**** BEGIN LOGGING AT Mon Apr 30 02:59:56 2007 Apr 30 02:59:58 some engineer from TaiWan and mainland of China are working on it Apr 30 03:00:59 the on-screen keyboard even had a button allocated for Chinese input, but it doesn't work :p Apr 30 03:01:22 en, cool, input method would be a problem i think Apr 30 03:01:56 *has Apr 30 03:02:36 i did not find any chinese fonts described in oe & openmoko, so i asked about that:) Apr 30 03:02:40 Readon: i'm not sure, Linux on the Zaurus pda's supports japanese input very well, although Japanese may be easier, i'm not sure Apr 30 03:04:44 it remains a long way. Apr 30 03:07:40 whoo, qemu works again Apr 30 03:08:46 Are there any open source handwriting software for Chinese ? Apr 30 03:09:49 I'm sure there's something Apr 30 03:09:54 I think the Agenda did Hangul Apr 30 03:12:23 hrm, nope, I was wrong Apr 30 03:14:55 i don't know, but gtk-im has something called "kanji pad" among other input methods Apr 30 03:15:12 * balrog-kun is off Apr 30 03:16:41 hrm, xmerlin seems to be a general stroke-recognition engine Apr 30 03:16:51 suitable for a graffiti-like interface Apr 30 03:17:09 I think the openmoko should use SCIM as a common input method platform Apr 30 03:17:52 eg: Chinese/Korea/Japanese input , Arab input , Thai input ... Apr 30 03:21:52 Readon , r u from Chengdu ? Apr 30 03:23:16 falls_huang: how could you know? Apr 30 03:23:33 I can see your IP and search it :) Apr 30 03:23:48 I'm from SiChuan too . hehe Apr 30 03:24:00 nice to meet you Apr 30 03:24:03 :) Apr 30 03:24:12 my gtalk : falls.huang@gmail.com Apr 30 03:44:42 anyone able to get a USB connection to the qemu-openmoko? Apr 30 04:04:50 oh, nm, wiki says that's not done yet Apr 30 04:29:15 openmoko: 03laforge * r1874 10/trunk/oe/packages/linux/ (2 files in 2 dirs): add 2.6.20.10 Apr 30 05:06:08 (script) wiki RecentChanges: [[Talk:Input_Method]] Apr 30 05:10:06 are there any plans to have a successor which has actual buttons instead of a touch screen? given the horrible touchscreens on the "chocolate" and the fact that touch screens mean you have to LOOK at the phone to dial (bad when driving), real buttons are must for any phone i get Apr 30 05:10:06 macogw: See http://wiki.openmoko.org/wiki/FAQ (Hopefully this link answers your question. This autotrigger for new nick and question during first 15 minutes done only once.) Apr 30 05:13:49 macogw: check planet.openmoko.org for hints about clues about future devices Apr 30 05:13:57 things are vague notions at best :-\ Apr 30 05:17:21 This sounds cool. Apr 30 05:20:24 woah gnome's embedded thing includes gconf? is there a good reason for gconf's existence? i've heard it called the windows registry ported to linux because it's so convoluted. kde doesnt have anything like it, so why do we gnome users have that confusing (scary) thing? Apr 30 05:20:42 id hate to try to modify gconf from a mobile device Apr 30 05:20:50 too small of a keyboard and no mouse Apr 30 06:17:27 nice. Apr 30 06:21:46 falls_huang: hi Apr 30 06:23:08 hi , jserv2 :) Apr 30 06:23:17 falls_huang: nice to meet you. Apr 30 06:23:39 falls_huang: I looked up the log, and see something regarding Chinese IM on openmoko. Apr 30 06:23:46 :) I'm studying openmoko now :) Apr 30 06:24:25 great! Apr 30 06:24:45 I had a meeting with FIC officer yesterday, maybe I will join FIC , they told me that they need me to develop openmoko in mainland of China Apr 30 06:25:02 so I must pay more attention to openmoko now :) Apr 30 06:25:17 Readon is another guy from mainland China. Apr 30 06:25:18 hmm Apr 30 06:25:31 falls_huang: cool! Apr 30 06:27:43 hi,jserv2 Apr 30 06:27:56 I wish that openmoko can be compatible with gpephone finally Apr 30 06:28:25 so, how about re-work on ptim? Apr 30 06:29:15 I think ptim has a good IM design for handheld devices. Apr 30 06:29:39 http://gpephone.linuxtogo.org/doc/components/manual-ptim.htm Apr 30 06:29:54 yes, scim is too big to mobile . I'm studying ptim now and not familiar with it :) Apr 30 06:30:38 falls_huang: I do hope to port my IM ( http://chewing.csie.net/ ) to openmoko or some devices. Apr 30 06:31:13 As far as I know, chewing IM is popular in Taiwan. :) Apr 30 06:31:59 Readon: hi Apr 30 06:32:36 * jserv2 notices that there are at least 4 persons speaking Chinese here. Apr 30 06:32:57 chewing is cool. But I think neo 1973 still need a handwriting input method . I can't find any opensource handwriting software for Chinese Apr 30 06:34:04 no, we only need to recognize the phonetic, then chewing IM will pick up the phrases. Apr 30 06:34:30 falls_huang: maybe you should talk to James Su. Apr 30 06:35:56 james su is suzhe ? Apr 30 06:36:13 falls_huang: yes Apr 30 06:36:17 falls_huang: http://openvanilla.org/wiki/zh/index.php?title=ChineseInformationProcessing Apr 30 06:36:27 falls_huang: I have updated some information there. Apr 30 06:36:38 thx ! Apr 30 06:37:36 falls_huang: I did work for Chinese handwriting before. Apr 30 06:37:57 but that was for Qtopia, and that is closed-source. Apr 30 06:38:12 i couldn't get fic-gta01 here. Apr 30 06:38:30 However, community involvement might help to drive something better. :) Apr 30 06:44:43 I hope that the community of gpephone and openmoko can develop a compatible system :) Apr 30 06:50:38 falls_huang: by the way, do you know the developers behind gpephone especially for IM framework? Apr 30 06:51:41 no Apr 30 06:52:21 pity Apr 30 06:54:16 there is a #gpephone channel IIRC Apr 30 06:55:46 good morning Apr 30 06:56:23 I just searched the mailing list of gpephone , It's seemed that "florian.boor at kernelconcepts.de " is the author of PTIM Apr 30 06:56:54 okay. Apr 30 06:57:21 I think the main problem is that we always write new input methods. Apr 30 06:57:45 for different kind of applications with different engineering efforts. Apr 30 06:58:52 so, we have to do programming for XIM, IIIMF, SCIM, OpenVanilla, UIM, etc. Apr 30 06:59:10 It is hard to exchange the IM resources. Apr 30 07:00:31 while migrating from desktop to handset, the problem is raising. Apr 30 07:08:30 openmoko: 03laforge * r1875 10/trunk/src/target/kernel/patches/ (20 files): Apr 30 07:08:30 openmoko: update patches to apply against 2.6.21[.1] Apr 30 07:08:30 openmoko: * Apr 30 07:14:30 openmoko: 03laforge * r1876 10/trunk/src/target/kernel/patches/s3c2410_udc-vbus_draw_pdata.patch: fix header file path (asm-arm instead of asm) Apr 30 08:27:53 openmoko: 03mickey * r1877 10/trunk/oe/packages/linux/ (8 files in 2 dirs): Apr 30 08:27:53 openmoko: * make 2.6.20.x recpipes use 2.6.20.x branch Apr 30 08:27:53 openmoko: * add new 2.6.21.1 recipe Apr 30 08:28:30 openmoko: 03mickey * r1878 10/trunk/oe/packages/linux/gta01-kernel_2.6.bb: update default kernel to 2.6.21.1 Apr 30 08:33:34 hi Apr 30 09:09:06 hyalo Apr 30 09:09:22 any kernel developers here ? Apr 30 09:09:22 kamathln: See http://wiki.openmoko.org/wiki/FAQ (Hopefully this link answers your question. This autotrigger for new nick and question during first 15 minutes done only once.) Apr 30 09:09:35 kamathln: some Apr 30 09:09:42 ok.. Apr 30 09:09:50 I got the FAQ trigger.. Apr 30 09:10:02 let me show it some respect and come back Apr 30 09:10:07 ;-) Apr 30 09:13:43 I did not get an answer from the FAQ: so here goes : are the usb devices the software can switch to are hrdwired or can we simulate other devices by hacking the kernel ? Apr 30 09:14:25 kamathln: they are the ones from usb gadget project Apr 30 09:14:30 www.linux-usb.org Apr 30 09:14:33 Ah! Apr 30 09:14:36 thankx Apr 30 09:18:33 which also means that they are gatewayed or simulated through software Apr 30 09:18:49 kamathln: yes Apr 30 09:20:06 and so to say, there can be a set of apps and drivers in open moko whcih can simulate keyboards, mice , complete hubs, etc.. Apr 30 09:20:32 kamathln: USB HID is on the agenda. :) Apr 30 09:20:41 counter Apr 30 09:20:42 http://wiki.openmoko.org/wiki/Neo1973 : >= 2007-05-10 (uncertainty is a guess) P1: 2 weeks 14:39:16 (14.611 +-3.5 days) (1051;187) Apr 30 09:20:45 hmmm, custom input device.... Apr 30 09:20:49 ok! thx Apr 30 09:20:54 kamathln: But still some higher items... Apr 30 09:20:56 exactly! Apr 30 09:21:35 * stefan_schmidt prefers bluetooth HID instead of using a cable. :) Apr 30 09:21:35 stefan_schmidt: "higher items" ? Apr 30 09:22:02 kamathln: There are still rough edges in kernel, bootloader, etc.. Apr 30 09:22:17 kamathln: Stuff that's needs to be done before Apr 30 09:22:38 yeah.. I just wanted to know if that is possible :-) Apr 30 09:22:43 am not in a hurry :-P Apr 30 09:22:49 kamathln: It is Apr 30 09:23:02 yeah thank you very much Apr 30 09:23:09 np Apr 30 09:23:28 when my friends ask , I can at least boast that I can do all these Apr 30 09:23:49 If I am not wrong I can install Open moko on my Motorola a1200 right ? Apr 30 09:24:05 I mean Motoming .. whatever you call it Apr 30 09:25:01 Linux neo1973 2.6.21.1-moko8 #1 PREEMPT Mon Apr 30 11:04:46 CEST 2007 armv4tl unknown :) Apr 30 09:25:17 oh ok.. I need the Neo1973 :-( Apr 30 09:25:45 hi Apr 30 09:26:56 I would just love to use Synergy on OpenMoko :-P Apr 30 09:26:59 http://synergy2.sourceforge.net/ Apr 30 09:27:29 heh, someone disables sound support in the 2.6.21.1 kernel Apr 30 09:27:49 I guess that means we really shouldn't use it as a phone :) Apr 30 09:29:28 is there a good understanding alsa mixers howto somewhere? i never figured out how to set the mixer for voip properly even on my desktop. Apr 30 09:33:26 quitte: not really as they change per chip Apr 30 09:36:09 :( mixers are the worst part i see at the moment for linux on the desktop.# Apr 30 09:36:33 that's why OHM was started Apr 30 09:36:43 ohm? Apr 30 09:37:30 thats pretty much ungooglable. can you give me an url? Apr 30 09:37:52 http://ohm.freedesktop.org/wiki/ Apr 30 09:38:00 thanks Apr 30 09:46:50 morning Apr 30 09:48:30 morning Apr 30 09:49:27 hi Mac Apr 30 10:05:42 counter Apr 30 10:05:42 http://wiki.openmoko.org/wiki/Neo1973 : >= 2007-05-10 (uncertainty is a guess) P1: 2 weeks 13:54:15 (14.579 +-3.5 days) (1052;187) Apr 30 10:54:33 counter Apr 30 10:54:34 http://wiki.openmoko.org/wiki/Neo1973 : >= 2007-05-10 (uncertainty is a guess) P1: 2 weeks 13:05:24 (14.545 +-3.5 days) (1053;187) Apr 30 11:09:08 (script) wiki RecentChanges: [[Wish_List_-_Hardware]] [[Wishlist:BuiltInScriptingLanguage]] [[Higher_level_language_options]] [[Current_events]] Apr 30 11:29:34 Hi guys Apr 30 11:30:35 Heya Cyphi^ Apr 30 11:32:11 how are you abraxa_ Apr 30 11:32:35 I'm feeling totally awesome, thanks =) Apr 30 11:32:36 You? Apr 30 11:32:56 same thanks : Apr 30 11:32:58 :) Apr 30 11:33:11 it's labours day .. atleast in Finland Apr 30 11:34:07 ..tomorrow Apr 30 11:34:10 That's tomorrow here in Germany Apr 30 11:34:18 tomorrow here too Apr 30 11:34:24 but today's the main party day Apr 30 11:34:30 here being finland Apr 30 11:42:43 * koen wonders is openmoko devels actually test the stuff they commit Apr 30 11:43:00 the .21 kernel has no working sound, gps or bluetooth Apr 30 11:45:29 koen: maybe it is 'work in progress' pushed to share? Apr 30 11:46:14 [10:28] CIA-16: openmoko: mickey * r1878 /trunk/oe/packages/linux/gta01-kernel_2.6.bb: update default kernel to 2.6.21.1 Apr 30 11:46:24 hrw: if that so, the above commit makes no sense Apr 30 11:46:52 koen: phase0 are expected to use 2007.04 snapshot as base anyway Apr 30 11:47:04 I don't hope so Apr 30 11:47:12 that snapshot is buggy as hell Apr 30 11:47:28 two weeks earlier or later is a lot better Apr 30 11:49:17 and laf0rge is offline for some hours Apr 30 11:49:44 he just said that he was going out for dinner now Apr 30 11:50:52 oh.. backlog not ended - sorry Apr 30 11:54:48 * koen tries to get a more meaningfull report out of the kernel Apr 30 12:08:08 koen - did the gps ever work? Apr 30 12:08:22 Elrond: you could at least power it on Apr 30 12:08:31 Ahh, yeah. :) Apr 30 12:08:38 http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=535 Apr 30 12:08:44 And that's now broken? Apr 30 12:08:54 *reading bug* Apr 30 12:08:54 with 2.6.21.1 you have a working touchscreen and X Apr 30 12:09:12 no gsm, no bluetooth, no gps, no sound, etc Apr 30 12:10:46 So a touchscreen/X11 with usb? ;o) Apr 30 12:10:54 Are the buttons working? ;) Apr 30 12:12:06 If I were much better at writing a testsuite, I'd werite one for the neo. ;-) Apr 30 12:13:45 koen: fwiw, there are some known miscompilations in gcc that are leading to not-working sound (see gcc bug 27363) Apr 30 12:14:01 rtp: the KConfig part is screwed up as well Apr 30 12:14:11 "kernel is broken" sounds like a joke bugreport Apr 30 12:14:13 rtp: and OE has teh cse patch for gcc 4.1.1 Apr 30 12:14:41 rtp: and that patch was applied for gcc 4.1.2, which I'm using Apr 30 12:15:00 (I don't wanted to add a mb() to the alsa driver) Apr 30 12:15:25 balrog-kun: I couldn't find a better description Apr 30 12:15:33 koen: ok. It was only a guess as I didn't know your compiler version :) Apr 30 12:15:38 ;) Apr 30 12:15:39 and I suspect I'll find more broken stuff Apr 30 12:16:26 so can anyone explain why such a broken kernel is the *default* now? Apr 30 12:17:02 koen - *duck* Because linux people love broken hot beelding edge crap? *run&duck* ;o) Apr 30 12:18:06 Elrond: I would be less annoyed if the openmoko people weren't such crybabies about using OE directly instead of a svn repo to keep things "stable" and "working" Apr 30 12:18:12 koen: did you look at your kernel config for the gta01_pm* stuff ? I suspect they're built-in, so you'll find no .ko file Apr 30 12:18:40 rtp: they don't show up in sysfs and generated modules previously Apr 30 12:18:54 (and that patch wasn't touched in the .21 update) Apr 30 12:19:00 rtp - The find should have found *.o then, no? Apr 30 12:19:37 Oh, there are .o! Apr 30 12:20:13 Elrond: that's why I said it was built-in :) Apr 30 12:26:02 /wc Apr 30 12:34:24 cu guys, going to have some beer and start to have a nice May 1st :D Apr 30 12:35:59 May 1st is tomorrow Apr 30 12:36:17 unless you live in the middle of the pacific Apr 30 12:36:39 loufoque, but May 2nd is work day, so you have to celebrate May 1st on 30th on April Apr 30 12:36:41 the party started two days ago here Apr 30 12:37:03 guaqua, the party started about a month ago here ;) Apr 30 12:37:18 where is your here, inz? Apr 30 12:37:33 May 1st in two hours here. Apr 30 12:38:34 guaqua, tampere Apr 30 12:38:49 otaniemi started ages ago aswell Apr 30 12:38:49 do you live in new zealand or something? Apr 30 12:39:00 but i've had other stuff to do so :) Apr 30 12:39:05 =) Apr 30 13:04:19 * koen mutters something about wasting some hours with .21 and still having no output from gps Apr 30 13:05:10 :/ Apr 30 13:05:13 koen - Did you finally find out, what's wrong with .21? Or did you get the *...* gps to give some bits to you? Apr 30 13:05:53 Elrond: the patchset it broken, so now I have to wait ~2 weeks for it to get fixed.... Apr 30 13:06:13 koen: why 2 weeks? Apr 30 13:06:16 koen - Just use a 2.6.20.1 kernel? Apr 30 13:06:25 XorA: average time-to-fix for openmoko Apr 30 13:06:41 Elrond: downgrading kernelmodules is a pain Apr 30 13:07:00 and I need to boot up a linux computer again for dfu-utils Apr 30 13:07:39 Why exactly is downgraduing modules a pain? Apr 30 13:09:08 (script) wiki RecentChanges: [[Buying_Interest_List]] [[Wish_List]] Apr 30 13:09:46 Elrond: because you have to tell the package manager to downgrade each module seperately Apr 30 13:10:54 *gna* Apr 30 13:12:21 Why can't mpdules be in one single package, which has all modules in it, and the package has the version-numer as part of the package-name? Apr 30 13:12:47 Elrond: which modules? Apr 30 13:12:54 Elrond: all? Apr 30 13:13:06 most of users does not need nfsd module Apr 30 13:13:12 meta-package? Apr 30 13:13:40 * guaqua thinking of meta-package hell Apr 30 13:14:04 hrw - Ahh, okay. Well, still putting the version in the package-name would allow having multiple kernel versions bootable. Apr 30 13:14:28 u-boot can boot kernels from jffs2, right? Apr 30 13:14:36 can but it is slow Apr 30 13:15:03 I don't reboot my phone 6 times a day, so I don't care about 'slow' Apr 30 13:15:24 what is slow in this case? Apr 30 13:15:43 booting Apr 30 13:15:51 meaning as in time Apr 30 13:15:53 or rather uboot mounting the filesystem and execing the kernel Apr 30 13:16:04 does it take minutes, hours? Apr 30 13:16:54 Probably minutes. Apr 30 13:17:00 depending on the jffs2 partition size, seconds to ~minutes Apr 30 13:17:15 But for getting a "this kernel really works" boot, it would be acceptable. Apr 30 13:27:48 Hi guys, Just wondering, been away for a while. How is phase two coming out. And have there been any new news on the phase 2.5 hardware refresh? Apr 30 13:27:48 morten_: See http://wiki.openmoko.org/wiki/FAQ (Hopefully this link answers your question. This autotrigger for new nick and question during first 15 minutes done only once.) Apr 30 13:28:56 morten_: http://lists.openmoko.org/pipermail/announce/2007-April/000012.html Apr 30 13:29:01 phase 1 and 1.5 you must mean :) Apr 30 13:29:32 guaqua, yeah, sorry Apr 30 13:34:59 hi there. Just a small question about the Neo1973 : does the wifi chip will be include in the first public release of the neo or in a possible second phase and an hardware refresh ? Apr 30 13:34:59 P1.5 will likely have WiFi (Atheros AR6K or Marvell 8686): http://gnumonks.org/~laforge/weblog/2007/03/25#20070325-returning_from_taipei http://wiki.openmoko.org/wiki/Neo1973 http://wiki.openmoko.org/wiki/FAQ#Q:_What_is_the_rationale_behind_the_exclusion_of_WiFi.3F (WiFi status for P1.5 will be known for sure when P1 becomes available and also what kind of discount for P1.5 is given for P1 buyers) Apr 30 13:35:41 Not the one available in maybe a couple of weeks Apr 30 13:44:56 mm, when I have wifi, I'll have a laptop around, I think/predict. Meaning I will not wait for P1.5 but go fo P1, still ! Apr 30 13:46:41 I'm drooling of the combination of gps and internet, that's an opening for som REALLY interesting applications. =) Apr 30 13:47:57 morten_: some thing already thought (somewhere out on the internet) : location based reminder Apr 30 13:47:59 s Apr 30 13:50:56 Yeah i know, but just think of many mokos connected in a network, it could give you superquick access to what's going on. lat's say there is somthing happening downtown. then he could post a notice on the exact possition, if he has a camera he can attach a picture, and send it to everyone he wants, maybe just his friends, or maybe public notice for everyone with a neo Apr 30 13:51:54 gps location should become a standard mms object what say ? Apr 30 13:54:53 kamathln, Yes, it should be, would make them much more precise when you want to say where somthing is, and with internet, you could find out where you're buddies are in real-time Apr 30 14:21:50 ug racing where the neo-network shows the route and determines teh wiener... =P Apr 30 14:22:02 (it's a forced typo) Apr 30 14:29:59 A lot of the most interesting applications require it to be pervasive though. Apr 30 14:30:26 Something installed on every phone, that the user can turn on with one click. Apr 30 15:09:08 (script) wiki RecentChanges: [[Wish_List]] Apr 30 15:20:53 counter Apr 30 15:20:54 http://wiki.openmoko.org/wiki/Neo1973 : >= 2007-05-10 (uncertainty is a guess) P1: 2 weeks 08:39:04 (14.360 +-3.5 days) (1054;187) Apr 30 15:21:01 hmm... Apr 30 15:35:24 TimRiker: greetings Apr 30 15:41:09 so, should I try to hack OpenMoko into a MOTOROKR E6? Apr 30 15:43:55 LuitvD: If you're capable then sure, why not Apr 30 15:44:22 I wonder what motorola would think about it :P Apr 30 15:44:54 I wonder why anyone should care ;) Apr 30 15:45:16 some support from moto would be nice :P especially when I do something wrong, turning the thing into a brick... Apr 30 15:45:35 don't think I'd be able to unbrick it whithout motorola's help Apr 30 15:45:52 LuitvD: first thing to find is the jtag port then Apr 30 15:45:59 yeah :P Apr 30 15:46:00 That's true but seeing how their past linux-based devices were all closed I doubt you'll be very welcome Apr 30 15:46:20 my thoughts exactly Apr 30 15:46:34 though e28 wasn't very collaborative either... Apr 30 15:47:10 That's why I'm thankful for having Sean and FIC =) Apr 30 15:47:11 and e28 was founded by a motorola guy who wanted more openness Apr 30 15:47:21 right...... Apr 30 15:47:47 though I'm almost certain I'm gonna buy that ROKR E6 Apr 30 15:47:52 :) Apr 30 15:48:01 openmoko or not, it's a fun device Apr 30 15:49:32 (updated wifi message:) Apr 30 15:49:33 wifi? Apr 30 15:49:33 P1.5 will likely have WiFi (Atheros AR6K): http://lists.openmoko.org/pipermail/announce/2007-April/000012.html http://wiki.openmoko.org/wiki/Neo1973 http://wiki.openmoko.org/wiki/FAQ#Q:_What_is_the_rationale_behind_the_exclusion_of_WiFi.3F (WiFi status for P1.5 will be known for sure when P1 becomes available and also what kind of discount for P1.5 is given for P1 buyers) Apr 30 15:50:12 lol Apr 30 15:50:30 read about that atheros :) nice chip Apr 30 15:55:12 openmoko: 03laforge * r1879 10/trunk/oe/packages/linux/gta01-kernel_2.6.bb: downgrade default kernel to 2.6.20.10 Apr 30 15:56:19 ah, it seems they finally tested the .21 kernel Apr 30 15:56:43 l Apr 30 16:14:42 cworth - *ping-again* ;) Apr 30 16:16:47 Elrond: what's up? Apr 30 16:18:40 openmoko: 03thomas * r1880 10/trunk/src/target/OM-2007/applications/openmoko-dialer/src/ (5 files): openmoko-dialer: Use MokoJournal to retrieve call history Apr 30 16:19:32 cworth - You asked, what I wanted to do, yesterday: I get 2d data in realtime, and want to plot it in real time. That's the basic thing. Apr 30 16:22:34 Elrond: What kind of plot, though? Apr 30 16:23:27 Elrond: I can imagine some particular kinds of data being usable with some particular display devices with a on-pixel-per-datapoint kind of thing, but not much of anything general. Apr 30 16:24:25 So, you might be interested in trying other kinds of ways to draw what you want with cairo, (which would allow you more flexibility to explore visualization options than if you stick to one-pixel-at-a-time). Apr 30 16:24:46 And if you are definitely sold on the one-pixel-at-a-time approach, then you might not even want to use cairo for that part. Apr 30 16:25:35 Well, those are measurement points. one-pixel-per-datapoint makes the most sense to me. Apr 30 16:25:57 consider them random(x), random(y) :) Apr 30 16:26:26 So drawing the points straight with xlib or gdk might make more sense? Apr 30 16:26:40 Elrond: But why a pixel? Why not a small circle, say? Apr 30 16:27:06 Or even a little radial gradient splotch? Apr 30 16:27:26 A small circle would also be fine. I just wanted to keep the amount of x11 traffic low. Apr 30 16:27:40 a little 3d phong-shaded sphere with transparency and reflections Apr 30 16:28:10 balrog-kun - tihi. ;-) 2000 of them don't make too much sense, really. ;-) Apr 30 16:28:47 Elrond: If you're talking about a 280+ DPI display device, (which is what I assume is of interest in a channel like this), then individual pixels really aren't that interesting of an object to draw---they are almost invisible. Apr 30 16:28:56 each one made up of 2000 little triangles :) Apr 30 16:29:39 Elrond: As for x11 traffic, you can draw your little "datapoint representation", (whatever it is), once to an xlib surface and then just copy it---that's a tiny amount of X11 traffic. Apr 30 16:29:51 cworth - Huh. ;-) We should have moved to a query. ;-) It's not related to a 300dpi device. More like 90dpi. ;) Apr 30 16:30:22 Elrond: Or if you need different colors, then render it as a mask once and composite different solid colors through it each time, (still a tiny amount of traffic). Apr 30 16:30:39 Ahh. :-) Apr 30 16:30:44 Elrond: Ah, OK. There's also #cairo for general cairo questions too. Apr 30 16:31:01 Huh. *feeling a little dumb* ;-) Apr 30 16:31:12 I probably should move there first. :) Apr 30 17:05:15 cworth: for which is release is the workqueue patch scheduled? Apr 30 17:08:06 koen: I need to check into that. I'm planning a 1.4.6 release right now, (for today, basically), so I want to check and see what the status of the workqueue stuff is. Apr 30 17:08:26 (I'm not sure if it's been merged in yet or not, and if not, why not.) Apr 30 17:09:29 (script) wiki RecentChanges: [[Buying_Interest_List]] Apr 30 17:09:53 cworth: I suspected you were going to release 1.4.6 soon :) Apr 30 17:10:40 cworth: judging from using my neo1973 and ipaqs, the workqueue patch doesn't seem to cause crashes Apr 30 18:16:06 Is it me or was the OM kernel reverted from r1876 (non-booting btw) to r1828? Apr 30 18:16:38 CoreDump|home: they switched that from 2.6.21.1 to 2.6.20.10 Apr 30 18:16:47 s/that/back/ Apr 30 18:16:54 ahh great Apr 30 18:16:56 thanks Apr 30 18:17:36 they should probably remove r1876 from the buildhost =) Apr 30 18:17:40 2.6.21.1 boots over here, but no bluetooth, gps, etc Apr 30 18:18:27 really? hmmm Apr 30 18:18:38 I build from OE, not from the overlay Apr 30 18:18:53 ah that's it then Apr 30 18:19:10 root@neo1973:~$ uname -a Apr 30 18:19:10 Linux neo1973 2.6.21.1-moko8 #1 PREEMPT Mon Apr 30 11:04:46 CEST 2007 armv4tl unknown Apr 30 18:21:17 hmm my 1876 was 2.6.20.7 *scratches head* Apr 30 18:21:33 anyway, 1828 is 2.6.10 and boots Apr 30 18:21:40 2.6.20.10 Apr 30 18:21:58 OE has 2.6.20 and that works Apr 30 18:23:45 I really would like to build OM / OE but that does not result in a "true" OM image so bug-fixing is kinda out-of-the questions as we have no control what is pulled from .dev and what not :\ Apr 30 18:23:45 s/\/with Apr 30 18:24:27 heh Apr 30 18:24:30 that's just FUD Apr 30 18:24:39 that's true Apr 30 18:24:54 I built both in the past and they are far from identical Apr 30 18:25:04 they are different Apr 30 18:25:12 but that doesn't stop you from fixing bugs Apr 30 18:25:39 even if it usually takes weeks (yes, weeks) for the bugfixes to get back into svn Apr 30 18:26:28 CoreDump|home: and mickey|zzZZzz keeps saying the svn will go away "real soon(TM)" Apr 30 18:26:51 It doesn't look that way to me =) Apr 30 18:27:10 * CoreDump|home kinda sees a OM branch w/ the old snapshot coming Apr 30 18:27:17 yeah Apr 30 18:27:19 me too Apr 30 18:27:27 but mickeyl keeps saying otherwise Apr 30 18:27:29 I could very well live with that =) Apr 30 18:27:46 I don't really care, I have a working phone with the moko gui Apr 30 18:28:19 but all the users complaining about bugs in the svn repo, that have been fixed eons ago in OE is annoying me Apr 30 18:28:48 until OM merges w/ OE there will always be two different images Apr 30 18:29:02 I'll be the first to discard the overlay when it comes to that ;) Apr 30 18:29:16 the difference being that with OE you can actually build usefull image :) Apr 30 18:29:56 koen: you should check a recent image from the buildhost. Quite usable (but dog-slow due to the theme) Apr 30 18:30:27 you mean an image with the .21.1 kernel? Apr 30 18:30:40 nah, as of today it is using 2.6.20.10 Apr 30 18:30:50 right Apr 30 18:30:55 the r1828 is what you want Apr 30 18:31:25 I don't want any stinking OABI binaries :) Apr 30 18:31:29 OABI is so last year Apr 30 18:31:30 =D Apr 30 18:31:37 true Apr 30 18:31:48 !pester mickey|zzZZzz Apr 30 18:31:49 mickey|zzZZzz : Are we there yet? .. Are we there yet? .. Are we there yet? Apr 30 18:31:53 besides, the angstrom image is faster than the "official" image Apr 30 18:32:21 (still slow) Apr 30 18:32:51 * CoreDump|home has some hopes for P1.5 getting a speed bump Apr 30 18:32:57 RP: pokylinux.org points to http://www.pimlicoproject.org/ Apr 30 18:33:22 counter Apr 30 18:33:22 http://wiki.openmoko.org/wiki/Neo1973 : >= 2007-05-10 (uncertainty is a guess) P1: 2 weeks 05:26:36 (14.227 +-3.5 days) (1055;190) Apr 30 18:33:28 Linux fic-gta01 2.6.20.10-moko8 #1 PREEMPT Mon Apr 30 18:03:44 UTC 2007 armv4tl unknow Apr 30 18:35:14 oh great, reverting the kernel fucked up image generation Apr 30 18:38:01 CoreDump|home: welcome to downgrade hell Apr 30 18:38:01 * koen already deleted the offending packages from deploy and the online feeds Apr 30 18:40:33 I kinda like the "only rebuild when upstream changed" thingy Apr 30 18:40:51 it's neat Apr 30 18:41:11 but I dislike the "develop in secret and don't send patch" approach Apr 30 18:41:48 well that's not exactly fair. They kinda announced what they did and the source is open..so Apr 30 18:42:08 they also say they have a 1.8 patch Apr 30 18:42:17 really? hmmm Apr 30 18:42:30 couldn't find one when I was looking Apr 30 18:42:36 that's the problem Apr 30 18:42:44 they *say* they have an 1.8 patch Apr 30 18:42:48 :\ Apr 30 18:43:00 hence my "develop in secret and don't send patch" comment Apr 30 18:43:04 * CoreDump|home sighs as 1.8 doesn't work for him anyway Apr 30 18:43:41 1.8 works great with a current .dev Apr 30 18:43:54 not for me. bitbake hangs on task 1 Apr 30 18:44:15 weird Apr 30 18:44:33 indeed Apr 30 18:44:44 didn't have a chance to look into it tho Apr 30 18:44:48 My local diff is not having any qt/e or opie software Apr 30 18:45:04 ? Apr 30 18:45:29 so it's not really .dev I'm building against Apr 30 18:45:59 well, opie's dead anyway.... Apr 30 18:46:45