| piggz | if anyone in the n950 club want to update their google latitude, you can now! (i hope ;) http://www.piggz.co.uk/pgzlatitude2_0_0_1_armel.deb | 00:00 |
|---|---|---|
| *** fiferboy has quit IRC | 00:00 | |
| Noobmonk3y | oooooooooo :) nice piggz | 00:01 |
| Noobmonk3y | is it me, or does the n950 have a pretty damn good battery life? | 00:01 |
| piggz | its better than the n900, but no symbian killer :) | 00:01 |
| Noobmonk3y | have not put a sim in it for 2 days, but it has been on with 4 apps running, still 70+% battery | 00:01 |
| Noobmonk3y | still has been connected to wireless and used a bit | 00:02 |
| javispedro | is it you, when you use it heavily it is slightly higher than n900 only | 00:02 |
| Noobmonk3y | lol true :P | 00:02 |
| javispedro | considering my n900 battery is two years old that's no compliment. | 00:02 |
| Noobmonk3y | my sgs2 is better then the n900 (not hard) but still down to 20% by bedtime normalyl | 00:02 |
| piggz | im sure if i took the sim out of my 950, it would have great battery life...its 3g internet that is the killer while @ work | 00:03 |
| Noobmonk3y | yeah gd point | 00:03 |
| *** willer_ has quit IRC | 00:04 | |
| piggz | right, im off for the night...feedback welcome if anyone test the latitude app | 00:08 |
| javispedro | for example I've had to try and remove the 10Hz tick in paused SDL applications... otherwise they suck the battery badly, for some reason even more than on n900.. | 00:10 |
| *** piggz has quit IRC | 00:10 | |
| mgedmin | crevetor, same url? | 00:22 |
| crevetor | mgedmin: yes | 00:22 |
| *** djszapi_ has left #harmattan | 00:23 | |
| mgedmin | so, you don't believe in version numbers? :) | 00:23 |
| crevetor | naaah ;) | 00:23 |
| mgedmin | hm, still seems wrong | 00:24 |
| mgedmin | seems 90 degrees flipped | 00:25 |
| crevetor | in portrait or landscape ? | 00:25 |
| mgedmin | so, in portrait, horizontally, screen top pointing north: arrow shows north correctly | 00:25 |
| mgedmin | turn 90 degrees so screen top points east: arrow now points to the south | 00:26 |
| crevetor | it does weird things when I tilt the device... | 00:26 |
| mgedmin | turn 90 degrees more, screen top points south: arrow shows north correctly | 00:26 |
| *** gri has quit IRC | 00:26 | |
| SpeedEvil | Doing it right is hard | 00:26 |
| mgedmin | turn 90 degrees more, screen top points west: arrow shows south again | 00:26 |
| SpeedEvil | you need to work out where the earths surface is, and project a notional compass onto it | 00:26 |
| SpeedEvil | And then have the display as a viewport onto this notional compass | 00:27 |
| SpeedEvil | (sort-of) | 00:27 |
| mgedmin | just write a set of unit tests :) | 00:27 |
| crevetor | hum... | 00:27 |
| mgedmin | and then play with minus signs until they pass | 00:27 |
| mgedmin | screen top points north-east: arrow now points east | 00:27 |
| crevetor | anyways got to go... | 00:27 |
| mgedmin | crevetor, can has source code please? | 00:28 |
| crevetor | hold on a sec | 00:28 |
| vandenoever | what is a good SVG editor for use with QML? inkscape save svg 1.2 which is too new | 00:28 |
| javispedro | too new for what? | 00:29 |
| vandenoever | for use with qml | 00:29 |
| javispedro | improbable | 00:29 |
| SpeedEvil | You start out with the geomagnetic model at your location, to get the dip angle of the field (ideally), then use that and then use the accel to get 'down', and work out the angle between down and the magnetic field, then go from there | 00:29 |
| javispedro | qt accepted svg 1.2 way back nearly 3 or 4 years ago | 00:29 |
| vandenoever | the files i make with inkscape and save as plain svg look bad | 00:29 |
| javispedro | bad in what way? | 00:29 |
| vandenoever | svg 1.2 is not a standard yet | 00:29 |
| vandenoever | transparent box with black text looks like black box on device | 00:30 |
| javispedro | vandenoever: tiny | 00:30 |
| vandenoever | right, svg 1.2 tiny would be fine | 00:30 |
| vandenoever | but that's not what inkscape purports to save | 00:31 |
| *** Noobmonk3y has quit IRC | 00:32 | |
| javispedro | which version of inkscape are you using? | 00:32 |
| vandenoever | 0.48 | 00:32 |
| crevetor | mgedmin: https://gitorious.org/qmlcompass/qmlcompass/trees/master | 00:33 |
| mgedmin | crevetor, thanks! | 00:33 |
| crevetor | Ok I'm off | 00:34 |
| mgedmin | *grumble* *grumble* github has better UI *grumble* | 00:34 |
| crevetor | see ya | 00:34 |
| crevetor | oh well.. | 00:34 |
| javispedro | vandenoever: I'm using r9654 (a bit later than 0.48 iirc) and it saves 1.1 save for soome stuff about flowed text | 00:34 |
| *** javispedro has quit IRC | 00:34 | |
| mgedmin | what's the git clone URL for that repo? | 00:34 |
| mgedmin | ah, I can click on "QMLCompass" and *then* gitorious will tel me | 00:34 |
| mgedmin | so, time to install Qt Creator, I guess | 00:35 |
| *** javispedro has joined #harmattan | 00:36 | |
| vandenoever | javispedro: i want normal text, not flowed text and when i create text box it's saved as flowed text for no apparent reason | 00:37 |
| mgedmin | oh dear, I remember seeing a web page describing the installation process somewhere... | 00:45 |
| mgedmin | there are so many web pages and so little space in my memory/bookmarks | 00:46 |
| mgedmin | I gather I have to do a custom install, since harmattan support is "experimental" and not enabled by default | 00:47 |
| javispedro | bah, I still prefer IRC to twitter. | 00:47 |
| mgedmin | will I need Qt Designer? I recall hearing on IRC that it "doesn't work" | 00:47 |
| javispedro | vandenoever: there was a FAQ item about that | 00:48 |
| javispedro | vandenoever: google a bit, it was something about right click then convert to sth | 00:48 |
| javispedro | the nth-char limit on twitter makes my arguments sound dumb. | 00:49 |
| *** aleksander_m has quit IRC | 00:51 | |
| DocScrutinizer | dang, this QMLcompass is nuts! | 00:52 |
| javispedro | QMLcompass? | 00:52 |
| javispedro | it probably just follows that qorientation sensor | 00:53 |
| DocScrutinizer | http://bit.ly/r75aw6 | 00:54 |
| DocScrutinizer | LOL | 00:54 |
| javispedro | btw since my aegis "lock-out" I've not been able to add my nokia account information again ("Service not available") | 00:54 |
| DocScrutinizer | eeew | 00:55 |
| javispedro | DocScrutinizer: you have? | 00:55 |
| DocScrutinizer | err, it started to work after reflash here | 00:55 |
| DocScrutinizer | initially I had "service not..." | 00:56 |
| * javispedro sighs | 00:56 | |
| javispedro | so it might actually really be a server failure? | 00:56 |
| mgedmin | which builds do I need? Harmattan Platform API or Meego 1.2 Harmattan API? | 00:58 |
| DocScrutinizer | lemme check if it works right now | 00:58 |
| rm_work | mgedmin: my stuff fails equally well on both :P | 00:58 |
| DocScrutinizer | maps err !!Maps just starts | 00:59 |
| mgedmin | whoa, FakeVim? sounds interesting | 00:59 |
| javispedro | DocScrutinizer: !!maps does start here indeed, and it shows background disabled map, then popups Nokia login prompt | 01:00 |
| javispedro | and clicking on it gets me "Service not available" in ~1 sec, too short for at imeout | 01:00 |
| DocScrutinizer | not here, just woks | 01:00 |
| *** NIN101 has quit IRC | 01:00 | |
| javispedro | that's how it used to work pre-lockout | 01:00 |
| javispedro | wonder if I will have to reflash again.. | 01:01 |
| rm_work | is there a package list anywhere? | 01:01 |
| rm_work | how do i actually find anything to install? | 01:01 |
| DocScrutinizer | well it recovered from fist lockout with 3 reboots... ;-P | 01:01 |
| rm_work | don't see a package manager | 01:02 |
| javispedro | 3 reflashes, haha | 01:02 |
| javispedro | rm_work: there's no extras-devel, no nokia store... | 01:02 |
| rm_work | also BTW, maps loads for me | 01:02 |
| javispedro | rm_work: http://wiki.meego.com/Community_Office/Community_device_program/Nokia/Development_repos | 01:02 |
| DocScrutinizer | I wonder where's that friggin accounts in settings | 01:03 |
| javispedro | DocScrutinizer: it's not in settings, but rather separate app | 01:03 |
| DocScrutinizer | lemme try search | 01:03 |
| mgedmin | this qt creator thing is fancy | 01:03 |
| mgedmin | and the emulation is in-effing-credibly slow | 01:04 |
| DocScrutinizer | ouch, that took a while. I felt aegis workin in background, til screen came up | 01:04 |
| mgedmin | iirc there's a way to have qt creator deploy to my n950? | 01:05 |
| rm_work | mgedmin: yeah and it works well | 01:06 |
| rm_work | mgedmin: there's docs somewhere i found for setting it up, it's pretty easy | 01:06 |
| mgedmin | tools -> options -> maemo -> add! | 01:06 |
| rm_work | you run the SDK app on the device, and it will give you the password it needs to deplot the ssh keys | 01:06 |
| DocScrutinizer | asks me for password when trying to sync | 01:06 |
| rm_work | *deploy | 01:07 |
| mgedmin | oh, I need to run something on the device | 01:07 |
| *** apol_ has joined #harmattan | 01:07 | |
| DocScrutinizer | but yeah, seems to work | 01:08 |
| * javispedro sighs | 01:08 | |
| javispedro | reflash time | 01:08 |
| mgedmin | hm | 01:09 |
| DocScrutinizer | :nod: | 01:09 |
| mgedmin | "Could not connect to host: Botan library exception: Botan: Decoding error: BER: Length field is too large" | 01:09 |
| mgedmin | using a pubkey that I've just deployed | 01:09 |
| DocScrutinizer | butanlib | 01:09 |
| mgedmin | it works when I use password auth | 01:09 |
| * mgedmin shrugs and checks /home/developer/authorized_keys | 01:09 | |
| * javispedro ponders a best way to solve qole's aegis problems | 01:10 | |
| javispedro | he would need to keep a list of digsigs for the chroot image's binaries | 01:11 |
| DocScrutinizer | javispedro: prior to reflash try the several reset options | 01:11 |
| javispedro | and when mounting it programatically load them into kernel | 01:11 |
| antman8969 | has anyone seen how to make a number only vkb without having to create a custom one? | 01:11 |
| javispedro | because from reading the code the validor keeps the hashlist per mounted device/superblock | 01:11 |
| DocScrutinizer | qole's problem? | 01:11 |
| *** seif_ has joined #harmattan | 01:11 | |
| DocScrutinizer | OMFG!°!! | 01:12 |
| javispedro | DocScrutinizer: author of EasyDebian -- a .deb with a large .ext2 image that is mounted, chrooted into, and contains a full Debian installation | 01:12 |
| DocScrutinizer | no way to attach external storage? | 01:12 |
| javispedro | of course you can attach external storage | 01:12 |
| javispedro | it's just that aegis will refuse to run anything from it | 01:12 |
| DocScrutinizer | hah | 01:13 |
| javispedro | (exception: "developer mode" and current_user=="user") | 01:13 |
| DocScrutinizer | thought as much | 01:13 |
| javispedro | otherwise, he might want to drop the idea of using images alltogether, now that / is ext4. | 01:13 |
| *** rcg has quit IRC | 01:13 | |
| *** rcg has joined #harmattan | 01:13 | |
| *** seif has quit IRC | 01:13 | |
| *** apol has quit IRC | 01:13 | |
| *** deimos has quit IRC | 01:13 | |
| *** SpeedEvil has quit IRC | 01:13 | |
| *** deimos has joined #harmattan | 01:13 | |
| mgedmin | woot, it runs! | 01:13 |
| mgedmin | I'm awesome | 01:14 |
| * DocScrutinizer got the feeling to be out of loop | 01:15 | |
| DocScrutinizer | javispedro: seen N950 "KIT light"? | 01:15 |
| javispedro | DocScrutinizer: got that feeling too, I am very busy this week with no time to follow all news... and next's week I might be away. | 01:16 |
| javispedro | s/'s// ;P | 01:16 |
| DocScrutinizer | javispedro: did a lil facy and messed with LEDs | 01:16 |
| javispedro | ah | 01:17 |
| javispedro | video? :) | 01:17 |
| DocScrutinizer | nah | 01:17 |
| DocScrutinizer | [2011-07-20 21:17:59] <DocScrutinizer> echo "open kbd, lock device | 01:17 |
| *** TSCHAKeee has joined #harmattan | 01:18 | |
| DocScrutinizer | was boted :-D | 01:18 |
| javispedro | heh | 01:18 |
| javispedro | btw | 01:18 |
| DocScrutinizer | bored | 01:18 |
| TSCHAKeee | arg, i am having issues deploying from Qt Creator. It installs the qml files alongside my application in /opt | 01:18 |
| javispedro | http://forum.meego.com/showpost.php?p=26750&postcount=5 (my post) | 01:18 |
| TSCHAKeee | but it's looking for the qml files in /home/developer | 01:18 |
| TSCHAKeee | does nayone know how to fix this? | 01:18 |
| javispedro | DocScrutinizer: early this week I tried to get fmrx working | 01:18 |
| javispedro | see that post for a few details | 01:18 |
| javispedro | no luck so far | 01:19 |
| javispedro | but I can confirm there's the tuner, it can tune, and it can report signal strength. But I don't know how to get the audio. | 01:19 |
| javispedro | (and signal strength greatly increases when HPs are connected) | 01:19 |
| * TSCHAKeee has no forehead ridges left from banging head against wall. | 01:19 | |
| TSCHAKeee | javispedro: maybe there's a hidden i2c port somewhere? ;) | 01:19 |
| DocScrutinizer | javispedro: nice :-D | 01:20 |
| javispedro | TSCHAKeee: there's probably a i2_s_ port somewhere ;) | 01:20 |
| rm_work | what is it with nokia putting radios in things and then hiding the fact that it exists? :P | 01:20 |
| DocScrutinizer | HAH | 01:20 |
| DocScrutinizer | now you hooked me | 01:21 |
| TSCHAKeee | so nobody has had issues with qml files loading from a QtCreator deployment? | 01:21 |
| rm_work | TSCHAKeee: i sat down and installed QtCreator and did the setup in like 5 minutes, and it worked fine for the hello world app | 01:21 |
| rm_work | and worked fine for my projects afterwards... | 01:21 |
| DocScrutinizer | rm_work: I'm ahppy about it, otherwise here you had to pay montly fee | 01:21 |
| mgedmin | yay, got qmlcompass to point north! | 01:21 |
| rm_work | i assume it's an issue with your config for QtCreator | 01:21 |
| rm_work | DocScrutinizer: what? montly fee to ... recieve radio signals? | 01:22 |
| mgedmin | in portrait mode | 01:22 |
| DocScrutinizer | rm_work: sure! | 01:22 |
| rm_work | DocScrutinizer: to WHO? cellphone company? | 01:22 |
| DocScrutinizer | err, the radio station | 01:23 |
| mgedmin | whee vim mode is just awesome | 01:23 |
| DocScrutinizer | actually it's a bit more complex | 01:23 |
| rm_work | umm | 01:23 |
| rm_work | how do they know you are recieving signals? | 01:23 |
| rm_work | >_> | 01:23 |
| DocScrutinizer | they got *spies* | 01:23 |
| rm_work | when you buy a radio from walmart do they require you to sign up and pay? >_> | 01:23 |
| DocScrutinizer | nope | 01:24 |
| mgedmin | ok, OrientationSensor is not the right thing to use | 01:24 |
| mgedmin | how do I discover if my app is in landscape or in portrait mode in QML? | 01:24 |
| rm_work | DocScrutinizer: ok, i officially completely do not understand what you're talking about then | 01:24 |
| *** Venemo has quit IRC | 01:24 | |
| DocScrutinizer | http://gez.de/ | 01:25 |
| DocScrutinizer | simple as that | 01:25 |
| DocScrutinizer | 99.6% are paying, the rest gets an occasional visit | 01:25 |
| rm_work | umm | 01:28 |
| rm_work | no, still totally not making sense | 01:28 |
| rm_work | it seems like they're trying to do something that is logistically impossible | 01:28 |
| rm_work | there is no real way to enforce this at all | 01:28 |
| rm_work | or even really get metrics on enforcement | 01:28 |
| DocScrutinizer | http://www.focus.de/kultur/videos/gez-spots-gebuehren-werbung-ade_vid_17975.html | 01:30 |
| rm_work | well heading home... | 01:30 |
| rm_work | i'll watch that when i get home | 01:30 |
| *** rm_work has quit IRC | 01:30 | |
| javispedro | DocScrutinizer: I extended the post, and added v4l packages: http://forum.meego.com/showpost.php?p=26750&postcount=5 | 01:33 |
| javispedro | it has instructions and brief dmesg dumps | 01:33 |
| DocScrutinizer | rm_you: they really get spies that run around and ring doors and watch windows to spot TV users that aren't on their list of paying customers | 01:33 |
| DocScrutinizer | rm_you: 2013 the state enforces the fee on each household | 01:34 |
| DocScrutinizer | so no more spy ringing your door | 01:34 |
| DocScrutinizer | rm_you: you should know first and 2nd channel, as the regional ones, are public hand | 01:35 |
| *** NIN101 has joined #harmattan | 01:35 | |
| DocScrutinizer | it was only like 1990 when we go tcommercial radio broadcast and TV at all here in Germany | 01:36 |
| *** andre__ has quit IRC | 01:36 | |
| DocScrutinizer | right now you have to pay for your PC connected to internet, as you *could* surf to URLs like www.tagesschau.de | 01:37 |
| DocScrutinizer | I'm wondering why they don't offer radio via landline phone call-in and then charge for each telephone | 01:39 |
| mgedmin | woot, fixed qmlcompass! | 01:39 |
| DocScrutinizer | obviously they did as well, so from 2013 on they charge *every* household, no matter if you got a radio / TV or not | 01:40 |
| DocScrutinizer | mgedmin: aaah, so no more weird rotations? | 01:40 |
| mgedmin | yep | 01:40 |
| DocScrutinizer | javispedro: cool | 01:40 |
| *** crevetor has left #harmattan | 01:40 | |
| javispedro | DocScrutinizer: feel free to reproduce, I'm sure it must be a stupid overlook on my part | 01:41 |
| DocScrutinizer | mgedmin: same ly.tyty URL? | 01:41 |
| mgedmin | no | 01:41 |
| mgedmin | :) | 01:41 |
| DocScrutinizer | :-S | 01:41 |
| *** vgrade has quit IRC | 01:41 | |
| mgedmin | crevetor's last version (same bit.ly url) works fine in portrait mode | 01:41 |
| DocScrutinizer | javispedro: a) which chip is the FMRX on? | 01:41 |
| javispedro | DocScrutinizer: wl1273 | 01:41 |
| mgedmin | correction: works fine while the orientation sensor returns TopUp | 01:42 |
| DocScrutinizer | b) do we have a I2S for that chip? | 01:42 |
| DocScrutinizer | umm then no | 01:42 |
| javispedro | if anyone has 1273 for that chip I'd appreciate. | 01:42 |
| javispedro | err | 01:42 |
| DocScrutinizer | ds | 01:42 |
| javispedro | s/1273/datasheet/ | 01:42 |
| DocScrutinizer | lemme check what I found "back when" | 01:42 |
| javispedro | technically, the chip has | 01:42 |
| javispedro | an "analog" path for fmrx | 01:43 |
| javispedro | and a "digital" path, which seems to be connected to a mcbsp something, which seems to be somekind of DMA (my knowledge of McBSP is nil at best) | 01:43 |
| *** vgrade has joined #harmattan | 01:43 | |
| DocScrutinizer | javispedro: | 01:44 |
| javispedro | the analog path -- no idea where it is connected. if it is at all. | 01:44 |
| DocScrutinizer | javispedro: | 01:44 |
| DocScrutinizer | Jul 9 17:12:43 (none) kernel: [ 0.214965] ALSA device list: | 01:44 |
| DocScrutinizer | Jul 9 17:12:43 (none) kernel: [ 0.214996] #0: dfl61-twl4030 (twl4030) | 01:44 |
| DocScrutinizer | Jul 9 17:12:43 (none) kernel: [ 0.214996] #1: dfl61-dac33 (tlv320dac33) | 01:44 |
| javispedro | DocScrutinizer: missing #2 ;) | 01:44 |
| DocScrutinizer | no #2 here | 01:44 |
| javispedro | check /proc/asound/cards | 01:44 |
| javispedro | aiui it is loaded as a module a bit later on the boot process | 01:45 |
| DocScrutinizer | yeah | 01:45 |
| *** meegoexperts_mob has joined #harmattan | 01:46 | |
| DocScrutinizer | Jul 9 17:12:43 (none) kernel: [ 0.272552] asoc: WL1273 BT/FM codec <-> omap-mcbsp-dai-2 mapping ok | 01:46 |
| javispedro | exactly | 01:47 |
| javispedro | this is the same way the std audio codec for speakers is connected, albeit on a different channel | 01:47 |
| DocScrutinizer | Jul 9 17:12:43 (none) kernel: [ 5.580261] wl1273_fm_module_init | 01:47 |
| DocScrutinizer | ls -l /dev/snd/ | 01:49 |
| DocScrutinizer | :-D | 01:49 |
| *** andre__ has joined #harmattan | 01:50 | |
| *** andre__ has quit IRC | 01:50 | |
| *** andre__ has joined #harmattan | 01:50 | |
| *** andre__ has quit IRC | 01:50 | |
| javispedro | well, no idea, and no documentation. | 01:52 |
| *** xerxes2 has joined #harmattan | 01:53 | |
| mgedmin | where's the deb that qt creator creates? | 01:54 |
| DocScrutinizer | javispedro: amixer -c2 | 01:54 |
| *** deimos has quit IRC | 01:54 | |
| *** rcg has quit IRC | 01:54 | |
| *** khertan has quit IRC | 01:54 | |
| *** dm8tbr has quit IRC | 01:54 | |
| mgedmin | oh, it's outside my build dir | 01:54 |
| DocScrutinizer | now say "thanks!" :-P | 01:54 |
| javispedro | DocScrutinizer: note my updated post, I edit several channels already ;P | 01:55 |
| DocScrutinizer | which channels? | 01:55 |
| javispedro | but those are not documented either, albeit I mostly know what the Mode, Audio, and Codec Mode mixers do by now. | 01:55 |
| DocScrutinizer | oh | 01:55 |
| DocScrutinizer | indeed quite some update on #5 | 01:56 |
| mgedmin | DocScrutinizer, you can try http://mg.pov.lt/qmlcompass_0_0_1-0mg1_armel.deb | 01:56 |
| mgedmin | now how do I send the patch to crevetor? | 01:56 |
| *** dm8tbr has joined #harmattan | 01:59 | |
| *** SpeedEvil has joined #harmattan | 01:59 | |
| * mgedmin creates a merge request on gitorious | 02:00 | |
| * DocScrutinizer *curses* missing c&p on e.g xterm | 02:01 | |
| *** khertan has joined #harmattan | 02:01 | |
| DocScrutinizer | HAH, I only need mg.pov.it I guess | 02:02 |
| *** rcg has joined #harmattan | 02:02 | |
| *** deimos has joined #harmattan | 02:03 | |
| DocScrutinizer | :sigh: | 02:04 |
| DocScrutinizer | obviously not | 02:05 |
| mgedmin | want a bit.ly link? | 02:05 |
| mgedmin | hey, why doesn't mg.pov.lt front page list all the irc logs I'm archiving? | 02:06 |
| mgedmin | 'cause I'm a lazy bastard, that's why | 02:06 |
| mgedmin | sleeptime | 02:08 |
| javispedro | some food for thought: | 02:11 |
| DocScrutinizer | mgedmin: please reduce decimals of accuracy to max 5 ;-P | 02:11 |
| javispedro | at the moment, thanks to aegis, webos guys can run a debian chroot more easily than we can on n950. | 02:12 |
| javispedro | (otoh, we can run custom kernels. they are stuck with stock) | 02:12 |
| *** meegoexperts_mob has quit IRC | 02:12 | |
| javispedro | or "we will be able to" run custom kernels ;) | 02:13 |
| DocScrutinizer | btw on 33.876676857567% accuracy this thing is outright useless | 02:14 |
| javispedro | how does it measure accuracy? | 02:14 |
| DocScrutinizer | nfc | 02:14 |
| * javispedro installs this qmlcompass knowing that my device is FULLY safe and SANBOXED thanks to Aegis.... | 02:15 | |
| DocScrutinizer | but at 66.6666666665348816% it starts to work | 02:15 |
| javispedro | NOT! | 02:15 |
| javispedro | mgedmin: was your patch supposed to lock orientation? | 02:16 |
| *** rcg has quit IRC | 02:16 | |
| DocScrutinizer | nope not here | 02:16 |
| javispedro | I think this is clearly one application that should be locked | 02:17 |
| DocScrutinizer | hmm | 02:19 |
| *** javispedro has quit IRC | 02:22 | |
| mgedmin | javispedro, no, I don't know how to do that | 02:22 |
| mgedmin | my diff is at https://gitorious.org/qmlcompass/qmlcompass/merge_requests/1 | 02:23 |
| mgedmin | I agree that locking is the right thing here | 02:23 |
| mgedmin | DocScrutinizer, accuracy was always 100% when I played with it :) | 02:24 |
| DocScrutinizer | I got 33.3333333432674408%, and 66.66666666667846746some% so far | 02:26 |
| mgedmin | qml/main.qml: text: "Accuracy : " + accuracy*100 + "%, angle : " + angle + "°" | 02:26 |
| DocScrutinizer | and 0% when I shocked the thing with a speaker magnet | 02:26 |
| mgedmin | how do I round things up in what is this, javascript? | 02:27 |
| DocScrutinizer | I'd suggest 1 bar for >33%, 2 bars for >66%, and 3 bars (full) for >95% | 02:28 |
| Elleo | Math.round() iirc | 02:29 |
| DocScrutinizer | also getting this as a plugin for camera would be really nice :-) | 02:29 |
| DocScrutinizer | eben printed into the picture taken | 02:30 |
| mgedmin | my sister was surprised Maps wasn't making use of the compass for rotation | 02:30 |
| mgedmin | or was it Drive? | 02:30 |
| DocScrutinizer | or in those additional info | 02:30 |
| DocScrutinizer | both | 02:31 |
| mgedmin | ok, Math.round worked | 02:32 |
| TSCHAKeee | http://www.youtube.com/watch?v=jKyryIUK1Ao <-- the new orbiter prototype coming up for the very first time on the n950 | 02:32 |
| mgedmin | building a .deb now ... | 02:33 |
| *** lbt has quit IRC | 02:34 | |
| mgedmin | how do I update the .deb version number from Qt Creator? | 02:34 |
| mgedmin | if I edit debian_harmattan/changelog directly and then press Build, it says "no changes, etc, nothing to be done" | 02:35 |
| mgedmin | oh, I need to Run, not just Build, if I want a .deb | 02:35 |
| mgedmin | DocScrutinizer, http://mg.pov.lt/qmlcompass_0_0_1-0mg2_armel.deb reports accuracy as 33% or 67% | 02:36 |
| mgedmin | you'll have to talk to the original author about bars | 02:36 |
| mgedmin | my sister's android phone has a compass app that doesn't show accuracy | 02:37 |
| mgedmin | but when you put a magnet next to it, pops up a box with text "abnormal magnetic field located, please (do something that I didn't read fast enough)" | 02:37 |
| mgedmin | ok, now I have to convince qt creator to build fbreader | 02:41 |
| mgedmin | *yawn* tomorrow | 02:42 |
| DocScrutinizer | night mgedmin | 02:42 |
| *** deimos has quit IRC | 02:53 | |
| *** MeeGoExperts has joined #harmattan | 02:54 | |
| *** smoku has left #harmattan | 03:04 | |
| *** NIN101 has quit IRC | 03:11 | |
| rm_you | grah i srsly need a development VM | 03:16 |
| rm_you | err wait | 03:16 |
| rm_you | QtCreator works in windoze doesn't it T_T | 03:16 |
| MohammadAG | yes | 03:16 |
| rm_you | T_T | 03:17 |
| rm_you | wow | 03:17 |
| rm_you | i was making this way too difficult | 03:17 |
| * rm_you fails | 03:17 | |
| MohammadAG | Can't imagine how clumsy cmd prompt will be, for cli stuff at least | 03:17 |
| rm_you | well | 03:19 |
| rm_you | it'll be bad but not having to reboot my desktop will make it worth it i hope | 03:19 |
| MohammadAG | contacted ddp today, got a standard answer :/ | 03:19 |
| rm_you | MohammadAG: problems? | 03:20 |
| MohammadAG | yeah, I start getting impatient after two weeks :P | 03:21 |
| *** TSCHAKeee has quit IRC | 03:29 | |
| *** TSCHAKeee has joined #harmattan | 03:29 | |
| npm | how do I solve "Error: 'Configured videosink video-output-bin is not working."' when playing video in http://nielsmayer.com/meego/qml/qmltube_1_0_6_armel.deb ?? | 04:03 |
| npm | i guess it's time to poke around w/ gconftool-2 | 04:04 |
| rm_you | what's the most basic container in QML? | 04:28 |
| rm_you | i guess the thing i don't understand is: we have these QML widget galleries installed that show all kinds of awesome widgets... but I don't get access to any of them in QtCreator? >_> | 04:29 |
| rm_you | i have 12 total elements | 04:29 |
| rm_you | that i can use | 04:30 |
| *** tonberry__ has quit IRC | 04:32 | |
| *** Termana has quit IRC | 04:36 | |
| *** Termana has joined #harmattan | 04:44 | |
| Termana | morning | 04:44 |
| rm_you | totally need someone who knows QML well >_> | 04:45 |
| rm_you | i guess so does everyone | 04:45 |
| rm_you | why do the first two elements i put on a window ALWAYS disappear? | 04:56 |
| rm_you | this is so frustrating | 05:11 |
| hiemanshu | Morning | 05:30 |
| hiemanshu | rm_you: eh? | 05:30 |
| hiemanshu | rm_you: what are you trying to do? | 05:30 |
| rm_you | make a super simple UI | 05:30 |
| rm_you | i want to have two spinners | 05:30 |
| rm_you | that's IT. | 05:30 |
| rm_you | that's the whole UI | 05:31 |
| rm_you | it's complicated slightly by lack of spinners | 05:31 |
| rm_you | and then slightly more by the first two things i add not showing up | 05:31 |
| rm_you | so i've now put two trash elements at the top of my app so they eat the "deadzone" or whatever it is | 05:31 |
| hiemanshu | do you use a layout? | 05:32 |
| rm_you | using a Page | 05:32 |
| rm_you | like... the default app | 05:32 |
| hiemanshu | do you use a layout? | 05:32 |
| hiemanshu | in no, read about anchor based layouts, works fine for most implementations | 05:33 |
| hiemanshu | s/in/if/ | 05:33 |
| infobot | hiemanshu meant: if no, read about anchor based layouts, works fine for most implementations | 05:33 |
| rm_you | i'm anchoring things inside the page | 05:33 |
| rm_you | do i NEED a layout container? | 05:34 |
| rm_you | i'm trying to figure out what the default "pane" or whatever is | 05:34 |
| hiemanshu | rm_you: can I see your code? | 05:34 |
| rm_you | every other UI toolkit i've used has some standard container | 05:34 |
| rm_you | sure, sec | 05:34 |
| rm_you | simultaneously watching Day9 SC2 :P | 05:34 |
| hiemanshu | well I am using anchors for my app, everything shows up just fine | 05:35 |
| rm_you | hrm k | 05:35 |
| rm_you | wait | 05:35 |
| rm_you | do elements stack on top of each other in QML? | 05:35 |
| rm_you | i've been assuming they nudge each other | 05:36 |
| hiemanshu | they stack up at times yes if dont have the anchoring done fine | 05:36 |
| hiemanshu | atleast it did for it | 05:36 |
| rm_you | so, what IS the "default container" element? | 05:37 |
| rm_you | like, something that does nothing besides contain other objects | 05:37 |
| hiemanshu | Rectangle? | 05:37 |
| rm_you | ah | 05:37 |
| rm_you | k | 05:37 |
| hiemanshu | wait lemme paste some code | 05:37 |
| hiemanshu | rm_you: http://fpaste.org/OjZq/ | 05:38 |
| hiemanshu | or http://fpaste.org/UwBi/ | 05:39 |
| antman8969 | rm_you, everything is a container | 05:41 |
| rm_you | lol | 05:41 |
| antman8969 | is something has anything nested inside of it in QML then it contrains it | 05:42 |
| rm_you | i noticed but it makes no sense having things inside a button :P | 05:42 |
| antman8969 | lol well depends. If you want a label anchored to the bottom you could have it "inside" the button (or just a sibling) | 05:42 |
| antman8969 | usually you don't need to create rectangles just for being containers is the point | 05:43 |
| antman8969 | and if you do, then an Item is better | 05:43 |
| antman8969 | and Item is a Rectangle, but without any visual appearance | 05:43 |
| rm_you | ah where is that | 05:46 |
| * rm_you looks | 05:46 | |
| rm_you | ah there | 05:46 |
| rm_you | working better... | 05:53 |
| rm_you | REALLY wish the designer worked tho | 05:53 |
| rm_you | for previews >_> | 05:53 |
| *** tonberryN950 has joined #harmattan | 06:02 | |
| tonberryN950 | whats the default root@localhost password for harnattan? | 06:02 |
| rm_you | rootme | 06:04 |
| tonberryN950 | cool.. thx pal | 06:05 |
| *** antman8969 has quit IRC | 06:07 | |
| rm_you | erm, how do i actually add a playlist to the media player | 06:08 |
| rm_you | what format does it use? | 06:08 |
| rm_you | obviously not plss or m3u | 06:08 |
| rm_you | s/plss/pls/ | 06:08 |
| infobot | rm_you meant: obviously not pls or m3u | 06:08 |
| rm_you | how does one increase the font size on a button? | 06:13 |
| hiemanshu | rm_you: set the font? | 06:24 |
| hiemanshu | rm_you: ButtonStyle | 06:25 |
| rm_you | it says font has no members | 06:25 |
| rm_you | ah | 06:25 |
| rm_you | ButtonStyle *instead of* Button? | 06:25 |
| hiemanshu | umm, no the Button will have a ButtonStyle element | 06:26 |
| rm_you | ah | 06:26 |
| rm_you | it doesn't | 06:26 |
| hiemanshu | rm_you: http://fpaste.org/J285/ | 06:26 |
| hiemanshu | thats an example giving on the l.d.n.o site | 06:27 |
| hiemanshu | s/.o/.c/ | 06:27 |
| infobot | hiemanshu meant: thats an example giving on the l.d.n.c site | 06:27 |
| rm_you | ah | 06:27 |
| rm_you | still trying to figure out how to make this work efficiently | 06:29 |
| tonberryN950 | was trying root the device to home screens in landscape, but its not working | 06:31 |
| *** apol_ has quit IRC | 06:55 | |
| rm_you | i want to make one of these: | 06:57 |
| rm_you | http://doc.qt.nokia.com/qt-components-symbian-1.0/qml-tumbler.html | 06:57 |
| rm_you | just need the dayColumn | 06:57 |
| rm_you | but it gives me syntax errors on the for loop | 06:57 |
| rm_you | http://pastie.org/pastes/2246575 | 07:04 |
| rm_you | so dumb | 07:05 |
| GAN900 | dumb | 07:05 |
| rm_you | every time i think i'm getting it, i go try one of their examples and it doesn't work | 07:05 |
| GAN900 | What now? | 07:05 |
| rm_you | see above | 07:06 |
| rm_you | http://doc.qt.nokia.com/qt-components-symbian-1.0/qml-tumbler.html | 07:06 |
| rm_you | my cut down version is: http://pastie.org/pastes/2246575 | 07:06 |
| rm_you | apparently you can't have a for-loop in the QML? but they are doing it... | 07:06 |
| rm_you | i get really frustrated when none of the SAMPLE CODE ever works for me | 07:08 |
| rm_you | i only have like two hours per day to work on coding, and i spend 90% of it banging my head against a wall with this kind of thing | 07:09 |
| rm_you | i want what should be a standard widget | 07:09 |
| rm_you | and in GTK i would have said "give me a spinner, set min to 0, max to 50, current to 20, done" | 07:09 |
| rm_you | i know this is supposed to be more generic (and thus more complicated) but it's just really annoying when I was expecting this to be EASIER | 07:10 |
| GAN900 | rm_you, connected at "dumb" | 07:14 |
| GAN900 | Yeah, I keep running into broken sample code, too. | 07:14 |
| GAN900 | com.nokia.meego had me tripped up for an hour. | 07:15 |
| GAN900 | rm_you, no components? | 07:16 |
| rm_you | very very few to choose from | 07:18 |
| *** djszapi_ has joined #harmattan | 07:18 | |
| djszapi_ | vgrade: still the same stage for kdelibs :( | 07:18 |
| djszapi_ | like last night, it got stuck there.. | 07:18 |
| *** DocScrutinizer has quit IRC | 07:33 | |
| *** DocScrutinizer has joined #harmattan | 07:33 | |
| *** khertan has quit IRC | 07:58 | |
| *** khertan_ has joined #harmattan | 07:59 | |
| *** khertan_ is now known as Khertan | 07:59 | |
| frals | rm_you: you cant write random js code in a qml block like that | 08:42 |
| *** hardaker has quit IRC | 08:43 | |
| frals | rm_you: should probably work if you do items: { forloophere; return daysList } | 08:50 |
| *** TSCHAKeee has quit IRC | 09:03 | |
| *** TSCHAKeee has joined #harmattan | 09:06 | |
| *** djszapi_ has quit IRC | 09:21 | |
| *** tonberryN950 has quit IRC | 09:30 | |
| *** spenap has joined #harmattan | 09:32 | |
| *** tonberryN950_ has joined #harmattan | 09:41 | |
| *** smoku has joined #harmattan | 09:45 | |
| *** aleksander_m has joined #harmattan | 10:00 | |
| *** achipa has quit IRC | 10:02 | |
| *** cpscotti has quit IRC | 10:06 | |
| djszapi | DocScrutinizer: hah, I have just gotten the answer, it is too much work to set up a search engine on the harmattan repo side, and right now it does not work with the available manpower :) | 10:15 |
| *** mzanetti has quit IRC | 10:18 | |
| tomma | are there any svg template for harmattan icons? | 10:28 |
| alterego | If not, wouldn't be to hard to make one by tracing the .png templates in inkscape :) | 10:33 |
| *** smoku has left #harmattan | 10:33 | |
| frals | wasnt there some huge asset with all the icons and stuff like that on the UI/UX guidelines site? | 10:33 |
| alterego | I think so, unless I'm getting confused with MeeGo UX guidlines. | 10:35 |
| * alterego chuckles | 10:35 | |
| Mek | there is a zip with some .ai files for icon templates, yes | 10:37 |
| Mek | (.ai files and .psd files, no free formats though) | 10:37 |
| *** gri has joined #harmattan | 10:38 | |
| alterego | Lame | 10:38 |
| alterego | Well, someone should create some SVGs and put them on the meego wiki. | 10:38 |
| alterego | I'll probably end up doing it myself at some point ;) | 10:38 |
| * gri would welcome this as I am too stupid to use Inkscape | 10:40 | |
| *** rcg has joined #harmattan | 10:56 | |
| *** lbt has joined #harmattan | 10:57 | |
| *** lbt has joined #harmattan | 10:57 | |
| djszapi | vgrade: do you have any idea what can go wrong /all/ the time after the build, before the real package creation process ? I think it is a real problem for Harmattan target in case of huge projects. | 10:57 |
| rcg | morning | 11:05 |
| rcg | anyone experienced with QML, ListViews in particular? | 11:05 |
| djszapi | shoot | 11:06 |
| rcg | am somehow missing how to get the highlighting/selection to work properly | 11:06 |
| *** achipa has joined #harmattan | 11:07 | |
| djszapi | rcg: this is my listview code from last year :) https://projects.kde.org/projects/playground/games/gluon/repository/revisions/master/entry/player/touch/Home.qml | 11:07 |
| rcg | http://pastebin.com/0aJvD2JZ | 11:07 |
| djszapi | (I am not sure how much it changed since that) | 11:07 |
| rcg | ^ using this code on n950.. the highlight box is always stuck at the top and selection does not work | 11:08 |
| rcg | djszapi: i see.. i'll have a loot at it... though, even some minimal example as pasted above does not work(?) | 11:08 |
| rcg | *look | 11:08 |
| frals | rcg: the highlight never moves unless you move it afaik | 11:10 |
| frals | ie unless you navigate the list with keyboard the selected delegate is always the same iirc | 11:10 |
| rcg | hmm ic.. but there should be a way to do selections via touch? in another example i also placed debug output to signals such as onCurrentIndexChanged etc. but does only get fired once when the list loads.. which btw also emits onHighlightChanged | 11:13 |
| frals | highlight moves with focus afaik, and it doesnt move when you scroll, unless you add the code in the delegate to change currentIndex | 11:16 |
| frals | what i usually end up doing is skip the highlight crap and just add a MouseArea in delegate and then have the background color of the delegate something like: color: mouseArea.pressed ? "white" : "black" | 11:17 |
| rcg | hmm ic.. that might be worth a try.. it's just i am pretty confused right now as the example on e.g. http://doc.qt.nokia.com/4.7-snapshot/qml-views.html kinda imply selection should work "out of the box".. and usually i would expect something like a ListView in a high level language like QML to do this selection stuff automatically or have this built in | 11:20 |
| gri | the qt-components have a default ListDelegate (which requires title and subtitle property) | 11:21 |
| frals | rcg: try navigating the listview with arrowkeys and see if it works, i suspect it will ;) | 11:21 |
| frals | but ye, i would expect it to do something similar for touch out of the box :< | 11:22 |
| rcg | http://pastebin.com/6gXy2Dt5 <-- this is the sample code i use to check this atm | 11:22 |
| rcg | hmm well.. even with the arrowkeys nothing happens?!? | 11:23 |
| frals | does the listview have focus? | 11:24 |
| rcg | hmm added "focus: true" to the list view.. still the same | 11:26 |
| rcg | scrolling the list up and down works btw. and in another example i listened to the onYChanged signal which gets emitted when the list is scrolled | 11:28 |
| *** tonberryN950_ has quit IRC | 11:32 | |
| rcg | hmm.. in the simple example even onYChanged is not emitted. but scrolling works. | 11:37 |
| rcg | http://pastebin.com/VwLsPmLd | 11:38 |
| rcg | anyway.. thanks a lot for your help all.. will let you know if i figure this out.. i feel like i am only missing a tiny bit here | 11:39 |
| *** smoku has joined #harmattan | 11:50 | |
| vgrade | djszapi, no ideas I'm afraid. lbt and X-fade on #meego are the admin s | 11:53 |
| *** andre__ has joined #harmattan | 11:58 | |
| vgrade | vgrade, lbt reports tat there is a publishing issue but does not have time to look until tonight | 12:01 |
| DocScrutinizer | djszapi: so you want to keep your account? | 12:02 |
| DocScrutinizer | is there a way to open Web form cmdline? on a given URL? | 12:04 |
| rcg | well.. using a mouse area in the delegate does the trick: http://pastebin.com/crgvszTQ thanks frals for the hint :) | 12:05 |
| DocScrutinizer | (I know there is, as there are appstarter bookmarks - just hopend for somebody knows the exact incantation) | 12:06 |
| djszapi | eh....there are "hidden" packages inside the nokia-binaries link, like this: http://harmattan-dev.nokia.com/pool/harmattan-beta/nokia-binaries/liba/libarchive/libarchive-dev_2.8.0-3+0m6_armel.deb This is funky since the that directory is not browsable by default, so you can easily waste hours to get the right dependencies ... | 12:06 |
| *** cpscotti has joined #harmattan | 12:12 | |
| djszapi | why the hell is libarchive binary ? | 12:14 |
| djszapi | seems, Nokia tend to not put every generated packages onto the website (like in case of xz-utils, liblzma-dev is missing) meaning that the community needs to duplicate the effort and drop that package usage. | 12:15 |
| *** lardman|gone is now known as lardman | 12:23 | |
| lardman | lcuk: did you ping me? | 12:24 |
| *** RST38h has joined #harmattan | 12:25 | |
| lardman | morning all | 12:25 |
| RST38h | moo lardman | 12:25 |
| lardman | hey RST38h | 12:25 |
| RST38h | So, all action basically moved from #maemo here? | 12:25 |
| lcuk | lardman, last night I did | 12:28 |
| lcuk | cannot remember what I wanted now though | 12:28 |
| lardman | lcuk: sorry, no internet at home yet (still!) | 12:28 |
| djszapi | lardman: use your N950 with mobile internet :) | 12:28 |
| lardman | RST38h: well related to N950, yep | 12:28 |
| djszapi | and hotspot mode :) | 12:28 |
| lardman | djszapi: I only get 2G, am out in the sticks | 12:28 |
| djszapi | uhh.. | 12:29 |
| lardman | There is a sum total of one wifi ap that I can detect in the village :D | 12:29 |
| lardman | I'm therefore hoping that dual band will work fine with my n router | 12:30 |
| lardman | ;) | 12:30 |
| DocScrutinizer | djszapi: (nokia tend not to put) now that's really bad, and I think this *must* change | 12:32 |
| djszapi | DocScrutinizer: the SDK team leader told me that they do not publish debian packages which comes from xz-utils or any other package which is not a dependency inside the SDK. | 12:33 |
| DocScrutinizer | djszapi: for me the general way Nokia acts clearly tells "they" don't really want "us" | 12:33 |
| djszapi | but it is no go in my not that humble opinion :) | 12:33 |
| * RST38h wonders if anyone in the world still knows what Nokia "wants" and where it is going | 12:34 | |
| RST38h | For SURE, I mean | 12:34 |
| djszapi | so to translate it this way: who cares about the community packages ? The point is that we provide a everything for our SDK, that is. Nothing else matters. | 12:35 |
| djszapi | not sure what made them making libarchive as binary. They most likely got it from debian anyway :D | 12:36 |
| *** Sicelo has quit IRC | 12:37 | |
| *** Sicelo has joined #harmattan | 12:37 | |
| DocScrutinizer | I'm almost tempted to have a long talk with my friend Harald Welte, what can be done | 12:39 |
| DocScrutinizer | e.g I don't think aegis and GPLv3 can coexist | 12:41 |
| Stskeeps | there's no gplv3 in the image | 12:42 |
| DocScrutinizer | that was sth to get checked then | 12:42 |
| DocScrutinizer | community is very forgiving as long as the players are showing good will | 12:42 |
| DocScrutinizer | I don't think Nokia actually qualifies for tagging them "showing good will" | 12:43 |
| DocScrutinizer | nobody really is going to complain about blobs to protect IP - like GFX or GPS drivers, whatnot | 12:46 |
| DocScrutinizer | but the general way Nokia acts seems way beyond that, hijacking linux and usurping to own it | 12:47 |
| djszapi | Cannot we add this link in some format to our sources.list file ? http://harmattan-dev.nokia.com/pool/harmattan-beta | 12:47 |
| djszapi | that would probably solve the "hidden" binary packages issue if you cannot find them on the webinterface. | 12:47 |
| djszapi | It is not a solution for the insane truncation though. | 12:48 |
| *** CaCO3 has joined #harmattan | 12:49 | |
| Mek | hmm, anyone know if any of http://www.developer.nokia.com/swipe/ux/pages/View_Headers.html is already possible with qml? or are those more libmeegotouch only widgets... | 12:52 |
| DocScrutinizer | landed! good by spaceshuttle | 12:58 |
| frals | djszapi: come on, that waas not the reply you got internally regarding the unpublished packages | 12:59 |
| *** lcuk has quit IRC | 13:01 | |
| *** lcuk has joined #harmattan | 13:01 | |
| frals | and since it seemed some bug plagued the publishing of the packages initially (which is why some sources were visible on google cache etc) i guess they disabled browsing nokia-binaries over http, should be fine to get them through apt in sb | 13:03 |
| DocScrutinizer | frals: then damn document that, nokia! | 13:05 |
| djszapi | frals: that was the answer I got from the SDK team leader, we can ask | 13:07 |
| DocScrutinizer | frals: quoting-> "we've been told aegis can get disabled completely" | 13:07 |
| djszapi | fragl: it has nothing to do with google cache | 13:07 |
| frals | " djszapi> seems, Nokia tend to not put every generated packages onto the website" | 13:08 |
| djszapi | * frals, sorry | 13:08 |
| DocScrutinizer | frals: It's not enough *somebody* can disable it, it's mandatory WE (the community developers) are empowered and told how to disable it | 13:08 |
| frals | djszapi: what was the last line from SDK team lead in the last email regarding that? | 13:08 |
| djszapi | frals: we had a private chat | 13:09 |
| frals | DocScrutinizer: this has nothing to do with aegis afaik | 13:09 |
| DocScrutinizer | sure, I know | 13:09 |
| frals | djszapi: ok, i went on the mailing list info which stated that | 13:09 |
| djszapi | frals: and the team lead did not answer in email... | 13:09 |
| DocScrutinizer | it's just in line wrt mindset and general "do no evil" policy | 13:09 |
| frals | isnt souyma the right person any more? | 13:09 |
| djszapi | well, I have been told Lengyel Marcell | 13:09 |
| djszapi | Hence I asked him. | 13:10 |
| frals | DocScrutinizer: i fully agree that aegis should be nuked from orbit, i guess djszapi can help with that ;) | 13:10 |
| frals | ah, maybe soumya is below him | 13:10 |
| djszapi | frals: No, I cannot. | 13:10 |
| DocScrutinizer | n he can't | 13:10 |
| DocScrutinizer | no* | 13:10 |
| djszapi | frals. I did not make decisions about the architecture, I was just a code monkey :) | 13:10 |
| Damion3 | can't you load a kernel module via an aegis signed .deb install which can then do whatever you like in kernel space including turning aegis off? | 13:10 |
| frals | djszapi: code monkeys know how to break the stuff they code, surely? ;-) | 13:10 |
| djszapi | cannot | 13:10 |
| djszapi | since there is an aegis validator | 13:10 |
| djszapi | Damion3 ^ | 13:10 |
| Damion3 | baked in to the kernel module loading stuff? | 13:11 |
| djszapi | frals: but who is the SDK team leader then ? I am confused | 13:11 |
| djszapi | I talked to Lengyel Marcell.... | 13:11 |
| frals | djszapi: i assume its marcell if that was what souyma told you :) | 13:11 |
| djszapi | yeah, so I said the truth then :D | 13:11 |
| frals | i thought it was her but seems i was wrong, i went on what souyma said | 13:11 |
| DocScrutinizer | frals: this is a clean implemetation of a security framework (TC) that has been developed and pentested for 10 some years | 13:12 |
| DocScrutinizer | it's probably stable | 13:12 |
| djszapi | frals: she/he could not help, just Marcell | 13:12 |
| frals | djszapi: ok, i was wrong then :) | 13:12 |
| Damion3 | if you're able to get dpkg to inject the correct checksums for your new .deb can't that include kernel modules to be loaded ? | 13:13 |
| DocScrutinizer | frals: there *might* be holes in config, but no vulnerabilities that developers of aegis would know about | 13:13 |
| djszapi | Damion3: it is not about dpkg and dpkg.real | 13:13 |
| djszapi | it is about the kernel validation module which works from the boot process... | 13:14 |
| DocScrutinizer | frals: e.g you might find a binary that allows privilege escalation as it has a poorly designed manifest | 13:14 |
| DocScrutinizer | frals: but there's definitely no joshua code in aegis | 13:14 |
| Damion3 | djszapi: the boot loader only lets you boot correctly signed kernels? and there is a fixed list of allowed modules that can be loaded? | 13:15 |
| *** vandenoever has quit IRC | 13:16 | |
| DocScrutinizer | Damion3: check how e.g GPG works | 13:16 |
| frals | its a shame the echo 0 > /etc/aegis/enable doesnt work on production devices, that would be neat :( | 13:16 |
| DocScrutinizer | Damion3: this gives you a bit of an idea how security frameworks work | 13:17 |
| flux | frals, indeed | 13:17 |
| DocScrutinizer | just TC aka aegis aka MSSF does to files and binaries what GPG does to mails | 13:17 |
| djszapi | frals: no | 13:18 |
| djszapi | that would not really be neat at all | 13:18 |
| djszapi | that would be the worst ever from user point of view. | 13:18 |
| Damion3 | DocScrutinizer: the signing, yeah, with essentially impossible to create a hash collision even with months of trying to generate a binary with matching hash | 13:18 |
| djszapi | Do not forget, Nokia wanted to make a user, customer device, not for us, developers. | 13:19 |
| Damion3 | due to factoring primes or something? | 13:19 |
| frals | djszapi: im aware, but from my pov as a developer it would make it so much faster to develop ;) | 13:19 |
| DocScrutinizer | Damion3: it's the same crypto stuff used, yes | 13:19 |
| Damion3 | but it's possible for joikuspot to create a wifi hotspot binary and kernel module, sign it and install it to a phone. When I run it the kernel modprobes their nat kludge in and voila the kernel is running their module. I don't see how somebody else can't just create a .deb and sign it. | 13:21 |
| Damion3 | does everything have to go via nokia? | 13:21 |
| DocScrutinizer | Damion3: basically yes, though nokia can issue signed root certs for other developers, like you are signing a pubkey of another person you know, in GPG keyring | 13:23 |
| DocScrutinizer | then this 3rd party can use this signed root cert to sign their stuff, with the permissions granted to root cert by Nokia | 13:23 |
| Damion3 | that'll explain why one of the partitions of the built in block device mentions certificates when you flash it | 13:23 |
| Damion3 | DocScrutinizer: that's give quite some trust to joikuspot | 13:24 |
| Damion3 | so I can click on any .deb via http and that gets autosigned by some aegis thing on the phone at install time | 13:25 |
| Damion3 | but that can only run as user, no root, no kernel modules unless signed by nokia? | 13:25 |
| Damion3 | s/nokia/nokia or somebody with a signed root cert/ | 13:26 |
| infobot | Damion3 meant: but that can only run as user, no root, no kernel modules unless signed by nokia or somebody with a signed root cert? | 13:26 |
| DocScrutinizer | Damion3: MSSF is really complex, I don't got it to wrap my head around it completely. I'm aware though of the basic design principles and mode of operation of TC | 13:27 |
| DocScrutinizer | for more read: | 13:27 |
| DocScrutinizer | ~aegis | 13:27 |
| infobot | i guess aegis is 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 | 13:27 |
| *** NIN101 has joined #harmattan | 13:27 | |
| djszapi | Damion3: you cannot load your own module... | 13:28 |
| djszapi | just in open mode with own kernel.. | 13:28 |
| djszapi | not even from ovistore, just from Nokia | 13:29 |
| Damion3 | djszapi: What about run stuff as root, mount something or use chroot? | 13:29 |
| djszapi | Damion3: you do not know how aegis works | 13:29 |
| djszapi | it is not about root or non root anymore, it is about credentials. | 13:29 |
| DocScrutinizer | Damion3: also http://www.gnu.org/philosophy/can-you-trust.html | 13:30 |
| Damion3 | yeah I noticed root not being able to kill user processes | 13:30 |
| Teo` | the main menu disappearing icons on rotation seems to be not random, I'd guess it might be a quite silly issue | 13:30 |
| Teo` | is that part proprietary or Free? | 13:30 |
| Damion3 | so I don't mean run as root sorry scratch that. I do however want to echo something in to /proc and do mounting and chroot | 13:30 |
| djszapi | Damion3: third time, you cannot load not Nokia modules in non-open mode with own kernel :D | 13:31 |
| Damion3 | I didn't mention loading kernel modules | 13:31 |
| Damion3 | well i did earlier | 13:31 |
| djszapi | oh you asked different thing.. | 13:31 |
| Damion3 | sorry yes | 13:31 |
| Damion3 | qemu+wine needs misc binfmt enabled and chroot | 13:32 |
| djszapi | forget root as "superuser". It is not that term in this context. It is a user device, not sysadmin | 13:32 |
| Damion3 | so chmod user /proc/sys/binfmt && echo "x86:elf" >/proc/sys/binfmt might work as a user | 13:32 |
| Damion3 | (those aren't the exact strings) | 13:33 |
| djszapi | again, it is credential based... | 13:33 |
| djszapi | not user/root/whatever related. | 13:33 |
| Damion3 | do how do I find it user or any user on the phone, has the correct credentials to do stuff like that ? | 13:34 |
| djszapi | accli | 13:34 |
| Damion3 | ahh :) | 13:34 |
| Damion3 | Current mode: normal | 13:36 |
| Damion3 | I'll need to find the crednetials | 13:37 |
| Damion3 | hmm <credential name="CAP::chown" /> | 13:37 |
| Damion3 | so "accli -t CAP::chown && echo yes" | 13:41 |
| Damion3 | means "user" doesn't have the chown capability | 13:41 |
| Damion3 | oddly I seem to be able to chown any file anywhere to any user | 13:41 |
| flux | damion3, you may need to have another token for the other user to actually be able to do it | 13:42 |
| flux | damion3, oh, actually I read that wrong :-) | 13:42 |
| flux | damion3, so you are 'user' or 'root'? as user you can chown anything as you like?-o | 13:43 |
| Damion3 | open shell, sat at busybox's ash shell in /home/user | 13:43 |
| Damion3 | id returns user | 13:43 |
| Damion3 | accli -t CAP::chown sets $? to 1 | 13:43 |
| Damion3 | touch blah && chown root blah | 13:43 |
| Damion3 | works | 13:43 |
| flux | fun :) | 13:43 |
| Damion3 | oh, now it doesn't | 13:44 |
| DocScrutinizer | beware, see http://forum.meego.com/showpost.php?p=25529&postcount=94 | 13:44 |
| Damion3 | that worked before :) | 13:44 |
| DocScrutinizer | aegis can be a bitch with superpowers | 13:45 |
| Teo` | now, this might be a frivolous question, but... is there a way to change the background image of the home screen? | 13:45 |
| Teo` | it's so black! | 13:45 |
| djszapi | Damion3: you should learn how to use accli | 13:45 |
| DocScrutinizer | Teo`: this is the right dresscode for FOSS funeral party | 13:45 |
| Damion3 | djszapi: okay, I have only just learnt about it :) no man page on the phone | 13:46 |
| Teo` | DocScrutinizer: for a moment there, I LOL'd then I realized you might be right :( | 13:46 |
| Damion3 | the help says -t tests for capabilities | 13:46 |
| Teo` | DocScrutinizer: you think they disabled it on purpose because it's a dev device? | 13:46 |
| DocScrutinizer | Teo`: yeah, my jokes often are poisoned | 13:46 |
| Damion3 | which it seems to do. I don't know why chown worked before. That must have been the previous firmware | 13:46 |
| DocScrutinizer | Teo`: this is a devel device for N9. N9 has AMOLED which eats power like my kids eat candy, on mostly white screens | 13:47 |
| Damion3 | ahhh and accli -I in develsh returns a massively increased list with multiple capabilities | 13:47 |
| Teo` | :-/ | 13:48 |
| Teo` | I sure hope they don't leave it just black and unchangeable | 13:48 |
| DocScrutinizer | I don't think theming is locked down anyway | 13:48 |
| djszapi | Damion3: you can also check the restok.conf out, if you do not like accli | 13:49 |
| djszapi | there are more ways of doing things ;) | 13:49 |
| DocScrutinizer | you probably just have to wait for OVI "Coming soon..." to download other themens X-P | 13:49 |
| Teo` | gah | 13:50 |
| DocScrutinizer | restok ?? o.O | 13:50 |
| Damion3 | what I want to know is, is there a capability for being able to place a value in a special kernel mounted binfmt_misc directory in /proc so I can point it at an installed binary, so executing x86 binaries works. | 13:50 |
| *** deimos has joined #harmattan | 13:51 | |
| DocScrutinizer | djszapi: what makes aegis brick the device? how does it work? | 13:51 |
| DocScrutinizer | http://forum.meego.com/showpost.php?p=25529&postcount=94 | 13:51 |
| Damion3 | wow, just typing this I realise that it interferes directly with the fundamental stuff like being able to execute signed binaries. | 13:51 |
| *** trx has quit IRC | 13:52 | |
| djszapi | Damion3: you can do, not sure I would do that, but it is technically possible. | 13:52 |
| DocScrutinizer | Damion3: http://forum.meego.com/showthread.php?t=3827 and the links therein are helpful with aegis | 13:53 |
| Damion3 | I guess if qemu-user-i386-static was installed signed to be able to execute then it should be happy itself being able to run other stuff | 13:54 |
| DocScrutinizer | djszapi: (brick) and especially *why* does it work that way. Where in aegis docs and whitepapers it says aegis will kill the system under certain conditions? | 13:54 |
| DocScrutinizer | I'm missing the rationale behind this behaviour of aegis | 13:55 |
| djszapi | DocScrutinizer: documentation is not my responsibility, SDK team :) | 13:55 |
| DocScrutinizer | sure sure, just thought you might know a few good pointers or hearsay | 13:56 |
| cpscotti | anyone knows how qtcAddDeployment works? or at least where's the doc (or source). The whole thing (qt creator's deployment.pri) looks like a baad&ugly workaround :/ | 13:56 |
| *** trx has joined #harmattan | 13:57 | |
| DocScrutinizer | djszapi: the general TC concept defines a state change of system from trusted (aka "normal") to tainted (aka "open") shall terminate all processes that requested a secure system, and clean out those processes' mamory content so no passwords or key or whatever can get compromised. A reboot would suffice for that, why does Aegis kill the system so you have to reflash? | 13:59 |
| DocScrutinizer | this feels utterly insane | 14:01 |
| DocScrutinizer | and an evil policy implemented by Nokia deliberately without any rationale from system PoV | 14:02 |
| DocScrutinizer | unless somebody provides a sound rationale for this behaviour | 14:03 |
| Damion3 | the links in the post you mention don't work for me so I'm assuming from what you're saying that those commands locked up the device and then it wouldn't boot. | 14:03 |
| DocScrutinizer | the links in http://forum.meego.com/showpost.php?p=25529&postcount=94 ? | 14:03 |
| Damion3 | was bash placed as /bin/sh ? Because that would obviously brick it :) | 14:04 |
| DocScrutinizer | no, it for sure wasn't | 14:04 |
| Damion3 | yes, 1st links says "The page isn't redirecting properly" | 14:04 |
| DocScrutinizer | this is a bug in OVI authentication | 14:04 |
| Damion3 | 2nd links says "Some Error" from http://paste.debian.net/122437/ | 14:04 |
| DocScrutinizer | humm | 14:04 |
| Damion3 | 3rd is the bash binary they were trying to setuid | 14:04 |
| Damion3 | it's not that important I'm just trying to see what you're saying. Are you saying it seems like Nokia are pusnishing attempts to created a setuid shell by making the device not boot rather than just saying not and causing a reboot to undo relaxed mode ? | 14:05 |
| DocScrutinizer | yeah share.ovi.com is borked once again | 14:05 |
| Damion3 | There is a phrase "never atribute to malice, what can equally be attributed to incompetence". A bit harsh, but I suspect the bricking wasn't planned | 14:06 |
| DocScrutinizer | Damion3: try to go to http://share.ovi.com | 14:07 |
| Damion3 | but then I don't get to see what those linked contained. maybe a message from nokia saying "we spot you're naught now reflash" | 14:07 |
| *** djszapi has left #harmattan | 14:07 | |
| DocScrutinizer | Damion3: then from there go to http://share.ovi.com/media/joerg900.public/joerg900.10170 | 14:07 |
| DocScrutinizer | might work | 14:07 |
| Damion3 | yes | 14:07 |
| Damion3 | would this happen with any extra binary added in to /bin ? | 14:08 |
| DocScrutinizer | I gather you got a nokia account as well, ad the ovi/nokia SSO has issues with redirecting for "complex" URLs | 14:08 |
| DocScrutinizer | and* | 14:08 |
| Damion3 | DocScrutinizer: yeah I'm using a different browser now. Multi login sso is hard :) | 14:09 |
| Damion3 | I should know, Google are still trying to get it right | 14:09 |
| DocScrutinizer | Damion3: (any binary) It seems an issue particular to that bash binary/.deb, but I don't know what aegis-exec -a CAP::setgid -a CAP::setuid <random-binary> will do - probably the result of bricking device will be similar | 14:10 |
| Damion3 | perhaps they dont' want any binary other than ones they approve of to be set with the setuid capability | 14:11 |
| Damion3 | although that doesn't make much sense if superuser doesn't mean anything anymore | 14:11 |
| Damion3 | so MSSF and aegis are different again to SElinux and none of these are the same as the filesystem specific setcap capabilities like granting binaries the ability to listen on ports <1024 and similar. | 14:12 |
| Damion3 | unix used to be so simple :) | 14:13 |
| Damion3 | ahha sys_chroot is in restok.conf | 14:14 |
| *** TSCHAKeee has quit IRC | 14:14 | |
| *** TSCHAKeee has joined #harmattan | 14:15 | |
| *** aleksander_m has quit IRC | 14:15 | |
| Damion3 | presumably this config is also signed and so can't be edited ;) People (hopefully) much smarter than me have already gone over all of these obvious stuff | 14:16 |
| Damion3 | of the 6 projects I'm working on. 4 of them do some quite nasty kernel/module interaction and are going to be a problem. And one of the other two might be awkward depending on what I can send over bluetooth. | 14:17 |
| Damion3 | it means I can't tinker. I need to do it properly and get Nokia to approve :( | 14:17 |
| Damion3 | well except I can tinker because the phone google seem to have can be put in open mode :-p | 14:18 |
| Damion3 | but it's not useful for others wasn't to try stuff out with me. | 14:18 |
| MohammadAG | two weeks in New Order :/ | 14:19 |
| Stskeeps | MohammadAG: i had like 12 | 14:19 |
| MohammadAG | Stskeeps, so I beat your record, yay(?) | 14:20 |
| Damion3 | ? | 14:20 |
| gri | Gah, startup performance of my app sucks even though I use MDeclarativeCache booster things :( Also experiencing funny flickering effects on device :( | 14:26 |
| DocScrutinizer | Damion3: hmm, I'm not sure this post looks *really* better now, but anyway... | 14:27 |
| *** aleksander_m has joined #harmattan | 14:30 | |
| *** lbt has quit IRC | 14:33 | |
| *** djszapi has joined #harmattan | 14:40 | |
| Damion3 | btw thanks that thread is like reading my own discoveries on the device I've been lent (until my dev one arrives). So I don't have any docs whatsoever with the one I have and it's apparently special in that I can put it in open mode. | 15:09 |
| Damion3 | (re: http://forum.meego.com/showthread.php?t=3827) | 15:09 |
| djszapi | Damion3: special device for open mode ? | 15:13 |
| kimju | most likely a internal prototype with r&d software. | 15:14 |
| djszapi | open mode has nothing to deal with specific device. It should work on both. | 15:14 |
| Damion3 | it's whatever nokia sent to google | 15:14 |
| *** lbt has joined #harmattan | 15:21 | |
| *** lbt has joined #harmattan | 15:21 | |
| kimju | btw, some info and examples about aegis: https://www.nixuopen.org/blog/2010/11/developing-for-meego-part-2/ and https://www.nixuopen.org/blog/2010/11/developing-for-meego-part-3/ | 15:22 |
| *** sebas has quit IRC | 15:23 | |
| djszapi | kimju: there were official examples FYI | 15:23 |
| djszapi | http://meego.gitorious.org/meego-platform-security/aegis-examples/trees/master | 15:24 |
| *** sebas has joined #harmattan | 15:25 | |
| djszapi | I am not sure it is the most up to date, we wrote quite a while ago | 15:26 |
| kimju | and then there is the harmattan security guide at developer.nokia.com | 15:27 |
| gri | my program can have a portrait and landscape statusbar at the same time! :( http://www.youtube.com/watch?v=f6YX0C8Ltz8 anyone has an idea? | 15:27 |
| djszapi | gri: I experienced the same yesterday :) | 15:28 |
| djszapi | kimju: actually that blog is a bit funky | 15:29 |
| gri | I'd bet if it was only meegotouch without qml, everything would be fine (like all the other account-ui plugins) | 15:29 |
| xerxes2 | if i want to port pychess (gtk based) to harmattan what do i use? | 15:29 |
| djszapi | or well, it is good enough, but it misses the Qt layer on the top of the security which can make the life easier as well | 15:30 |
| spenap | has anyone managed to use the share interface (libshare-ui)? | 15:30 |
| achipa | spenap: shoot | 15:31 |
| kimju | djszapi, anyway, I found it interesting to read as it was written from slightly different point of view than the official documents | 15:32 |
| spenap | achipa, that was the question. I installed and used it trying to follow this example: https://meego.gitorious.org/meego-sharing-framework/share-ui/blobs/master/examples/link-share/page.cpp | 15:32 |
| spenap | but it said "wrong interface" and nothing showed up | 15:32 |
| djszapi | kimju: from an official person :D | 15:32 |
| kimju | more like an tutorial than a reference.. | 15:33 |
| achipa | spenap: what's your SHAREUI_DBUS_SERVICE ? | 15:33 |
| achipa | spenap: scratch that, I would just use ShareUiInterface* interface = new ShareUiInterface(); | 15:34 |
| achipa | and it should go to the right place | 15:34 |
| spenap | he, there's no such constructor, achipa | 15:34 |
| spenap | or at least | 15:35 |
| spenap | that's what I got | 15:35 |
| spenap | if you check libshare-ui-dev | 15:35 |
| spenap | the ShareUI/shareinterface.h (or something like that, I'm writing it from memory) | 15:35 |
| spenap | only has an explicit constructor requiring all the args | 15:35 |
| spenap | I was asking looking for some success stories :), I tried this two days ago, so cannot check the exact errors at the moment | 15:36 |
| djszapi | kimju: I think the interesting bit is also that, it has been published a while ago. I thought these things are secret, mainly that time back. | 15:36 |
| *** gri has quit IRC | 15:39 | |
| *** khertan1 has joined #harmattan | 15:40 | |
| khertan1 | Hello | 15:40 |
| achipa | spenap: worksforme ? | 15:40 |
| achipa | (with that exact constructor) | 15:40 |
| DocScrutinizer | kimju: how does https://www.nixuopen.org/blog/2010/11/developing-for-meego-part-2/ apply to harmattan? | 15:41 |
| djszapi | most part (99%) is irrelevant to Harmattan. | 15:43 |
| DocScrutinizer | I mean it's about netbook, RPM and all that | 15:43 |
| DocScrutinizer | djszapi: thought as much | 15:45 |
| * DocScrutinizer closes browser window without creating a bookmark for it | 15:45 | |
| kimju | there is some generic description of the framework. the examples are of cource not directly applicable | 15:46 |
| DocScrutinizer | >> Minimal changes to the current model (enforcement phase) Good level of flexibility and granularity, easy to understand concept (KISS)<< http://selinuxproject.org/~jmorris/lss2010_slides/LinuxCon_access_control_v3.pdf MUHAHAHA *cough* | 15:48 |
| djszapi | kimju: I do not see any generic point between meego and the maemo security model nowadays. It was a try in the beginning to make an upstream implementation of aegis with smack, but they radically went far away from each other. I am not sure you should read so old articles in the mobile industry, might be a bit oudated :) | 15:52 |
| spenap | ouch, achipa | 15:52 |
| spenap | HARMATTAN_X86 target and installing libshare-ui-dev? | 15:53 |
| DocScrutinizer | /me wonders what >> source with higher trust level SW source trust is based on the source Quality Assurance level << means in reality | 15:56 |
| DocScrutinizer | meh | 15:56 |
| * DocScrutinizer wonders what >> SW source trust is based on the source Quality Assurance level << means in reality | 15:56 | |
| * DocScrutinizer wonders what's the email of the "source Quality Assurance" evaluation department in Nokia | 15:58 | |
| frals | in practice i guess it means something like trusted level: on device >> ovi >> cobs apps? | 15:58 |
| achipa | spenap: actually, just head-on from qtcreator and the harmattan platform api target | 15:58 |
| spenap | mmm | 15:58 |
| spenap | achipa, ShareUiInterface(const QString &service, const QString &path, const QDBusConnection &connection, QObject *parent = 0); <- this is what /usr/include/ShareUI/shareuiinterface.h reads | 15:59 |
| DocScrutinizer | frals: the question was "who is assigning >trust levels< to sources?" | 15:59 |
| achipa | spenap: didn't check - it worked for me so never bothered to look into the header | 15:59 |
| *** djszapi has left #harmattan | 16:00 | |
| *** fiferboy has joined #harmattan | 16:00 | |
| spenap | achipa, thanks anyway. I'll go back to that later and see what I get. | 16:01 |
| achipa | spenap: there was one caveat, to HAVE TO SET THE MIMETYPE, otherwise it falls flat on it's face | 16:02 |
| *** hardaker has joined #harmattan | 16:03 | |
| DocScrutinizer | anyway it seems resonable to postpone *any* bugreports against aegis until N9 got mass deployed and you also found a proper exploit for the bug :-) (preferably an exploit that can't get defeated by deploying an aegis-update) | 16:10 |
| Damion3 | if these security measures mean movie studios and game developers relax and trust the device I see this as a good thing. Same for governments wanting an ultra secure in-the-field device to replace the crackberry | 16:12 |
| DocScrutinizer | haha | 16:12 |
| DocScrutinizer | governments are more interested in an immutable backdoor in Skype | 16:13 |
| DocScrutinizer | which for sure is enforcable via aegis | 16:13 |
| Damion3 | I still don't care :) *la la la* (blocks eyes and ears) | 16:13 |
| DocScrutinizer | also makes sure not everybody could use that backdoor | 16:13 |
| Damion3 | if it means they like the device and it's popular then good | 16:14 |
| DocScrutinizer | nonsense | 16:14 |
| Damion3 | you don't want it to be popular? | 16:14 |
| DocScrutinizer | please read | 16:14 |
| DocScrutinizer | ~aegis | 16:14 |
| Damion3 | Of course for me, it just so happens I like to and am involved in faffing about with quite low level kernel-module-esque stuff which it gets in the way of :) | 16:14 |
| infobot | [aegis] 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 | 16:14 |
| DocScrutinizer | esp 2nd link | 16:15 |
| DocScrutinizer | also http://www.gnu.org/philosophy/can-you-trust.html | 16:15 |
| Damion3 | oh dear | 16:15 |
| Damion3 | sandal wearing GNU/debian fans rant like this | 16:15 |
| Damion3 | "They believe that it may cause consumers to lose anonymity in their online interactions" - interesting. Might be true. I STILL don't care :) | 16:16 |
| DocScrutinizer | I'm not allowed to write what I'd like to answer | 16:16 |
| RST38h | http://www.youtube.com/watch?v=I25UeVXrEHQ | 16:16 |
| *** seif_ has quit IRC | 16:18 | |
| Damion3 | also I'd like a device to be trusted by movie studios. That doesn't mean they should trust it, as it could be just as crackable by the wrong types of people. I don'tcare how naïve movie studioes are I just want them to trust the device and thus publish for it. In the hope it makes the device do well and be popular enough to succeed. | 16:18 |
| RST38h | Let us put it this way: | 16:19 |
| RST38h | If the studios do not trust my device, do not publish for it, or force me to jump through incredible loops to get content, I will go and torrent the same content. | 16:19 |
| DocScrutinizer | ok I found an answer that doesn't sound too rude: please consider buying a Mediaplayer device, or a winP7 or Android based one instead | 16:19 |
| Damion3 | aegis seems like a genuine effort to succeed in its goal, which is unneccessary when all you need to do is convince less technical types that the security measures in place are enough. | 16:19 |
| flux | damion3, well, let them put their secret keys into some secure place. but let us have a mode where we don't need those keys, but do everything else. | 16:20 |
| flux | I'm pretty sure the bootloader is quite solid anyway. it was for n900 already, wasn't it? | 16:20 |
| flux | iow there is a secret place somewhere. | 16:20 |
| Damion3 | flux: yes. | 16:20 |
| * RST38h yawns: this Aegis stuff is not new, Damion3. Other prople and companies have tried before. They have all failed. | 16:20 | |
| RST38h | There are clear technical reasons WHY they failed | 16:20 |
| flux | xbox360 is not cracked yet, is it? | 16:20 |
| RST38h | flux: Cracked, uses XTEA | 16:20 |
| Damion3 | PS3 took about 4 years | 16:21 |
| RST38h | Had to physicall hack it though | 16:21 |
| *** razvanpetru has joined #harmattan | 16:21 | |
| RST38h | Let me see... SD cards have "secure" features that nobody uses. Hard drives have "secure" areas that nobody uses. | 16:22 |
| Damion3 | I think we're sort of agreeing. What was needed was some vague handwavey effort which claimed to be secure to enable the publishers (who's business model is wrong anyway, we agree on that) to trust it. However some people are working (too hard) on making their wishes realised. | 16:22 |
| DocScrutinizer | >>You might think you can find out what nasty things a treacherous-computing application does, study how painful they are, and decide whether to accept them. Even if you can find this out, it would be foolish to accept the deal, but you can't even expect the deal to stand still. Once you come to depend on using the program, you are hooked and they know it; then they can change the deal. Some applications will automatically download | 16:22 |
| DocScrutinizer | upgrades that will do something different—and they won't give you a choice about whether to upgrade.<< | 16:22 |
| RST38h | BluRay/DVD go to great lengths to "protect" their content, both have been hacked, + people are using torrents anyway | 16:23 |
| flux | rst38h, well, physically hacking is a valid way, but I guess we're not talking about that here | 16:23 |
| RST38h | Freaking Symbian Foundation pretty much killed most Symbian development by introducing Aegis-like model | 16:23 |
| flux | I'm not sure how it would affect pay-per-view applications, if someone can retrieve the keys | 16:23 |
| flux | (or if it would affect) | 16:24 |
| RST38h | How many *more* hardware security blunders do you need? :) | 16:24 |
| flux | rst38h, everybody's doing it, so it's a race of not being the worst I guess :) | 16:24 |
| RST38h | flux: Actually, no | 16:24 |
| RST38h | flux: Apple is not doing it. Instead, they persuaded publishers to strip DRM | 16:24 |
| razvanpetru | no Symbian was much worse | 16:25 |
| RST38h | flux: And use lower prices to push pirates out of the picture | 16:25 |
| razvanpetru | you pretty much had to have a publisher ID (200$/year) to publish apps | 16:25 |
| flux | rst38h, well, that simplifies the software & managemnt a lot, not to mention UX :) | 16:25 |
| flux | UX is such a hot word. can't they use that to eradicate this silliness? | 16:25 |
| DocScrutinizer | RST38h: actually hoping for hackers to crack security is an elitist thinking only hackers can afford. Normal users are pretty much on peril of us hackers counteracting security measures and that's a bad thing to start with. | 16:31 |
| *** apol_ has joined #harmattan | 16:31 | |
| Damion3 | I don't understabd "on peril of us hackers" | 16:32 |
| Damion3 | assume english isn't my 1st language | 16:32 |
| Damion3 | (it is, I'm just lame ;)) | 16:32 |
| Damion3 | http://testyourvocab.com/ | 16:32 |
| DocScrutinizer | ~dict peril | 16:33 |
| infobot | Dictionary 'peril' (2 of 9): \Per"il\, v. t. [imp. & p. p. {Periled}or {Perilled}; p. pr. & vb. n. {Periling} or {Perilling}.] To expose to danger; to hazard; to risk; as, to peril one's life. [1913 Webster] ;; put in a dangerous, disadvantageous, or difficult position . | 16:33 |
| fiferboy | I really need to learn the dict protocol for my application | 16:33 |
| fiferboy | Maybe I could just pipe IRC output to it :) | 16:33 |
| DocScrutinizer | probably english isn't my premium language (I prefer assembler) | 16:33 |
| *** leinir has joined #harmattan | 16:33 | |
| lardman | perilled? | 16:34 |
| DocScrutinizer | s/peril/in a dependent state / | 16:34 |
| lardman | missing an em- ? | 16:34 |
| *** qgil_ has joined #harmattan | 16:35 | |
| lardman | this is the problem with using an American dictionary of course ;) | 16:35 |
| lardman | afternoon qgil_ | 16:35 |
| DocScrutinizer | qgil_: hi! | 16:35 |
| Damion3 | so to picka synonym: users are on danger of us hackers | 16:36 |
| Damion3 | I know what peril means, I meant the sentense didn't scan | 16:36 |
| qgil_ | hi there, I didn't know there was a #harmattan channel | 16:36 |
| Damion3 | and yes I used to say z80 asm was my 1st language :) | 16:36 |
| *** w00t has joined #harmattan | 16:37 | |
| lardman | qgil_: well harmattan talk was sort of taking over the #maemo channel | 16:37 |
| DocScrutinizer | Damion3: I know what you meant, it seems though you want to go on a meaning-of-words level though for this discussion, and I won't join in | 16:37 |
| qgil_ | lardman: not complaining | 16:37 |
| qgil_ | btw any Chess lovers willing to practice some Qt & Quick skills for a MeeGo mobile app focusing on handset? | 16:37 |
| Damion3 | I'mn not tolling I didn't scan what your sentense meant. Unless it was "users are in a situation where they have to rely on hackers to crack a security mechanism in order to get a useful device, which is a bad situation to start in" | 16:38 |
| DocScrutinizer | so evidently you got the meaning of what I said | 16:38 |
| Damion3 | okay then sorry | 16:39 |
| lardman | qgil_: sounds like we need a wiki page somewhere with app suggestions? | 16:39 |
| Damion3 | it really didn't scan well | 16:39 |
| DocScrutinizer | qgil_: seems like we'd love a word from you which should be the common community repo to throw in *all* the little tools we happen to build | 16:40 |
| Damion3 | I'd have s/peril of us/on a perilous situation of having to wait for us/ | 16:41 |
| Damion3 | I'll stop, I did evidently grok it | 16:41 |
| DocScrutinizer | qgil_: for now google seems our best friend to find if somebody else built midnight commander for harmattan | 16:41 |
| qgil_ | lardman: we are not looking for an app, we are looking for developers to join http://wiki.maemo.org/Miniature | 16:41 |
| qgil_ | DocScrutinizer: do you mind having this discussion at http://forum.meego.com/showthread.php?t=3671 ? | 16:42 |
| qgil_ | IRC is too volatile for my taste | 16:42 |
| lardman | qgil_: ah my mistake, but again, a page listing things that people would like ported (preferably requests from active devs in the project) might still be useful | 16:42 |
| DocScrutinizer | sure, no problem | 16:42 |
| qgil_ | especially when the answers rely in more than one person | 16:43 |
| qgil_ | lardman: yes, and actually I believe someone created such page already | 16:43 |
| lcuk | lardman, meego has the ITP (Intent to port) page: http://wiki.meego.com/Build_Infrastructure/Community_Builder/ITP | 16:45 |
| DocScrutinizer | qgil_: from a cursory read this thread is about a repo similar to maemo-extras. What we discussed here yesterday was a mere developer dump repo similar to extras-devel_just_'worse', that's not meant for end users | 16:48 |
| rm_you | morning qgil_! | 16:48 |
| * rm_you runs off to work | 16:48 | |
| qgil_ | DocScrutinizer: still, this discussion is relevant there | 16:49 |
| DocScrutinizer | qgil_: ...where we 250 devels would be well aware of all the risk this introduces | 16:49 |
| DocScrutinizer | qgil_: ok, will continue to read more closely. I hope though the ones participating in yesterday's discussion will also look there and contribute what they think | 16:50 |
| qgil_ | the current workflow is supposed to be all around Community OBS: devel = your home repo, testing is that common repo where packages clash together and stable is where only the end-user ready stuff lands | 16:50 |
| qgil_ | also note that these decisions are more on the side of the Harmattan OSS community developers than on the opinions of a Nokia employee that doesn't have to code on any app (yet) | 16:51 |
| qgil_ | in other words, don't expect me to give the ultimate answer | 16:51 |
| DocScrutinizer | qgil_: I'm really the noob in this discussion, I'm just interested in adding a repo (however to establish that) and then get mc by doing apt-get install mc, without having to mess with google to find it's been djszapi who built it and it's in *his* private repo | 16:52 |
| qgil_ | DocScrutinizer: as for toay there is noob ready answer to your question, but you are invited to join the discussion, sharing your problems and proposing solutions | 16:53 |
| qgil_ | (gosh, I thought my % of IRC typos would improve after the holidays) | 16:53 |
| DocScrutinizer | qgil_: http://mg.pov.lt/harmattan-irclog/%23harmattan.2011-07-20.log.html#t2011-07-20T14:44:17 | 16:54 |
| qgil_ | X-Fade (Niels Breet) is the guys coordinating all this setup | 16:55 |
| qgil_ | I see he is not even on this channel, but he is following that forum discussion | 16:55 |
| DocScrutinizer | ok | 16:55 |
| DocScrutinizer | qgil_: thanks a lot, so it's x-fade then | 16:56 |
| lardman | lcuk: That's not quite the same though | 16:56 |
| *** rm_work has joined #harmattan | 16:59 | |
| khertan1 | import com.nokia.extras 1.1 is not installed ... oups | 17:00 |
| khertan1 | is it available for qtcreator ? | 17:00 |
| qgil_ | DocScrutinizer: also not that this is not a Harmattan exclusive discussion, since this problem is common to any MeeGo OSS community developers | 17:00 |
| qgil_ | note | 17:01 |
| khertan1 | oups ... wrong channel, anyway ... hello everyone :) | 17:02 |
| DocScrutinizer | hmm, for a repo providing .deb for harmattan it seems to be, nevertheless - while the general problem for sure is common to whole FOSS community | 17:02 |
| mgedmin | but but but Harmattan packages aren't installable on Meego devices and vice versa | 17:02 |
| mgedmin | am I mistaken? | 17:02 |
| DocScrutinizer | hi khertan1 | 17:03 |
| lcuk | mgedmin, depends how they are written | 17:03 |
| mgedmin | my question was more about packagin | 17:04 |
| lcuk | some developers (like venemo) are creating real cross platform versions | 17:04 |
| RST38h | "Nokia posts massive loss, blames 'ambiguity'" (C)TheRegister | 17:04 |
| mgedmin | can I install a meego .rpm package on a n950? | 17:04 |
| lcuk | yes, his puzzle-master app is installable on pretty much every OS he has worked with | 17:04 |
| khertan1 | mgedmin: not currently ... maybe someone will port alien tool :) | 17:04 |
| DocScrutinizer | lcuk: but the harmattan developer toolbox repo is exactly *not* supposed to be cross-platform | 17:05 |
| lcuk | but qt is | 17:05 |
| * RST38h innocously asks why not create harmattan-specific repo at maemo.org | 17:05 | |
| DocScrutinizer | so what? | 17:05 |
| khertan1 | and the dev toolbox ... is mostly Qt :) | 17:05 |
| DocScrutinizer | I once more missed a parallel thread? :-D | 17:06 |
| khertan1 | RST38h: lost obs capability | 17:06 |
| RST38h | khertan: I can build with local OBS | 17:06 |
| DocScrutinizer | khertan1: I'd expect it to be mostly commandline anyway | 17:06 |
| khertan1 | someone ... don't remember who was working on how to build package for both system at the same time : debian style and rpm one | 17:06 |
| Stskeeps | khertan1: that's entirely possible | 17:06 |
| RST38h | khertan: + I am sure someone can hack together a scrpipt that would send built packages from the community OBS into the repo | 17:06 |
| khertan1 | Stskeeps: yep of course ... obs can do both, i was talking about using maemo current architecture | 17:07 |
| khertan1 | RST38h: oh ... yes also ... it s an other solution | 17:07 |
| *** razvanpetru has left #harmattan | 17:07 | |
| *** hardaker has quit IRC | 17:08 | |
| * khertan1 like more and more qtsdk ... just need python support ... need to look at qtcreator documentation to made plugin. | 17:08 | |
| DocScrutinizer | khertan1: when I'm looking for midnight commander I don't give a f* about if it's cross-platform or Qt, I just want to save compiling it myself and rather use the binary djszapi built 2 days ago | 17:08 |
| khertan1 | DocScrutinizer and where he build them ? | 17:09 |
| *** hardaker has joined #harmattan | 17:09 | |
| khertan1 | cobs ? | 17:09 |
| DocScrutinizer | assuming I trust in djszapi not doing anything that makes my house get nuked when I start that binary | 17:09 |
| Stskeeps | DocScrutinizer: i installed stuff by just clicking deb in browser | 17:09 |
| khertan1 | DocScrutinizer: installed git by downloading deb on the c-obs rzr home | 17:10 |
| * mgedmin sees "Ogg Vorbis" among the supported audio formats at https://www.developer.nokia.com/Devices/Device_specifications/N9/ and feels happy and fuzzy | 17:10 | |
| DocScrutinizer | Stskeeps: yes, but that'S exactly the point, WHERE to will you browse with your browser to click on a .deb | 17:10 |
| DocScrutinizer | ? | 17:10 |
| khertan1 | DocScrutinizer community obs | 17:10 |
| Stskeeps | DocScrutinizer: repo.pub.meego.com/home:/djszapi/ probably | 17:10 |
| Stskeeps | or djszapi:/ | 17:10 |
| DocScrutinizer | exactly | 17:10 |
| DocScrutinizer | this SUCKS | 17:10 |
| khertan1 | DocScrutinizerhttps://build.pub.meego.com/ | 17:10 |
| khertan1 | DocScrutinizer: why ? | 17:11 |
| khertan1 | there is a search feature | 17:11 |
| khertan1 | works well | 17:11 |
| DocScrutinizer | >>I'm just interested in adding a repo (however to establish that) and then get mc by doing apt-get install mc, without having to mess with google to find it's been djszapi who built it and it's in *his* private repo | 17:11 |
| Stskeeps | DocScrutinizer: well that's what apps.meego.com is supposed to solve.. | 17:11 |
| DocScrutinizer | ok, hope it eventually will establish this | 17:12 |
| khertan1 | DocScrutinizer: did you know the plateform and his architecture is still in devel and this is the purpose of n950 ? :) | 17:12 |
| DocScrutinizer | though - as mentioned above - [2011-07-21 15:48:31] <DocScrutinizer> qgil_: from a cursory read this thread is about a repo similar to maemo-extras. What we discussed here yesterday was a mere developer dump repo similar to extras-devel_just_'worse', that's not meant for end users | 17:13 |
| DocScrutinizer | khertan1: are you trolling? | 17:13 |
| qgil_ | DocScrutinizer: this extras-devel_just_'worse' would be part of the infra and QA process to publish OSS community apps for end users so yes, it is relevent - trust me :)= | 17:14 |
| kimju | I still suggest (as I did yesterday) that while waiting for the apps and apps-testing repos to get created and working, you just create a home:user:harmattan-testing-temp and collaborate there. when the official things get working, migrate there. | 17:15 |
| Stskeeps | it's really easy(TM) | 17:15 |
| khertan1 | DocScrutinizer no, i just suggest to not be angry or too much aggressiv about community repository not ready | 17:15 |
| Stskeeps | (what kimju suggests) | 17:15 |
| DocScrutinizer | khertan1: I'm all happy and relaxed | 17:16 |
| khertan1 | :) | 17:16 |
| DocScrutinizer | khertan1: just trying to scratch an itch probably a lot of devels face | 17:16 |
| qgil_ | khertan1: fwiw I didn't perceive any anger or aggressivity from DocScrutinizer | 17:16 |
| DocScrutinizer | right NOW I have to add | 17:16 |
| khertan1 | qgil_: yep i probably not choose the right word ... sorry :) | 17:16 |
| smoku | suggestion. see http://software.opensuse.org/ - we could have the same search engine for C-OBS | 17:17 |
| khertan1 | s/word/words | 17:17 |
| DocScrutinizer | smoku: search engine sounds awesome | 17:17 |
| khertan1 | DocScrutinizer sorry i didn't use the right words | 17:17 |
| DocScrutinizer | np | 17:17 |
| qgil_ | guys, all this discussio is really #meego relevant and there you have X-Fade, lbt and others working with the OBS on a daily basis | 17:17 |
| DocScrutinizer | :-) | 17:17 |
| khertan1 | smoku: https://build.pub.meego.com/search/search <<< it s the same type of search, isn't it ? | 17:18 |
| DocScrutinizer | ok, I'm handing the microphone to whoever feels more competent about this topic. As mentioned above I'm the noob here about repo topics | 17:18 |
| smoku | khertan: compare to http://software.opensuse.org/search?q=player&baseproject=ALL&exclude_debug=true | 17:18 |
| kimju | back to technical things. could someone with fairly clean device run 'dpkg -l > foo' and pastebin the results? I've already littered mine with local versions of packages, would like to have something to compare against | 17:19 |
| smoku | notice 1-Click installer links | 17:19 |
| khertan1 | smoku: yeah, indeed the result presentation is better | 17:19 |
| qgil_ | just making sure you are aware of http://apps-beta.meego.com/ | 17:20 |
| DocScrutinizer | probably that person with higher competence would be kimju then | 17:20 |
| qgil_ | DocScrutinizer: do me a favor and write a post with your concerns at that forum thread I posted above - thanks! | 17:21 |
| * lbt hopes the forum/email happens soon ... I'd love to join in having done more than a little work on repos etc,,,, | 17:22 | |
| kimju | DocScrutinizer, if you want I can walk you through setting such repo and how to push / accept things there. it's easy, the hard part would be deciding what to accept there. | 17:23 |
| khertan1 | Does someone here notice slow ssh on rm-680 ? | 17:24 |
| Stskeeps | khertan1: think it's terminal artifact | 17:24 |
| khertan1 | the one on the device ? | 17:24 |
| khertan1 | or the client i use to connect to it ? | 17:25 |
| lbt | DocScrutinizer: did you read the 'Surrounds' post I made a while back? | 17:25 |
| khertan1 | did you know any tips to improve this ? | 17:25 |
| rm_work | oh, for a "repository package search" there is always PackMan | 17:25 |
| rm_work | i could look at modifying it maybe | 17:26 |
| lbt | kimju: you too ? | 17:26 |
| rm_work | to look at Harmattan repos | 17:26 |
| kimju | khertan1, is the opening of connection slow or the connection after that? | 17:26 |
| rm_work | err | 17:26 |
| rm_work | PackRat | 17:26 |
| kimju | lbt, ? | 17:26 |
| rm_work | http://ageofikon.com/packrat/ | 17:26 |
| khertan1 | kimju: both ... many connection refuse until one works | 17:26 |
| khertan1 | s/refuse/timeout | 17:26 |
| kimju | is your network ok? | 17:26 |
| khertan1 | it s happen on 2 different network | 17:27 |
| lbt | kimju: search the wiki for Apps and Surrounds... I'm about to reboot :) | 17:27 |
| khertan1 | the one at work i could not be sure, but the one at home have no problem | 17:27 |
| rm_work | DocScrutinizer: would updating Packrat to look at Harmattan user repos help you? | 17:28 |
| rm_work | DocScrutinizer: I think I could do that fairly easily | 17:28 |
| *** gri has joined #harmattan | 17:28 | |
| lcuk | khertan1, over my wifi connecting via ssh is faster than connecting to other machines | 17:28 |
| khertan1 | kimju: same tips as meego ce on n900 ... pinging continuously google.com seems to fix the problem | 17:28 |
| DocScrutinizer | rm_work: if it omplements a one-step way to install "things" from a number of different authors, then yes | 17:29 |
| rm_work | DocScrutinizer: well, look at it | 17:29 |
| rm_work | http://ageofikon.com/packrat/ | 17:29 |
| DocScrutinizer | not exactly apt-get install, but maybe even nicer | 17:30 |
| rm_work | it generates .install files | 17:30 |
| rm_work | i don't know if those work in Harmattan | 17:30 |
| DocScrutinizer | awesome | 17:30 |
| rm_work | do they? | 17:30 |
| DocScrutinizer | they *should* afaik | 17:30 |
| rm_work | ok i will make a clone of this and see if i can make it look at harmattan | 17:30 |
| khertan1 | kimju: did you got your package list ? | 17:31 |
| rm_work | i just need a list of repositories | 17:31 |
| kimju | khertan1, not yet | 17:31 |
| rm_work | where was the user repo list? | 17:31 |
| DocScrutinizer | kimju: there's your technical solution :-D | 17:31 |
| Stskeeps | guys, for good measure, making a extras-devel of sorts is easy. make a OBS project, then have people click "submit package" on their package, find your repo name, write a reason and it shows up under "my requests" for repo maintainer | 17:31 |
| Stskeeps | :P | 17:31 |
| khertan1 | kimju: i got pastbin error :) | 17:31 |
| khertan1 | hihi | 17:31 |
| Stskeeps | you can add maintainers under 'users' in the project | 17:31 |
| khertan1 | kimju: http://pastebin.com/UcJQCMRD | 17:32 |
| lbt | of course you could at least *try* to learn from the mess in maemo extras-devel | 17:32 |
| lbt | and extras-devel was not what you want ... you want a simple place to put shared libs | 17:33 |
| * khertan1 didn't see extras-devel as a mess | 17:33 | |
| lbt | where you can reasonably rely on them | 17:33 |
| lbt | khertan1: ask xerxes2 | 17:33 |
| lbt | x-fade even | 17:33 |
| kimju | khertan1, thanks. | 17:34 |
| lardman | lbt: version changes breaking things you mean? | 17:34 |
| lbt | extras-devel exposed you to every experimental package from every developer | 17:34 |
| lbt | you got random version changes | 17:34 |
| lbt | it was a mess | 17:34 |
| lbt | sure some people can cope - most can't | 17:34 |
| DocScrutinizer | lbt: this is strictly developer only | 17:34 |
| khertan1 | lbt: but it s not supposed to be something to use | 17:34 |
| lardman | well that was more an issue of people not bothering to promote as the comments from the QA people were so offhand and rude, imho | 17:34 |
| khertan1 | i see it more as a test to promote to testing | 17:34 |
| lbt | you just don't grok the OBS | 17:35 |
| khertan1 | lardman: also | 17:35 |
| lbt | everyone gets their own PPA | 17:35 |
| rm_work | anyone have a link to the community repos list? | 17:35 |
| lbt | we have a 'shared pkgs' area called Surrounds | 17:35 |
| lbt | we have a development version called Surrounds:Testing | 17:35 |
| khertan1 | and who push to surrounds package ? | 17:36 |
| lbt | that's where all the shared (eg python) pkgs | 17:36 |
| lbt | the maintainer | 17:36 |
| khertan1 | i mean ... you will got also version change for your app (as a developper) | 17:36 |
| lardman | rm_work: http://wiki.meego.com/Community_Office/Community_device_program/Nokia/Development_repos | 17:36 |
| lbt | you get that through your PPA khertan1 | 17:36 |
| rm_work | thanks lardman | 17:36 |
| lardman | PPA sounds fine, as long as it can be enabled and searched quickly and easily | 17:37 |
| lbt | app version changes are reasonably 'private' | 17:37 |
| rm_work | now i need to figure out how to parse that page with Perl :P | 17:37 |
| lbt | lardman: we make that work | 17:37 |
| lbt | enabled is trivial | 17:37 |
| lbt | and as a PPA owner you can have *loads* of PPAs | 17:37 |
| lbt | you own the namespace | 17:37 |
| lardman | lbt: this all needs to be explained somewhere | 17:37 |
| lbt | lardman: it does | 17:38 |
| lardman | so that people can be pointed at it | 17:38 |
| lbt | read my blog from 6 months ago :) | 17:38 |
| lbt | it was pre-device so no-one really paid enough attention | 17:38 |
| lbt | but the framework is all in place | 17:38 |
| lardman | lbt: needs to be either on the wiki, or linked from the wiki from the migrating from N900 or similar | 17:38 |
| lbt | give me a few mins to reboot | 17:38 |
| lardman | sure, I certainly didn't pay attention :) | 17:38 |
| lardman | cool | 17:38 |
| *** lbt has quit IRC | 17:38 | |
| DocScrutinizer | I'm sure qgil_ will kick us all for not discussing this on #meego :-D but anyway seems we're getting really close now. At least I feel this topic might got sorted by tomorrow | 17:41 |
| lcuk | DocScrutinizer, better to be discussing it and having the context to use it under | 17:42 |
| lcuk | as lbt said, when this was heavily discussed last time there was no device to assist with motivation | 17:42 |
| lardman | I am coming round to the idea of PPAs, as long as one can search for a package you need (e.g. a library on which you want to rely) | 17:43 |
| DocScrutinizer | lardman: searchability basically was my concern, yes | 17:44 |
| lcuk | lardman, the idea of ppas is that once you find that package you bring it into your ppa | 17:44 |
| lardman | the one question I do have is how different PPAs are prioritised when you install something - if I have an older version of a lib that I've pulled into my PPA from the original dev, what happens if they update and their PPA is also enabled on my device? | 17:44 |
| rm_work | so are ALL the repositories here? http://repo.pub.meego.com/home:/ | 17:44 |
| lardman | rm_work: at a guess, yes | 17:45 |
| rm_work | anyone have any idea why 75% or so (at a glance) are duplicated, one with a colon at the end and one without? | 17:45 |
| Stskeeps | rm_work: home:cvm vs home:cvm:blah | 17:45 |
| rm_work | wow most of these aren't proper apt repos | 17:46 |
| rm_work | >_> | 17:47 |
| rm_work | damn i thought that'd be a sweet shortcut to getting all of the repositories but it looks like we actually DONT want 90% of them because they are trash | 17:47 |
| Stskeeps | rm_work: you know, we have rpm repos to | 17:47 |
| Stskeeps | o | 17:47 |
| Stskeeps | :P | 17:47 |
| rm_work | ah | 17:47 |
| rm_work | i maintain my statement then :P | 17:48 |
| lardman | Stskeeps: any ideas about my question a few lines up? | 17:48 |
| rm_work | short of starting the never ending format war, let's just say i dislike RPM | 17:48 |
| *** lbt has joined #harmattan | 17:49 | |
| *** lbt has joined #harmattan | 17:49 | |
| lardman | wb lbt | 17:49 |
| lbt | hey | 17:49 |
| Stskeeps | lardman: which one? | 17:49 |
| lardman | I'll repaste for lbt too | 17:49 |
| Stskeeps | rm_work: i've learnt to like rpm personally and i've done deb | 17:49 |
| lardman | the one question I do have is how different PPAs are prioritised when you install something - if I have an older version of a lib that I've pulled into my PPA from the original dev, what happens if they update and their PPA is also enabled on my device? | 17:50 |
| lbt | that's a zypper issue... | 17:50 |
| lbt | ah | 17:50 |
| lardman | ok | 17:50 |
| lbt | we're in #harmattan | 17:50 |
| lbt | not #meego | 17:50 |
| lardman | well package manager related anyway | 17:50 |
| lbt | so it's depends on pinning | 17:51 |
| lbt | otherwise they are peers (iirc) | 17:51 |
| lbt | http://wiki.meego.com/Community_Office/Task_Forces/MeeGo_Surrounds_and_Extras | 17:51 |
| lbt | FYI | 17:51 |
| khertan1 | hum ... i got segfault on a qml code. Someone have idea how to found the problem on device knowing that i'm doing the code part in python and it s not supported by qtcreator | 17:51 |
| lbt | http://mer-l-in.blogspot.com/2011/01/meego-community-development-apps.html | 17:51 |
| * lardman has a read | 17:52 | |
| lbt | also ... much of this is aimed at a 'grand vision' ... I'm not expecting that to be the first and last step on the journey | 17:53 |
| mgedmin | khertan1, gdb? | 17:54 |
| lardman | so what is the name of the harmattan obs thingie that one can build againsy? | 17:55 |
| kimju | lardman, http://forum.meego.com/showpost.php?p=25005&postcount=29 | 17:56 |
| lardman | oh ok | 17:56 |
| lardman | thanks | 17:57 |
| kimju | works for me, but the publishing of the repo took a while. | 17:57 |
| lardman | no that's fine, was just checking the wiki page and see it's been added there too | 17:58 |
| khertan1 | X Error: BadAccess (attempt to access private resource denied) 10 | 17:59 |
| khertan1 | Extension: 135 (MIT-SHM) | 17:59 |
| khertan1 | Minor opcode: 1 (X_ShmAttach) | 17:59 |
| khertan1 | Resource id: 0x3000002 | 17:59 |
| khertan1 | < any idea ? | 17:59 |
| khertan1 | oups ... too much line | 17:59 |
| mgedmin | tha'ts not a segfault | 18:00 |
| mgedmin | that's an X error | 18:00 |
| mgedmin | an interesting one | 18:01 |
| mgedmin | that I've no clue how to debug, sorry! | 18:01 |
| khertan1 | mgedmin: yeah sometime i got this error ... sometime segfault | 18:01 |
| mgedmin | interesting | 18:01 |
| mgedmin | fbreader segfaults when you minimize it | 18:01 |
| DocScrutinizer | if in doubt, blame aegis ;-P | 18:04 |
| khertan1 | lol | 18:04 |
| *** willer_ has joined #harmattan | 18:04 | |
| khertan1 | it s an aegis problem !!! there is the word "denied" and "private" | 18:04 |
| khertan1 | :) | 18:04 |
| Mek | bah, qtmobility's qml Map widget is so broken... it randomly doesn't render half the MapObjects I add to it, it randomly crashes on orientation changes... | 18:06 |
| mgedmin | life on the bleeding edge | 18:07 |
| frals | khertan1: what are you doing when it crashes? | 18:09 |
| gri | khertan: Did you try the MDeclarativeCache? Do you also have the impression it doesn't actually boost anything on the device? | 18:10 |
| khertan1 | view.setSource('qml3/main.qml') | 18:10 |
| mgedmin | neat | 18:11 |
| khertan1 | frals: view is a QDeclarativItem | 18:11 |
| lcuk | khertan1, does your app work on the netbook devices? | 18:11 |
| *** Clint has joined #harmattan | 18:11 | |
| frals | Khertan: item? not view? | 18:11 |
| lcuk | so that the meego ideapad or similar can do source editing | 18:11 |
| khertan1 | frals: view ... sorry :) | 18:13 |
| khertan1 | lcuk: for the moment it didn't work at all on harmattan :) | 18:13 |
| lcuk | lol! | 18:13 |
| khertan1 | got strange error | 18:13 |
| khertan1 | this is the problem i'm trying to resolve | 18:13 |
| frals | khertan1: does it work with a simple qml file like Rectangle { color: "green" }? | 18:14 |
| khertan1 | now i ve a more complete error : http://pastebin.com/HX2zMEP3 | 18:14 |
| khertan1 | frals: yep | 18:14 |
| khertan1 | and works in the qemu simulator | 18:14 |
| frals | khertan1: could you share your qml? :) | 18:15 |
| *** CaCO3 has quit IRC | 18:18 | |
| khertan1 | frals: yep of course | 18:24 |
| khertan1 | but i found the error | 18:24 |
| khertan1 | stupid ... mistake | 18:24 |
| khertan1 | view .... was ... garbage collected | 18:24 |
| khertan1 | :) | 18:24 |
| frals | khertan1: haha :) | 18:24 |
| mgedmin | whee! | 18:26 |
| *** M4rtinK2 has joined #harmattan | 18:26 | |
| lardman | catch you tomorrow everyone | 18:28 |
| *** lardman is now known as lardman|gone | 18:28 | |
| DocScrutinizer | lardman|gone: don't forget to DL chanlog! | 18:31 |
| DocScrutinizer | :-D | 18:31 |
| lardman|gone | docscrutinizer: lucky I saw the flashing icon there ;) | 18:31 |
| lardman|gone | will do! | 18:31 |
| *** CaCO3 has joined #harmattan | 18:31 | |
| kimju | hmm.. "apt-cache show mp-harmattan-rm680-pr" - this seems to be some metapackage, depending on all the system components belonging to this release (1.2011.22-6). but it seems to depend on specific versions of the components with =, not with the more relaxed >=. | 18:38 |
| kimju | I'd like to update xkb-data to my own version, containing "developer" hwkb-layouts (including at least |, ~ and ^ for english base and for finnish usage, a & o umlauts). I can replace the package with dpkg --purge --force-all & reinstall, but if I change its version number then the mp-harmattan-rm680-pr stays in broken state and apt keeps complaining. | 18:38 |
| mgedmin | kimju, consider using dpkg-divert instead | 18:39 |
| kimju | and reusing the original version number for the edited package would be wrong | 18:39 |
| mgedmin | that way you don't have to remove/upgrade xkb-data | 18:39 |
| mgedmin | instead you install a new package that nicely replaces some of the files originally belonging to xkb-data | 18:39 |
| *** achipa has quit IRC | 18:40 | |
| kimju | I'm not sure if aegis allows me to do that. | 18:40 |
| mgedmin | do what, exactly? | 18:40 |
| kimju | divert a file included in an official package. | 18:41 |
| mgedmin | hmm | 18:41 |
| kimju | with a package from non-trusted source. | 18:41 |
| DocScrutinizer | anyway the issue is well known from fremantle CSSU | 18:41 |
| mgedmin | I still think it's worth a try | 18:41 |
| kimju | in this specific case I think it's possible to also create an additional file containing the new layouts. | 18:42 |
| mgedmin | aegis didn't forbid me from editing files belonging to the meego-keyboard package | 18:42 |
| mgedmin | I'm not that familiar with aegis, but I get the impression that it protects binaries more than it protects other stuff | 18:43 |
| DocScrutinizer | kimju: previously on fremantle the suggested solution been to uninstall the metapackage all together | 18:43 |
| kimju | will that break upcoming SSUs? | 18:44 |
| DocScrutinizer | which also has undesirable side effects | 18:44 |
| DocScrutinizer | yes | 18:44 |
| mgedmin | reading http://dz015.wordpress.com/2011/07/05/the-good-the-bad-and-the-ugly-or-python-security-fw-and-harmattan/ | 18:44 |
| mgedmin | I would be cautious about dpkg-divert'ing a binary | 18:44 |
| mgedmin | well, an executable file | 18:45 |
| mgedmin | (one that requests extra security tokens) | 18:45 |
| mgedmin | but data files? should be safe | 18:45 |
| DocScrutinizer | >>>After installing the package, do NOT modify the installed files if they request a token. Security FW will discover an unexpected change in the file and lock the device (ops! reflash). <<< HAHA | 18:49 |
| DocScrutinizer | WTF aegis? | 18:49 |
| kimju | can you point to an example of simple data package diverting a file? | 18:49 |
| kimju | I'll try if that solution works. | 18:50 |
| mgedmin | kimju, http://gedmin.as/debs/mg-keymap_0.9.5-1.tar.gz | 18:50 |
| mgedmin | I used it a long time ago for exactly that purpose: diverting one of keymap files | 18:51 |
| mgedmin | or was it Compose? | 18:51 |
| mgedmin | anyway, x.org reshuffled all their xkb files and my package no longer installs cleanly | 18:51 |
| mgedmin | but the principles are still valid | 18:51 |
| kimju | ok, thanks.. lets see what happens.. | 18:53 |
| mgedmin | I wonder if debhelper has a way to do it more easily than manually writing preinst and postrm scripts | 18:53 |
| DocScrutinizer | wouldn't you think that A) aegis should forbid editing rather than getting upset after editing happened? and B) wasn't it *absolutely enough* to not allow to execute the "binary" after it got edited - WTF has aegis to "lock the device"? | 18:53 |
| mgedmin | DocScrutinizer, I completely agree | 18:53 |
| DocScrutinizer | as I see it this is a mayor SECURITY FLAW of aegis, as any simple rogue malware could make your device lock and destroy all your system plus probably all your data | 18:55 |
| kimju | easy DoS | 18:55 |
| DocScrutinizer | how's that going to protect the user? in *any* way? from whatever? | 18:56 |
| mgedmin | it's supposed to protect MAFIAA from users "hacking" their devices, I think | 18:57 |
| DocScrutinizer | of course | 18:57 |
| DocScrutinizer | but that's really too much | 18:58 |
| *** achipa has joined #harmattan | 18:58 | |
| *** w00t has left #harmattan | 18:59 | |
| cpscotti | oh my.. the pinch zoom in the task switcher is SO beautiful and so useless at the same time.. can't stop doing it.. :D | 18:59 |
| DocScrutinizer | see, I can edit the file anyway, when mounting the fs to another aegis-free system. So why needs the device to lock, rather than just forbid execution of the edited file | 18:59 |
| mgedmin | oops, accidentally hit 'close all' in task switcher | 19:00 |
| mgedmin | thought it was 'cancel' | 19:00 |
| DocScrutinizer | I'd maybe even accept a instant reboot, but locking is *EVIL* | 19:00 |
| cpscotti | mgedmin, haha.. yep.. There's no explicit cancel.. and "done" is a bad name. | 19:01 |
| DocScrutinizer | no other purpose than punishing devels | 19:01 |
| DocScrutinizer | eh? | 19:01 |
| cpscotti | mgedmin, you can click outside of all tasks to cancel it too ... but I think it should cancel if you clicked in one specific task | 19:01 |
| mgedmin | I half-hoped I could cancel and switch to a task by double-tapping | 19:02 |
| cpscotti | that would be cool actually | 19:02 |
| DocScrutinizer | mgedmin: ?? | 19:03 |
| mgedmin | DocScrutinizer, what? | 19:03 |
| DocScrutinizer | I do't get the whole purpose/idea | 19:03 |
| mgedmin | neither do I | 19:04 |
| mgedmin | I'm still half-hoping (again) that maybe this "device is locked" thing is a bit of an exaggeration | 19:04 |
| DocScrutinizer | "done" cancels kill-mode as does taping on background. Killing multiple individual apps in kill-mode is ok to me | 19:04 |
| mgedmin | and that maybe it just doesn't let you launch the app, and "suggests" a reflash in the sense that it's a well-defined way for a non-technical user to return to a known good state? | 19:04 |
| mgedmin | I'm not curious enough to try and lock my device | 19:05 |
| mgedmin | oh, we're talking about _task switching_ now | 19:05 |
| mgedmin | sorry! | 19:05 |
| mgedmin | :) | 19:05 |
| DocScrutinizer | mgedmin: definitely device-locked means you have to reflash (unless it recovers on 3rd boot as it once did for javispedro) | 19:05 |
| DocScrutinizer | but only once | 19:06 |
| DocScrutinizer | mgedmin: http://forum.meego.com/showthread.php?p=25529#post25529 | 19:07 |
| DocScrutinizer | (I dang should find out how to resize that image :-S ) | 19:08 |
| mgedmin | scary | 19:09 |
| *** smoku has quit IRC | 19:19 | |
| khertan1 | WUHOUHOUWHOWIWHOU !!! Basic feature of KhtEditor QML edition is now working ... !!! syntax highlighting, open, save, ... | 19:21 |
| mgedmin | congrats! | 19:21 |
| khertan1 | good evening everyone time to go back home | 19:22 |
| khertan1 | bye | 19:22 |
| *** khertan1 has left #harmattan | 19:22 | |
| *** seif has joined #harmattan | 19:23 | |
| *** spenap has quit IRC | 19:24 | |
| kimju | mgedmin, at least local (manual) diversion works, even with reinstalling the official package. next step: to package the edited symbols file. | 19:28 |
| mgedmin | cool! | 19:28 |
| *** DocScrutinizer is now known as joerg_rw | 19:30 | |
| *** joerg_rw is now known as DocScrutinizer | 19:31 | |
| *** aleksander_m has quit IRC | 19:31 | |
| rm_work | hey DocScrutinizer | 19:35 |
| rm_work | DocScrutinizer: can you check out http://ageofikon.com/prh/ and tell me if it is generating .install files correctly? | 19:36 |
| rm_work | and/or, if it is actually listing all the packages | 19:36 |
| DocScrutinizer | I don't know if it's listing all pkgs ;-) | 19:36 |
| rm_work | lol | 19:36 |
| rm_work | well | 19:36 |
| DocScrutinizer | but I'll check anyway | 19:36 |
| rm_work | yeah, try :P | 19:36 |
| rm_work | just a first review | 19:37 |
| rm_work | i have no idea if those .install files will be right at all | 19:37 |
| DocScrutinizer | what OS version shall I select? | 19:38 |
| rm_work | err | 19:38 |
| rm_work | ./ ? | 19:38 |
| rm_work | http://ageofikon.com/prh/ | 19:38 |
| rm_work | new link | 19:38 |
| rm_work | not to the old packrat | 19:38 |
| rm_work | i just copied the whole thing and started hacking at it | 19:39 |
| DocScrutinizer | duh same window title, so I got tricked into closing new window and doing F5 on old one | 19:39 |
| rm_work | for example search for "cutesoma" | 19:42 |
| DocScrutinizer | "Download failed \n Couldn't download that file. Type application/x-install-instructions not supported" | 19:43 |
| rm_work | at what point | 19:43 |
| rm_work | trying to download the .install? | 19:43 |
| DocScrutinizer | at clicking on the download icon | 19:43 |
| rm_work | or, after the package manager opened the .install? | 19:43 |
| rm_work | erm | 19:43 |
| rm_work | oh | 19:43 |
| rm_work | that sounds like a browser issue | 19:43 |
| DocScrutinizer | yup | 19:43 |
| rm_work | the browser is blocking (or doesn't know about) that mimetype | 19:44 |
| DocScrutinizer | yup | 19:44 |
| rm_work | well can you look at an install file on your desktop or something? | 19:44 |
| rm_work | do you know what they are supposed to look like? | 19:44 |
| tomma | i think it contains repo and package to install | 19:44 |
| rm_work | does ANYONE here know how to verify the correctness of a .install? | 19:44 |
| rm_work | yeah but | 19:44 |
| DocScrutinizer | I doubt I ever seen an install file for harmattan | 19:44 |
| rm_work | hrm, so does harmattan even support .install? | 19:44 |
| rm_work | how would we install? | 19:45 |
| rm_work | there is a deb link | 19:45 |
| tomma | i dont think harmattans current manager supports it? | 19:45 |
| rm_work | if you notice | 19:45 |
| rm_work | if you expand the package there is a "Download: <versionnumber>" | 19:45 |
| rm_work | where that is a link to the deb | 19:45 |
| DocScrutinizer | you can install plain .deb right away though | 19:45 |
| rm_work | so | 19:45 |
| rm_work | how should i handle that | 19:45 |
| rm_work | should i replace the download icon with the deb? | 19:45 |
| rm_work | that'd be a dependency nightmare if there's no way to add the repo nicely tho | 19:46 |
| DocScrutinizer | sounds kinda resonable | 19:46 |
| DocScrutinizer | I guess aegis simply doesn't allow adding repos | 19:46 |
| rm_work | unfortunately my phone has zero connectivity here | 19:46 |
| gri | Are those "X Error: BadMatch" and BadDamage messages normal when creating a QDeclarativeView? (in scratchbox and on device) | 19:46 |
| rm_work | so i can't test anything | 19:46 |
| rm_work | in general though, does this help you? | 19:47 |
| DocScrutinizer | looks very nice, yes | 19:47 |
| rm_work | k i will have to clean it up a lot more | 19:47 |
| DocScrutinizer | even found mc :-D | 19:47 |
| rm_work | but usable for now i guess | 19:47 |
| rm_work | you want to put it on a wiki somewhere or something? | 19:47 |
| rm_work | i have to go to a meeting in a moment | 19:48 |
| DocScrutinizer | I guess it should go to wiki for sure | 19:48 |
| DocScrutinizer | no idea where to though | 19:48 |
| rm_work | yeah me either | 19:48 |
| rm_work | right now it is a manual update by my running the perl script | 19:48 |
| rm_work | but i will put it on cron in a bit | 19:48 |
| DocScrutinizer | let's postpone the wiki thing for later when it really works | 19:48 |
| rm_work | k | 19:48 |
| rm_work | well | 19:48 |
| rm_work | i would like people to test it | 19:48 |
| rm_work | and tell me what they want / what is broken | 19:48 |
| DocScrutinizer | you'll reach a lot of those that actually can test here in this chan | 19:49 |
| rm_work | also, keep an eye open for a way to make .install (or similar proceedure, automatic adding of a repo) work | 19:49 |
| DocScrutinizer | :-) | 19:49 |
| rm_work | yeah, truth :P | 19:49 |
| rm_work | back in an hour or so | 19:50 |
| DocScrutinizer | also afk for dinner etc | 19:50 |
| rm_work | what does fms go by in IRC? | 19:56 |
| rm_work | would be easier if i could ask him some stuff | 19:56 |
| rm_work | ack, right, meeting, bbl | 19:56 |
| *** rm_you_N950 has joined #harmattan | 20:22 | |
| rm_you_N950 | well, webirc works | 20:23 |
| rm_you_N950 | kinda | 20:23 |
| *** ChanServ sets mode: +o DocScrutinizer | 20:31 | |
| *** rm_you_N950 has quit IRC | 20:32 | |
| *** wazd has joined #harmattan | 20:34 | |
| *** DocScrutinizer changes topic to "A cozy little place for pure harmattan device and development discussions | Developers check this: http://library.developer.nokia.com/ | http://wiki.meego.com/N950_landing_page | MeeGo N9(|50) CE on its way, MOSLO still missing, discuss in #meego-arm please | Chanlogs: See ~logs, http://mg.pov.lt/harmattan-irclog/" | 20:34 | |
| alterego | I really miss having a dedicated camera button. | 20:37 |
| alterego | but the camera is so quick, kinda makes up for it I suppose. | 20:38 |
| *** Scifig has joined #harmattan | 20:38 | |
| wazd | http://zhil.in/genesis.svg <- PicoDrive Harmattan icon, anyone? :) | 20:39 |
| *** ChanServ sets mode: -o DocScrutinizer | 20:39 | |
| DocScrutinizer | alterego: notice however flash is by far not as quick as camera | 20:40 |
| alterego | forgivable :p | 20:41 |
| DocScrutinizer | alterego: and finally you found the camera yes it is quick, but until you *get there* - hah, I beat that with N900 hands down | 20:41 |
| alterego | I keep it open in task switcher now. | 20:41 |
| alterego | bbiab | 20:42 |
| wazd | awww, nobody's interested in harmattan icons *turns on sad violine* :( | 20:43 |
| *** Scifig has quit IRC | 20:48 | |
| DocScrutinizer | wazd: your selling is no good | 20:50 |
| DocScrutinizer | wazd: nice icon - what's been the question? | 20:51 |
| DocScrutinizer | maybe my problem is I got no clue what's PicoDrive | 20:51 |
| tomma | DocScrutinizer, i would say that getting to camera is faster as camera application starts a lot faster than in n900 | 20:52 |
| DocScrutinizer | tomma: maybe, when you actually *know* all the swipes and scrolls and whatnot you have to do to unlock, and get to camera | 20:53 |
| tomma | there is this shortcut even from lock screen | 20:53 |
| DocScrutinizer | tomma: on N900 I open the cam lens slider - done | 20:53 |
| DocScrutinizer | wut? from lockscreen? | 20:53 |
| DocScrutinizer | indeed, never occured to me | 20:54 |
| npm | yay... i got video playback working .... http://nielsmayer.com/meego/qml/qmltube_1_0_6_armel.deb | 20:55 |
| npm | it *was* aegis afterall | 20:55 |
| npm | added <credential name="GRP::video" /> to qmltube.aegis and now it works | 20:56 |
| tomma | for video playback? | 20:56 |
| npm | YES! | 20:56 |
| DocScrutinizer | toldya, when in doubt, blame aegis | 20:56 |
| tomma | =) and x11vnc doesn't need anything to access framebuffer | 20:57 |
| npm | my suspicions were right, despite djzapi saying otherwise | 20:57 |
| npm | anyways, please check out my port of cutetube-qml | 20:58 |
| rm_work | npm: instructions to install? | 20:59 |
| tomma | is this debug build? | 20:59 |
| tomma | rm_work, dpkg -i <package> | 20:59 |
| rm_work | lol yes well | 21:00 |
| rm_work | was wondering about anything besides just installing the deb | 21:00 |
| rm_work | does aegis complain, etc | 21:00 |
| tomma | it does not if there is working manifest | 21:00 |
| rm_work | also, someone please look at http://ageofikon.com/prh/ | 21:00 |
| npm | tomma: no it's a regular build | 21:01 |
| rm_work | i want an mplayer build | 21:01 |
| wazd | DocScrutinizer: well, it's Sega Genesis Emulator for maemo :) | 21:01 |
| npm | rm_work "wget http://nielsmayer.com/meego/qml/qmltube_1_0_6_armel.deb ; dpkg -i qmltube_1_0_6_armel.deb" | 21:02 |
| tomma | npm, you have whole repository packed there? | 21:02 |
| npm | http://wiki.meego.com/Tubelet-and-cutetube-port for details | 21:02 |
| leinir | go to site in browser, click on deb, download, tap deb once completed, say "yes, i don't care it's not coming from Nokia", wait for thing to finish installing... ;) | 21:03 |
| npm | tomma: what do you mean? | 21:03 |
| tomma | atleast a lot of .svn folders =) | 21:03 |
| npm | oh yeah there's a weird bug | 21:03 |
| npm | which i have in both my packages | 21:03 |
| npm | qmlcreator seems to be including my .svn dirs | 21:03 |
| npm | they shouldn't be there/ | 21:03 |
| *** mzanetti has joined #harmattan | 21:04 | |
| npm | (ignore for now, yes it makes package bigger) | 21:04 |
| npm | but if you have suggestions on how to make DEPLOYMENTFOLDERS in http://ytd-meego.googlecode.com/svn/trunk/playground/qmltube/qmltube.pro not include my .svn dirs, i'd love to hear about it | 21:06 |
| npm | note i haven't checked in the latest changes that built the above .deb | 21:06 |
| npm | since i just got it built w/ video working/ | 21:06 |
| tomma | aah... never used that template =) | 21:06 |
| tomma | you can always put qml files to resources | 21:07 |
| tomma | but nice looking app | 21:07 |
| rm_work | ok have to walk outside to get network, will be right back (trying to get a video to play) | 21:07 |
| npm | it's not mine | 21:07 |
| npm | i actually don't like the UI. i just ported it to linux/meego and harmattan | 21:07 |
| npm | my goal is to use some of the infrastructure that works (like the video playing stuff and the OpenGL additions that make it work on harmattan) | 21:08 |
| npm | since the original author also has an n950 i'll let him do his thing and support his own prog | 21:09 |
| tomma | =) | 21:09 |
| DocScrutinizer | wazd: now I know why you are the graphics expert of maemo community :-D | 21:09 |
| *** cpscotti has quit IRC | 21:11 | |
| rm_work | wazd: that icon looks very strange O_o but i do not know what it is referencing | 21:11 |
| DocScrutinizer | s/graphics/design/. | 21:11 |
| DocScrutinizer | I think it looks unique and kinda nice, with a very own appeal | 21:11 |
| wazd | rm_work: http://www.axess.com/twilight/console/detail/genesisb.jpg | 21:12 |
| DocScrutinizer | and I guess if I'd know PicoDrive I'd probably recognize the icon immediately | 21:12 |
| npm | i have been sending patches, which have been dutifully ignored ( https://garage.maemo.org/tracker/index.php?func=detail&aid=6932 ) | 21:12 |
| wazd | DocScrutinizer: well, it's the whole series of icons for maemo/meego emulators | 21:12 |
| wazd | DocScrutinizer: like http://zhil.in/nes.svg <- NES, http://zhil.in/snes2.svg <- SNES and so on | 21:13 |
| DocScrutinizer | wazd: HAH the 6 buttons | 21:13 |
| npm | by the way, how do you force the virtual keyboard to pop-down? force focus on some other widget? | 21:13 |
| DocScrutinizer | obvious | 21:13 |
| wazd | DocScrutinizer: well, I agree it looks strange for those who absolutely don't know what it is :) | 21:14 |
| DocScrutinizer | I didn't say that | 21:14 |
| DocScrutinizer | actually the very first one had quite some appeal to me | 21:15 |
| npm | by the way, what happens if you change a QML file on an installed program, will aegis lock it up, or only if you change the binary? | 21:15 |
| wazd | DocScrutinizer: cool :) | 21:15 |
| npm | or does it also automatically checksum the QML files | 21:15 |
| DocScrutinizer | npm: up to anybody's guess | 21:18 |
| DocScrutinizer | npm: I'd say test it and prepare for reflash | 21:18 |
| DocScrutinizer | oh wait, we got NO proper backup tool :-/ | 21:19 |
| DocScrutinizer | so you better script all your setup and customization | 21:20 |
| DocScrutinizer | oh wait, we don't know how to do any of that either | 21:20 |
| DocScrutinizer | maybe modulo the wget X && dpkg -i X bits | 21:21 |
| DocScrutinizer | QED: developers are wasting 80% of their time fighting aegis | 21:22 |
| *** seif has quit IRC | 21:22 | |
| DocScrutinizer | while not really needing any of the "added security" it offers | 21:22 |
| *** M4rtinK2 has quit IRC | 21:23 | |
| npm | i'm not wasting my time fighting aegis. i just spent a little time learning about it and its a good and necessary idea for mobile handsets | 21:23 |
| DocScrutinizer | hell there's not even a single commercial app on our harmattan that would "need" aegis, like yoghurtsport or angry birds or skype or... | 21:23 |
| DocScrutinizer | npm: ooh, also for developer devices? | 21:24 |
| npm | well nobody is going to put a commercial app involving money, secure comms, etc on the device w/o it | 21:24 |
| npm | and developers need to get used to working with it | 21:25 |
| DocScrutinizer | :nod: that's why we got no angry birds, no skype, no joikuspot on N900 - it had no fscknf aegis | 21:25 |
| * mzanetti used skype a lot on the N900 | 21:25 | |
| npm | well it took me one day to figure out how to write an aegis file that got video working... there's probably an aegis confug that progs lneed | 21:26 |
| DocScrutinizer | and yes, I need to errrr WUT? I give a f* about aegis until my app works as it should, then I will test it with aegis | 21:26 |
| npm | why wouldn't skype from maemo be portable to harmattan? | 21:27 |
| Mek | there is skype for harmattan, just not on these developer devices | 21:27 |
| npm | oh. cool. where? | 21:28 |
| DocScrutinizer | not for you, fool | 21:28 |
| DocScrutinizer | for the trustworthy developer only! | 21:28 |
| npm | ? | 21:28 |
| DocScrutinizer | "we" are BOT trustworthy | 21:28 |
| DocScrutinizer | NOT* | 21:28 |
| Mek | yeah, nokia-internal/subcontractors only at the moment.. but it will be on the actual released devices | 21:28 |
| *** smoku has joined #harmattan | 21:29 | |
| DocScrutinizer | so our devices *can not* run angry birds or skype | 21:29 |
| DocScrutinizer | as well as can not disable aegis | 21:29 |
| DocScrutinizer | which is kinda paradox | 21:29 |
| Mek | the device can run it, you just 'don't have access to the packages... | 21:29 |
| kimju | remember that this is still pre-release software.. | 21:29 |
| GeneralAntilles | I just want GTalk. | 21:30 |
| kimju | those packages might come available with the real release.. | 21:30 |
| DocScrutinizer | I heard the image won't flash | 21:30 |
| npm | well we're supposed to be developing apps :-) and probably expected to buy an n9 for all the product stuff | 21:30 |
| Mek | "the image"? | 21:30 |
| hiemanshu | we should have all the apps on the actual release, or I am just going to send the N950 back, and duct-tape a BT KB to the N9 | 21:30 |
| DocScrutinizer | the image that provides the commercial apps to internals | 21:30 |
| npm | by the way, would it be a violation of my agreement to enable twitter capabilities? | 21:30 |
| hiemanshu | npm: until you dont speak about it, who will know? :P | 21:31 |
| Mek | there are about 50 or so images made for various configurations each night, and among those since recently are also images for non-R&D n950s | 21:31 |
| Mek | and otherwise still just apt-get installing angry-birds and skype was no problem :) | 21:32 |
| DocScrutinizer | hmm, ok then | 21:32 |
| npm | i think it's just missing some keys, which i can steal from other distros per https://bugs.meego.com/show_bug.cgi?id=14852 | 21:32 |
| MeeGoBot | Bug 14852 maj, High, ---, robert.bradford, VERI FIXED, [REG]Facebook option is blank in status panel | 21:32 |
| npm | (which is what i do for regular meego netbook tablet too) | 21:33 |
| DocScrutinizer | still I wonder why aegis can't get disabled on CE devices, while there's evidently nothing on those devices that needs this kind of restrictive hard protection | 21:33 |
| npm | http://lists.meego.com/pipermail/meego-dev/2011-April/482408.html http://lists.meego.com/pipermail/meego-dev/2011-April/482414.html http://lists.meego.com/pipermail/meego-dev/2011-April/482416.html | 21:34 |
| hiemanshu | I understand the need to have aegis on the device, so it would work on the N9 too with it enabled, but dev mode should have an option to disable aegis, so you can test it without having to create a new package everytime | 21:34 |
| DocScrutinizer | hiemanshu: exactly | 21:35 |
| Mek | yes, i agree | 21:36 |
| hiemanshu | I would prefer not wasting time porting opencv on the device, creating a aegis manifest, creating a package installing it, everytime I fix something in the code | 21:36 |
| npm | isn't there a "developer mode" | 21:36 |
| DocScrutinizer | all his aegis "protection" is just a nuisance for normal developing workflow, and the aegis-selfdistruct is the obsolute WTF of the year | 21:36 |
| npm | or is that just "develsh" | 21:37 |
| hiemanshu | npm: its just to enable develsh, and devel-su | 21:37 |
| npm | oh so that's how to become root. i've been sshing in as root :-) | 21:38 |
| hiemanshu | lol | 21:38 |
| npm | well its only my 3rd day w/ device | 21:38 |
| npm | and i got 1.5 programs running | 21:39 |
| npm | and yes i'm still battling aegis on the other. | 21:39 |
| npm | i'll start complaining by DocScrutinizer if i cant fix it after a while :-) | 21:39 |
| npm | s/by/like | 21:39 |
| DocScrutinizer | I'm taking bets on how many apps will be available for N9 that were not available for N900 for the only reason N900 had no "protection" via aegis. I bet that'll be exactly 0.0 | 21:40 |
| hiemanshu | DocScrutinizer: naah, it will be -100 | 21:41 |
| DocScrutinizer | more like it | 21:41 |
| hiemanshu | most people wont port their apps because of the forced QML move and/or aegis | 21:41 |
| npm | how forced is the qml move... i haven't taken a survey of the libraries. whoah where's libgtk+ etc? | 21:43 |
| *** smoku has left #harmattan | 21:43 | |
| *** smoku has quit IRC | 21:43 | |
| DocScrutinizer | haha | 21:43 |
| npm | but doesn't "meego compliance" require it? | 21:43 |
| hiemanshu | naah | 21:44 |
| npm | i thought the claim was that harmattan was "api compatible" or somesuch w/ meego | 21:44 |
| *** smoku has joined #harmattan | 21:44 | |
| DocScrutinizer | I'd think if gtk was available, probably somebody already had ported xchat and I *might* actually use the device | 21:44 |
| DocScrutinizer | yeah QML API compatible | 21:45 |
| hiemanshu | DocScrutinizer: I have a ugly-ish version of quassel2go running :P | 21:45 |
| DocScrutinizer | and QtMobiliy | 21:45 |
| DocScrutinizer | ooh, with meego. Sorry no idea. Thought you said "with maemo" | 21:46 |
| npm | well it is "MeeGo 1.2 Harmattan" | 21:47 |
| npm | maybe file a bug report? | 21:48 |
| wazd | somebody's already porting QuitIM | 21:49 |
| wazd | qutim* | 21:49 |
| wazd | So I guess we'll have multi-protocol chat after all :) | 21:50 |
| RST38h | no way to write plugins for the existing harmattan chat? | 21:52 |
| frals | its telepathy so i assume existing ones should more or less work | 21:52 |
| gri | I don't think the accounts api is compatible to setup an account for the telepathy plugin | 22:00 |
| *** wazd has quit IRC | 22:01 | |
| GeneralAntilles | Silly | 22:03 |
| GeneralAntilles | Community support is one of the biggest strengths of the platform, I don't get crippled plugin interfaces. | 22:03 |
| RST38h | General, have you got yours finally? | 22:03 |
| GeneralAntilles | Got it Monday afternoon. | 22:04 |
| RST38h | ah | 22:04 |
| GeneralAntilles | Wishing for better Components docs (or some well-commented sample code that actually works) and a software update. . . . | 22:05 |
| tomma | use those components .qml as reference | 22:10 |
| *** CaCO3 has quit IRC | 22:12 | |
| GeneralAntilles | Trying to get a ButtonRow in my toolbar (for page tabs) to not either explode the toolbar or give me a black screen. | 22:13 |
| frals | GeneralAntilles: error messages and code? | 22:14 |
| fiferboy | GeneralAntilles: Anchoring left and right on the toolbar worked for me | 22:14 |
| GeneralAntilles | frals, don't currently have my head in the code. | 22:14 |
| frals | GeneralAntilles: ok | 22:14 |
| GeneralAntilles | Gonna try again tonight and upload it if I can't figure it out. | 22:14 |
| frals | i do a fair bit of qml at work so might be able to provide some help | 22:15 |
| *** seif has joined #harmattan | 22:17 | |
| gri | nice qt-components bugs I discovered today: there is no Separator component; the undocumented MoreIndicator does not support theme inversion; the ScrollDecorator is invisible when the page has inverted theme; ListDelegate's selection ignores theme inversion; there is no basic Header element for showing a icon, title and subtitle | 22:18 |
| gri | Where should I file this? meego? Nokia? nowhere? | 22:18 |
| fiferboy | Ooo, header with icon, title, and subtitle would be useful | 22:19 |
| gri | I did it myself to have the same look as other UIs in accounts-ui | 22:20 |
| achipa | gri: bugreports.qt.nokia.com | 22:20 |
| *** cpscotti has joined #harmattan | 22:22 | |
| *** seif has quit IRC | 22:22 | |
| *** aleksander_m has joined #harmattan | 22:26 | |
| *** gri_ has joined #harmattan | 22:32 | |
| *** mzanetti has quit IRC | 22:34 | |
| *** gri has quit IRC | 22:35 | |
| *** aleksander_m has quit IRC | 22:37 | |
| *** mzanetti has joined #harmattan | 22:44 | |
| *** mzanetti_ has joined #harmattan | 22:57 | |
| *** mzanetti has quit IRC | 22:58 | |
| *** mzanetti_ is now known as mzanetti | 22:58 | |
| *** javispedro has joined #harmattan | 22:59 | |
| *** javispedro has quit IRC | 22:59 | |
| *** javispedro has joined #harmattan | 22:59 | |
| javispedro | moo. | 23:00 |
| Stskeeps | moo | 23:00 |
| DocScrutinizer | moo | 23:02 |
| infobot | baah | 23:03 |
| javispedro | infobot: genius. | 23:03 |
| infobot | An arbitrary precision integer and multiple precision floatingpoint calculator. URL: http://www.5z.com/jirka/genius.html | 23:03 |
| *** crevetor has joined #harmattan | 23:04 | |
| DocScrutinizer | interesting :-D | 23:04 |
| crevetor | alterego: it's that time of the day again. Any ideas ? | 23:04 |
| javispedro | btw. | 23:05 |
| javispedro | Anyone did custom "account types" in Fremantle? | 23:05 |
| javispedro | (that is MSN, etc.) | 23:05 |
| javispedro | I wonder how much different it is from Harmattan. | 23:05 |
| javispedro | cause I believe final device won't have XMPP support. At all. Only GTalk. | 23:05 |
| crevetor | what do you mean "did" ? | 23:06 |
| javispedro | developed | 23:07 |
| javispedro | Account stuff "should" be open, and we have the xmpp telepathy backend installed, so it should not be a hard task. | 23:11 |
| javispedro | (and in the meanwhile we could use such a plugin for gtalk access ( ;) ) ) | 23:12 |
| *** meegoexperts_mob has joined #harmattan | 23:12 | |
| DocScrutinizer | wow that last post almost broke my lexem analyzer / parser | 23:13 |
| javispedro | too shallow stack? | 23:15 |
| *** rlinfati has joined #harmattan | 23:16 | |
| rlinfati | Hi, some one can see for me the bug QDesktopServices::openUrl(url) not working in Maemo6 (https://bugreports.qt.nokia.com/browse/QTBUG-9629) on https://projects.maemo.org/bugzilla/show_bug.cgi?id=163554 ?? | 23:17 |
| rlinfati | i have the same problem :( | 23:18 |
| DocScrutinizer | javispedro: more like local context leaking to outer frame on stack | 23:19 |
| DocScrutinizer | some implicit escaping kicked in on ;) and was still active on ) | 23:21 |
| DocScrutinizer | which somewhat caused the emoticon regex to act greedy | 23:22 |
| javispedro | there, backlogs read. | 23:29 |
| * javispedro _loves_ logs. | 23:29 | |
| *** _NIN has joined #harmattan | 23:36 | |
| *** NIN101 has quit IRC | 23:38 | |
| *** fiferboy has quit IRC | 23:39 | |
| GeneralAntilles | I hate reading backlog. | 23:43 |
| DocScrutinizer | I hate backlog building up while I read backlog | 23:45 |
| GeneralAntilles | Yeah, that too. | 23:48 |
| *** Scifig has joined #harmattan | 23:51 | |
Generated by irclog2html.py 2.15.1 by Marius Gedminas - find it at mg.pov.lt!
