**** BEGIN LOGGING AT Mon Jan 10 02:59:58 2011 Jan 10 03:35:05 mickeyl/Heinervdm: Okay, flashed the firmware but still seeing the error 512 in the log: http://pastebin.com/xWgqHH4S Jan 10 06:50:44 GNUtoo|laptop, pespin: heh, you guys should sleep more at night... and less now :P Jan 10 06:52:32 GNUtoo|laptop: I pulled current kernel trees yesterday... did not build yet though Jan 10 06:52:36 hope to do that today Jan 10 06:52:44 if daywork permits Jan 10 07:53:23 moin Jan 10 07:53:30 mrmoku: you're also using bitbake master? Jan 10 07:53:48 mrmoku: bitbake-1.10 shouldn't be so bad with branch changes Jan 10 07:54:16 mrmoku: only bitbake-master has now branch as part of persistence db key.. so LOCALCOUNT is reset to 0 with every branch change Jan 10 07:59:34 JaMa: hmm... let me check Jan 10 08:00:18 JaMa: using bitbake of you chroot Jan 10 08:00:50 which says 1.11.0 Jan 10 08:01:02 hmm strange, if you didn't upgrade it since downloading chroot Jan 10 08:01:18 it's from bitbake master but some old rev Jan 10 08:01:35 I had the problem that revs for projects that were on gdbus branch (phonefsod for example) have been sticky Jan 10 08:02:47 it continued to build the tip of gdbus though it should have built master Jan 10 08:04:43 mickey|zzZZzz, OutputStream.write_async doesn't use a length parameter anymore. i think the .length parameter sets the io_priority parameter. Jan 10 08:07:36 freesmartphone.org: 03mok 07cornucopia * rfa6fb20a3cbe 10/fsodeviced/conf/openmoko_gta/fsodeviced.conf: Jan 10 08:07:36 freesmartphone.org: fsodeviced: disable wifi + bluetooth in openmoko_powercontrol for gta02 by default Jan 10 08:07:36 freesmartphone.org: Signed-off-by: Klaus Kurzmann Jan 10 08:07:57 JaMa: so, fsodeviced should be fixed too now Jan 10 08:08:08 pending the srcrev bump for that commit Jan 10 08:08:23 looks like we're getting closer to sync :-) Jan 10 08:08:52 so should I wait for another cornucopia bump before build? :) Jan 10 08:09:15 * mrmoku bumps Jan 10 08:09:22 * JaMa waits Jan 10 08:10:51 bumped Jan 10 08:23:52 building Jan 10 08:24:14 and recipe for 2.6.37 pushed Jan 10 08:25:33 nice Jan 10 08:25:43 very curious if that will fix my WS :) Jan 10 08:31:51 probably not. larsc patch look like mine Jan 10 08:32:17 but if it will, it would be interesting... Jan 10 08:33:28 gena2x: don't destroy my illusions ;) Jan 10 08:33:42 will tell you when tested Jan 10 08:35:24 it is question what is better do not have illusions and get nice surprise then something unexpectdly works, or have illusions destroyed... but yes, please tell :) Jan 10 08:35:40 :P Jan 10 08:36:26 mrmoku: I'll build it in shr-kms again Jan 10 08:36:53 mrmoku: it boots ok, but I had problem with xorg ts and no time to debug why.. Jan 10 08:37:14 ok Jan 10 08:38:18 I will be busy with looking at n900 kernel and fixing last gdbus fallouts anyway Jan 10 08:39:56 * JaMa has new gf and now with 11y son.. so less time for shr and mostly away from home... Jan 10 08:40:31 heh Jan 10 08:40:40 and again JaMa is too fast for me :P Jan 10 08:40:45 my oldest sun has 10y :) Jan 10 08:42:38 hehe Jan 10 08:48:21 *yawn* Jan 10 08:51:34 mrmoku: missing fso-specs bump? | Requested 'fso-specs >= 2011.01.09.1' but version of fso-specs is 2010.12.13.1 Jan 10 08:53:51 moin mickey|office Jan 10 08:54:19 mrmoku: bumping Jan 10 08:55:26 JaMa: oh, forgot that one Jan 10 08:56:04 mickey|office, OutputStream.write_async doesn't use a length parameter anymore. i think the .length parameter sets the io_priority parameter. Jan 10 08:58:18 oh Jan 10 08:58:32 then my last commit in libfsobasics needs fixing Jan 10 08:59:12 yep. i don't want to push it because it depends on the vala version Jan 10 08:59:35 but only fsobasics/utitlities.vala is affected Jan 10 09:01:03 with which version did they change that? Jan 10 09:05:29 commit 02abf1661fd5fcec9600ca2d1726c660b8ea59a9 Jan 10 09:05:29 Author: Evan Nemerson Jan 10 09:05:30 Date: Sat Oct 23 00:57:03 2010 -0700 Jan 10 09:05:30 gio-2.0: use uint8[] for buffers instead of void* Jan 10 09:09:06 ah, quite a while ago Jan 10 09:58:40 freesmartphone.org: 03morphis 07next * ra8e8b8f86c21 10msmcomm/libmsmcomm-next/ (11 files in 5 dirs): libmsmcomm-next: add first phonebook subsystem messages Jan 10 09:58:41 freesmartphone.org: 03morphis 07next * r6b99aa32172c 10msmcomm/msmcomm-specs/src/ (Makefile.am phonebook.vala): msmcomm-specs: add initial interface definition for phonebook service Jan 10 10:38:26 morning GNUtoo|laptop Jan 10 10:38:42 hi Jan 10 10:39:04 how can I charge my nokia n900 with fso, does charging needs bq27xxx? Jan 10 10:39:14 no Jan 10 10:39:18 ok Jan 10 10:39:25 adding the fso plugin to the conf should be enough Jan 10 10:39:28 so I only need to activate the plugin Jan 10 10:39:33 ah it's not by default Jan 10 10:39:39 *it's not activated Jan 10 10:39:44 ok I'll add it then Jan 10 10:39:47 k Jan 10 10:40:44 maybe we should change the default config for it Jan 10 10:40:58 if we do we must warn n900 users Jan 10 10:41:05 because they're used to: Jan 10 10:41:08 to not use the script? Jan 10 10:41:10 sh charge.sh.txt Jan 10 10:41:14 yes Jan 10 10:41:19 if they do both Jan 10 10:41:24 I wonder what happen Jan 10 10:42:05 i'd think nothing Jan 10 10:42:06 a mail to shr-user should suffice... don't think there are many shr/n900 users anyway Jan 10 10:42:11 true Jan 10 10:42:16 and yes, I'd think it's not fatal anyway Jan 10 10:42:17 yes that is sufficent Jan 10 10:42:32 or warn every dev individualy trough irc Jan 10 10:42:36 or send a mail Jan 10 10:42:47 for a short period of time it's not fatal Jan 10 10:43:01 I mean it disconnects usbnet and do a strange noise Jan 10 10:43:15 I'm unsure about the strange noise tough Jan 10 10:44:09 strange noise??? Jan 10 10:45:15 I had that once Jan 10 10:45:18 disconnecting usbnet is what I've observed when touching the charger node, that's why i'm leaving it alone until we have a proper means to detect wallcharger Jan 10 10:45:20 I don't remember why exactly Jan 10 10:45:50 so it might be the charging script alone disconnecting? Jan 10 10:45:53 and I remember that removing the usb cable made it stop Jan 10 10:45:58 a buzz Jan 10 10:45:59 mrmoku: yes Jan 10 10:46:06 very small buzz Jan 10 10:47:47 mrmoku, big question Jan 10 10:47:55 for nokia900 Jan 10 10:48:05 the dvfs branch....does it contain a dvfs patch? Jan 10 10:49:07 http://git.freesmartphone.org/?p=linux-2.6.git;a=shortlog;h=refs/heads/nokia900/dvfs Jan 10 10:49:13 * GNUtoo|laptop doesn't see a dvfs patch Jan 10 10:49:36 * GNUtoo|laptop sees the n900 support patches on top if 2.6.37 Jan 10 10:49:49 and 2.6.37 mainline doesn't have dvfs Jan 10 10:50:01 GNUtoo|laptop: no Jan 10 10:50:29 GNUtoo|laptop: as the kernel did not work back then... I started with the nokia one only Jan 10 10:50:40 never came to rebase upon the dvfs stuff Jan 10 10:50:52 I hope to be able to do that today Jan 10 10:51:00 in the evening though Jan 10 10:51:20 and the pull request for the smartreflex framework is for 2.6.38 Jan 10 10:52:18 so, what I want to do is to take the nokia kernel and rebase it upon the omap-pm tree, which has the DVFS stuff Jan 10 10:53:27 ok Jan 10 11:02:21 HI everybody Jan 10 11:02:48 playya_: you spoke to me about desimlocking the pre 2 right ? Jan 10 11:13:48 what is the smartreflex fw for? Jan 10 11:14:41 mickey|office: to be used for dynamic voltage frequency scaling Jan 10 11:16:07 ah Jan 10 11:16:18 http://elinux.org/OMAP_Power_Management/SmartReflex Jan 10 11:16:21 thanx Jan 10 11:16:24 yw Jan 10 11:16:38 speaking about that... Jan 10 11:16:49 will we use suspend or try to not use it on n900? Jan 10 11:19:54 mickey|office: dunno, can't we have both? :P Jan 10 11:20:16 well Jan 10 11:20:27 not using it will make some things nicer Jan 10 11:20:37 it depends on what we gain wrt. standby time Jan 10 11:20:48 and what level of activeness we want during standby Jan 10 11:20:57 anyways, not so important now Jan 10 11:21:23 standby time under maemo without suspending is quite impressive I think Jan 10 11:21:32 so I for one would prefer to not supsend it Jan 10 11:21:38 *nod* Jan 10 11:21:50 on the other hand... if somebody wants it to suspend... then he probably should be able to do so Jan 10 11:24:53 mickey|office: what would Suspend() do without actually suspending? Jan 10 11:26:00 mrmoku: suspending resources, i.e. shutting them down or putting in low action mode (i.e. preventing unsolicited messages coming in since no one is looking at them anyways) Jan 10 11:26:27 messages, i.e. signal strength updates Jan 10 11:26:38 or location updates Jan 10 11:26:42 or waht not Jan 10 11:27:42 so writing mem to the sysfs node should be quite easily to do depending on a config option Jan 10 11:28:25 playya_: did you push my libeflvala change? Jan 10 11:28:43 mrmoku: if that's all, yes Jan 10 11:29:22 mickey|office: did not really try that yet, but should be Jan 10 11:29:47 on the other hand though, we might want to introduce more fine grained resource handling in the future anyways, i.e. specifying what you want to do with the resource, and this in turn could give information to the resource broker (fsouaged) which level of suspend to choose... Jan 10 11:30:30 I will check if suspend/resume works on n900 when I have a new kernel Jan 10 11:30:43 right Jan 10 11:30:57 unfortunately i don't believe it will fix forwarding Jan 10 11:31:01 otherwise i could use it, too Jan 10 11:31:54 my mail to the ofono list has not been passed through as it seems :/ Jan 10 11:32:12 and on #ofono no one seems to be able to answer that question Jan 10 11:32:15 I have to fix the lowlevel plugin, I know Jan 10 11:32:41 sure, but i'm afraid that is unrelated to forwarding Jan 10 11:33:46 I'm more optimistic :) Jan 10 11:33:50 hehe Jan 10 11:33:52 good for you ;) Jan 10 11:34:00 and perhaps for us all Jan 10 11:34:15 my work on the context manager is progressing faster than i thought Jan 10 11:34:22 i guess it'll be done by the end of this week Jan 10 11:34:27 at least the location part Jan 10 11:34:30 great Jan 10 11:34:35 so i can do something different afterwards Jan 10 11:35:19 I have some daywork pressure this week Jan 10 11:35:53 after that I hope to get something done with libisi Jan 10 11:36:01 ok, sounds good Jan 10 11:38:50 * mrmoku shower Jan 10 12:44:39 mrmoku: seems like bluetooth resource is still handled twice and fsodeviced fails (with current SRCREV + conf) Jan 10 12:56:40 JaMa: hmm... mickey|office ^^ Jan 10 12:57:21 JaMa: it was rfkill itself alone getting it twice due to btusb adding another bogus rfkill for bluetooth Jan 10 12:57:33 anyway... have to run now to a client Jan 10 12:57:46 JaMa: and mickey|office fixed it that it should not segfault anymore Jan 10 12:57:49 bbl Jan 10 13:01:35 mrmoku, mickey|office: yes, looks like double rfkill http://paste.pocoo.org/show/318258/ Jan 10 13:02:45 anyone remembers why we have PREFERRED_PROVIDER_virtual/libusb0 = libusb intead default libusb-compat? Jan 10 13:20:52 freesmartphone.org: 03morphis 07next * r84ef0dd66f18 10msmcomm/ (11 files in 6 dirs): msmcomm: implement phonebook ready unsolicited response message Jan 10 15:31:59 SHR: 03Martin.Jansa 07libphone-ui-shr * r00b9cbd8240a 10/src/view/ (contact-list-common.c message-list-view.c): adapt to elementary genlist_item_insert_before API change from r55869 Jan 10 15:35:25 heyho Jan 10 15:40:06 JaMa, hi, do you now if upgrading from feeds has some problems right now? Jan 10 15:42:16 nothing changed in feeds for last few weeks Jan 10 15:42:24 only tests feeds are changing Jan 10 15:42:28 and still have some issues Jan 10 15:43:04 ah ok, then the upgrades I have pending are safe ones ;) Jan 10 15:43:07 pespin: btw emtooth2 really depends only on elementary? I guess at least glib-2.0 and libeflvala? Jan 10 15:43:34 JaMa, yeah true. I had a look at iliwi to copy the deps but didn't find them Jan 10 15:43:44 JaMa, so I think you should add them to Iliwi too Jan 10 15:44:13 ok Jan 10 15:44:58 JaMa, I'll add them to my local recipe and send it as patch to ml with correct rev when I fix a bug I expect to fix today ;) Jan 10 15:45:16 because emtooth2 recipe is still not in OE right? Jan 10 15:45:35 I have it prepared.. no need to send anything ;) Jan 10 15:45:51 just ping me with rev Jan 10 15:46:15 JaMa, ok. Did you added the OE_EXTRACONF + = ""--enable-fso"" ? Jan 10 15:46:33 nope Jan 10 15:46:37 It's missing in the recipe I sent to ml ;) Jan 10 15:46:43 * check_data_file_clashes: Package emtooth2 wants to install file /usr/share/applications/emtooth.desktop Jan 10 15:46:46 But that file is already provided by package * emtooth Jan 10 15:46:49 * check_data_file_clashes: Package emtooth2 wants to install file /usr/share/pixmaps/emtooth.png Jan 10 15:46:52 JaMa, yeah, use newest rev Jan 10 15:46:52 But that file is already provided by package * emtooth Jan 10 15:46:54 it's fixed there Jan 10 15:46:56 why not drop emtooth1 at all? Jan 10 15:47:16 JaMa, because they are different apps with different deps Jan 10 15:47:20 and to it just as upgrade in emtooth_svn.bb? Jan 10 15:47:55 if you don't plan to maintain emtooth1 then I would like to "reuse" recipe Jan 10 15:48:25 JaMa, uhm ok, as you wish then. Jan 10 15:48:54 but I thought emtooth could be useful for someone. No vala,glib needed. Jan 10 15:49:06 ok as you wish then :) Jan 10 15:49:16 I think I'd keep it for the moment Jan 10 15:49:37 can you use some newer version than 0.0.2 in configure.in? Jan 10 15:49:39 if it fails to compìle/run in the future I might remove it. Jan 10 15:49:53 JaMa, uhm why? Jan 10 15:50:15 JaMa, use latest revision atm -> rev 105 Jan 10 15:50:27 it should be in sync with PV in recipe Jan 10 15:50:38 and there was 0.1.. Jan 10 15:50:48 so to sync it it would need to go backwards Jan 10 15:52:21 JaMa, uhm ok, I'll set it to 0.1 in configure.in Jan 10 15:52:24 use rev 106 Jan 10 15:53:11 why not 123? Jan 10 15:53:17 that's what I have now Jan 10 15:53:45 argh yeah, forgoto to svn up and "svn info" showed me 105 now sorry ;) Jan 10 15:55:30 JaMa, no sorry. forgot it and the git import failed :/ Jan 10 15:56:20 argh gnome-keyring is now working now Jan 10 15:56:24 *not Jan 10 15:56:34 playya_: pushed it only to OE then.. Jan 10 15:58:07 �� Jan 10 15:58:50 JaMa, sorry you'll have to wait a moment, I don't know why gnome-keyring doesn't like my password anymore and I can't commit in svn >.< Jan 10 16:01:12 ok.. will push it with 0.0.2 in PV and then bump later Jan 10 16:01:22 who speaks Russian Jan 10 16:02:16 Putin? Jan 10 16:02:55 yes yes yes" Jan 10 16:03:58 so who already know Russian Jan 10 16:04:46 JaMa, commited ;) Jan 10 16:04:58 in rev 124 Jan 10 16:05:44 pespin �� Jan 10 16:06:19 your name pespin Jan 10 16:06:43 your money Jan 10 16:06:44 xD Jan 10 16:07:37 any money Jan 10 16:07:57 it Th Jan 10 16:08:55 �� Jan 10 16:09:12 iiiiiiiiiiiiiiiiiiiiiiiiiiiiiiiii Jan 10 16:09:45 :'( Jan 10 16:12:26 STALKER: stop doing it Jan 10 16:12:31 STALKER: or i'll ban you Jan 10 16:12:44 STALKER: zabanyu! Jan 10 16:13:31 STALKER: what do you want? BTW, most folks use utf-8, so your (koi8-r?) messages are not readable. Jan 10 16:21:59 PaulFertser: he asked "who speak russian" "pespin what is you name" "aaaaaa" in koi8-r. nothing really provocating, seem just new to irc... Jan 10 16:22:13 gena2x: ok Jan 10 16:22:33 gena2x: thanks Jan 10 16:24:26 gena2x, well, the name and all those chars doesn't help hehe ;) Jan 10 16:25:30 :) Jan 10 16:37:57 JaMa, it's 56026 Jan 10 16:39:08 playya_: OE has now 56024.. will drop that patch with next bump, thanks Jan 10 17:08:04 damn. seem i bricked my fr :( Jan 10 17:08:35 gena2x: :-O ? Jan 10 17:09:04 ah, no, it's ok :) Jan 10 17:09:10 just hm... Jan 10 17:09:30 let me check and i'll write somethings strange :) Jan 10 17:11:13 ok, sorry for buzz Jan 10 17:11:40 just it automagically attempted to boot from nand on inserting battery Jan 10 17:12:04 and failed as i reformat flash Jan 10 17:12:27 which seem destroyed a bit with v37 Jan 10 17:12:58 but ok, my bad, after some lucky aux+power i finally got my NOR Jan 10 17:13:33 NOR rules \o/ Jan 10 17:20:06 anyway, after my v37 expirement i lost everything :). ubi on nand got completely corrupted, then i tried to boot from sd and mount it debian on sd crashed because of failure of reading ubi and put all /etc files on /lost+files with individual codes Jan 10 17:21:33 even after i decided to flash fresh qtmoko it doesn't flash with many "unrecoverable error:" Jan 10 17:22:09 now cleaned whole nand in rewriting everything... Jan 10 17:23:21 i mean it actually flashed, and even boot, but after few steps hang completely and secon boot were not so successfull :) Jan 10 18:29:51 one quick question. is shr development dead ? Jan 10 18:31:34 freesmartphone.org: 03mickey 07cornucopia * r290a1513f13c 10/fsotdld/ (4 files in 3 dirs): Jan 10 18:31:34 freesmartphone.org: fsotdld: add dummy location provider, configurable via config: Jan 10 18:31:34 freesmartphone.org: [fsotdl.location_provider_dummy] Jan 10 18:31:34 freesmartphone.org: accuracy = 100 Jan 10 18:31:34 freesmartphone.org: frequency = 6 Jan 10 18:31:34 freesmartphone.org: latitude = 50.0 Jan 10 18:31:35 freesmartphone.org: longitude = 8.0 Jan 10 18:31:35 freesmartphone.org: 03mickey 07cornucopia * r8a5c7bf28d37 10/libfsobasics/fsobasics/smartkeyfile.vala: libfsobasics: smartkeyfile: add doubleValue for config Jan 10 18:33:30 far from it, it's alive and kicking, but sometimes more action happens on ports, sometimes more on UI and apps, sometimes more on middleware Jan 10 18:34:07 Ar|stote|is, hi, not at all Jan 10 18:34:13 ive been visiting the blog now and then and the last post i find is from may 2010. Thats why ive been wondering what happened Jan 10 18:34:36 the folks are not the most active bloggers, right Jan 10 18:34:36 Ar|stote|is, the thing is that there is no update currently because we're very actively migrating to gdbus Jan 10 18:35:24 Ar|stote|is: yeah, the blog is quite dead Jan 10 18:35:31 not development though Jan 10 18:36:03 bleh i just wish i knew that there were updates. not read an entire article on the blog :P ill be traveling to barcelona in a month and wanted to toy around with my freerunner a while. Jan 10 18:36:30 but you get my hopes up. Ill try and install the latest testing build and see what happens. Jan 10 18:38:44 Ar|stote|is, basically a lot happen on irc Jan 10 18:39:00 but it happens at the hours where everybody is on irc Jan 10 18:39:08 not when everybody is eating or sleeping Jan 10 18:39:13 there are also some mailing lists Jan 10 18:40:34 well then starting tomorrow ill hang around here to see changes etc Jan 10 18:40:37 mickey|zzZZzz, btw I know you don't care anymore about htcdream(that has a great keyboard) but could you take it into account when doing your gps fsotldt stuff? Jan 10 18:40:41 and see how it is going Jan 10 18:40:56 mickey|zzZZzz, specifically the vala standalone activator Jan 10 18:41:20 also how will navit work on nokia900? Jan 10 18:41:21 and ill have to try and locate a replacement for my aux buttong and resolder it for the fun part reflashing Jan 10 18:41:42 Ar|stote|is, ah I've mine broken too, it was hard to resolder but I did it Jan 10 18:42:02 else you could get a microsd Jan 10 18:42:06 Ar|stote|is: I already resoldered mine twice Jan 10 18:42:08 and flash QI Jan 10 18:42:13 or something like that Jan 10 18:42:20 i have lost the broken button. i have to find a replacement to resolder again Jan 10 18:42:29 i think i already have qi installed Jan 10 18:42:32 ok Jan 10 18:42:37 so use the microsd Jan 10 18:42:40 or.... Jan 10 18:42:43 you could do that Jan 10 18:42:47 get shr on the microsd Jan 10 18:42:52 and flash from the microsd Jan 10 18:42:57 with nandwrite etc... Jan 10 18:43:02 mdt-utils Jan 10 18:43:05 i thought i may send it to some company to fix it and apply the two patches on my a5 Jan 10 18:43:23 so: Jan 10 18:43:27 untar SHR on the microsd Jan 10 18:43:42 get the ubifs or jffs2 images on the microsd somewhere as simple files Jan 10 18:43:50 ssh into the booted freerunner Jan 10 18:43:54 cat /proc/mtd Jan 10 18:43:59 to see the partition mapping Jan 10 18:44:03 then flash your images Jan 10 18:44:10 shut down the freerunner Jan 10 18:44:15 remove the microsd Jan 10 18:44:21 else it boots on the microsd(QI) Jan 10 18:44:26 and boot on NAND Jan 10 18:44:31 automatically Jan 10 18:45:58 bbl Jan 10 18:46:04 well i have one month of exams so everything will be done after that. ill ask again then if i have any problems :P Jan 10 18:46:12 thanks for the info and the good news Jan 10 18:50:44 lindi-: (restart navit) gpsd won't actually release the device for 60 seconds, so navit has plenty of time to re-request it. Jan 10 18:51:22 mrmoku: .37 is already built in shr-kms Jan 10 18:55:51 freesmartphone.org: 03mickey 07cornucopia * r08b6ddf9d569 10/fsotdld/src/plugins/provider_location_dummy/plugin.vala: fsotdld: provider_location_dummy: deliver values read from config Jan 10 18:55:53 freesmartphone.org: 03mickey 07cornucopia * r643a97b7a9ad 10/fsotdld/src/plugins/contextmanager/plugin.vala: fsotdld: contextmanager: handle location updates with varying accuracy Jan 10 18:57:20 GNUtoo|laptop: i don't think any special care is necessary, isn't the "activator" started/stopped on-demand and provides a resource? Jan 10 18:57:30 yes it is Jan 10 18:57:37 good, then you just need a proper gpsd config Jan 10 18:57:40 but it doen't do parsing etc.... Jan 10 18:57:47 only activation Jan 10 18:58:04 for instance it was impossible use in ogpsd Jan 10 18:58:22 because ogpsd did parsing and activation in the same file Jan 10 18:58:47 if it works with gpsd, it will work with fsotdld Jan 10 18:58:58 it didn't work with gpsd Jan 10 18:59:02 *ogpsd Jan 10 18:59:32 good, it might just work then Jan 10 18:59:39 ok Jan 10 18:59:44 what about n900? Jan 10 18:59:49 navit won't work? Jan 10 18:59:52 no way Jan 10 19:00:07 ah ok Jan 10 19:00:10 that's very far away Jan 10 19:00:20 fsotdld will provider location Jan 10 19:00:23 but not through gps :D Jan 10 19:00:29 too complicated to make a wrapper like that: Jan 10 19:00:46 libisi->gpsd->fsotlldt Jan 10 19:00:59 *fsotdld Jan 10 19:01:31 too complicated is relative. if someone wants to write a gpsd plugin for the n900 using libisi, then it might just work with fsotdld Jan 10 19:01:41 if not, then we will take a shortcut Jan 10 19:01:46 and feed it directly into fsotdld Jan 10 19:01:47 ok Jan 10 19:03:25 mickey|zzZZzz: so we are now using gpsd as daemon talking to GPS device? Jan 10 19:03:56 yes that's great for NMEA Jan 10 19:04:17 dos1: you can, if you want. Jan 10 19:04:43 it's not a necessity. fsotdld will contain some limited support for talking to GPS devices directly Jan 10 19:04:52 but the plugin to talk to gpsd is available and working Jan 10 19:04:52 mickey|zzZZzz: i was rather asking from architecture side ;) Jan 10 19:05:09 well... i still don't like gpsd Jan 10 19:05:09 mickey|zzZZzz: ok then, i understand Jan 10 19:05:36 but it seems to work and contains a lot of protocol support Jan 10 19:05:49 so that's good enough for me Jan 10 19:06:46 when i have my example map view done i'll blog about the new context stuff Jan 10 19:06:49 bbl Jan 10 19:07:11 ok Jan 10 19:11:59 PaulFertser: 60 seconds is too tight margin :) Jan 10 19:12:30 lindi-: why? Jan 10 19:20:09 PaulFertser: if I make a typo in the configuration navit doesn't start Jan 10 19:20:23 PaulFertser: it can take more than 60 seconds to fix that typo and I don't want to loose fix because of that :) Jan 10 19:20:49 PaulFertser: in short, I think we need to support different policy options ("always on", "always off", "on when needed", ...) Jan 10 19:21:02 lindi-: then you can use some "echo '?WATCH={"enable":true}' | telnet localhost 2497" trick or something. Jan 10 19:21:46 PaulFertser: yeah but it won't work for always off case Jan 10 19:22:26 PaulFertser: what if gpsd just told the hook how many clients it has? Jan 10 19:22:33 PaulFertser: then the hook could implement the policy Jan 10 19:23:00 PaulFertser: your simple version would be just if [ $1 -gt 0 ]; then om gps power 1; else om gps power 0; fi Jan 10 19:23:11 PaulFertser: but it could also ask the user or something equally esoteric Jan 10 19:23:57 lindi-: the hook is executed with ACTIVATE when there's one client and with DEACTIVATE when it's zero. Jan 10 19:25:37 PaulFertser: yeah but that keeps the policy inside gpsd Jan 10 19:27:06 lindi-: but hooks should be optional, so the gpsd policy basically shouldn't be changed, it can only be extended. Jan 10 19:27:46 PaulFertser: maybe Jan 10 19:27:58 PaulFertser: any idea how policy could be made configurable as well? Jan 10 19:28:57 lindi-: not yet Jan 10 19:32:07 lindi-: what about using external daemon for policy management? Jan 10 19:32:36 lindi-: i mean gpsd requests the resource from it, or releases. But it's up to the external daemon to decide if it's really the time to turn gps device off. Jan 10 19:32:58 PaulFertser: it would be useful if the policy stuff actually saw what clients are asking for it Jan 10 19:33:30 lindi-: but how can one identify a client? By the tcp socket it used to connect to gpsd? Jan 10 19:34:07 PaulFertser: yeah that's a problem Jan 10 19:35:00 lindi-: what about a tiny libgps-using app which you can use as a daemon and start it when you want to keep the device "always on"? Jan 10 19:35:22 lindi-: that way you just do it before messing with navit config. Jan 10 19:36:33 PaulFertser: sure but that won't help with the always off case still? Jan 10 19:36:41 PaulFertser: I might want to keep navit running but stop gps Jan 10 19:37:07 lindi-: then just remove the device from gpsd devices list via its unix control socket. Jan 10 19:39:37 PaulFertser: hmm Jan 10 19:39:50 sounds bit hackish :) Jan 10 19:40:14 lindi-: well, you say i do not want this device to be used by gpsd. So you just remove it from its control. Sounds logical enough to me. Jan 10 19:40:17 :) Jan 10 19:40:39 PaulFertser: what if I want to log all data that is received from gps? Jan 10 19:41:20 PaulFertser: so I need to have a client that is always connected and listening Jan 10 19:41:26