**** BEGIN LOGGING AT Tue Apr 20 03:00:04 2010 Apr 20 04:41:01 budfive: pong Apr 20 04:44:14 DocScrutinizer: Hey. You use an N900, right? Apr 20 04:44:28 DocScrutinizer: I'm going to get that or the pre. Wondering if you had comments Apr 20 04:44:39 well, yes and no Apr 20 04:45:12 budfive: for this topic yes. Apr 20 04:45:41 DocScrutinizer: what do you mean? Apr 20 04:45:56 the N900 is a real linux phone out of the box, has a decent large screen Apr 20 04:46:11 (my N900 is at repair ATM) Apr 20 04:46:26 DocScrutinizer: the n900 runs x? has a shell? the pre doesnt by default, i believe Apr 20 04:46:27 N900 has a decent kbd Apr 20 04:47:10 the pre kbd looks like it might be too small, but I've never seen one in person Apr 20 04:47:18 N900 has X, shell, users, upstart, Apr 20 04:48:01 downsides? Apr 20 04:48:11 heavy Apr 20 04:48:15 expensive Apr 20 04:49:11 but it works ok past that? Apr 20 04:49:27 aah, and of course N900 has 32GB builtin, plus uSD Apr 20 04:49:38 oh right. the pre has a lot less Apr 20 04:49:40 it's awesome Apr 20 04:49:47 so is heavy/expensive the main downside then? Apr 20 04:49:56 yes Apr 20 04:50:05 the pre plus has more ram i think. this is the only obvious advantage i saw for the pre offhand Apr 20 04:50:23 how much ram? Apr 20 04:50:31 pre plus has 512mb Apr 20 04:50:34 original pre had 256 Apr 20 04:50:36 wow Apr 20 04:50:58 yes, n900=256 Apr 20 04:51:10 ever hit that ceiling? Apr 20 04:52:12 sure, it swaps Apr 20 04:52:50 well, yeah Apr 20 04:52:54 ok Apr 20 04:52:57 oh, one more thing Apr 20 04:53:24 are there many "normal" people using it? Does it have an "app" ecosystem like android/apple/etc? Apr 20 04:53:52 I live in the US, so nobody has even heard of this thing. Sorry for being ignorant Apr 20 04:53:53 yes, and yes Apr 20 04:54:05 np Apr 20 04:54:50 http://www.ovi.com/services/ Apr 20 04:54:52 zeroedout: moin Apr 20 04:54:57 DocScrutinizer: ok, thanks. Will definitely get one i think. I've been holding out for the pre plus to come out for at&t in the us, but it's taking a while and may never happen anyway Apr 20 04:55:01 moin to all :) Apr 20 04:55:13 moin Apr 20 04:56:10 budfive: make sure you're not hostage of a contract with a non-N900-compatible carrier Apr 20 04:56:31 t-mo works ok iirc Apr 20 04:56:35 DocScrutinizer: what do you mean? Apr 20 04:56:47 DocScrutinizer: I'm not in a contract (using the gta02 right now) Apr 20 04:57:04 at+t (?) has stupid frequencies the N900 doesn't support Apr 20 04:57:04 DocScrutinizer: my current provider isn't compatible with n900 3g, but i'll move to tmobile us at some point Apr 20 04:57:14 DocScrutinizer: yep. i'm all too familiar with that Apr 20 04:57:36 DocScrutinizer: it'll still work, but without the 3g, right? Apr 20 04:57:47 right Apr 20 04:57:50 afaik Apr 20 04:58:11 actually EDGE is amazingly fast as well Apr 20 04:58:57 DocScrutinizer: right. I believe I get a better deal with at&t, so probably won't move immediately Apr 20 04:59:31 DocScrutinizer: the providers play games where unlimited data plans have different costs depending on which device they think you're using. So if they think you don't have a smartphone, the data is half the price Apr 20 04:59:48 DocScrutinizer: and i don't think they actually check Apr 20 05:00:31 budfive: when you get a N900 (I really suggest you do), check FM-TX radio and headset detection immediately. And treat USB receptacle with care, it's weak Apr 20 05:01:07 what's fm-tx? Apr 20 05:01:19 you can transmit over fm radio? Apr 20 05:01:22 (0.5 price) yeah, heard that several times in #maemo Apr 20 05:01:26 seems to work Apr 20 05:01:39 (fm-tx) yes Apr 20 05:01:51 for car stereo e.g. Apr 20 05:01:52 DocScrutinizer: at times one of these two don't work? Apr 20 05:02:03 yes Apr 20 05:02:47 ok. thank you very much. will work on getting one tomorrow Apr 20 05:03:29 welcome. And you'll not regret it :-) Apr 20 05:04:32 it's by far the most outstanding exciting mobile device I ever had the pleasure to own Apr 20 05:05:31 aah, btw the PHONE apps are mediocre at best Apr 20 05:07:29 I've been using the GTA02. Suspect it'll be a step up. :) Apr 20 05:07:44 sure, you bet it is Apr 20 05:10:48 ouch, forgot ONE thing :-/ Apr 20 05:12:44 listening Apr 20 05:13:21 budfive: it doesn't support USB hostmode (yet?) -> http://talk.maemo.org/showthread.php?t=31921&page=68 bottom Apr 20 05:14:17 i see. that is a downer. and it's not clear if the hardware can do it? Apr 20 05:22:23 well, I'm accepting bets. Mine is 99% it can do it Apr 20 05:23:21 so the driver is insufficient then? The kernel sources are available, I assume? Apr 20 05:23:45 yes, kernel is completely FOSS I've been told Apr 20 05:24:11 but it's missing some drivers it seems Apr 20 05:24:56 for hostmode. A gifted hacker should be able to patch the existing drivers though Apr 20 05:25:37 any other missing drivers? Apr 20 05:27:02 hmm, not really afaik. Battery charging is closed source in userland Apr 20 05:27:55 but it works, and I'm pushing Nokia to get a sane replacement / fix / patch, to implement sysnodes like in OM Apr 20 05:28:25 GPS is closed source in userland aiui Apr 20 05:29:14 and it is really poor for now, when you got no connectivity for A(gps) Apr 20 05:30:29 http://stskeeps.subnetmask.net/maemo.org/openness/pr1.1/ Apr 20 05:30:36 for the apps Apr 20 05:31:20 aha. good to see Apr 20 05:32:08 earlier i've heard vague rumors about the whole userland moving to qt at some point. Has any of that been pushed out yet? Apr 20 05:32:11 even their clock UI can't be distributed? ;) Apr 20 05:32:26 :) Apr 20 05:32:43 honestly, if you don't dream of an iPhone, then there's for sure no smarter more amazing device you might find than the N900 Apr 20 05:33:14 DocScrutinizer: you sold me an hour ago :) Apr 20 05:33:34 budfive: qt is next major OS revision, yes Apr 20 05:35:20 (qt) actually that's exactly same situation we've seen for OM OS Apr 20 05:37:09 the major userland will move to qt with maemo6 (or meego), but still it shall support all gtk based apps as well Apr 20 05:37:40 right. if it's just a linux box then it should just work if all the libraries are still there Apr 20 05:48:54 it *is* a true linux box Apr 20 05:52:08 last i hear it lacked xauth :) Apr 20 05:52:53 (we were trying to figure out if you could make an emergency call even if you had accidentally run "xtrlock" to lock the simulate an X application that grabs keyboard/mouse focus) Apr 20 05:54:41 err? sorry, don't get it Apr 20 05:55:36 hm. ffalarms are not alarming. 8( Apr 20 05:55:44 :( Apr 20 05:55:47 well ok many routers don't include such basic utilities eitehr Apr 20 05:56:19 DocScrutinizer: it depends on what this true linux box means :) Apr 20 06:02:50 lindi-: I think about what I conceive to be the common sense meaning of linux-box Apr 20 06:04:05 mine includes xauth if X is included Apr 20 06:04:34 hmm, so what stops you from installing it? Apr 20 06:05:14 i guess that's a different issue entirely :) Apr 20 06:05:29 but it lacked a compiler too i think Apr 20 06:06:39 huh? Apr 20 06:07:04 DocScrutinizer: well just try building xauth and see if you can. you have bought the device after all :) not me Apr 20 06:07:12 lindi-: sorry, I think you completely lost me Apr 20 06:08:16 DocScrutinizer: for me "true linux box" means that if there's X there's also xauth. afaik xauth wasn't available from the repos and afaik there was no easy way to compile it on the device Apr 20 06:08:34 but of course true linux box is very vague :) Apr 20 06:08:48 and for sure I'll not dive into compiling anything I don't feel I need it, just to understand your objections Apr 20 06:08:59 s/objections/concerns Apr 20 06:09:10 I can think of a thinclient that only runs X and does not have xauth or other x binaries Apr 20 06:09:43 this one is just a terminology issue, nothing necessarily bad about n900 Apr 20 06:10:50 I'd consider the lack of a mandb package a real deficit. But I'm sure you could build your own mandb pkg as well Apr 20 06:12:56 and a system with fucking busybox, that even borks initscripts when replacing that crap by a decent shell + coreutils, yeah it's debatable if you may call that a true linux Apr 20 07:33:38 smallest gta01 images 55M for jffs2 (is it small enough?) Apr 20 07:36:01 JaMa: i'm not sure if it was 57MB or 53MB Apr 20 07:59:28 JaMa: it shoud work, 55M means 12,000 free NAND PAGES to fill in the first boot Apr 20 08:01:41 Heinervdm: I've checked gta01 dir and didn't find smaller one.. but strange I expected few more there Apr 20 08:02:07 Heinervdm: ah you mean the limit not "smallest image" Apr 20 08:03:21 smallest I see there is 63M Apr 20 08:03:22 JaMa: yes limiut Apr 20 08:29:19 http://shr-project.org/trac/wiki/FSOSHRUDCON%2710 Apr 20 08:46:23 what do you think about switch from xserver-kdrive-common to xserver-common Apr 20 08:46:26 ? Apr 20 08:47:19 I'm testing it ATM and with few patches to add our custom logic (nocursor, kernel ts calibration) from kdrive-common we can move easily Apr 20 08:47:52 but maybe it's time to think about non-root X again Apr 20 08:48:27 I haven't looked at the recipe, xserver-drive-common doen't do much so a switch shouldn't be that hard Apr 20 08:49:22 yes it's really easy if we want to keep everything as it is Apr 20 08:51:24 Heinervdm: linuxhotel is available? Apr 20 08:52:11 mrmoku: Serdar just called them Apr 20 08:53:08 mrmoku: i think we should fill that page with some initial content and then we can announce it on the ml this evening Apr 20 08:53:09 Heinervdm: great Apr 20 08:53:20 already added myself Apr 20 08:53:31 mickey|office: http://shr-project.org/trac/wiki/FSOSHRUDCON%2710 Apr 20 08:55:02 There are FSI or FIS there at the same time, so there aren't that much rooms left Apr 20 08:57:05 * JaMa added himself to garden camp :) Apr 20 08:57:45 :) Apr 20 09:04:24 and checking train connection reveals time about 9,5h :/ and with car about 7h Apr 20 09:06:19 i just have 3h by train Apr 20 09:13:07 I'm thinking about train to Munchen, buying bmw5 there and continue to Essen :) Apr 20 09:13:45 JaMa: hehe Apr 20 09:14:40 JaMa: if your train seriously passes via München and you want to spare the money for the bmw... I'm going by car from München to Essen... my VW-Bus won't be as much fun as a new BMW5 though :P Apr 20 09:17:46 lol Apr 20 09:18:09 I hope for you the trains will not be as crowded then as they're now Apr 20 09:18:31 well normally I would have to switch train in Berlin (or bus prague-Nurnberg), but I want to buy BMW5 for so long.. that it would be nice opportunity :) Apr 20 09:18:47 Iff that volcano doesn't stop, flying might be impossible for a long time... Apr 20 09:20:19 They are flying again today Apr 20 09:20:53 poor TAsn if they wont fly again in July .. :) Apr 20 09:21:50 Heinervdm: some volcano experts say the real eruptions might yet come Apr 20 09:21:59 JaMa: well TAsn can still travel on his board :P Apr 20 09:22:34 hehe :) Apr 20 09:22:46 surfboard? Apr 20 09:22:54 new worldrecord: longest surf trip Apr 20 09:23:01 Wonka: yep Apr 20 09:24:03 mickey|office: ping Apr 20 09:24:28 maybe kite+board will be better for inland :) Apr 20 09:27:12 mmh, 6h by train for me Apr 20 09:27:17 hmm mickey is sleeping in his office *G* Apr 20 09:43:20 lp0 on fire Apr 20 09:43:20 lp0 off-line Apr 20 09:43:57 hehe, wiki says it's not common in modern Linux systems, but today I had it many times in dmesg.. Apr 20 10:28:11 TAsn: i'm sorry for a nitpicking question but is it possible to get all the results of a query in one call? I know GetMultipleResults needs number but i do not want to do an extra call to get the amount of results. Moreover if there's only one, GetMultipleResults doesn't return anything. Apr 20 10:47:34 PaulFertser: don't think so... at least getting the count first is what we do too Apr 20 10:48:00 TAsn: maybe you could change GetMultipleResults to return the complete thing when given a count of 0 ? Apr 20 10:48:41 mrmoku: yes, or -1. Apr 20 10:49:47 yup -1 is probably even better Apr 20 10:50:44 And make it consitently return 1 result when asked for 1. Apr 20 11:09:14 procmail works fine now. my notmuch gives those mails "unseen, inbox" tags and notmuchsync renames them to ":2,S" after reading. So I don't need further work on it. Apr 20 11:09:19 oops Apr 20 11:29:52 ahoy Apr 20 11:30:20 as just wrongly mentioned in #openmoko: i noticed a possible bug in dfu-util: after "usb_claim_interface()" a "usb_set_altinerface(_rt_dif.dev_handle, 0)" call seems to be necessary to actually switch to DFU-Mode when the USB-Device still uses its default interface. usb_claim_interface() doesn't send anything over the bus, but the device needs to know when to serve DFU-Requests (like the following dfu_get_status() request). it works w Apr 20 11:30:48 but it doesn't work if a device offers a DFU-Device Descriptor in normal runtime mode (to serve appDETACH requests) Apr 20 11:35:46 rigid: can you please mail that to openmoko devel list with Cc to Stefan Schmidt? Apr 20 11:36:37 PaulFertser: sure Apr 20 11:37:04 i thought he might be here anyway, and i'm a bit lazy when it comes to e-mail ;) Apr 20 11:45:08 rigid: he's not an active member of OM community anymore, unfortunately. Apr 20 11:49:21 PaulFertser: ah, is I see now he wrote a mail yesterday to step-up as new maintainer for dfu-util so he's the right person... i guess there won't be a #dfu-util channel in the near future ;) Apr 20 11:49:44 he could join here anyway without beeing active, tho Apr 20 11:49:51 i guess Apr 20 11:49:54 he does from time to time Apr 20 11:50:25 ah... i'll also idle here (until the next reboot or kick :) Apr 20 12:35:25 TAsn: building 48160 with some more fixes from dh now Apr 20 12:36:25 spaetz: seems like you need to bump libphone-ui-shr SRCREV a bit (failed build) Apr 20 12:37:39 or maybe not.. :) Apr 20 12:38:28 mrmoku: this should be compatible with current EFL rev? http://git.shr-project.org/git/?p=libphone-ui-shr.git;a=commit;h=25a03a38e5784c198b79d12c9be33039bb351d6a Apr 20 12:42:39 JaMa: uugh, how do I fix that? Apr 20 12:42:49 some edje problem? Apr 20 12:42:56 or liphone-ui-shr? Apr 20 12:43:58 rint(msg.format_message_as_text()) to output it like notmuch show Apr 20 12:44:00 oops Apr 20 12:45:57 mrmoku: there was commit/revert/commit, but seems working in shr-u, but fails in shr-t Apr 20 12:46:35 spaetz: it was reverted because this issue, but then commited again (I think after EFL bump), not sure why it fails now when you have the same EFL etc Apr 20 12:46:56 JaMa: yep, that needs some current EFL Apr 20 12:47:24 but don't I have the current one? Apr 20 12:47:54 I see JaMa is at it :) Apr 20 12:48:29 hehe fighting with screen :) Apr 20 12:48:38 JaMa, spaetz: 47811 was it Apr 20 12:49:58 hmm clean&build and still fails http://tinderbox.openembedded.net/packages/553341/ Apr 20 12:50:33 JaMa: too few arguments is the other way round Apr 20 12:50:38 libphoneui too old Apr 20 12:51:22 ahh, JaMa do you update the libphoneui srcrev or should I do it? Apr 20 12:51:51 spaetz: please do it Apr 20 12:56:54 JaMa: pushed, starting a rebuild Apr 20 12:58:34 mrmoku: wasn't also shr-specs or something needed in newer SRCREV? Apr 20 12:58:45 or it was different issue? Apr 20 12:59:19 OE git is slow and SHR buildhost is slow, but pulling git from the buildhost...... oh my Apr 20 12:59:37 it's stuck since eternity Apr 20 13:00:35 yes it gets slower and slower :/ Apr 20 13:00:42 fatal: The remote end hung up unexpectedly Apr 20 13:00:48 hah. lies. I expected that! Apr 20 13:05:14 mrmoku: any reason why you also compile statically by default? Apr 20 13:05:24 /usr/lib/phoneui/modules/shr.la installed but not shipped in any package Apr 20 13:05:41 why not disable if we don't need it anyway... Apr 20 13:07:39 spaetz: no idea about that stuff... you should talk to TAsn :) Apr 20 13:13:34 hehe Apr 20 14:07:35 TAsn: you broke it :( Apr 20 14:08:18 TAsn: http://paste.debian.net/69895/ Apr 20 14:10:43 PaulFertser: I have no idea about python... but res = self.db_handler.get_content([num_id, ], None) misses something after the num_id? Apr 20 14:15:15 num_id, indicates a tuple Apr 20 14:15:22 mrmoku: hm, not exactly. It's one of the latest commits (that i asked for) that started to unconditionally use .get method even if the value's None. Apr 20 14:15:38 mickey|office: do not bother, TAsn will fix it. Apr 20 14:15:44 k Apr 20 14:15:46 :) Apr 20 14:16:27 mickey|office: i was discussing opimd api with TAsn lately and it appeared it lacks some obviously needed things, so it was a quick patch to satisfy me, well, it had a side-effect :) Apr 20 14:16:39 ah Apr 20 14:16:53 mickey|office: btw, fsogsmd debug/non-debug issue, what do you think about it? Apr 20 14:17:18 mrmoku: is it reported? ;) Apr 20 14:18:25 PaulFertser: this is somewhat annoying; I'd appreciate someone trying to debug where we overrun the Calypso and how Apr 20 14:18:52 PaulFertser: i have added some additional minimum command interval timing to libgsm0710mux, someone should play with that as well Apr 20 14:19:16 mickey|office: I played with timeouts in the default config to no avail Apr 20 14:19:26 mickey|office: are there some others not in the config? Apr 20 14:21:05 the minimum command timeouts are relatively new Apr 20 14:21:11 added them when i was in austria, i think Apr 20 14:21:23 mickey|office: in fact i haven't seen any obvious sign of modem communication failure in the strace log of a non-working case. I think i'll try to provide you with it so you can have a quick look. Apr 20 14:22:22 mrmoku: device_sendpause_threshold = 500 Apr 20 14:22:31 ah well Apr 20 14:22:33 250 Apr 20 14:22:55 mickey|office: and another unrelated issue: has anyone actually tried GPRS with fsogsmd? Because with noauth it works good for me when i start it but pppd becomes zombie when i turn pdp off. Probably debian's version of glib is too old... Apr 20 14:23:03 if that doesn't make a difference, then it's not modem communication, but something else in libgsm0710mux Apr 20 14:24:06 PaulFertser: or it's a bug in pppd that only occurs when it's controlled by a parent process Apr 20 14:25:22 mickey|office: i know how you "love" pppd but ahem hm heh many folks need gprs and need it soon, not in a several months time when you'll rewrite pppd parts you need ;) Apr 20 14:29:28 i'm not going to rewrite it Apr 20 14:30:32 i have at least one modern platform where i need it as well Apr 20 14:30:46 so i'm all ears finding out what happens and fixing it Apr 20 14:32:04 Hehe Apr 20 14:32:22 mrmoku told me it's only you who can actually fix that ;) Apr 20 14:33:54 mickey|office: right atm i enjoy writing elisp too much to go back to the debugging hell, and also i run fsogsmd on the device atm which severly limits my debugging possibilities. Apr 20 14:41:21 i can give it a try when mrmoku "releases" an image with fsogsmd Apr 20 14:41:35 my selfcompiled images no longer boot on my FR, so i have a hard time testing anything Apr 20 14:42:03 mickey|office: ohh... that 250 seems to help :D Apr 20 14:42:33 * mrmoku rebooting to verify Apr 20 14:42:53 that would confirm my suspicion that we overrun the modem Apr 20 14:43:05 hence explaning why DEBUG helps Apr 20 14:43:21 250 is of course ridicuously slow Apr 20 14:43:26 so we would need to find the sweet spot Apr 20 14:43:38 ok, if reboot works too I will try 100 Apr 20 14:43:45 (basically it's the minimum amount of time to wait between two consecutive commands in milliseconds) Apr 20 14:45:18 hmm no :( Apr 20 14:45:32 mickey|office: after reboot it did not work Apr 20 14:46:15 2010-04-20T14:45:35.013265Z [ERROR] TiCalypsoModem <4C>: Can't open main channel; open returned false Apr 20 14:46:40 hmm Apr 20 14:46:44 that's how it breaks? Apr 20 14:46:48 yep Apr 20 14:46:57 i thought it stalls during operation Apr 20 14:47:07 i.e. before registration Apr 20 14:47:12 but after channel opening Apr 20 14:47:50 another thing I observed before is that when I killed fsogsmd fsousaged went away too Apr 20 14:48:23 hmm Apr 20 14:48:30 could you test the autoON? Apr 20 14:48:36 like, move fsogsmd out of the filesystem Apr 20 14:48:38 err Apr 20 14:48:40 fsousaged, that is Apr 20 14:48:43 then reboot Apr 20 14:48:54 if fsogsmd can't find fsousaged it will autoenable the resource Apr 20 14:49:01 ok Apr 20 14:49:07 just to be sure that this doesn't change the situation Apr 20 14:49:17 i'd like to be 100% sure that it's fsogsmd/libgsm0710mux Apr 20 14:52:12 Hm, i wonder if it's gsm consortcium folks that decided on starting timestamp not on unix epoch... Apr 20 14:58:09 mickey|office: on first boot autoON worked... at least it successfully got a channel Apr 20 14:58:35 it did not register because phonefsod did not send a SetFunctionality as GSM was not there for it Apr 20 14:58:38 rebooting now Apr 20 15:00:45 mickey|office: worked again Apr 20 15:01:48 hmm, interesitng Apr 20 15:02:25 interesting is also that I have no usb network without fsousaged... how could that be related? Apr 20 15:03:41 * mrmoku does another reboot to be sure Apr 20 15:06:34 that's related like that: Apr 20 15:06:45 resources automatically enable themselves if fsousaged is not found Apr 20 15:06:57 the UsbHost resource served by fsodeviced will kill your USB client Apr 20 15:06:59 including usb hostmode Apr 20 15:07:01 correct Apr 20 15:07:02 yeah, ok Apr 20 15:07:13 we need a way to disable this automatic enabling per resource Apr 20 15:07:19 on third boot it got its channels fine too Apr 20 15:07:21 hmm Apr 20 15:07:29 * mrmoku removes fsousaged back in place Apr 20 15:07:31 is this without the timeout now? Apr 20 15:07:44 the 250 is still in the config Apr 20 15:07:47 hmm Apr 20 15:07:49 remove that Apr 20 15:07:51 ok Apr 20 15:08:06 the bug may be a combination bug, who knows Apr 20 15:09:50 * mrmoku rebooting Apr 20 15:12:34 mickey|office: hehe, without that 250 it goes even further and unlocks the SIM :) Apr 20 15:12:57 yeah, w/ 250 everything will take ages Apr 20 15:12:59 would have done that with 250 too maybe and I was just impatient :P Apr 20 15:13:00 yeah Apr 20 15:13:08 as this will lead to 4 commands per second :) Apr 20 15:13:15 and we issue about 50 or 60 on startup Apr 20 15:13:22 :) Apr 20 15:13:34 so no calypso overrun... Apr 20 15:13:38 yeah Apr 20 15:13:41 it strongly looks like that Apr 20 15:13:44 now Apr 20 15:26:02 mickey|office: hmm.. I moved fsousaged back in place... got no channel... killed fsogsmd... requested GSM and it worked fine Apr 20 15:29:22 heyho Apr 20 15:29:36 PaulFertser, will test. Apr 20 15:31:29 PaulFertser, what exactly did you do? Apr 20 15:31:40 I mean Apr 20 15:31:55 what was the content of the Apr 20 15:32:04 call Apr 20 15:32:24 TAsn: i can pass it to you, but it's irrelevant, you saw the traceback ;) Apr 20 15:33:30 ok my bad. Apr 20 15:33:30 :) Apr 20 15:34:14 TAsn: http://paste.debian.net/69909/ Apr 20 15:34:46 have fun :P Apr 20 15:34:50 freesmartphone.org: 03tom 07framework * rc09a1a7f5b87 10/framework/subsystems/opimd/ (db_handler.py pimd_generic.py): opimd: Fixed a crash on Update. Apr 20 15:34:52 and tahnks. Apr 20 15:38:45 TAsn: thanks for caring :) Apr 20 15:39:12 I don't care, I'm a bug fixing robot :P Apr 20 15:40:25 all hail the bug fixing robot Apr 20 15:40:56 TAsn: btw, is it the right way to mark calls as not new and messages as read? Apr 20 15:42:16 PaulFertser, Calls: New. Messages: MessageRead :( Apr 20 15:42:19 I want to fix that. Apr 20 15:42:58 Those are relics from the past. Apr 20 15:43:06 Hm, org.freesmartphone.PIM.Calls.UpdatedCall seem to be not in the specs at all. Apr 20 15:43:07 Same goes for Sender/Recipient in messages. Apr 20 15:43:18 PaulFertser, maybe :) Apr 20 15:45:12 mrmoku: i guess i need the exact image you have Apr 20 15:45:18 so i can take a look on my own Apr 20 15:47:50 mickey|office: btw, is it just me or fsogsmd doesn't send Network.Status signal on strength change? Apr 20 15:51:25 correct Apr 20 15:51:28 that's by design Apr 20 15:51:46 we have Network.SignalStrength IIRC Apr 20 15:51:54 (available on some modems) Apr 20 15:52:06 mickey|office: Yep, but it's illogical to have the value returned by two different signals, no? Apr 20 15:52:45 ? Apr 20 15:52:51 Status reports the full status Apr 20 15:52:53 (i.e. i wouldn't say anything if this wasn't in the struct returned by Network.Status signal) Apr 20 15:53:03 Signal reports the signal strength Apr 20 15:53:09 the latter being only available on some modems Apr 20 15:53:14 Probably i misunderstand something but i somehow do not like it... Apr 20 15:53:36 I can explain this i think Apr 20 15:54:05 this is actually one part where the GSM shines through Apr 20 15:54:16 usually there is no unsolicited result code for signal strength changes Apr 20 15:54:23 so you have to poll, if you want to know the strength Apr 20 15:54:29 mickey|office: i expect a signal that notifies me of data change to be called every time any of the data it embeds changes. Apr 20 15:55:21 we could issue a full status update on %CSQ Apr 20 15:55:32 but this would be a bit over the top IMO Apr 20 15:55:40 So if you got the non-standard unsol telling it changed, and this data is embebedded in Network.Status, i'd expect a to get Network.Status again. Apr 20 15:55:58 hmm Apr 20 15:56:01 this i don't like somehow Apr 20 15:56:16 telling all the stuff that didn't change Apr 20 15:56:17 mickey|office: but do you at least "feel" what i'm talking about? Apr 20 15:56:34 yes, sure Apr 20 15:56:38 signal handling is a special case Apr 20 15:56:47 because of the AT command heritage Apr 20 15:57:17 i added the strength signal only on request Apr 20 15:57:25 i didn't like it either Apr 20 15:57:32 perhaps it should go Apr 20 15:57:35 mickey|office: that's ok, but can you probably remove the Strength property from Network.Status? Apr 20 15:57:48 hmm Apr 20 15:57:59 then you are missing a critical info for displaying a top panel Apr 20 15:58:22 since the signal strength signal is optional Apr 20 15:59:08 mickey|office: if i want to know signal strength i'll ask for it explicitly once and then will wait for the signal. Apr 20 15:59:47 please put that on the standards mailing list Apr 20 15:59:53 /msg MemoServ READ 1 Apr 20 15:59:54 Ok, i'll try. Apr 20 15:59:57 need to find out what the others think Apr 20 15:59:57 shit lol Apr 20 16:00:35 TAsn: ;p Apr 20 16:00:43 sicu, I love memoserv ;P Apr 20 16:01:10 TAsn: yeh, appreciate the reminder =] Apr 20 16:01:29 Anyhow, look at the delivered siglaunchd.conf, it's cool for auto-answering calls. Apr 20 16:01:53 sicu, I still don't have time for docs/wiki/whatever Apr 20 16:01:56 but it's cool. Apr 20 16:02:06 sicu, You can use it for reverse-ssh to your home. Apr 20 16:02:30 TAsn: nice :) Apr 20 16:02:38 Or maybe when you loose the phone you can call it and make it send you by sms the gps location + cell location Apr 20 16:02:44 And I don't know what :) Apr 20 16:02:57 indeed, plenty of usecases for this Apr 20 16:03:13 * lindi- has had that for a year Apr 20 16:03:15 sicu, One thing I'm not sure about is the state. Apr 20 16:03:24 lindi-, I wrote that also a year+ ago. Apr 20 16:03:35 I ran it for several months so that a test call is done every two hours Apr 20 16:03:46 lindi-, what software did you use? Apr 20 16:03:47 then i concluded that my SIP provider is too unreliable :) Apr 20 16:03:58 I didn't find anything useful so I had to do it on my own. Apr 20 16:04:05 TAsn: http://iki.fi/lindi/openmoko/answering-machine runs a hook based on caller id Apr 20 16:05:07 cool. in what I wrote you can regex on incoming numbers no verify they are allowed, and then request a password entered by dtmf. Apr 20 16:05:31 TAsn: while we are at it, can you explain me a little how outgoing messages are handled? What will happen if it's not sent, what about message receipts etc? Apr 20 16:05:39