**** BEGIN LOGGING AT Thu Mar 26 02:59:57 2009 Mar 26 03:03:25 snakes! Mar 26 03:03:42 on.. a plane! Mar 26 03:07:08 on a moko :( Mar 26 06:05:41 Ainulindale: posted to the mailing list. Happy now? :-) Mar 26 06:52:59 hi Mar 26 06:53:38 i just flashed my gta02 with FSO5.1 and that looks pretty well Mar 26 06:53:44 very reliable Mar 26 07:11:06 good morning Mar 26 07:19:43 spaetz, mrmoku, Ainulindale: my request: have vala-settings be called shr-settings, if it is accepted we go with it... and open an repo for it. Mar 26 07:23:17 morning Mar 26 07:23:34 ptitjes: ok for me Mar 26 07:27:54 FSOSHR'09 will happen in essen Mar 26 07:32:27 fressen! Mar 26 07:32:58 spaetz: and have dos1 moving to Vala along to you Mar 26 07:33:11 :) Mar 26 07:47:39 vala power!! Mar 26 07:47:41 :D Mar 26 07:58:01 * mrmoku passes raster a big fat Schweinebraten Mar 26 08:00:13 mmmm Mar 26 08:00:20 schweinbraten Mar 26 08:00:23 lecker Mar 26 08:14:32 mmmh, tasty! Mar 26 08:15:06 ptitjes: :). He did great work, so I don't want it to be pushed aside (and him) Mar 26 08:16:03 and I cannot promise that I will always be able to invest loads of time into that project. Mar 26 08:16:20 (just to be upfront honest about things) Mar 26 08:16:55 but moving away from python on an embedded system, does sound like the correct approach to me Mar 26 09:19:54 hi Mar 26 09:22:34 morgen Mar 26 09:39:39 mickey|zzZZzz: you broke libeflvala Mar 26 09:39:57 where is the image after make image? Mar 26 09:42:29 * dent got himself a CPU upgrade to cut down OE build times :) Mar 26 09:42:41 I hope it helps... Mar 26 09:42:42 serdar1: tmp/deploy/glibc/images/om-gta02 Mar 26 09:42:44 serdar1: tmp/deploy/glib/image or something Mar 26 09:42:53 ahh mrmoku was quicker :) Mar 26 09:43:01 and correcter :) Mar 26 09:43:02 and more precise :P Mar 26 09:43:04 :-) Mar 26 09:43:09 more correct, rather Mar 26 09:43:15 hehe Mar 26 09:43:17 morning Mar 26 09:43:23 morning Mar 26 09:43:49 * mrmoku had some coffee 5 mins ago... and is awake :-) Mar 26 09:44:00 just fixed libeflvala to a ref that is working. No coffee yet, that must explain it Mar 26 09:44:16 current head is broken Mar 26 09:44:20 hope you got the correct ref then ;) Mar 26 09:44:26 s/ref/rev/ Mar 26 09:44:27 mrmoku meant: hope you got the correct rev then ;) Mar 26 09:44:33 so do I :-). just testing Mar 26 09:44:53 BTW, should we move current unstable to testing again and start stabilizing it? Mar 26 09:45:06 a lot happened since then, I guess Mar 26 09:45:41 * mrmoku is putting his Ainulindale hat on and says: Mar 26 09:45:45 we should ask the list :-) Mar 26 09:46:03 bah. they won't do the work Mar 26 09:46:19 (sorry if that sounds elitist). I expect to be spanked by Ainulindale for this Mar 26 09:46:36 :) Mar 26 09:47:08 :-) anyway... I'm away this weekend... might be better to do it not before monday... if we do it Mar 26 09:47:44 on the other hand... when I'm not here people will spank you... So we could also do it short before I leave :P Mar 26 09:48:22 SHR: 03Sebastian 07shr-overlay * r9b58b698475e 10/openembedded/conf/distro/include/shr-autorev-unstable.inc: Merge branch 'master' of git+ssh://git@shr.bearstech.com/shr-overlay Mar 26 09:48:32 SHR: 03Sebastian 07shr-overlay * r1bff4463b72f 10/openembedded/conf/distro/include/shr-autorev-unstable.inc: fix libeflvala ref to a working tree. sorry Mar 26 09:48:33 SHR: 03Sebastian 07shr-overlay * r5f9dabec749d 10/openembedded/conf/distro/include/shr-autorev-unstable.inc: fix libeflvala ref as current HEAD is broken Mar 26 09:50:09 no, not that urgent. I don't have time over the weekend anyway. Mar 26 09:50:22 just in general Mar 26 09:50:57 Hmm. Is anyone else making a C(++) phone ui for fso? Mar 26 09:52:57 Rakshasa! Mar 26 09:53:06 I'm trying to make some lib for FSO Mar 26 09:53:20 but now I'm stuck fighting OE Mar 26 09:53:34 or, well... I have time to work on it only over night or some weekends, so I'm slow Mar 26 09:53:59 ehm, I mean a C++ lib to use FSO Mar 26 09:54:12 based on dbus-c++ and boost Mar 26 09:55:01 dent: Hey, that's nice. I'm taking the top-down approach and am writing the ui in Elementary first. Mar 26 09:56:00 what I want to produce it an easy-to-use C++ binding for FSO Mar 26 09:56:38 but afaik dbus-c++ is sync only now... i.e. for async one would have to use threads... Mar 26 09:56:38 ever tried to remove the bluetooth gadget in the illume shelf? Makes my enlightenment crash reliably Mar 26 09:56:51 dent:Tell me how it goes, so I can use it. That would make my job that much easier. Mar 26 09:57:08 so far there's not much + there are some problems I'm facing Mar 26 09:57:26 have a look at git://zub.lamer.la/libframeworkd-c++.git Mar 26 09:57:35 (not overly inventive name... yeah :) Mar 26 09:58:34 current issue is: boost.signals is not thread safe... I need boost.signals2, btu this is not yet in released boost. to use it, I need latest boost + signals2, but latest boost is nto in OE and building boost is a pain Mar 26 09:59:36 spaetz: yep Mar 26 09:59:37 the code in the repo uses boost::signals, btu that is not thread safe and miltithreaded code with that might explode... Mar 26 10:01:40 There's no alternative to using boost? Mar 26 10:02:34 I looked into libsigc++, but that seems to not be thread safeeither Mar 26 10:02:47 and I don't feel like implementing it myself... losts of work + error prone Mar 26 10:03:12 aaand, I think I'd need boost anyway eventually... i.e. shared_ptr and friends Mar 26 10:03:38 I'll try to get latest boost compiled in OE... hopefully I manage over weekend Mar 26 10:04:21 Ok. good luck with that. I've been put down by boost compilation so many times I've lost count =P Mar 26 10:04:28 lol :) Mar 26 10:04:45 Don't remember though why I needed it. =P Mar 26 10:04:47 they now have experimental CMake support... that sounds promising Mar 26 10:05:09 but just by-hand cmake build with OM toochain failed anyway Mar 26 10:05:38 hmm. Mar 26 10:05:52 Hopo you manage to get it running. Mar 26 10:05:59 hope... Mar 26 10:06:19 so do I... if I manage to get it all working somehow, I'll spam the list, maybe there will be some c++ interested ppl Mar 26 10:06:26 :) Mar 26 10:07:17 hm... another approach would be to ignore singals altogether... and clients derive from base classes supplied by lib Mar 26 10:07:20 but that imho sucks Mar 26 10:08:11 Can't be that many, considering that even c++ compilation is broken in fresh toolchain install. =( Mar 26 10:08:36 broken? I just added some C++ include paths Mar 26 10:08:43 or is there some more sutff? Mar 26 10:08:55 Hmm. How long time ago? Mar 26 10:09:06 a month? Mar 26 10:09:26 when I built the toolchain Mar 26 10:09:53 Okay. I installed mine just a week ago, and I had to fix the setup-env script and the .la's. Mar 26 10:10:21 I used the prebuilt version though. Maybe I should build the toolchain myself too. Mar 26 10:11:42 aha Mar 26 10:12:25 well... I'm using cmake for the builds w the toolchain... la's come into play with autotools/libtool only(?) Mar 26 10:15:48 Yep. I'm used to using autofoos, so I think I'll be using them in the future too... Mar 26 10:16:14 I envy you... I never managed to grasp them :-/ Mar 26 10:16:26 and when they choke, Im lost Mar 26 10:16:43 mrmoku, spaetz: thanks Mar 26 10:17:36 It's not the easiest and most clear way to setup building. true. Mar 26 10:20:09 spaetz, mrmoku: no migration as is Mar 26 10:20:13 first a changelog :-) Mar 26 10:20:21 Do we have a solution for UTF8/Dictionnary? Mar 26 10:24:45 Ainulindale: well... we could ask the list if people prefer a UTF8 working dictionary, that is slow as hell and won't let you use UTF8 chars for messages and contacts... or a fast one that does not work correctly for UTF8 chars ;) Mar 26 10:26:38 mrmoku: Well we don't have to ask that Mar 26 10:26:42 As currently we don't support UTF8 Mar 26 10:26:56 So if the new unstable is more reliable and doesn't introduce new weird stuff Mar 26 10:26:59 I'd say go for it Mar 26 10:27:07 Just write a short changelog for the blog and ML :-) Mar 26 10:28:35 Ainulindale: There is one thing we should investigate before... somebody complained to not receive all SMS in unstable Mar 26 10:31:42 Ok then, that has to be verified :-) Mar 26 10:31:49 (That's what I'm talking about) Mar 26 10:31:53 Correct path to do that is Mar 26 10:32:06 Post about the new unstable on the ML/blog and ask for feedback about what you think might break Mar 26 10:56:34 freesmartphone.org: 03mickey 07libeflvala * r6a3f0fb6456c 10/ (tests/testelementary.vala vapi/evas.vapi): vapi: add init and shutdown to evas.vapi Mar 26 10:58:06 mickey|zzZZzz: aha, I already wanted to complain :) Mar 26 10:58:52 mickey|cafe o/ Mar 26 10:59:15 mickey|cafe: just so you know SHR-WugBranler is here and active, and I can give you the conf to set it up for FSO Mar 26 10:59:27 (because right now it's hosted at home) Mar 26 11:03:32 (it's the bot for trac) Mar 26 11:04:34 mickey|cafe: Hotel room at FSOSHRUDCOn or sleeping bag in the seminar room or camping? :) Mar 26 11:05:04 Ainulindale, mrmoku, ptitjes, DocScrutinizer, raster: Same question for you. Mar 26 11:06:17 stefan_schmidt: I'm old and I need some comfort. I'd like a hotel room Mar 26 11:06:52 mickey|cafe: That was my guess already. noted. Mar 26 11:06:59 * mrmoku will sleep in his car :-) Mar 26 11:07:22 :) Mar 26 11:07:40 * mirko-paroli throws a happy good morning in to the room Mar 26 11:07:51 stefan_schmidt: for me it's hotel room Mar 26 11:08:13 Ainulindale: ok, noted Mar 26 11:08:18 stefan_schmidt: thanks a lot for handling that Mar 26 11:08:29 mrmoku: ok, so no hotel room for you. :) Mar 26 11:08:30 I'm noting that you have a good point. Mar 26 11:08:41 ;) Mar 26 11:08:42 stefan_schmidt: a parking lot though :-) Mar 26 11:08:49 All the people having good points will be rewarded by spanking. Mar 26 11:08:57 morning mirko-paroli Mar 26 11:09:32 hi mickey|cafe, back in GER ? Mar 26 11:09:41 ya, for a bit Mar 26 11:09:42 *phew* Mar 26 11:09:47 * mickey|cafe exhausted Mar 26 11:10:01 mickey|cafe: tell me about it, got in a 4 this morning Mar 26 11:10:25 heh Mar 26 11:10:30 we are crazy Mar 26 11:10:40 By the way mickey|cafe Mar 26 11:10:50 yes, but it's fun :) Mar 26 11:11:01 ptitjes asked if it would be any good to do tee-shirts for the "core" team for FSOSHRUDCON Mar 26 11:11:18 Just to remember it and make sure that everybody knows who we are :-) Mar 26 11:11:26 Ainulindale: ya, that'd rock Mar 26 11:11:43 So we'll need mensurations then :-> Mar 26 11:11:50 (measurements?) Mar 26 11:12:23 L usually fits here Mar 26 11:12:59 * mrmoku checks the label of his t-shirt... Mar 26 11:13:14 XXL Mar 26 11:13:57 * stefan_schmidt uses M for his body :) Mar 26 11:15:08 well XL will suffice :-) Mar 26 11:15:57 * stefan_schmidt is away for lunch to fill up his stomach for M. later. Mar 26 11:18:17 mickey|cafe: if you have some time a quick runthrough of how one would write and eflvala app with glib and e loop would be great Mar 26 11:18:44 i will when seamless integration is done Mar 26 11:18:48 so far I just don't have any glib loop and only 1 thread. But e.g. receiving dbus signals won't work this way Mar 26 11:18:49 that's what I'm working on atm. Mar 26 11:19:01 you can see the first bits in the repository Mar 26 11:19:09 I'm folloing it :) Mar 26 11:19:48 i will write a simple example that does something with FSO over dbus to illustrate Mar 26 11:20:00 But e.g. I don't seee yet how I would pass signals from one thread to the other Mar 26 11:20:12 that's what i mean with quick runthrough Mar 26 11:20:33 but I'l wait a few more commits and ask again ;-) Mar 26 11:22:28 that's done with my CommandQueue Mar 26 11:22:45 ya, give me a bit more time Mar 26 11:22:49 and then it'll show Mar 26 11:23:05 the basic idea is a queue and a fifo Mar 26 11:23:12 over the queue you push commands Mar 26 11:23:19 the fifo you use to indicate that there's a new command waiting Mar 26 11:23:23 so everything goes polling-free Mar 26 11:23:28 I see, sounds good Mar 26 11:23:51 ya, i developed this technique 7 years ago when there was no other polling-free inter-thread-comm-framework Mar 26 11:23:57 now 7 years later there's still nothing :D Mar 26 11:24:03 so I'm porting it to glib now Mar 26 11:26:16 mickey|cafe: pipe() + select() ? Mar 26 11:26:23 (polling free!) Mar 26 11:26:24 :) Mar 26 11:26:29 yep Mar 26 11:26:50 really hoped there's something built-in glib these days Mar 26 11:26:55 but apparantly not Mar 26 11:27:11 aaah Mar 26 11:27:21 mind u Mar 26 11:27:24 Qt is much better in that regard Mar 26 11:27:27 pthread_cond_wait() should allow for this Mar 26 11:27:34 they have thread-save Qt signals Mar 26 11:27:43 being able to "cross" the thread boundary Mar 26 11:27:51 aaah Mar 26 11:36:57 mickey|zzZZzz: still want to switch off charging on gta02 while having usb connected? Mar 26 11:37:05 mickey|cafe: still want to switch off charging on gta02 while having usb connected? Mar 26 11:40:08 mickey|zzZZzz: ok, thanks for the update Mar 26 11:51:59 * mwester wonders if there's an updated testing to test yet Mar 26 11:57:33 mirko-paroli: ? Mar 26 11:58:03 mickeyl: When we met in Braunschweig you said you'd like to test Mar 26 11:58:11 hoe long the 02 runs Mar 26 11:58:33 do a echo 0 > /sys/class/power_supply/usb/device/usb_curlim Mar 26 11:58:39 aah, that one Mar 26 11:58:42 completely forgot about that Mar 26 11:58:44 thanks Mar 26 11:58:50 and it stops charging while being connected to usb Mar 26 11:58:55 I guess that should land in the neo-specific powerclass Mar 26 11:59:09 hi mickey Mar 26 11:59:11 you can check if it is charging with a cat /sys/class/power_supply/battery/uevent Mar 26 11:59:15 hey pwgen Mar 26 11:59:32 if you want to start charging again do a echo 500 > /sys/class/power_supply/usb/device/usb_curlim Mar 26 11:59:47 ok, that's handy Mar 26 12:00:11 mickey: is there an kexecboot image for the neo or should i build my own ? Mar 26 12:00:43 meeeh Mar 26 12:00:45 mickey: .. btw my 512 mb sd works fine .. Mar 26 12:00:45 pwgen: no idea, there should be enough infrastructure in OE to do that and I've read somethign about kexecboot on the OM lists Mar 26 12:00:54 pwgen: at least, good. Mar 26 12:02:15 ptitjes? Mar 26 12:02:42 its mentioned at the qi doc . but it seems there is no build yet ..(:-(( Mar 26 12:02:53 raster: can I do something like smart_callback_add( "elm,state,toggle,on", cb_gsmpower_on ); on a toggle? Mar 26 12:04:21 mirko-paroli: did you answer to stefan about the hotel? Mar 26 12:04:39 Ainulindale: ehm...I only caught the last part of the discussion Mar 26 12:04:59 so not sure what the question was Mar 26 12:05:01 :( Mar 26 12:05:18 spaetz: no - but u can do it for "changed" Mar 26 12:05:33 spaetz: seel Elementary.h Mar 26 12:05:36 it has docs there Mar 26 12:05:41 EAPI Evas_Object *elm_toggle_add(Evas_Object *parent); Mar 26 12:05:42 ... Mar 26 12:05:46 /* smart callbacks called: Mar 26 12:05:47 * "changed" - the user toggled the state Mar 26 12:05:47 */ Mar 26 12:05:58 they list standard styles - if it supoprts styles Mar 26 12:05:59 and signals Mar 26 12:06:06 well smart callback signals Mar 26 12:06:20 mirko-paroli: hotel or camping for you? Mar 26 12:06:34 raster: same question for you although I know you'd rather be in the hotel Mar 26 12:06:39 Ainulindale: I like sleeping in beds, so Hotel Mar 26 12:06:47 stefan_schmidt: free highlight Mar 26 12:06:56 yep, but changed is not called when something sets state_set () :) Mar 26 12:07:08 camping? dude. i've flown all the way to europe. i'm hella-not going to stuff myself into a sleeping bag in a tent! Mar 26 12:07:08 :) Mar 26 12:07:16 mirko-paroli: anyway it's not that expensive Mar 26 12:07:19 on the other hand the code shows that signals for cat /sys/class/power_supply/usb/device/usb_curlim Mar 26 12:07:19 raster: :-) Mar 26 12:07:22 stefan_schmidt: +2 for rooms Mar 26 12:07:22 500 Mar 26 12:07:23 oops Mar 26 12:07:25 root@om-gta02 ~ $ cat /sys/class/power_supply/battery/uevent Mar 26 12:07:26 hmm Mar 26 12:07:33 seems this scale cache is coming up a treat Mar 26 12:07:37 its getting lots of hits Mar 26 12:07:49 signals for on/off are issued even when I manually do a state_set() Mar 26 12:09:52 p: 586, h: 112267, m: 12750, n: 26725 Mar 26 12:10:03 spaetz: for toggles? Mar 26 12:10:45 hmm Mar 26 12:10:47 shouldnt Mar 26 12:10:50 i need to check into that Mar 26 12:10:53 ok Mar 26 12:11:09 12750 cache misses (going the old drawing path of scale on the fly) Mar 26 12:11:14 and 112267 hits Mar 26 12:11:20 nice Mar 26 12:11:28 ie use cached scaled copy Mar 26 12:11:35 this is scrolling around in illumes app list Mar 26 12:11:57 586 is the # of cached copies of images it's had to populate Mar 26 12:12:08 (over time) Mar 26 12:12:17 raster: moin! :) Mar 26 12:12:26 and 26725 is the # of draw ops where it doesnt need scaling at all Mar 26 12:12:41 thats pretty decent in a real life test Mar 26 12:12:48 lets tyr elementary Mar 26 12:13:33 stefan_schmidt: just updated the wiki page for the basic info about who would like a room Mar 26 12:14:03 Ainulindale: great, thanks Mar 26 12:14:13 * stefan_schmidt is just back from lunch Mar 26 12:16:07 p: 913, h: 336692, m: 29237, n: 248697 Mar 26 12:16:22 913 scaled sections from src images populated Mar 26 12:16:33 248697 draw ops that dont need scaling Mar 26 12:16:38 29237 scache misses Mar 26 12:16:46 cache misses (thus old style long-path draw) Mar 26 12:16:58 and 336692 cache hits Mar 26 12:16:58 thats pretty nice Mar 26 12:17:18 indeed Mar 26 12:17:19 good work Mar 26 12:17:21 ptitjes: you should add Nicolas to the list if he's not already in it Mar 26 12:17:29 for smooth scaling AND alpha blendign a cache hit can be this level of speedup: Mar 26 12:17:33 Image Blend Smooth Same Scaled: 32.56 428.39 Mar 26 12:17:35 ie 10x Mar 26 12:17:40 well more than Mar 26 12:17:41 14x Mar 26 12:18:07 for solid (but smooth) scaling (no alpha) Mar 26 12:18:11 Image Blend Smooth Solid Same Scaled: 325.84 1228.33 Mar 26 12:18:20 3.7x Mar 26 12:18:35 blending without smooth - Mar 26 12:18:36 Image Blend Nearest Same Scaled: 215.63 434.75 Mar 26 12:18:37 about 2x Mar 26 12:18:42 and Mar 26 12:18:43 Image Blend Nearest Solid Same Scaled: 1074.81 1205.86 Mar 26 12:18:56 for no blending, nearest (non smooth) so about 1.2x Mar 26 12:19:07 this smells to me of a win Mar 26 12:19:45 so far cant detect any leaks Mar 26 12:19:51 seems to clean up after itself appropriately Mar 26 12:19:59 it doesnt know to free orig pixels tho Mar 26 12:20:00 hehe, hopefully Mar 26 12:20:07 i want it to do that Mar 26 12:20:11 eg u load a 1600x1200 image Mar 26 12:20:15 but its a "wallpaper" Mar 26 12:20:22 u only ever use it scaled to 640x480 Mar 26 12:20:28 its never used at other sizes Mar 26 12:20:34 (common for bg's) Mar 26 12:20:47 so i'd like it to free the original 1600x1200 pixels once it has a cached copy it's using Mar 26 12:20:56 (cached scalec copy) Mar 26 12:20:59 i need to add that logic Mar 26 12:21:09 sounds sensible ;) Mar 26 12:21:16 thats what's missing - this should make things like the app list scroll much faster Mar 26 12:21:19 as would anything else Mar 26 12:21:25 as not scaling costs come down Mar 26 12:21:43 and no one has to special-case and pre-scale their stuff as its all picked up and handled on the fly Mar 26 12:21:43 :) Mar 26 12:22:19 (just lettign u guys know as it affects u guys a lot with etk, elementary, edje evas etc. being used) Mar 26 12:22:30 so u basically get some nice optimisations/speedups for free Mar 26 12:23:49 for free always good ;D Mar 26 12:23:59 well Mar 26 12:24:03 it does cost some extra ram Mar 26 12:24:08 so far its a separate cache Mar 26 12:24:13 fixed right now @ 4mb Mar 26 12:24:56 bearable Mar 26 12:25:07 raster: yep for toggles. mmh but re-reading the code. "changed" should actually be called Mar 26 12:25:16 i've put in code to attempt to stop cache flopping Mar 26 12:25:24 it cached images going in and out of cache all the time Mar 26 12:25:27 in http://svn.enlightenment.org/svn/e/trunk/TMP/st/elementary/src/lib/elm_toggle.c state_set emits edje_object_signal_emit(wd->tgl, "elm,state,toggle,on", "elm"); Mar 26 12:25:29 as the cache thrashes Mar 26 12:25:50 will retest is "changed" is being issued Mar 26 12:25:59 spaetz: state vs action Mar 26 12:25:59 :) Mar 26 12:26:04 maybe its theme Mar 26 12:26:06 need to check Mar 26 12:26:31 aaah Mar 26 12:26:33 the theme emits Mar 26 12:26:48 thats why Mar 26 12:26:51 need to fix that Mar 26 12:27:37 k Mar 26 12:29:04 * PaulFertser has just spent like 4 hours just to get dummy glib dbus gobject application working. "I'm not like them, But I can pretend... I think I'm dumb or maybe just happy" Mar 26 12:30:08 Ainulindale: Who wanted to announce FSOSHRUDCON at om-devel? We have 7 seats free. Anyone else we want to have in for sure? Mar 26 12:30:11 PaulFertser: ;D Mar 26 12:30:37 hmm, DocScrutinizer is no longer on the list. Mar 26 12:30:50 huh? Mar 26 12:30:57 DocScrutinizer: You will come? Mar 26 12:31:05 huuh? Mar 26 12:31:15 stefan_schmidt: 7? Mar 26 12:31:16 * stefan_schmidt speaks slowly Mar 26 12:31:18 Why 7? Mar 26 12:31:21 context plz Mar 26 12:31:30 DocScrutinizer: Will you attend FSOSHRUDCON? Mar 26 12:31:31 damn did I erase him? Mar 26 12:31:39 Ainulindale: could be Mar 26 12:31:41 stefan_schmidt: though... he does not need a hotel room... DocScrutinizer never sleeps :P Mar 26 12:31:46 stefan_schmidt: I intended to Mar 26 12:31:47 Yeah I erased DocScrutinizer Mar 26 12:31:52 DocScrutinizer: hotel room? Mar 26 12:31:57 Oh nevermind, he deserves it Mar 26 12:31:58 * stefan_schmidt adds him back Mar 26 12:32:13 stefan_schmidt: Ok then complete "Deubeuliou" insteads of "Deubeulio" please Mar 26 12:32:21 Ainulindale: ok Mar 26 12:32:36 probabaly yes. tho no double room Mar 26 12:32:39 Ainulindale: 7, now 6, because I git the info that we have space for 20 people. Mar 26 12:32:51 Maybe 22 Mar 26 12:33:10 DocScrutinizer: single. hmm let me see what I can do Mar 26 12:33:28 stefan_schmidt: and we're already 14? Mar 26 12:33:30 prolly gonna get a regular room Mar 26 12:33:49 DocScrutinizer: Ah you're playing single then? Mar 26 12:33:56 Pfff! :-) Mar 26 12:34:06 You don't want to opensource your room! Mar 26 12:34:18 my gf doen't want to ancompany ;) Mar 26 12:35:06 well neither is mine, but I'll share the room Mar 26 12:35:13 It'll be less expensive Mar 26 12:35:20 Ainulindale: there's some business secrets that should be kept undisclosed XD Mar 26 12:35:20 Ainulindale: yes, already 14 Mar 26 12:36:49 DocScrutinizer: heh :-) Mar 26 12:36:59 DocScrutinizer: Such as your wooden leg? Mar 26 12:37:01 (Woops!) Mar 26 12:37:51 more like my jeckil&hide nature ;) Mar 26 12:37:53 spaetz: nice mail by the way Mar 26 12:38:07 DocScrutinizer: well Mar 26 12:38:10 You'll see me in the morning Mar 26 12:38:14 Hopefully after my first coffee Mar 26 12:38:27 And you won't be so afraid of your own nature Mar 26 12:39:01 Ainulindale: you are welcome. I planned to do something like this anyway. Mar 26 12:41:58 spaetz: well I know I'm annoying with that but I genuinely think this is a must do :-) Mar 26 12:42:42 yes it is annoying and I agree that it is a must do :-) Mar 26 12:42:56 BTW, I found a cool .sid file as a ring tone :) Mar 26 12:43:11 * spaetz will write a mail to the list asking for feedback Mar 26 12:43:56 Well I'm already worrying about this new ringtone Mar 26 12:44:03 Did you talk about that with mickeyl? Mar 26 12:44:10 Just to know if I have to say no right away Mar 26 12:44:36 you better should. No I didn't ask him. But as it is a genuine 80's sound, he might even like it :) Mar 26 12:44:57 * mrmoku would even fork SHR to get a different ringtone :P Mar 26 12:45:09 heh :-) Mar 26 12:45:18 http://sspaeth.de/uploads/tmp/2112.sid Mar 26 12:45:30 Anyway we should deliver three or four of them as a default Mar 26 12:45:50 From an abandonware C64 computer game (Llamatron 2112) Mar 26 12:46:54 * stefan_schmidt thinks we need something discreet no noise... Mar 26 12:47:22 hehe Mar 26 12:47:24 (I can't listen to sid files) Mar 26 12:48:14 it's freaking annoyance. no standard player dan playback Mar 26 12:48:30 yum install sidplay Mar 26 12:48:53 totem played it just fine Mar 26 12:49:04 I just needed to teach ubuntu to use totem for it Mar 26 12:49:13 Ubunwhat? Mar 26 12:49:22 hehe Mar 26 12:49:29 <[HtR]_0xEF> Hi. Anyone tried using Bluetooth headset with freerunner? Want to use it all the time because the other side hears noise all the time.. Mar 26 12:49:32 *cough* Mar 26 12:49:56 ah ok listening to it Mar 26 12:49:58 it's awful Mar 26 12:50:02 Please kill me already. Mar 26 12:50:07 * mrmoku shrugs Mar 26 12:50:17 Don't you guys have any tastE? Mar 26 12:50:18 [HtR]_0xEF: http://wiki.openmoko.org/wiki/Manually_using_Bluetooth#Once_Again.2C_Bluetooth_Headset_on_Freerunner Mar 26 12:50:29 (FYI my ringtone is "willie the pimp" from frank zappa) Mar 26 12:50:51 <[HtR]_0xEF> PaulFertser: Aha I tried all three ways lol and nothing helped me.. Mar 26 12:51:05 Ainulindale: cooooool Mar 26 12:51:19 [HtR]_0xEF: all three ways? I'm telling you the only correct way. Mar 26 12:51:21 one of my favs Mar 26 12:51:51 j.l.ponti Mar 26 12:51:58 DocScrutinizer: heh Mar 26 12:52:03 Ponty is a jerk though Mar 26 12:52:40 possibly Mar 26 12:53:11 By the way DocScrutinizer Mar 26 12:53:14 <[HtR]_0xEF> PaulFertser: I mean I've tried three sections named "Headset Audio", "Bluetooth Headset on Freerunner", "Once Again, Bluetooth Headset on Freerunner". I tried to setup A2DP and it did not work too aughh.. Mar 26 12:53:14 I found a gift for you Mar 26 12:53:16 http://cgi.ebay.fr/Balancoire-damour-BONDAGE-FETISH-BDSM-SEX-GAY_W0QQitemZ180335842330QQcmdZViewItemQQptZFR_CP_Beaute_Bien_%C3%AAtre_massage_th%C3%A9rapies_naturelles?hash=item180335842330&_trksid=p3913.c0.m22 Mar 26 12:53:28 At your age it's necessary to have something like that to spice things up Mar 26 12:53:52 [HtR]_0xEF: either you provide the details how exactly the third method didn't work for you, or i won't be able to help you. Mar 26 12:53:57 hehe Mar 26 12:54:05 "Love Swing", the name sounds great Mar 26 12:54:50 by the way stefan_schmidt, alphaone Mar 26 12:55:30 would you know how it's possible for a student having an engineering degree (abitur + 6) from russia to go to Germany to study things such as psychology or sociology, not starting from the first year? Mar 26 12:55:38 I.e. validate some ECTS Mar 26 12:55:50 (it's for my girlfriend's sister) Mar 26 12:56:16 Ainulindale: pretty sure you wouldn't save any time Mar 26 12:56:21 eh, not really, sorry Mar 26 12:56:32 It's hard enough to get credits if you have studies a similar topic Mar 26 12:56:42 well she studied this kind of things for our diploma Mar 26 12:56:44 that's why Mar 26 12:56:52 (she can prove it) Mar 26 12:56:56 <[HtR]_0xEF> PaulFerster: sorry I did not try the thir way because of this: "I managed to connect my bluetooth headset to the Freerunner and route calls to it. However, I couldn't get an actual voice over it, just crackling." I've decided not to try it if it's not possible to hear anything but crackling. Mar 26 12:56:59 Ainulindale: You would need to talk to the University and check whether they will acknowledge previous lectures elsewhere Mar 26 12:57:02 Ainulindale: me neither. It's even difficult inside one country how this things are handled :/ Mar 26 12:57:13 alphaone: ok I'll tell her that then Mar 26 12:57:17 Sheesh Mar 26 12:57:24 Students are really not the same they were back in my days Mar 26 12:57:29 Back in my days they knew stuff Mar 26 12:57:36 Ainulindale: That will be different from University to University Mar 26 12:57:36 That's out Mar 26 12:57:43 Now they just lie lazily around developing in python Mar 26 12:57:46 Not even in a proper language Mar 26 12:57:47 <[HtR]_0xEF> Is there any way to filter out the noise (bzzzzzz) that the other side hears while using the phone? Mar 26 12:58:05 Ainulindale: proper language like french? ;-) Mar 26 12:58:18 [HtR]_0xEF: only hardware buzz fix or bluetooth. You'd try the third method. It should work. YMMV. Mar 26 12:58:19 Meh! Mar 26 12:58:20 [HtR]_0xEF: where do you live? Mar 26 12:58:48 [HtR]_0xEF: if he starts asking you about what you're wearing, be wary Mar 26 12:58:49 htrxxxxx: do buzzfix. sorry can't answer your nick Mar 26 12:59:19 <[HtR]_0xEF> alphaone: Eastern Europe Mar 26 12:59:46 <[HtR]_0xEF> Ainulindale: I doubt he'll get here where I live lol )) Mar 26 12:59:58 He's resourceful Mar 26 13:00:06 Ainulindale: Well, I was getting worried about " "Love Swing", the name sounds great Mar 26 13:00:06 by the way stefan_schmidt, alphaone" Mar 26 13:00:36 haha Mar 26 13:06:33 DocScrutinizer: by the way if you like Zappa Mar 26 13:06:39 Did you go and see ZPZ? Mar 26 13:06:46 (Zappa Plays Zappa) Mar 26 13:06:52 nope Mar 26 13:06:56 (Dweezil playing his father's songs) Mar 26 13:06:59 alas not Mar 26 13:07:12 Well last year, or the year before that, they were touring with Steve Vai & Terry Bozzio Mar 26 13:07:22 and after that Ray White (or Ike Willis, can't recall) Mar 26 13:07:27 I went to see them, it was pretty nice :-) Mar 26 13:07:44 (DISCOOOOO BOY!) Mar 26 13:07:50 [HtR]_0xEF: Well, as soon as I sort things out you could send it to me and I can perform the buzz-fix Mar 26 13:08:38 By the way, I have two devices to fix myself. Mar 26 13:08:43 I have to remember to bring them. Mar 26 13:08:47 alphaone, stefan_schmidt, mickeyl Mar 26 13:08:55 We should formally talk about the written material for the convention Mar 26 13:09:11 Sure, bring them Mar 26 13:09:13 What we want to write, divide things up & all Mar 26 13:09:27 And put them online afterwards for all to see :-) Mar 26 13:10:18 mickeyl: might possibly do a free speech not writing down too much ;-) Mar 26 13:10:43 :) Mar 26 13:10:49 * mickeyl loves keynote-style presentations Mar 26 13:10:52 huge pictures Mar 26 13:10:57 little text Mar 26 13:10:59 Well I think if we can have some presentations, some written material, it'll be good to put that online to show what we're doing an all Mar 26 13:11:09 It's an incitative for us Mar 26 13:11:13 FSO lacks that a lot IMHO Mar 26 13:11:22 (tutorials, architecture schemas, etc...) Mar 26 13:11:32 yep Mar 26 13:11:43 we talked about that SHR wise with ptitjes Mar 26 13:11:44 <[HtR]_0xEF> PaulFerster: So I'll have to switch to the FSO framework.. Maybe that was the problem -- I was using QTX framework..\ Mar 26 13:11:46 By the way mrmoku Mar 26 13:11:56 Would you like to do a talk yourself? Mar 26 13:12:12 Maybe talk about how we should continue our overlay without impairing the OE migration? Mar 26 13:13:25 We should really talk about all the subjects Mar 26 13:13:32 Maybe raster would be interested in talking too? :-> Mar 26 13:14:14 ping him Mar 26 13:14:42 I keep doing that but as soon as I do it he hides Mar 26 13:15:10 hehehe Mar 26 13:15:17 I thought australian folks would be more kangaroo like than ostrich like Mar 26 13:15:59 he can't hide his german heritage Mar 26 13:16:05 * DocScrutinizer hides Mar 26 13:17:24 Ainulindale: no thanks... I'm better in listening ;) Mar 26 13:17:35 mrmoku: ok :-) Mar 26 13:18:07 ah morphis Mar 26 13:18:10 :) Mar 26 13:19:28 morphis: Mar 26 13:19:46 heyho Mar 26 13:19:47 raster: you'd like to do a talk at tfff-sos-shrudetalcon Mar 26 13:19:50 morphis: is there a way of reliably closing the kbd in elm ? Mar 26 13:19:50 ? Mar 26 13:19:58 morphis: in py-elm ofc Mar 26 13:20:45 mirko-paroli, you can require the keyboard Mar 26 13:20:46 * raster wakes up Mar 26 13:20:55 DocScrutinizer: i have something to talk about? Mar 26 13:20:59 i could Mar 26 13:21:01 That's the question Mar 26 13:21:02 morphis: the suspects keyboard_mode_set(gui.elementary.ELM_WIN_KEYBOARD_OFF) and eyboard_mode_set(0) don't seem to work Mar 26 13:21:04 dunno Mar 26 13:21:05 mirko-paroli, you want to disable it or to close? Mar 26 13:21:10 Would you be interested in talking about something raster? Mar 26 13:21:11 morphis: close it Mar 26 13:21:11 hm Mar 26 13:21:27 Ainulindale: maybe - as long as i dont have to prepare too much for it Mar 26 13:21:43 raster: Ainulindale asked if you have some talk and paper for it Mar 26 13:21:49 i am thinking maybe something tutorial-based? Mar 26 13:21:54 morphis: I also thought of unfocusing the entry, but unfocus doesn't work either Mar 26 13:21:56 raster: why not? Mar 26 13:21:57 DocScrutinizer: oh.. an existing one? Mar 26 13:22:03 with keyboard_mode_set(gui.elementary.ELM_WIN_KEYBOARD_OFF) you set the keyboard mode (like char, alphanum etc.), closing should be done by keyboard_mode_set Mar 26 13:22:05 nah Mar 26 13:22:07 raster: I was just interested in knowing if you were willing to talk about something Mar 26 13:22:13 sure Mar 26 13:22:14 Related to the subject of the convention of course Mar 26 13:22:24 WHERE'S MY COFFEE?? Mar 26 13:22:29 morphis: k, I think I do both Mar 26 13:22:50 Ainulindale: i was thinking of givint a talk on fishing! Mar 26 13:22:51 :) Mar 26 13:22:53 giving Mar 26 13:22:53 :) Mar 26 13:23:01 morphis: keyboard_win_set(0) ? Mar 26 13:23:01 Yeah I was wondering about that too Mar 26 13:23:12 mirko-paroli, I currently see, there was a bug in pyhon-element which makes keyboard_win_set not working Mar 26 13:23:26 morphis: ah ok Mar 26 13:23:28 mickeyl: by the way Mar 26 13:23:34 Yesterday I was talking with alphaone about a use case Mar 26 13:23:38 hehe, FR used for bait Mar 26 13:23:44 Which justified the existence of org.shr.Usage Mar 26 13:23:44 mirko-paroli, I am currently commiting the change to svn Mar 26 13:23:47 morphis: was just curious if it was me or the bindings being off a little Mar 26 13:23:52 morphis: superb Mar 26 13:23:54 mickeyl: that is, a lack of functionnality in ousaged Mar 26 13:23:59 * mirko-paroli heroes morphis once again :) Mar 26 13:24:12 We have no way currently to sever a link between a requester and its resource if we're not the requester itself Mar 26 13:24:19 Or if we have, I don't know about it Mar 26 13:24:38 mirko-paroli, :D Mar 26 13:25:05 doesn't sound like we want that Mar 26 13:25:06 why? Mar 26 13:25:25 mirko-paroli, checkout rev 39725 Mar 26 13:25:28 mickeyl: well if we have two users for a resource Mar 26 13:25:35 morphis: sry for being a pain about this, I am trying to understand cython so that I can send patches rather than calling for help ... Mar 26 13:25:52 and we want to have a master settings app which allows to stop a specific resource usage Mar 26 13:25:55 no matter :) Mar 26 13:25:57 morphis: k, I'll have to ask angus to build an image than Mar 26 13:26:00 We can kill all the resource users Mar 26 13:26:03 But not a specific one Mar 26 13:26:22 (here, the use case was that ophonekitd requests the GSM resource, if I want to make him release it, I have to send it a signal somehow) Mar 26 13:26:30 mirko-paroli, you want to build a new image for this little change? Mar 26 13:26:34 Ainulindale: ??¿? Mar 26 13:26:43 sounds like correct behaviour Mar 26 13:26:48 (I would find it more practical to be able to say to FSO that the resource usage should be stopped) Mar 26 13:26:50 requesting someone to release it Mar 26 13:26:56 And then, ophonekitd would handle that event Mar 26 13:26:59 DocScrutinizer: was? Mar 26 13:27:04 morphis: well, no, I would switch to the first elm app as well Mar 26 13:27:14 ah Mar 26 13:27:24 and replace the current phoneLogs with its newer and more shiny version :) Mar 26 13:27:49 well, you will get a signal if you use the resource policy stuff Mar 26 13:27:57 I'll pass that to mickeyl to answer ;) Mar 26 13:28:02 mickeyl: yeah but that will stop all the resource usage, right? Mar 26 13:28:14 (if I set it to disabled) Mar 26 13:28:16 Ainulindale: yes. Mar 26 13:28:25 mickeyl: and if I only want to stop one of the users, then? Mar 26 13:28:31 I have to act as if I was him Mar 26 13:28:33 then you have to ask him Mar 26 13:28:39 And that may prove to be not practical Mar 26 13:28:42 why do you want to stop it in the first place? Mar 26 13:28:50 I had an use case I forgot about Mar 26 13:28:58 lol Mar 26 13:29:50 Ainulindale: Hehe Mar 26 13:30:14 well I talked about that with alphaone yesterday so he might be able to tell you what I meant Mar 26 13:30:37 mickeyl: and anyway, if we don't have any mean to stop a resource usage given the user, maybe a hash or something and the resource Mar 26 13:30:48 that means every layer on top of FSO should implement something like that Mar 26 13:30:53 Seems to me a bit redundant Mar 26 13:31:03 (in architectures like SHR, that is) Mar 26 13:31:42 like i said, i still don't see the use for it Mar 26 13:31:47 so i would say noone needs to implement that Mar 26 13:32:09 Well in SHR if I want to release the GSM resource in ophonekitd (that is to say only the call handling) Mar 26 13:32:17 I have to release it as ophonekitd Mar 26 13:32:27 Let us say I want to have GPRS but not basic GSM Mar 26 13:32:52 If I set the policy to disabled Mar 26 13:33:06 It's a hard stop for every user Mar 26 13:33:37 (I should come up with a slightly more elaborate use case, I know) Mar 26 13:35:43 Ainulindale: your use case yesterday was the camera me thinks Mar 26 13:35:59 yeah but I can't recall the point of it Mar 26 13:36:17 * mrmoku didn't get the point in the first place :P Mar 26 13:36:37 well anyway any distro based on FSO with a central GSM handling daemon will have this issue Mar 26 13:36:43 If they want to stop the resource usage of that daemon Mar 26 13:36:50 They will have to implement a signaling system to handle the release Mar 26 13:37:00 I don't think it's convenient as the framework should be able to handle that IMHO Mar 26 13:38:15 I'm still not convinced. But make an API proposal and then lets talk Mar 26 13:39:05 Heh :-) Mar 26 13:39:13 Ainulindale: Maybe have more usage methods than enabled and disabled? We talked about that some time ago already... Mar 26 13:39:17 Yes Mar 26 13:39:34 I just think that Mar 26 13:39:38 Like being able to request GPRS and GSM-Phone Mar 26 13:39:39 As we can have several users for a resource Mar 26 13:39:53 We may need ways to handle these users Mar 26 13:40:16 well, i think it's conceptually wrong to disable something if you're not the user Mar 26 13:40:25 (I'm not sure my idea is pertinent but I find the fact I had to write org.shr.usage not convenient) Mar 26 13:40:29 mickeyl: On that I agree Mar 26 13:40:31 so i don't see why there is a problem with asking the user to release it Mar 26 13:40:33 Though I think the use case is a bit constructed... Mar 26 13:40:38 That's why I was thinking about something more, that is a change of ownership Mar 26 13:40:43 Transfer of ownership maybe? Mar 26 13:41:06 So something as an handshake over events for a software to say it's giving its ownership away Mar 26 13:41:30 huh? Mar 26 13:41:41 (Wait a minute) Mar 26 13:42:05 Now you're talking about making sure events/signals are handled by an app? Mar 26 13:44:33 if there's no template in posix we probably don't need it ;-) Mar 26 13:46:26 freesmartphone.org: 03mickey 07framework * r527a16a240cc 10/etc/freesmartphone/opreferences/schema/phone.yaml: opreferencesd: add schema entries for ring-vibration and message-vibration Mar 26 13:49:44 THere again Mar 26 13:49:45 So Mar 26 13:49:47 I was saying Mar 26 13:49:58 The problem conceptually speaking is ownership Mar 26 13:50:04 A requesters "owns" his resource Mar 26 13:50:16 So what is lacking, IMHO, is a way to transfer this owernship Mar 26 13:50:21 s/ern/ner/ Mar 26 13:50:21 Ainulindale meant: So what is lacking, IMHO, is a way to transfer this ownership Mar 26 13:50:31 Being able to say that someone/something is taking over the ownership Mar 26 13:50:41 That could be handled by "events" in the form of a handshake Mar 26 13:50:56 THat's only what I was saying Mar 26 13:51:43 In how far is that different from one releasing a resource and the other one requesting it? Mar 26 13:52:12 mrmoku: because you have to tell to the first one to release the resource Mar 26 13:52:16 and meanwhile you'll lack the service Mar 26 13:52:48 so telling the first one to release implies architecture/development Mar 26 13:52:57 and the lack of service implies my whining Mar 26 13:53:00 so the other one could request the resource first and then tell the first owner that he can release it, no? Mar 26 13:53:13 mrmoku: yeah but you still have to write the piece of code to tell him to release Mar 26 13:53:48 This might be harmful when powering up a device is power consuming Mar 26 13:53:58 usually you pass a handle for things like that Mar 26 13:53:59 (Only example that comes to mind is a camera, I don't know if this is the case) Mar 26 13:54:19 In fact, to make a metaphor Mar 26 13:54:36 What I would like is the same thing as in GSM Mar 26 13:54:41 Being able to roam across cells Mar 26 13:55:02 Here I would like t o be able to transfer the ownership on a resource over users, using a "handshake" between processes Mar 26 13:55:13 The power consumption issue/lack of service is a good example I think Mar 26 13:55:27 mickeyl: (are you still there? :-) ) Mar 26 13:55:32 something like ChangeResourceOwner(old_dbus_client, new_dbus_client)... Mar 26 13:55:42 I wouldn't do it that way Mar 26 13:55:44 I would do a Mar 26 13:55:44 which can be called by anyone Mar 26 13:55:52 "RequestOwnership(owner,resource)" Mar 26 13:55:59 Then it would raise an event for the owner asking him to answer Mar 26 13:56:13 And on approval, an event on the new requester to confirm the ownership Mar 26 13:56:18 (hence "handshake") Mar 26 13:56:52 Because right now we know how to share resources Mar 26 13:57:02 We don't know how to give resources away using the framework Mar 26 13:57:43 But again, this is only an idea which would make things a lot simpler to any daemon implementing framework Mar 26 13:58:09 In order for example to be able to actively stop a daemon handling and passing it over to another one Mar 26 13:58:26 I'm sure ptitjes would find a use case for that Mar 26 13:58:38 (I'm too tired to think about any) Mar 26 13:58:44 I don't Mar 26 13:58:51 You don't what? Mar 26 13:59:00 find any usecase Mar 26 13:59:09 That's because you're a hardware guy =) Mar 26 13:59:22 I have a stupid example Mar 26 13:59:28 If I have a recording software for a camera Mar 26 13:59:28 you're sure I am? Mar 26 13:59:43 And a software to handle video calls Mar 26 14:00:02 In order to reduce the CPU consumptions for image processing Mar 26 14:00:13 You might want to be able to transfer the ownership of the resource Mar 26 14:00:18 Between the two things Mar 26 14:00:21 If a video call comes in Mar 26 14:00:22 nope Mar 26 14:00:23 (for example) Mar 26 14:00:29 Nope what? :-) Mar 26 14:00:52 we need priority to request a resource, admittedly Mar 26 14:01:15 I'm not sure about that at all Mar 26 14:01:18 but we don't want to pass ownership the way you suggested Mar 26 14:01:23 Resource sharing as to be there Mar 26 14:01:24 * morphis is moving to another room ... Mar 26 14:01:45 But to me we would need to be able to stop one of the users' usage Mar 26 14:01:58 camera e.g. can't be shared reasonably Mar 26 14:02:06 If we don't do that, we need to implement a releasing system on top of any daemon implementing the resource Mar 26 14:02:16 (if we want to stop the handling in this daemon) Mar 26 14:02:49 so a videocall has to allocate resource camera on a higher prio than snapshot app Mar 26 14:03:49 this will suspend or revoke resource usage from snapshot, while allocation for videocall succeeds Mar 26 14:04:52 DocScrutinizer: and then when this software doesn't need the resource anymore? Mar 26 14:05:36 what is "this software"? there's been two of them in my example Mar 26 14:06:37 the second one Mar 26 14:07:08 videocall simply freeing the resource Mar 26 14:07:24 that doesn't answer to my question Mar 26 14:07:28 the question is how to handle this in snapshot Mar 26 14:07:31 would the first one get its resource again? Mar 26 14:08:03 we need to think about that Mar 26 14:08:07 (what you are talking about here is not priority, but mutual exclusion) Mar 26 14:08:12 (IMHO) Mar 26 14:08:32 Which may be useful in the end, but not sure now Mar 26 14:08:50 there's a couple of valid solutions to that. none of them comprises direct communication between the two apps Mar 26 14:10:27 most simple signaling to "first app" would be a EAGAIN until "second app" is freeing resource again Mar 26 14:12:16 well this doesn't solve my use case anyway Mar 26 14:12:29 What if I want to tell to the first app to release its resource on an external point of view? Mar 26 14:12:35 which usecase? Mar 26 14:12:43 Say, in a control panel or settings app **** ENDING LOGGING AT Fri Apr 03 18:33:43 2009