 factor? Jun 19 18:55:45 --best ? Jun 19 18:55:48 -9 Jun 19 18:55:49 -9 Jun 19 18:56:10 even more amazing Jun 19 18:56:38 I'd have expected gzip to boost performance way more on longer files Jun 19 18:57:03 seems dict and window are ridiculously small Jun 19 18:57:19 Moo all Jun 19 18:57:26 remoo Jun 19 18:57:28 yes - it's possible that's not the best order Jun 19 18:57:29 Any news while I have been away? Jun 19 18:57:39 nah Jun 19 18:58:10 some RM-680 powertop-alike screendump Jun 19 18:58:31 we see it has 512MB and 1GHz Jun 19 18:59:01 Taaasty Jun 19 18:59:04 and obviously been configured to use same amount of swap Jun 19 19:00:03 you also could read out some of the possible speedsteps of that CPU, but I guess that's particularly boring Jun 19 19:00:03 hi Jun 19 19:00:18 how to play mov file on maemo? Jun 19 19:00:35 Sorting by ebay site - 1.0M - I need to work out how to sort by URL. Jun 19 19:01:20 SpeedEvil: bored? ;-D Jun 19 19:01:59 or are you trying to find a metrics for me, to compare net info in tmo vs meego forum? Jun 19 19:02:17 X-D Jun 19 19:02:27 DocScrutinizer: It's a log from a program called wwwoffled which is a web proxy - directly accessing the URL is not quite trivial. These are random browsing logs from a period in 2009. Jun 19 19:02:57 sure Jun 19 19:03:09 I miss the general purpose Jun 19 19:03:34 I'm trying to make metrics for a compressing web proxy. So I can inspire someone to actually write teh thing. Jun 19 19:03:44 To save on everyones bandwidth. Jun 19 19:03:49 * DocScrutinizer idly ponders about his dupfile killer app Jun 19 19:03:59 AAAHHH now! Jun 19 19:04:47 For example - nice graphs saying 'with a cache size of 130M, with the example dataset, the average reduction over time is: Jun 19 19:05:00 well, there's a thing called compressed html iirc Jun 19 19:05:03 Rather than handwavey postulation on no data. Jun 19 19:05:09 yes - it's not quite the same. Jun 19 19:05:20 And would give similar results to gzipped html Jun 19 19:05:41 http://wiki.openmoko.org/wiki/Server:WebProxy - I suspect I've posted before. Jun 19 19:06:12 The key is diffing against the best matching page the client has downloaded in the last 250M (or whatever) - not simply compressing. Jun 19 19:06:34 indeed Jun 19 19:06:46 sound rationale Jun 19 19:07:26 though building a dict based on that 250MB would probably work even better Jun 19 19:07:40 Sometimes, yes. Jun 19 19:08:03 given it's a comprehensive dict Jun 19 19:08:28 so gzip is out Jun 19 19:08:31 At some point, you have to give up and accept the fact that you're not going to be able to do it properly due to serverside CPU limits. Jun 19 19:08:59 :nod: Jun 19 19:09:39 I think the dict of gzip is using some LRU algo to purge tuples Jun 19 19:09:50 arr are these tuples at all? Jun 19 19:09:55 prolly Jun 19 19:11:20 the problem is you need to stay in sync, and recover from datastream dropouts Jun 19 19:11:23 ~seen Venemo Jun 19 19:11:28 venemo <~Timur@fedora/Venemo> was last seen on IRC in channel #meego, 3d 21h 32m 34s ago, saying: 'user0, yeah, I did need to get my device to R&D mode when I used the rescue initrd'. Jun 19 19:11:48 even md5 sum over a 250MB cache are too heavy I guess, on mobile Jun 19 19:11:59 The mobile does nothing. Jun 19 19:12:07 Oh - right Jun 19 19:12:46 And yes - the mobile only changes the cache based on a complete transaction that it has notified the server of. Jun 19 19:12:52 Or something. Jun 19 19:13:26 hmm, that's possible but still icky and tricky Jun 19 19:14:05 what if your notifications doesn't make it thru to server? what if the ACK gets lost? Jun 19 19:14:07 'last seen update' sent with next request Jun 19 19:14:58 doesn't help mobile out, as you don't want to keep several transactions open to rollback them, at mobile Jun 19 19:15:31 yes, you can resend transaction codes to make sure server is in sync Jun 19 19:15:54 You only commit finished transactions, and then notify the server of your state on next request - server has to rollback Jun 19 19:16:51 hmm, a very very complicated protocol Jun 19 19:17:03 yes - it's going to be moderately complex. Jun 19 19:17:44 I wonder if the protocol overhead will eat up the net savings on compression Jun 19 19:18:01 Shouldn't. That's not the fun part though. Jun 19 19:18:12 Mobile says 'send me ebay.com'. Jun 19 19:18:25 and 'I want compressed pictures and stuff' Jun 19 19:19:13 Server then diffs ebay.com with the cache, and then does the same with all of the files that the mobile is going to request shortly as they are new files it hasn't seen, and sends it in one lump. Jun 19 19:19:40 cookies are annoyin Jun 19 19:19:56 LOL Jun 19 19:20:41 probably any embedded scripts also are annoying Jun 19 19:22:04 The evil nasty hack version might run a similar browser as the client on the server over the request. That's probably going too far though. Jun 19 19:36:55 so, any demo of the hoped-for device would still mean a long wait time b4 it would be available for the masses, is that right? Jun 19 19:37:49 so i should still follow thru with my n900 order :) Jun 19 19:41:19 sirdancealot: I'd wait until Tuesday at least. If it is what we hope, there might be a glut of N900s on ebay ;-) Jun 19 19:41:47 indeed Jun 19 19:41:54 ic. thanks Jaffa Jun 19 19:42:30 What's the source of Thursday? Jun 19 19:42:37 sirdancealot: What if (and this'd be crazy time), Nokia said on Tuesday "we'll sell these for $199 to anyone with a maemo.org or meego.com account" Jun 19 19:42:50 SpeedEvil: Thursday? Jun 19 19:42:55 err Jun 19 19:42:55 though... If I were to get me a N9 (which I'm definitely not ;-P), I'd wait to ebay my N900 until I hold the new device in my greedy hands Jun 19 19:42:57 tuesday Jun 19 19:43:03 DocScrutinizer: Indeed Jun 19 19:43:09 Nokia Connections Jun 19 19:43:23 Jaffa, I'll eat my shirt. Jun 19 19:43:29 SpeedEvil: http://www.itwire.com/your-it-news/mobility/47833-nokias-sneak-peek-of-new-market-disputing-device-coming-june-21 and http://twitter.com/#!/WeAreMeego/status/82254724727975936 and http://www.nokiaconnection.net/ Jun 19 19:43:33 some market disrupting device Jun 19 19:43:37 Jaffa, then i should get one of those accounts quick :) Jun 19 19:43:53 Jaffa: Ah Jun 19 19:44:01 Jaffa: Thanks Jun 19 19:45:31 Jaffa: (we sell to maemo.org account) I'd say "Nokia you're nuts, I ask for device price getting calculated by the formula (500 - ) Jun 19 19:45:54 DocScrutinizer: Genius. So they give me a device and a load of cash? Great :-) Jun 19 19:46:03 me too ;-P Jun 19 19:46:29 at least most of the trolls wouldn't get a subsidized device Jun 19 19:47:05 DocScrutinizer: Aye. Most of the trolls only have talk.maemo.org accounts, so let's limit it to garage.maemo.org accounts (i.e. the Midguard DB) Jun 19 19:47:13 and when I think about it, I'd prefer NOBODY gets a subsidized device, when the alternative is trolls getting them way too cheap Jun 19 19:47:43 DocScrutinizer: I'd certainly prefer targetted giveaways to general subsidisation (if it included trolls) Jun 19 19:47:52 Anyway, bed time Jun 19 19:47:57 Night! Jun 19 19:48:02 cya Jaffa Jun 19 19:48:03 night Jun 19 19:48:41 garage account sounds fair Jun 19 19:49:11 though iirc they did the min-karma-needed thing before Jun 19 19:49:36 wich seems a more general approach Jun 19 19:50:32 I'd think they should have had a device queue since long Jun 19 19:51:16 they are missing out on feedback of a small crowd of smart knowledgeable people Jun 19 19:52:01 evidently their internal evaluation and tester process didn't always perform up to expectations Jun 19 19:53:22 anyone know how electric guitars work? Jun 19 19:53:31 but then this once more would need a mindset based on cooperation and trust into FOSS community, rather than "serving them" Jun 19 19:53:37 Fields Jun 19 19:53:42 wondering if something like a pocket amp app would be doable Jun 19 19:53:54 MohammadAG: huh? Jun 19 19:54:03 electromagnetism? Jun 19 19:55:07 MohammadAG: you ask if the headset/AV mic input is capable of dealing with signals from an E-guitar pickup? short answer: No Jun 19 19:56:01 no, I Jun 19 19:56:10 fucking hildon taking focus Jun 19 19:56:25 I'm asking how electric guitars send signals Jun 19 19:56:32 or what type of signals those are Jun 19 19:56:58 low voltage high impedance electrical waves in a cable ;-) Jun 19 19:57:55 in a way similar to a classical passive microphone, though still sufficiently different so you can't use a mic input for E-guitar Jun 19 19:58:51 which probably each band can tell you - you can't use the vocals amp for guitar Jun 19 19:59:47 at least that's what I've been told - I'm no musician Jun 19 20:00:17 you're an engineer, you probably know more about this :P Jun 19 20:00:43 I think impedance is sufficiently different so this usually doesn't pan out Jun 19 20:01:31 some numbers from my arse: guitar:100k-Ohm, mic:300Ohm Jun 19 20:02:26 this time by rear might not have as good numbers as it usually does Jun 19 20:02:49 short: I may be wrong here Jun 19 20:04:39 Useless. Jun 19 20:10:12 <_trine> dont electric guitars have a pickup to catch the string vibrations Jun 19 20:10:42 <_trine> nothing to do with impedance Jun 19 20:12:25 GAN900: huh? Jun 19 20:12:42 _trine: what do you think is a pickup? Jun 19 20:13:03 what do you think is the name of it'S electrical properties? Jun 19 20:13:10 MohammadAG: http://en.wikipedia.org/wiki/Pickup_(music_technology)#Pickup_sound Jun 19 20:13:21 DocScrutinizer, your ass. Jun 19 20:13:43 <_trine> the pickups output could be any impedance Jun 19 20:13:49 <_trine> you wanted it to be Jun 19 20:13:52 Amp input expected to be ~1M Ohm for guitar amps, while voltage can be significantly higer than with mics Jun 19 20:14:24 GAN900: My ass been excellent as usual Jun 19 20:14:52 so im writing an identica client ... wondered what select * from itemsres would be importantfeatu Jun 19 20:15:02 wow Jun 19 20:15:08 love the random paste there Jun 19 20:15:09 sorry Jun 19 20:15:14 let me try that again Jun 19 20:15:35 im writing an identica client and wondered what features you think important Jun 19 20:15:37 :) Jun 19 20:15:48 <_trine> the required impedance would only be the impedance which would match the amplifier you wanted to use so you would get maximum power transfer Jun 19 20:16:32 _trine: orly. OK, I guess my EE is worth a crap then Jun 19 20:17:14 <_trine> is orly a word Jun 19 20:17:30 in my experience the amp get built to specs of picup, not vice versa Jun 19 20:17:54 <_trine> well that is irrelevant Jun 19 20:18:21 indeed, which part of your explanation isn't - for mohamnmad's problem? Jun 19 20:18:51 <_trine> only that they match is relevant to a point however for power you would choose a lower impedance Jun 19 20:19:21 sure, yes. no doubt Jun 19 20:19:57 _trine, Orly is the name of a white owl Jun 19 20:20:00 <_trine> maximum power transfer occurs when the impedance is matched Jun 19 20:20:12 I don't choose anything, I have to choose a pickup that sounds nice, or even a guitar that sounds nice and comes with a pickup that *has* a certain impedance Jun 19 20:20:31 _trine: and cake is best when baked Jun 19 20:21:07 <_trine> I have no idea what the characteristic impedance is of a guitar pickup Jun 19 20:21:15 <_trine> is Jun 19 20:21:27 aah, good. Now we got something to discuss Jun 19 20:22:02 <_trine> but whatever it is it would need to match the amplifier Jun 19 20:22:43 <_trine> when I was a lad Jun 19 20:23:01 well, isn't that what I said? Modulo you don't 'match' impedance for audio frequency, rather you make sure input impedance of amp >> source impedance Jun 19 20:23:06 <_trine> we could buy head phones which were 20000 ohms Jun 19 20:23:17 <_trine> now they are 4 and 8 ohms Jun 19 20:24:02 where >> usually means factor 10 at least Jun 19 20:24:40 * _trine opens a bottle of white wine and slurps Jun 19 20:24:49 the impedance matching is relevant for RF Jun 19 20:25:04 to get a good SWR Jun 19 20:25:33 <_trine> stand wave ratio does not make a lot of difference Jun 19 20:25:39 <_trine> standing Jun 19 20:25:53 <_trine> no power is lost Jun 19 20:26:48 for LF you want a most neutral transmission which is more like probing than like power transmission, which means the probe (=amp) has to have an impedance high enough to not change the properties of device under test (guitar) Jun 19 20:27:53 so while pickups are typically 100kR, the amp has in imput impedance of 1MR or higher Jun 19 20:28:46 <_trine> not necessarily Jun 19 20:28:56 and the voltage is really high with ~1V RMS Jun 19 20:29:24 _trine: sorrry, I start to get bored and call "Off Topic" Jun 19 20:29:34 <_trine> good idea Jun 19 20:31:15 <_trine> I still have not got pyrit yet :( Jun 19 20:31:27 MohammadAG: anyway, you can't possibly connect an E-guitar to N900 Jun 19 20:31:47 the strings are supposed to be slightly magnetic Jun 19 20:32:01 could n900 pick the vibrations up? Jun 19 20:32:11 no Jun 19 20:32:37 no sensor for that Jun 19 20:33:16 you *could* try optical pickup, but the video cam is way too slow for that Jun 19 20:33:33 <_trine> could you not arrange to vibrate the touch screen Jun 19 20:33:37 as is proxy sensor, and ALS Jun 19 20:34:43 hey, how's about using the best builtin sensor we got for that: the microphone ;-P Jun 19 20:35:24 really, sounds like Friday afternoon Jun 19 20:36:24 <_trine> we're only idly chatting ,,, chewing the cud Jun 19 20:36:41 we could try to exploit FMTX and FMRX to build a doppler radar X-P Jun 19 20:39:04 * DocScrutinizer wonders idly if he enjoyed more the rant about N9 quad core CPU, than discussing EE basics and Oz pickups Jun 19 20:54:16 MohammadAG: anyway, you can't possibly connect an E-guitar to N900 damn, I was expecting to have some fun with that Jun 19 20:55:24 MohammadAG: you at very least would need some simple transformer, or better a buffer amp powered by micbias Jun 19 20:56:04 I could provide schematics if you're inclined :-) Jun 19 20:56:08 DocScrutinizer, well I was actually expecting mic input to work fine, but oh well :p Jun 19 20:58:00 actually a FET amp could work fine for this usecase Jun 19 20:58:24 woukd probably consist of <10 components Jun 19 21:00:48 I'd even manage to build that thing into the 6.3mm female connector of the needed adapter Jun 19 21:01:17 that's however EE weekend fancy Jun 19 21:03:33 no need, there are pocket amps, was just wondering if the N900 would be enough instead of carrying one extra piece of HW :p Jun 19 21:03:48 :nod: Jun 19 21:04:18 anyway your pocket amp rarely supports sox 8-) Jun 19 21:04:52 * DocScrutinizer tinks sox >> a suitcase full of effect boxes Jun 19 21:05:30 but then again the latency... oh well Jun 19 21:05:57 you're back to cubase ;-P Jun 19 21:24:02 DocScrutinizer, just saw your post saying the battery applet wouldn't work if we removed BME and implemented stuff in kernel space, https://gitorious.org/community-ssu/status-area-applet-battery :P Jun 19 21:24:30 hmm Jun 19 21:24:34 so what? Jun 19 21:25:08 I don't think I said exactly that Jun 19 21:26:52 it's not about reimplementing a bat applet that ships around broken hal and reads bat status directly from gauge. It's more about reinplementing hald-addon-bme to put the rest of system back to mormal operation Jun 19 21:27:54 breaking hal will break an unknown but huge number of apps in turn Jun 19 21:29:10 so main prerequisite to replace bme is to get a replacement for hal-addon-bme, not to code a new battery applet that can work without hal Jun 19 21:31:27 don't get me wrong, a working bat applet is a nice thing, for e.g. hostmode. But its purpose to keep system architecture intact is rather limited Jun 19 21:33:29 I heard meego has libbme now, which should give quite some clues about how to re-implement for bme replacement all the interfaces that used to talk to bme Jun 19 21:34:30 probably hal-addon-bme not found there, as meego might have nuked hal, sth we can't reasonably do for maemo Jun 19 21:35:09 which is the whole crux Jun 19 21:36:50 you are better off to rewrite hal-addon-bme so it talks to gauge directly rather than to bme. Rather than reimplementing all the bits that rely on hal for battery monitoring Jun 19 21:38:18 by following the applet replacement path you end with a system that *looks* like maemo but is completely incompatible in the end Jun 19 21:39:08 we could as well switch to meego then Jun 19 21:40:55 in fact a meego battery-eye would work better on that suggested bme-free architecture than the maemo app of same name Jun 19 21:50:33 hi Jun 19 21:50:39 stskeeps already had written a hald-addon-bme replacement for mer. He might give some hints if he doesn't decide to be a grinch Jun 19 21:50:47 can I format my micro sd to ubifs? Jun 19 21:50:58 would it be better then ext3? Jun 19 21:51:15 ubifs is designed for raw flash devices. Jun 19 21:51:22 The microSD is not a raw flash device. Jun 19 21:51:34 It's got a little CPU in it that pretends it's a block device. Jun 19 21:51:41 so ext3 is my best option? Jun 19 21:51:54 actually the microSD *might* already use ubifs internally ;-) Jun 19 21:52:00 Dunno - just that ubnifs is not an option at all Jun 19 21:52:12 o ok Jun 19 22:00:47 hi .. i've noticed that facebook as a chat account is missing now.. how can i get it back? Jun 19 22:02:03 btw. got the cssu installed.. Jun 19 22:02:41 Hello #maemo Jun 19 22:04:09 how to .mov file in maemo? Jun 19 22:05:12 play* Jun 19 22:11:59 juozapas: kmplayer Jun 19 22:18:38 Hurrian: thx, i will try Jun 19 22:19:37 so apparently, epitaph is still begging for donations for his "tune-up" script\ Jun 19 22:19:55 eh... Jun 19 22:21:24 anyone notice that the added a few words to the description in the signature? Jun 19 22:21:55 "Nokia N900 Smartphone Performance Efficient Enhance Optimization Tune-Up Utilities" -- sure looks like SEO Jun 19 22:23:03 i love people who base their work on others' stuff and ask for donations Jun 19 22:23:12 libgiomm-2.4.so.1.2.0 an other lib*.so Jun 19 22:23:21 OH SHIT NIGGA WHAT ARE YOU DOING Jun 19 22:23:54 they're all *.sh scripts, what the hell is it doing depending on GNOME libraries? Jun 19 22:24:14 4.0M ./opt/n900-tune-up/coreutils-gnu_8.4-2maemo1_armel.deb Jun 19 22:24:17 tf? Jun 19 22:24:25 wtf ?!even Jun 19 22:24:43 it installs coreutils. wtf? Jun 19 22:25:24 paste the scripts I guess Jun 19 22:25:25 * DocScrutinizer digs for his tmo trolls basher hat Jun 19 22:27:17 4.0K ./usr/sbin/gainroot Jun 19 22:27:20 dohoho Jun 19 22:27:24 wasn't that the guy who 'invented' mv and ln, or sth along that line and level knowledge? Jun 19 22:28:10 sth like ""chmod +x, secret sauce pro tip"" Jun 19 22:28:12 personally, i find it embarrasing to stomp over package maintainer scripts as seen in tune-up Jun 19 22:28:38 he should've learned regular expressions Jun 19 22:29:10 he's already ripping off code, should'nt be too hard for him to rip Transitions Control Jun 19 22:29:31 huh? Jun 19 22:30:06 ripping community code and scripts, repackaging as his own Jun 19 22:30:11 also: >I want 5 EUR and you can pay with Paypal. Jun 19 22:30:17 >5 EUR Jun 19 22:30:25 now he's going too far Jun 19 22:30:41 without that, its technically nagware Jun 19 22:30:56 repackaging things as your won, and then asking for money, that's clearly a ban and bashing Jun 19 22:31:42 got a pointer to the stolen stuff? Jun 19 22:31:57 plus to original? Jun 19 22:32:07 I feel temped to do sth about it Jun 19 22:33:01 usr/lib/libmlockall -> mlocker (it's in HIS SIG!) Jun 19 22:33:35 8.0K ./usr/share/policy/etc/rx51/syspart.conf => http://talk.maemo.org/showthread.php?t=71115 Jun 19 22:34:04 i'm yet to examine the other scripts Jun 19 22:46:43 Hurrian: kmplayer doesn't play mov also Jun 19 22:47:38 kden, use handbrake to convert Jun 19 23:29:30 it seems texrat is already at it, regarding this "optimizer" Jun 19 23:30:33 DocScrutinizer: you mean the web one - or have I come in partway through Jun 19 23:31:36 nah, this luser that claims to optimize your N900 beyond awesomeness, as he learned about how to use busybox cli commands Jun 19 23:31:51 epitaph Jun 19 23:32:22 http://talk.maemo.org/showpost.php?p=1032281&postcount=343 Jun 19 23:32:41 ah Jun 19 23:32:48 >>Epitaph, your claims to be a victim ring hollow. I'm not sure what your goal is with this "application" (and I use the term loosely) but it doesn't seem in any way honorable.<< Jun 19 23:33:17 period Jun 19 23:36:57 youre stolen 1h of my time by making me read this crappy thread, thank you Jun 19 23:37:27 s,youre,youve Jun 19 23:49:07 sorry, but same here :-) Jun 19 23:58:16 though I had to go thru the nice thread http://talk.maemo.org/showthread.php?t=71115 to finally end up at epitaph's BS Jun 19 23:59:32 but maybe it's been targeted at Hurrian ? ;-D Jun 20 00:02:29 * DocScrutinizer wonders if it's maybe time to publish a few .deb in tmo, that do *real* stuff, real nasty stuff ;-) Jun 20 00:03:14 nagware with a 25min sleep, LAAAAME Jun 20 00:03:43 DocScrutinizer: how is it targeted at me? Jun 20 00:04:10 you came up with this dude epitaph :-) Jun 20 00:04:16 ah Jun 20 00:04:44 well, epitaph's been bitching hard in the kernel-bfs thread about cgroups too. Jun 20 00:04:53 ooh Jun 20 00:05:13 abill_uk's son? Jun 20 00:06:03 nah Jun 20 00:06:15 abill_uk never contributed anything useful Jun 20 00:06:41 * DocScrutinizer ponders to publish a cracker .deb to rip out the nagware part of epitaph's "optimizer" X-P Jun 20 00:07:24 Hurrian: what's epitaph's useful contribution? Jun 20 00:07:43 his older posts were less... clueless and inflammatory Jun 20 00:08:26 * Hurrian ponders to publish a Super Tools DEB - 300MB, and all it does is increment every osso- app by 999 versions Jun 20 00:08:42 I think his very first posts were along the line of "secret sauce tip: use chmod +x to make your scripts start without " Jun 20 00:09:57 and iirc even that had ome bug in it Jun 20 00:10:14 http://talk.maemo.org/showthread.php?p=1032329#post1032329 Jun 20 00:10:21 for the 9000'th time Jun 20 00:10:31 nah, you won't trap me again ;-D Jun 20 00:10:57 nuf of tmo for today Jun 20 00:12:39 so many people are clamoring for nitdroid D: Jun 20 00:12:41 tmo is like salt, or arsenic: you mustn't take too much of it a day Jun 20 00:13:17 it's more like arsenic, that place is TOXIC Jun 20 00:13:42 poor fools, didn't they see they'd get a better droid phone when buying a device that's made for and shipped with genuine andridiot? Jun 20 00:14:36 ~tmo Jun 20 00:14:36 hmm... tmo is http://talk.maemo.org, or too much off-topic, or not ~t-mo Jun 20 00:14:40 actually, s/Maemo/iOS/g s/N900/iPhone 4/g s/N810/iPod Touch/g the past 2 years and it looks EXACTLY like "the other forums" Jun 20 00:14:55 I liked the former def better Jun 20 00:15:18 which was like "trolls, morons, " Jun 20 00:15:59 whatever happened to the community being solely focused on maemo+device? Jun 20 00:16:39 dunno, many gave up on the trolls flooding the place Jun 20 00:16:40 the newer members seem to be... a lot more "enthusiastic" for the HEs on the N900 Jun 20 00:17:20 HE? Jun 20 00:18:09 hacker editions? which would imply a lot of those lusers consider themselves to be hackers Jun 20 00:18:14 hacker editions for the older devices, Community Editions/Developer Editons for the N900 Jun 20 00:18:20 a scary idea Jun 20 00:18:29 such as MeeGo and Android Jun 20 00:18:45 hopefully the same users wont get too bitchy over Cordia+MeeGo Jun 20 00:19:08 who cares about meego, really? Jun 20 00:19:38 ...OMG UNOFFICIAL OPEN SOFTWARE DEVELOPED ON WONKY CLOSED SOURCE SOFTWARE HAS A SHITLOAD OF BUGS Jun 20 00:19:49 I actually can't see any benefit from meego-arm for the average user, neither instanly nor long term Jun 20 00:20:32 Doc: there won't be any benefit from running MeeGo as long as it's /that/ slow on the N900 Jun 20 00:21:02 it won't, nah CAN NOT get faster than maemo is Jun 20 00:21:18 so where's the benefit, again? Jun 20 00:22:07 meego been meant to kick off the future generation of linux devices from Nokia. Evidently there's no such thing like a future linux phone series Jun 20 00:23:11 working vsync (works by wsegl flag though...), btrfs root (a lot faster than ext4 root on the N900) and 2.6.37 (hey, it's a newer kernel! we Jun 20 00:23:23 *we're no longer restricted to the ancient 2.6.28 Jun 20 00:23:32 the whole purpose of meego-arm is a mere self-preservation and continuing useless work to not have to face it's useless Jun 20 00:24:18 Doc: and MeeGo seems to be preinstalled mostly on x86 tablets. Jun 20 00:24:30 I find this kubuntu stuff very interesting. They are targeting now telephony functions for 11.10. Definitly will keep an eye on that, however, I think they will "steel" most stuff from meego, but anyway. Jun 20 00:24:35 couldn't care less, really Jun 20 00:25:36 I can't think of a single itsy bit that meego would allow me to do on N900, which I can't do today Jun 20 00:25:43 au contraire Jun 20 00:26:02 I can not understand why meego it's so much ram. Jun 20 00:26:09 it absolutely academic Jun 20 00:26:20 *eats Jun 20 00:26:21 NIN101, bloat? Jun 20 00:26:24 NIN101: because of MeeGo Handset UX Jun 20 00:26:40 qt-err-foo Jun 20 00:26:47 yes of course, but damn... Jun 20 00:26:50 this js-fake thing Jun 20 00:26:57 QML? Jun 20 00:26:58 ooh Jun 20 00:27:00 yep Jun 20 00:27:04 :/ Jun 20 00:27:35 once again: designed and tailored for laptops, not for embedded Jun 20 00:30:01 honestly, if anyone wanted a big system upgrade, it would be a rebuilt meego-adaptation-n900 system Jun 20 00:30:10 none of that RPM nonsense Jun 20 00:30:24 DEB packaging, hildon desktop+matchbox Jun 20 00:31:52 i still don't understand the switch to RPM Jun 20 00:31:55 I'm quite ignorant about the packaging. Ideally I never touch it anyway, so who gives a f* if it's *.deb or *.rpm? Jun 20 00:32:25 it's like driving a BMW for a few years and then deciding "yeah, that's all nice and good, but i'd much rather be driving a ford focus" Jun 20 00:32:43 pff Jun 20 00:33:04 then again people's decisions never really make sense Jun 20 00:33:17 Only insane wack-jobs care about package management format. Jun 20 00:33:34 I simply say you ideally don't even know what's your package management called Jun 20 00:33:34 It's fundamentally not an important part of the OS. Jun 20 00:33:37 i'm just wondering "why bother switching to something worse"? Jun 20 00:33:52 speedevil: you mean package signing and verification aren't important? Jun 20 00:33:58 waste a bunch of resources and investment on a non-improvement Jun 20 00:33:58 SpeedEvil: +++ Jun 20 00:34:01 it's a large reason of why i stopped using arch Jun 20 00:34:05 Hurrian: In some cases, no. Jun 20 00:34:09 it's THE reason why I don't use arch Jun 20 00:34:44 Signing is rather unimpiortant if you get your packages over https - for example Jun 20 00:34:53 yes, but what I hack the server? Jun 20 00:35:17 yeah, and BMW is using shock absorbers that are painted blue, that's why I never ride a BMW Jun 20 00:36:01 yo Jun 20 00:36:04 sup Jun 20 00:36:47 letting aside signing is a part of RPM as well, why would you care? is it *you* to worry about how _maintainers_ get their shit together? Jun 20 00:37:04 does anyone use NITDroid on eMMC? Jun 20 00:37:32 *BURP* Jun 20 00:37:42 HtheB, i used to, but NITDroid burns my hand. Jun 20 00:37:49 why? Jun 20 00:37:59 yeah, a good question :-P Jun 20 00:38:09 did you use N12? Jun 20 00:38:10 asked best in #nitdroid Jun 20 00:38:18 (everybody on #nitdroid is sleeping, so... dont blame me :P) Jun 20 00:38:19 it seems to be working on some high frequency all the time and on stock voltage Jun 20 00:38:29 Hurrian: not on the latest versions Jun 20 00:38:47 those fools even OC by default Jun 20 00:38:48 but somehow, when I install the kernel, I cant boot up any OS anymore Jun 20 00:39:03 I always end up with flashing the N900 Jun 20 00:39:25 DocScrutinizer: no, you CAN oc it, it isn't on by default Jun 20 00:39:34 well, now you know why there's that part in /topic Jun 20 00:39:45 uhuh Jun 20 00:39:51 been told differently Jun 20 00:39:59 :D ahah Jun 20 00:40:11 neh, the problem is that my old phone didnt had such problem Jun 20 00:40:13 HtheB: read /topic, follow directions Jun 20 00:40:14 I got a new one Jun 20 00:40:24 maybe new hardware rev? Jun 20 00:40:30 no Jun 20 00:40:34 (later version? ) Jun 20 00:40:40 no Jun 20 00:40:53 can you also say "yes"? Jun 20 00:40:59 no Jun 20 00:41:01 :D heh Jun 20 00:42:49 stupid kernel panics Jun 20 00:43:21 honestly it's somehow annoying when you get roughly equal amounts of users hyping here how awesome alternative OS XY, and others whining here how much it sucks and breaks the device. And both completely unrelated to maemo Jun 20 00:44:18 yeah Jun 20 00:44:20 I know Jun 20 00:45:10 though I dont wine about sucking or something like that xD Jun 20 00:45:21 I just wanted to ask if someone has any experience about the latest kernels Jun 20 00:45:32 since somehow it didnt work on my (new) N900 Jun 20 00:45:45 mine broke some time ago, brought it to nokia Jun 20 00:45:54 they sent me a phone back..... Jun 20 00:45:58 Guess which one .... -_- Jun 20 00:46:34 E7 Jun 20 00:46:36 no Jun 20 00:46:39 N8 ¬¬ Jun 20 00:46:41 I went mad Jun 20 00:46:46 ewww Jun 20 00:46:52 Mine is in repair somewhere, I hope it does not happen to me :/ Jun 20 00:46:54 told them that IF they were going to replace my phone, that I want the E7 Jun 20 00:47:04 after some arguing, they accepted it Jun 20 00:47:10 they will send me an E7 Jun 20 00:47:26 though, I'm going to sell it :P Since I already bought a NEW N900 in the shop where I work Jun 20 00:47:32 We still have a couple of them left Jun 20 00:47:39 If I get something different than N900, I'm going to sell it and buy a new one. Jun 20 00:47:41 (bought it for €255) Jun 20 00:48:10 NIN101: im sure u will get a different phone Jun 20 00:48:12 what was broken? Jun 20 00:48:13 the USB? Jun 20 00:48:16 yep Jun 20 00:48:29 u get a different device... Jun 20 00:48:33 100% Jun 20 00:49:32 seems Nokia redefined N900 as broken-by-design and is basically doing a global swap Jun 20 00:49:40 yeah Jun 20 00:49:54 why does the usb broke any way? Jun 20 00:50:03 a pity, for sure also thanks to OCers Jun 20 00:50:07 Doc: unsurprising, they evidently barely spent any money on hardware quality control Jun 20 00:50:17 BS Jun 20 00:50:44 I didnt see any other device where the USB broke off o_O Jun 20 00:51:01 well, that's a design flaw Jun 20 00:51:16 isnt it just the same? Jun 20 00:51:28 i mean, just put some more iron on it lol Jun 20 00:51:37 so it will stick for a longer time :P Jun 20 00:51:39 probably paired with MP issues causing some components not 100% correctly soldered Jun 20 00:51:39 not iron, lead+silver Jun 20 00:51:59 hmmm Jun 20 00:52:10 I think I might even buy a new N900 :( Jun 20 00:52:20 Just in case something happens to my N900 now Jun 20 00:52:30 I dont want to have it swapped again :( Jun 20 00:52:36 I think I might try to find the money to get me another 1 or 2 spare Jun 20 00:52:40 HtheB, can't you just have it repaired? Jun 20 00:52:47 Hurrian: they wont Jun 20 00:52:51 or before it breaks, fix the dry joint Jun 20 00:52:54 yourself Jun 20 00:53:02 That could be an option Jun 20 00:53:13 a reasonable suggestion Jun 20 00:53:20 thats too bad, i like the n900... Jun 20 00:53:23 desolder, resolder with 100% lead, solder to ground plane Jun 20 00:53:44 in fact, why not replace with with a microusb-AB port and fix it Jun 20 00:53:50 though, if I can buy a NEW N900 for €255, and get a swap for E7 (new price €600), then... I can sell the E7 for like €450 Jun 20 00:53:51 DocScrutinizer: sorry, disappeared for a sec, but YES it matters to me, as I *am* a package maintainer Jun 20 00:53:53 (and am not about to solder that delicate parts... :) Jun 20 00:54:01 I am (or was) a Maemo/Meego Developer Jun 20 00:54:04 which means I can make some profit :p Jun 20 00:54:04 not just some user Jun 20 00:54:08 and buy a new n900 lol Jun 20 00:54:56 and having experience with both Deb and RPM, I can confidently tell you my opinion all day long about which one is superior >_> but it'd be a waste of breath Jun 20 00:55:09 or, a waste of finger movement, in this case Jun 20 00:55:18 Hurrian: you mean the on like on the N8? Jun 20 00:55:22 yes. Jun 20 00:55:59 instead of using a female-female adapter, you just have to use the Nokia OTG cable Jun 20 00:56:05 saves a bit of spage Jun 20 00:56:07 *space Jun 20 00:56:53 then i guess you'll just have to use a tiny dremel head to carve the rectangle in the N900's shell Jun 20 00:57:46 hmm Jun 20 00:58:43 heh Jun 20 00:58:55 well, since my n900 is still intact, lets have it this way :P Jun 20 00:59:50 on a side note, i'm working on a way to have the entire thing white Jun 20 01:00:09 white? Jun 20 01:00:14 a white N900. Jun 20 01:00:18 nice :D Jun 20 01:00:28 which leaves the problem, i'll need to somehow paint the digitizer Jun 20 01:00:45 well, if you leave it black, it will be like the Samsung Galaxy S Jun 20 01:00:46 * Hurrian thinks decals Jun 20 01:01:08 sometimes white sux, Jun 20 01:01:17 * HtheB pukes when he thinks about the icrap 4 white Jun 20 01:01:42 HtheB: the white iP4 is actually thicker, due to the UV penentrating the phone Jun 20 01:01:45 i lol'd Jun 20 01:01:54 yeah Jun 20 01:01:59 Hurrian: HtheB: http://talk.maemo.org/showthread.php?p=547991#post547991 Jun 20 01:02:17 DocScrutinizer: :D Jun 20 01:02:20 Exactly what I mean :D Jun 20 01:02:46 nox-: ^^^ Jun 20 01:02:52 if they just do it like that at the Nokia repair center.... they would survive the usb connector breaking thing Jun 20 01:03:09 HtheB: it's not in the service manual, they probably wont do it Jun 20 01:03:30 lol Jun 20 01:03:31 lcuk: lol you found me Jun 20 01:03:47 DocScrutinizer: I lol'd about this: Advice . protect camera lens when solder Jun 20 01:04:01 I think he pwnd his own camera lens Jun 20 01:04:23 you're actually supposed to remove the camera when working on the board Jun 20 01:06:07 no, never heard about that Jun 20 01:07:38 docscrutinizer, lv1+2 service manuals Jun 20 01:07:56 and it's indeed a good idea to think of soldering on those small devices like an operation and cover the complete irrelevant area with some protective material Jun 20 01:08:00 however, us mere mortals don't have the camera removal tool Jun 20 01:08:09 yep Jun 20 01:08:22 :/ Jun 20 01:08:26 I never seen they recommend to always remove the cam module though Jun 20 01:09:16 given the proximity, it'd probably be best to remove it Jun 20 01:09:31 iirc it's just mentioned as one step of a complete disassembly of the device, which doesn't mean you always need to do it Jun 20 01:09:49 why do u need to remove the camera completely? Jun 20 01:09:53 will it burn? Jun 20 01:10:03 :shrug: Jun 20 01:10:12 * DocScrutinizer wouldn't Jun 20 01:10:14 burn it Jun 20 01:10:35 I'd do it when using hot air gun though Jun 20 01:10:51 or IR preheating Jun 20 01:11:37 for fixing USB I definitely would want to save the connector from a mating cycle Jun 20 01:12:05 just covering the lens seems enough Jun 20 01:20:19 anyone did the battery mod? Jun 20 01:23:22 accourding to the website, they already sold 340 batteries Jun 20 01:23:59 so around 170 people are doing the mod :P Jun 20 01:24:13 There are reportedly cells that fit in the n900 and get 1600mAh Jun 20 01:24:42 As in I know someone that's measured the capacity Jun 20 01:24:53 which batt? Jun 20 01:25:11 I forget the details Jun 20 01:25:14 someone is doing some tests, and the best that he found (yet) was 1510 Jun 20 01:25:18 Will ask tomorrow Jun 20 01:25:37 http://talk.maemo.org/showthread.php?t=65568 Jun 20 01:25:51 I think it was from that thread - friend ordered - and got one that gives 1590 on the first couple of cycles Jun 20 01:26:17 hmmm Jun 20 01:27:27 3 cheers for tmo awesomeness to drop pictures as they are only external links and prone to vanish with original URL going down Jun 20 01:27:30 :-((( Jun 20 01:28:21 http://cgi.ebay.com/2-x-Battery-BL-5J-BL5J-Nokia-5800-5230-N900-X6-GIF-/170545202999?pt=PDA_Accessories&hash=item27b5494737 Jun 20 01:28:22 lol Jun 20 01:29:12 1820mah my ass Jun 20 01:30:44 Hurrian: http://www.ofroj.com/wp-content/uploads/classipress/for-sellnokia-n97-32-785399444.png Jun 20 01:31:44 Hurrian: also this http://cdn1.iofferphoto.com/img/item/199/845/700/rfld.jpg Jun 20 01:34:44 White n900 Jun 20 01:36:02 or how about a GOLD n900? http://cgi.ebay.com/Nokia-N900-N-900-Housing-Case-Cover-Stylus-Gold-/170629661144?pt=PDA_Accessories&hash=item27ba5201d8 Jun 20 01:37:43 why not make it out of real gold? Jun 20 01:38:10 . . . . Jun 20 01:38:16 if I was rich :( Jun 20 01:39:39 poor antenna performance - but oh well, those who want to cary a N900 made of gold probably never need to actually use it Jun 20 01:39:55 Gold is also heavy Jun 20 01:40:11 Carbon fibre would at least have some virtues. Jun 20 01:40:25 could probably knock 10 grams off the weight Jun 20 01:40:37 doesn't matter, a pakistany gets paid for sole purpose to carry it the whole day ;-P Jun 20 01:41:17 lol Jun 20 01:41:27 nokia is going to make use of a new material Jun 20 01:41:34 the one they found a couple years ago Jun 20 01:41:38 have you ever heard somebody complaining about weight of his Rolex, or about it not being accurate ;-D Jun 20 01:42:18 lol Jun 20 01:42:48 it probably wouldn't matter if it doesn't run at all Jun 20 01:43:09 Nokia will use Graphene Jun 20 01:43:10 http://conversations.nokia.com/2011/06/14/graphene-nokias-future-super-material/ Jun 20 01:43:20 omfg Jun 20 01:44:15 While being thin, it’s the strongest material ever tested, having a breaking strength 300 times greater than steel and is also the lightest material ever, best intrinsic conductor and super-flexible, too. Jun 20 01:44:46 Graphene was found by a dutch guy who won a nobel prize for finding this out Jun 20 01:45:02 (inventing it) Jun 20 01:45:46 http://www.youtube.com/watch?v=IX-gTobCJHs#at=18 Jun 20 01:45:49 interesting demo video Jun 20 01:46:10 Amorphous metal casings are actually plausible Jun 20 01:46:20 'metallic glass' Jun 20 01:46:59 A new process has been worked out where you melt the metallic glass with a massive pulse of electricity, then form it in a few milliseconds Jun 20 01:47:08 So it's not hot long enough to recrystalise. Jun 20 01:48:30 graphene bla blah, roll up display, roll up battery, bla blah blah.... Jun 20 01:48:34 pfff Jun 20 01:49:57 :D Jun 20 01:49:58 :D Jun 20 01:50:22 if nokia uses this, its just.... "normal" Jun 20 01:50:30 SpeedEvil: I see how you heat the stuff up in microseconds, given you got Thor himself on your side, but how the hell are you cooling it similarly fast? Jun 20 01:50:31 if apple use this, it's a NEW REVOLUTION.... Jun 20 01:50:33 Well - Nokia is working on the roll-up (and blow away) company Jun 20 01:50:33 =_= Jun 20 01:50:43 DocScrutinizer: Smack it with two bits of metal Jun 20 01:50:56 DocScrutinizer: Which are chilled Jun 20 01:51:02 aaah, ok Jun 20 01:51:28 Only works for thin sections of course - but that's ideal for things like phone cases Jun 20 01:51:49 * user0 is back! Jun 20 01:52:04 * HtheB is going away! Jun 20 01:52:10 * SpeedEvil farted. Jun 20 01:52:22 * user0 looks around Jun 20 01:52:39 HtheB: you can't build roll up LiIon today, not with graphene and not with 100 times the power / volume Jun 20 01:52:44 * HtheB dies Jun 20 01:53:05 im not a scientist so, I donno :P Jun 20 01:53:23 HtheB: you also have a hard time building roll up LCD displays Jun 20 01:53:36 they can already do that DocScrutinizer ;) Jun 20 01:53:36 Roll up LCD is easy. Jun 20 01:53:40 and even for OLED it's hard Jun 20 01:53:41 (though a bit crunchy) Jun 20 01:53:51 samsung made it already Jun 20 01:53:54 DocScrutinizer : flashing doesnt seeem to remove the lock code at boot Jun 20 01:54:11 i tested it for you last time Jun 20 01:54:12 HtheB: citation needed - I dare to not believe that Jun 20 01:54:21 DocScrutinizer: ok hang on :) Jun 20 01:54:30 user0: it can :) just be sure you flash the emmc first Jun 20 01:54:48 user0: yeah, we had a bit of problems to decrypt your 4 line report Jun 20 01:55:26 HtheB: what's that now with flashing eMC first? Jun 20 01:55:37 DocScrutinizer : wasnt sure if you read what i typed last time Jun 20 01:55:49 the room was like some bot idling lobby last time Jun 20 01:55:54 you're the first one to claim it depends on eMMC flashing done first Jun 20 01:56:04 DocScrutinizer: so what wasnt clear about my "4 line report" ? Jun 20 01:56:14 DocScrutinizer: well... the lock code should go after normal flashing Jun 20 01:56:21 alost everything :-D Jun 20 01:56:29 :D Jun 20 01:56:33 HtheB: listen and learn Jun 20 01:56:37 ._. Jun 20 01:56:47 it's always better if u flash the emmc too :P Jun 20 01:56:48 lol Jun 20 01:56:49 anyway Jun 20 01:56:50 http://www.physorg.com/news/2011-05-foldable-crease.html Jun 20 01:56:59 user0: you said you got a device - same one or new one? Jun 20 01:57:11 same one Jun 20 01:57:11 http://ic.tweakimg.net/ext/i/imagenormal/1305278052.jpeg Jun 20 01:57:15 k Jun 20 01:57:44 it wasnt asking for a code at boot after they sent it back Jun 20 01:57:50 HtheB: pff, amoled Jun 20 01:58:06 >.> Jun 20 01:58:11 ok, so you john'd the lock code? Jun 20 01:58:13 much better then lcd if u ask me Jun 20 01:58:19 yes Jun 20 01:58:28 then i changed it, and flashed the device Jun 20 01:58:29 then you reflashed Jun 20 01:58:33 what exactly? Jun 20 01:58:52 rootfs Jun 20 01:59:01 :nod: Jun 20 01:59:20 so...it still asked for a code Jun 20 01:59:20 what happened? Jun 20 01:59:26 only this time i knew it Jun 20 01:59:33 :nod: Jun 20 02:00:16 and you had to disable it then, or did it magically stop to ask for lock code after first boot? Jun 20 02:00:18 maybe its just my device, i dont know Jun 20 02:00:47 it was disabled Jun 20 02:01:01 it just asked for the lock code after flashing the device Jun 20 02:01:07 though you enabled it prior to flashing, I suppose Jun 20 02:01:08 didnt ask for any lock code after that Jun 20 02:01:32 nope Jun 20 02:01:52 oh, you changed the lock code, but didn't actually enable it? Jun 20 02:02:07 "ask for lock code on boot" or whatever Jun 20 02:02:08 yaaay, Fallout on N900 ! Jun 20 02:02:35 correct Jun 20 02:02:43 i just changed it and left it disabled Jun 20 02:02:48 user0: what's been the image you flashed? Jun 20 02:03:36 RX-51_2009SE_20.2010.36-2_PR_COMBINED_MR0_ARM.bin Jun 20 02:03:40 thanks Jun 20 02:03:48 WEIRD story, WEIRD Jun 20 02:04:27 I mean how many users changed they lockcode, then disabled it and reflashed rootfs. No such report ever before Jun 20 02:05:10 well its working perfectly atm Jun 20 02:05:18 you repeated the procedure to test if it is reproducable for you (your device)? Jun 20 02:06:02 also, would you mind to send me a dd copy of your mtd1? Jun 20 02:06:38 yes Jun 20 02:06:47 a few you times actually Jun 20 02:07:14 sorry? Jun 20 02:07:55 yes i flashed it more than once to see if it still asked for a lock code Jun 20 02:08:08 and let me check the size of that file you're asking about Jun 20 02:08:13 what kind of a lock code? Jun 20 02:08:14 and did it same weird thing second time you flash? Jun 20 02:08:17 the connection here sucks Jun 20 02:08:26 yes Jun 20 02:08:33 WEIIIIRD Jun 20 02:09:50 wasnt surprised when weird stuff started happening Jun 20 02:09:55 it got the device for a low price Jun 20 02:10:05 393216 Bytes (393 kB) Jun 20 02:10:19 says dd Jun 20 02:10:45 dd if=dev/mtd1 od=whatever Jun 20 02:10:50 of= even Jun 20 02:13:14 though: Jun 20 02:13:20 IroN900:~# cat /proc/mtd |grep conf Jun 20 02:13:21 mtd1: 00060000 00020000 "config" Jun 20 02:13:45 size: 00060000 whatever that means Jun 20 02:14:13 ~6 * 65535 Jun 20 02:14:13 393210 Jun 20 02:14:19 mhm Jun 20 02:14:54 ~6 * 65536 Jun 20 02:14:54 393216 Jun 20 02:16:11 so NOLO is 'only' 100k size Jun 20 02:16:25 ;-P Jun 20 02:16:50 DocScrutinizer : what info does this file contain ? Jun 20 02:17:46 locale, serial#, IMEI(?), some bme cruft, hw revision, WLAN bands allowed (US/EU/WORLD...) Jun 20 02:18:23 I'm not angry if you don't feel like sending it to me, thoughI think the risk is low Jun 20 02:19:11 however I'd suggest not to pusblish it to the world Jun 20 02:19:20 ;-) Jun 20 02:19:48 :D Jun 20 02:20:11 problem is the file's not plain text :P Jun 20 02:20:20 yep, indeed Jun 20 02:20:45 you can attach to a mail nevertheless, just like you'd do with a photo Jun 20 02:20:47 more weird stuff btw :P Jun 20 02:21:38 you just need to create a true file out of the /dev/mtd1, by using dd Jun 20 02:21:48 the date and time went back to 6am 2009 after keeping it off for a few hours Jun 20 02:21:58 normal Jun 20 02:22:22 means your backup button cell is as wrecked as anybody else's Jun 20 02:22:53 actually it'd be surprising if it were *not* Jun 20 02:23:08 http://www.youtube.com/watch?v=noe3kR8KqJc Jun 20 02:23:09 lol Jun 20 02:23:21 you know when you remove the battery theres this sticker with some info about the phone Jun 20 02:23:31 yes Jun 20 02:23:50 it had "made in finland" Jun 20 02:24:02 hmm, one of the early ones Jun 20 02:24:13 might explain something Jun 20 02:24:33 maybe you got a pre-summit device Jun 20 02:24:40 it seemed as if that sticker has another sticker under it Jun 20 02:24:41 they allegedly are "different" Jun 20 02:24:51 cat /proc/cpuinfo now :) Jun 20 02:25:03 so i removed the first one slowly and indeed there was another sticker under it Jun 20 02:25:13 with "made in korea" Jun 20 02:25:17 LOL Jun 20 02:25:27 now WTF? Jun 20 02:25:30 lol Jun 20 02:25:39 someone trolled you, nokia service centers are supposed to take that off Jun 20 02:25:51 fail Jun 20 02:26:00 eh, they take it off if they need to replace it Jun 20 02:26:05 IroN900:~# cat /proc/cpuinfo |grep evis Jun 20 02:26:07 CPU revision : 3 Jun 20 02:26:08 Revision : 2101 Jun 20 02:26:16 referbished device ? Jun 20 02:26:27 apprently that's the water ingress label Jun 20 02:26:40 anyway, im going to sleep :) nite all Jun 20 02:26:51 its almost 4.30 over here :( Jun 20 02:27:00 user0: possibly Jun 20 02:27:04 good night Jun 20 02:27:52 user0: might have seen a mainboard swap Jun 20 02:29:01 actually it'd be interesting to ask other users that got a PCB swap for fixing USB Jun 20 02:34:38 DocScrutinizer : the correct IMEI is on the one under btw Jun 20 02:35:02 They can copy the IMEI Jun 20 02:36:27 the first sticker has an imei that doesnt even work on nokia's site Jun 20 02:36:34 the one under does and is the correct imei of the device Jun 20 02:36:50 weird shit indeed :P Jun 20 02:37:48 * SpeedEvil wishes he could alter the IMEI Jun 20 02:38:29 weird weird weird Jun 20 02:39:10 maybe it's a professionally stolen device? Jun 20 02:39:36 thats what i was thinking Jun 20 02:39:55 and they nessed up / ruined something on unlocking it Jun 20 02:40:32 how did you query the real IMEI? Jun 20 02:40:38 settings Jun 20 02:40:38 ~#06#? Jun 20 02:40:53 try *#06# in dialer please Jun 20 02:41:18 same thing Jun 20 02:44:39 cat /dev/mtd1 |strings ;#< might give some clues what happened with that device... Jun 20 02:53:59 command's giving an error Jun 20 02:54:13 or im just too sleepy to even type it correctly Jun 20 02:54:31 anyways i'll be back tomorrow Jun 20 02:54:32 g'night **** ENDING LOGGING AT Mon Jun 20 02:59:57 2011