**** BEGIN LOGGING AT Sun Jul 22 02:59:58 2007 Jul 22 03:06:00 Yay! Jul 22 03:06:08 Payment confirmation! Jul 22 03:06:32 brain, RT#? Jul 22 03:06:38 24xx Jul 22 03:06:46 fuckin a Jul 22 03:06:49 09 Jul 22 03:07:41 huh. billing on saturday evening. wow Jul 22 03:09:33 * sagacis sniffs Jul 22 03:09:35 ckuethe: evening? around here it's already night Jul 22 03:10:04 In fact, I should already be sleeping Jul 22 03:10:09 Sunday afternoon here. Funny thing timezones :) Jul 22 03:10:25 But trying to make openmoko compile with ARM EABI is _fun_ Jul 22 03:12:12 Now we wait for shipping. It is a very exciting time Jul 22 03:13:18 compbrain: you guys should lower your expectations. If you are expecting it to only ship next week, and it ships sooner, you can feel "wow, it's fast!" (instead of "why is it taking so swearword long?") Jul 22 03:13:51 (but then, having a really high ticket number helps) Jul 22 03:15:40 cesarb: Yeah, I've got classes to distract me during the week Jul 22 03:15:47 I only get anxious on weekends :) Jul 22 03:17:54 cesarb: GMT-0700 Jul 22 03:18:08 ckuethe: GMT-0300 here Jul 22 03:20:25 yes, i had a high ticket number too ... i resigned myself to the notion of "it'll get here when it gets here" Jul 22 03:25:02 I'm also waiting for a new laptop - so its a familiar theme Jul 22 03:42:03 (script) wiki RecentChanges: [[P1_Owners]] [[User:Chooseopen]] [[Hardware:AGPS]] [[OpenMoko_Local_Groups:_Adelaide]] [[OpenMoko_Local_Groups]] [[Minimo]] Jul 22 03:57:46 jaebird_mobile: heya Jul 22 03:57:56 hey Jul 22 03:58:50 irc is cool on mobile devices :) Jul 22 04:14:05 counter? Jul 22 04:14:06 (last update 2007-07-21 06:20) http://londerings.novalis.org/wlog/index.php?title=Neo1973_counter : Order https://direct.openmoko.com/ ; Shipping prob. 75% in 3 days 19:45:48 (3.823±2.5 days) (1768;249) Jul 22 04:14:18 grrr Jul 22 04:21:18 Woohoo, payment received on my openmoko order :) Jul 22 04:21:55 and shipping is cheaper :) Jul 22 04:24:58 * sagacis starts openly sobbing Jul 22 04:30:02 hmm i still have not been billed Jul 22 04:30:31 daMaestro, whats your ticket? Jul 22 04:30:37 1964 Jul 22 04:31:03 hm.. yeah, strange Jul 22 04:31:16 i would have thought they were working in order Jul 22 04:32:54 would have, but there's a 24xx that's charged too\ Jul 22 04:33:00 i think they just don't like me Jul 22 04:33:59 you got/sent the YES_I_DO email? Jul 22 04:35:25 I've not been billed and I'm 1912 Jul 22 04:35:37 strange Jul 22 04:35:52 * sagacis said "YES_I_DO" on the 13th, right after getting the mail Jul 22 04:36:03 I think they are just randomly charging... Jul 22 04:36:14 Maybe the network? I used mastercard Jul 22 04:36:19 visa here Jul 22 04:37:12 sagacis is sad Jul 22 04:37:17 afaik, i said YES_I_DO also Jul 22 04:39:27 10,000 years without an open cell phone... savor the final days. ;-) Jul 22 04:48:03 maybe they divided up the list and have multiple people working at it from different points Jul 22 04:48:08 that could explain it Jul 22 04:49:18 No "at" command on the moko? Jul 22 04:59:20 yeah i sent YES_I_DO 18 minutes after receiving it Jul 22 04:59:55 maybe they don't like orange or don't like visa cards or assume that the database is ordered when it isn't. Jul 22 05:04:22 hi all Jul 22 05:42:23 (script) wiki RecentChanges: [[MokoMakefile]] Jul 22 06:03:44 anyone got suggestions for qemu optimizations? Jul 22 06:07:01 galexande, it might be visa... i ordered black Jul 22 06:42:43 (script) http://openmoko.com changed: sizeof(diff -u)=1113 Jul 22 06:55:07 anyone done any qemu optimizations? Jul 22 07:07:31 bkruse_home: yeah, I put it on a faster machine >_> Jul 22 07:07:34 * rtyler ducks Jul 22 07:08:34 bkruse_home: I used the same method, it speeds up make openmoko-devel-image too. Jul 22 07:08:42 heh Jul 22 07:09:21 rtyler: Compile of openmoko (first time takes hours)? Jul 22 07:09:31 didn't for me Jul 22 07:09:47 or well, the qemu bits, I haven't built all of openmoko from scratch yet Jul 22 07:11:23 rtyler: Ah, from scratch, and it depends on the hardware, my sempron takes a good 12-14 hours, core2duo does it in 5. Jul 22 07:40:46 * gcb77 took a trip to work today and uploaded the VMWare disk image Jul 22 07:40:52 WOOT! Jul 22 07:41:20 Then I found out my hosting comp is using old Apache, you cant download files over 2GB Jul 22 07:41:31 GAH! Jul 22 07:41:43 (script) wiki RecentChanges: [[QT2410/kr]] [[QT2410/zh_cn]] [[QT2410]] [[Category:Development_Board]] [[P1_Owners]] [[Wish_List_-_Hardware]] [[Wish_List]] Jul 22 07:43:19 I'm setting up an FTP acount... should be ready in a few minutes Jul 22 07:49:01 Any luck making hello world app? Jul 22 07:49:24 been busy working on other projects unfortunately, got sick of fighting QEMU Jul 22 07:49:30 lol Jul 22 07:49:56 Um, I have to clear out my build dir and rebuild from scratch. Jul 22 07:50:15 had to do that several times Jul 22 07:50:38 rm -rf build should do it. Jul 22 07:51:01 make clobber Jul 22 07:51:32 then another few hours of rebuild Jul 22 07:51:57 Shouldn't the remove buld dir work? Jul 22 07:52:14 haven't tried that Jul 22 07:52:36 I tried removing individual packages, and couldn't get it to build again Jul 22 07:53:32 gcb77: You would have to remove the stamps. Jul 22 07:55:26 ok, ftp is ready: ftp.gcbnet.dev user: moko pw: openmoko Jul 22 07:55:49 lol, got that backwards... ftp.gcbdev.net Jul 22 07:56:16 yeah, I don't think dev is a TLD? Jul 22 07:58:21 I think that'll only let one person in at a time Jul 22 07:58:56 Um, might do. Jul 22 07:59:18 nvm, there we go, let's see if an ls works >_> Jul 22 08:01:10 * rtyler shakes fist at dreamhots Jul 22 08:01:12 dreamhost as well Jul 22 08:01:59 I'm curios what the download speed is like Jul 22 08:02:07 230 User moko logged in Jul 22 08:02:07 Remote system type is UNIX. Jul 22 08:02:07 Using binary mode to transfer files. Jul 22 08:02:07 ftp> ls Jul 22 08:02:07 200 PORT command successful Jul 22 08:02:09 425 Unable to build data connection: No route to host Jul 22 08:02:14 * rtyler cries Jul 22 08:02:31 grr Jul 22 08:03:08 try get mokodev.vmdk.bz2 Jul 22 08:04:19 ftp> get mokodev.vmdk.bz2 Jul 22 08:04:19 local: mokodev.vmdk.bz2 remote: mokodev.vmdk.bz2 Jul 22 08:04:19 200 PORT command successful Jul 22 08:04:24 * rtyler waits for anything to happen Jul 22 08:06:29 ftp> get mokodev.vmdk.bz2 Jul 22 08:06:29 local: mokodev.vmdk.bz2 remote: mokodev.vmdk.bz2 Jul 22 08:06:29 200 PORT command successful Jul 22 08:06:29 425 Unable to build data connection: Connection timed out Jul 22 08:06:30 ftp> Jul 22 08:06:35 * rtyler screams and kicks dreamhost Jul 22 08:07:57 ~lart dreamhost Jul 22 08:07:57 That is strange... I can get to it just fine from here, but not from other places Jul 22 08:07:57 * apt executes killall -KILL dreamhost Jul 22 08:07:57 * mokobot hits dreamhost with an anvil and laughs with a contralto voice ... Haha Ha HA Ha Jul 22 08:08:17 lol Jul 22 08:08:18 gcb77, why not bittorrent? Jul 22 08:08:30 dreamhost hates torrent Jul 22 08:08:31 apt: that's a neat trick, does mokobot have a man page somewhere? :P Jul 22 08:08:41 dreamhost also hates customers Jul 22 08:08:45 * rtyler chuckles Jul 22 08:08:49 If you think dreamhost is bad, try bluehost Jul 22 08:09:11 rtyler: can you get to www.gcbdev.net Jul 22 08:09:20 but of course Jul 22 08:09:48 anyone know of a good way to split a file into multiple archives? Jul 22 08:10:02 aforementioned anvil hitting :P Jul 22 08:10:11 lol Jul 22 08:10:49 * gcb77 types anvil[TAB][TAB]... pauses [TAB][TAB]... nothing Jul 22 08:12:15 lol Jul 22 08:13:08 gcb77, rar? Jul 22 08:13:22 gcb77: RAWR! Jul 22 08:13:38 * gcb77 types rar[TAB][TAB]... pauses [TAB][TAB]... nothing Jul 22 08:14:00 tar -M is an option Jul 22 08:14:06 Tar can do it, I'm not sure on the command. Jul 22 08:24:24 roflmao Jul 22 08:24:42 " Jul 22 08:24:43 C'mon I am #3560 and 54 years old. Will I live to receive it ? I' afraid Jul 22 08:24:43 my last words will be: "Did it arrive ? No ? Aaaaaargh." (dies) Jul 22 08:24:45 " Jul 22 08:25:00 lol Jul 22 08:25:14 ScaredyCat: my number is 3824, can you leave me yours in your will? Jul 22 08:25:38 minie has been paid for :D Jul 22 08:25:42 I'm 1826 Jul 22 08:25:57 that ^^ was posted to the mailing list Jul 22 08:27:14 I've suspended my subscription to the community mailing list... To many people talking only about "I'm number ## when will I receive it ?" Jul 22 08:27:27 It got kind of boring... Jul 22 08:27:40 this is why these things should be randomised. ;) Jul 22 08:28:02 yeah Jul 22 08:29:25 they really got 3500+ orders? Jul 22 08:29:37 no Jul 22 08:29:43 bah! :) Jul 22 08:29:47 "Payment received"!! (rt #229x) Jul 22 08:30:09 ticket numbers are generated when email hits the rt system, so that includes spam Jul 22 08:30:19 hmm Jul 22 08:30:28 * Markinoko is desperate Jul 22 08:33:01 Then they started at 1820. Jul 22 08:33:36 * aloril2 wonders what is biggest rt number in payment received mails Jul 22 08:34:33 * ewanm89_laptop still can't order :( Jul 22 08:35:07 aloril2, they just confirmed my payment Jul 22 08:35:22 aloril2, #229x Jul 22 08:35:24 based on jeddy3's number: probable has been CANCEL mails and/or close to 1 order/ticket number and/or spam rt numbers Jul 22 08:35:25 * Markinoko doesn't care since he uses the emulator and _actually_ works on some developement projects instead of staring at his mailbox Jul 22 08:36:23 Markinoko: I'm actually developing too. Jul 22 08:36:41 Rebuilding from cratch atm. Jul 22 08:37:09 s/cratch/scratch/ Jul 22 08:37:09 ewanm89_laptop meant: Rebuilding from scratch atm. Jul 22 08:37:12 I had RT#1832... got blank YES_I_DO e-mail ( I still responded YES_I_DO ), and nothing so far Jul 22 08:37:20 ewanm89_laptop: well, I know some of the people here are of course. Jul 22 08:37:41 But I can't even order the neo. Jul 22 08:38:23 ewanm89_laptop: I want a GTA02. waiting silently seemed like a good option to me... ;-) Jul 22 08:39:03 * hads wants both Jul 22 08:39:14 * ewanm89_laptop want both. Jul 22 08:39:34 * ewanm89_laptop finds emulator difficult to work with for development. Jul 22 08:39:46 I just wanna get jack running and get rid of pulseaudio Jul 22 08:40:11 Why's that? Jul 22 08:40:13 !!!!! Jul 22 08:40:17 I got my sms! Jul 22 08:40:18 Markinoko: Probably not to difficult. Jul 22 08:40:30 mmazur: did you text JOKE1 to 77423? Jul 22 08:40:33 * rtyler chuckles Jul 22 08:40:36 30 minutes ago Jul 22 08:40:47 My bank smsed me with the cc card charge! Jul 22 08:41:23 ewanm89_laptop: well I have jack compiled for the target. the package installs fine, but I'll probably need some patches submitted to the jack devs. doesn't exacly run too well now Jul 22 08:41:59 hads: pulseaudio doesn't have half of jack capabilities. sorry... Jul 22 08:42:16 No need to apologies to me. Just curious. Jul 22 08:42:21 :-) Jul 22 08:43:01 pulseaudio like most audio servers is good for "I want several apps to acces the sound card at the same time without crashing" Jul 22 08:43:26 jack is the same and has some sort of conrtolable patch bay Jul 22 08:43:38 Yeah Jul 22 08:43:56 from my experience jack is good for I Jul 22 08:44:03 So many different projects. Jul 22 08:44:07 "I want several apps to acces the sound card at the same time and crashing" Jul 22 08:44:20 s/acces/access/ Jul 22 08:44:21 ewanm89_laptop meant: "I want several apps to access the sound card at the same time and crashing" Jul 22 08:45:02 ewanm89_laptop: that was probably with a very old jack version ;-) Jul 22 08:45:17 ewanm89_laptop: I run my whole home studio with jack Jul 22 08:45:26 Markinoko: So the version in fiesty is old? Jul 22 08:45:47 ewanm89_laptop: and I seen jack handle quite a few thousand channels... Jul 22 08:46:03 ewanm89_laptop: what revision number? Jul 22 08:46:10 ewanm89_laptop: and which app crashed ? Jul 22 08:46:20 it might be my crappy sound cards. Jul 22 08:47:14 oh... and it's ubuntu... we've had a lot of trouble with the debian like packagers... although I do like debian like distros I must say they are sometimes a big bother for linux audio Jul 22 08:47:16 Oh, and I got my mail! Jul 22 08:47:33 mmazur: congrats :-) Jul 22 08:48:43 Um, I wonder how well it would work on gentoo on the same hardware. Jul 22 08:49:08 ewanm89_laptop: what's the hardware ? Jul 22 08:50:44 rtyler: http://www.gcbdev.net/openmoko/ Jul 22 08:51:35 WAHOO! Jul 22 08:51:39 * rtyler starts download Jul 22 08:52:19 * ewanm89_laptop is also interested, but doesn't have vmware running yet. Jul 22 08:54:31 seems there is no reason to tick the fast delivery option... Jul 22 08:54:37 heh Jul 22 08:56:21 raster said he picked his one up by himself in Taiwan. that might be the fastest way :p Jul 22 08:56:34 yeah, but then you have to go to Taiwan Jul 22 08:56:44 * rtyler coughs and sneezes after playing with pidgeons Jul 22 08:57:50 taiwan is nice Jul 22 09:00:11 gcb77: did you zero your disks first vmware ones) Jul 22 09:00:58 s/first vmware ones/first. The vmware ones I mean/ Jul 22 09:00:59 ScaredyCat meant: gcb77: did you zero your disks first. The vmware ones I mean) Jul 22 09:01:10 started out with growable disks Jul 22 09:02:12 done with disc one, onto second one :D Jul 22 09:02:23 wha, that was fast! Jul 22 09:03:31 bluehost used to throttle to ~100k/sec Jul 22 09:05:56 hi Jul 22 09:07:53 gcb77: I work at a reasonable big startup in San Francisoc, we have big pipes :P Jul 22 09:09:38 rtyler: :) Jul 22 09:12:38 gah, almost got the other one, ran out of space :P Jul 22 09:12:42 * rtyler starts nuking files Jul 22 09:13:33 doh, the way they are right now, you have to use tar to re-assemble, then bunzip Jul 22 09:13:47 I'm working on a tar+bzip Jul 22 09:17:36 strange, we got #1952 and still no CC charged Jul 22 09:18:05 apparently numbering sequence != shipping sequence Jul 22 09:18:12 wtf :) Jul 22 09:18:15 Markinoko: More watching there inboxes. Jul 22 09:19:32 ewanm89_laptop: hehe. seems like todays hot topic is "what-the-algorithm-behind-the-sequence-number". facinating ;-) Jul 22 09:19:53 lucky dip, i think. Jul 22 09:26:02 I wonder if our group order joined those blackholed... Jul 22 09:32:59 I can say it's not the YES_I_DO email order... because I was away when they sent those out, and was probably the last to answer Jul 22 09:33:33 ScaredyCat, I've answered 4 days after getting the email, so don't sweat it :) Jul 22 09:34:10 I have paid for mine... Jul 22 09:34:45 but I was away for 5 days (time zone diffs) Jul 22 09:36:10 Like I've said -- I've answered to the TILL_DEATH_DO_US_PART 4 days after receiving it. And got my cc charged a few hours ago. Always look on the bright side... ;) Jul 22 09:42:17 mmazur: what was your rt # ? Jul 22 09:42:43 (script) wiki RecentChanges: [[P1_Owners]] [[SH1_FAQ]] [[Getting_Started_with_your_Neo1973/fr]] Jul 22 09:44:32 do you think that they are actually packing them as they charge? Just seems to be rather slow. Jul 22 09:44:50 between charges Jul 22 09:45:10 packing is slow Jul 22 09:45:29 imagine about 3-5 mins per one order Jul 22 09:45:34 they need to hire some packing-wenches Jul 22 09:45:42 it's 3000 mins Jul 22 09:45:55 s/some/some busty/ Jul 22 09:46:40 3-5 mins per box.. ? wtf how slow are you! ;) Jul 22 09:47:01 print sticker, charge CC, pack the box... Jul 22 09:47:11 charge cc first Jul 22 09:47:15 i mean a time to finish whole proccess Jul 22 09:47:34 ok, but only one at a time then? Jul 22 09:48:01 if there is one person :) Jul 22 09:48:07 I mean you could have 1 person doing the charging, then when it's ok, hand over to the packer/printer Jul 22 09:48:27 which could be anyone... Jul 22 09:48:38 :o Jul 22 09:48:42 no, in fic it must be a kernel hacker! Jul 22 09:48:45 storm's a brewin' Jul 22 09:49:59 yeah, well.... :x Jul 22 09:51:09 they're going to need some busty packing wenches for GTA02, or they're boned Jul 22 09:51:37 and when GTA02 is shipped, they can send the busty packing wenches to me \o/ Jul 22 09:51:44 aren't they using distributors for gta02 ? Jul 22 09:51:49 I hope so :) Jul 22 09:51:53 * ScaredyCat too Jul 22 09:53:41 ScaredyCat: packing can be rather involved Jul 22 09:54:05 * LarstiQ prods mmazur Jul 22 09:54:38 it's not that invloved. Jul 22 09:54:59 sure if you didn't bother to prepare, like FIC.. it'll be slow Jul 22 09:57:33 Ah. Jul 22 09:57:34 I imagine there's some poor soul just checking the cards... on a Sunday.... Jul 22 09:57:49 it's probably LaForge Jul 22 09:57:54 the poor bastard Jul 22 09:57:58 mmazur: http://wiki.openmoko.org/wiki/P1_Owners ;) Jul 22 09:58:10 2280 Jul 22 09:58:15 mmazur: woot Jul 22 09:58:22 ewon: they'll be shipping out of ca.us i think, so you'll be at least in the same state - convenient for busty wench delivery Jul 22 09:58:26 ScaredyCat: or they could be iso 9000 certified :) Jul 22 09:58:26 well, it should at least be a little less stressful ewon ;) Jul 22 09:59:22 iso 9000 is a crock of shit Jul 22 09:59:45 well, the motivation behind it is reasonable, but the implementation is so, so painful. Jul 22 09:59:56 Yay, I've been charged (1851) Jul 22 10:00:02 I've worked in places with certification where they couldn't tie their own shoelaces Jul 22 10:00:42 ScaredyCat: now, did they need a document to tell them how to find the document to tell them how to tie their own shoelaces? Jul 22 10:01:07 maybe they've split orders to few groups Jul 22 10:01:26 and few of these groups are faster :) Jul 22 10:01:37 yes, but they cvouldn't find the document to tell them how to find the document to tell them how to find the document to tell them how to tie their own shoelaces Jul 22 10:01:56 ynezz: each kernel hacker gets to implement his own charging/shipping solution for his group...? Jul 22 10:02:03 lol Jul 22 10:02:09 cjb_ie: indeed Jul 22 10:02:19 it's a competition! Jul 22 10:02:39 ponts for accuracy and thoughput Jul 22 10:02:43 points Jul 22 10:04:32 someone should sort p1 owners by orderid Jul 22 10:04:55 aye Jul 22 10:05:02 * LarstiQ volunteers ScaredyCat Jul 22 10:05:08 that's strange - i got order-id 2177 but i am not listed - how is the order there? Jul 22 10:05:16 naa, that's a pyton programmers job ;) Jul 22 10:05:30 mdt: you have to add it yourself Jul 22 10:05:59 ah, understand. :D sry. but i got no other mail but the YES_I_DO request... Jul 22 10:06:09 ah bug in select, jesus Jul 22 10:06:11 :) Jul 22 10:06:17 only vey early orders seemed to get that mdt Jul 22 10:06:49 s/vey/very/ Jul 22 10:06:49 ScaredyCat meant: only very early orders seemed to get that mdt Jul 22 10:07:18 ScaredyCat: feh, I'll take your challenge ;P Jul 22 10:07:30 if you haven't had a 'We've charged your credit card' email, they haven't processed you yet Jul 22 10:07:37 ah, i see, all the orders with 'Payment Confirmed' are smaller order ids. that makes sense... Jul 22 10:07:43 LarstiQ: Good job ;) Jul 22 10:08:18 * ScaredyCat stars at the rain :( Jul 22 10:08:28 s/stars/stares/ Jul 22 10:08:32 * LarstiQ glares at the sunlight. Jul 22 10:08:42 dodgy keyboard, honest Jul 22 10:09:21 at least here, I can be pretty sure it wont flood :) Jul 22 10:10:14 You completed that task yet LarstiQ ? Jul 22 10:10:35 is there a way to participate in development? seems to be quit close the community... i wrote about some arm and gsm expirience to laforge but got no reply. Jul 22 10:10:54 ScaredyCat: no, busy fixing (other) bugs atm. Jul 22 10:11:17 he's very busy and under lots of pressure atm mdt... Jul 22 10:11:24 mdt: you can write software fine Jul 22 10:11:37 you can use the mailing lists too.... Jul 22 10:11:38 i already do Jul 22 10:11:49 theres specific ones for the gsmd etc Jul 22 10:14:13 thing is, i am about writing a software, currently running on greenphone. i don't use x11, gtk,... but i need all the other stuff like gsmd, dbus, whatever. i don't know if the arch of openmoko allows that Jul 22 10:15:24 mdt: there are images available (or you can build them) to test with Jul 22 10:16:07 yes i read about that and all the trouble with it ;) did not start because i thought this runs much smoother at a later time :) Jul 22 10:16:38 is framework-devel the right list to discuss architectural questions? Jul 22 10:16:45 or openmoko-apps? Jul 22 10:17:30 gcb77: has created a vmware image Jul 22 10:17:37 that might help Jul 22 10:17:44 gcb77: this is just the disk? do I need to create a VMWare Image around it or something? Jul 22 10:18:36 ScaredyCat, why a vmware image? as a linux development platform containing everything needed? Jul 22 10:18:48 yes Jul 22 10:19:27 isn't there a working step by step discription to succeed on a already running linux system? Jul 22 10:19:36 yes Jul 22 10:19:49 i didn't find one that works for me :/ Jul 22 10:19:58 but a) not everyone has the time b) not everyone runs linux Jul 22 10:20:08 mokomakefile Jul 22 10:20:22 try that if you don;t want the vmware image Jul 22 10:21:00 so long as you have the deps mokomakefile will work fine Jul 22 10:21:25 and the hdd space Jul 22 10:21:30 the problem I've had with getting everything up to speed is getting an emulator like QEMU to work with networking >_> Jul 22 10:21:32 and finaly you get a image for qemu? Jul 22 10:22:48 * ScaredyCat hugs his Dutch Tivo Jul 22 10:22:53 farscape :) Jul 22 10:23:26 mdt: you can Jul 22 10:23:32 mdt: you can also just do native development Jul 22 10:23:49 mdt: as far as having the right dependencies goes that should be fine Jul 22 10:24:12 LarstiQ, if i have a neo.... Jul 22 10:25:01 mdt: then what? Jul 22 10:25:18 [12:20] LarstiQ: mdt: you can also just do native development Jul 22 10:25:24 ? Jul 22 10:25:58 mdt: heh, two different ways of interpreting native :) Jul 22 10:26:06 * LarstiQ meant on the host and not the target Jul 22 10:26:07 oups... scripts of mokomakefile require bash but don't say so.... :D Jul 22 10:26:23 mdt: yeah, a known problem Jul 22 10:26:36 LarstiQ, ah, you meant you can run x86 mode or arm emu in qemu.... i see Jul 22 10:26:47 right Jul 22 10:27:06 that's nasty require a bash but... hm. okay. Jul 22 10:29:10 mdt: I've had a try at fixing that but gave up Jul 22 10:30:24 LarstiQ, is you don't do so from the begining it s a lot of work :/ Jul 22 10:30:40 devel-core does not exist :/ Jul 22 10:31:03 mdt: aye, and it's not in the moko scripts themselves but in one of the dependencies iirc Jul 22 10:31:12 so meh Jul 22 10:31:52 the wiki does not mention the debian version that is required... Jul 22 10:32:26 will it work with unstable? Jul 22 10:32:26 as in, suite? I'd think that etch would work, let me check the version of monotone it carries. Jul 22 10:32:29 mdt: duh :) Jul 22 10:32:36 hehehe, that's cool : http://rafb.net/p/J4a0jA51.html Jul 22 10:33:24 mdt: I think etch will work as well, but unstable certainly should. Jul 22 10:34:34 so how much space is needed? Jul 22 10:36:07 7.5gb Jul 22 10:36:18 oups... ;) Jul 22 10:38:10 ...and takes how long? Jul 22 10:39:44 7 hours Jul 22 10:40:03 well, on my dual xeon 2.4ghz 2 gb ram 308 mins Jul 22 10:40:21 but 7 hours is a good guideline, for the first run Jul 22 10:43:14 Takes about 5 on my core2duo Jul 22 10:46:16 Might take a bit longer now I've added INHERIT "rm_work" Jul 22 11:00:02 right, i better not even think of trying on my little crusoe. you guys will probably release twice in the time it takes to compile once :-) Jul 22 11:01:49 cjb_ie: But updates are real quick. Jul 22 11:02:18 Bur someone broke something again; Jul 22 11:02:21 http://pastebin.ca/629493 Jul 22 11:12:02 Solution to my problem: http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=641 Jul 22 11:42:25 did anyone manage to get some sound out of the emulator ? Jul 22 11:42:43 (script) wiki RecentChanges: [[P1_Owners]] [[OpenMoko:Community_Portal]] [[Getting_Started_with_your_Neo1973/fr]] [[User:Phlintphour]] [[OpenMoko_Local_Groups]] Jul 22 11:48:51 Markinoko: No, haven't managed sound yet. Jul 22 11:56:25 I have, but it's horribly laggy and slow Jul 22 12:22:49 god damned emulator... crashes my whole system Jul 22 12:50:13 Wee. I got charged. Jul 22 12:50:18 hehe Jul 22 12:50:19 gratz Jul 22 12:50:31 :-) Jul 22 12:50:45 Hm, I wonder what I'll have to do to get this x86 build set up Jul 22 12:51:46 read the wiki Jul 22 12:51:50 and than ask Jul 22 12:51:57 if something is unclear Jul 22 12:53:11 or Jul 22 12:53:25 download a pre-prep'd environment Jul 22 12:54:01 daxxar, whats your rt? Jul 22 12:54:36 2233 Jul 22 12:54:44 "DO NOT expect to be able to use it as an everyday phone for several months." from the wiki..., how long do you think it will take untill you can make/receive phonecalls/sms ? Jul 22 12:54:57 I think make/receive phonecalls is already working Jul 22 12:55:10 SMS isn't. Jul 22 12:55:13 ok Jul 22 12:55:14 But they mentioned that e.g. power-management is crappy/lacking, so battery life is short Jul 22 12:55:26 daxxar right Jul 22 12:55:32 6 hours or so Jul 22 12:55:41 didnt look at clock Jul 22 12:55:47 woglinde: I'm getting a weird make-error that rwhitby doesn't understand at all. :-S Jul 22 12:55:55 6hours of usage or idletime ? Jul 22 12:56:25 hm I think thats idenpend Jul 22 12:56:32 cant remeber Jul 22 12:56:56 The power managment sucks. Jul 22 12:57:11 And the CPU doesn't really save power unless you do active stuff with it. Jul 22 12:57:21 It doesn't use much less power when idle than active, forex. Jul 22 12:57:32 make[1]: *** read jobs pipe: No such file or directory. Stop. Jul 22 12:57:32 uh *g* Jul 22 12:57:33 ahh ok Jul 22 12:57:41 However. Jul 22 12:57:59 Based on the datasheet of the CPU and other bits, in principle, it could be a lot better. Jul 22 12:59:12 http://wiki.openmoko.org/wiki/Neo1973_Power_Management Jul 22 12:59:41 woglinde: This is my /home/moko/build/tmp/work/i686-linux/quilt-native-0.45-r1/temp/log.do_stage.32236: http://rafb.net/p/SZTesT79.html Jul 22 13:00:05 Only thing resembling an error I can see is "make[1]: *** read jobs pipe: No such file or directory. Stop.", which I can't make sense of Jul 22 13:00:17 oehm Jul 22 13:00:26 there's a power mode that turns the cpu 'off' isn;t there.. ie it sits and waits for an event trigger from gsm/gps chips Jul 22 13:00:26 In principle, it seems like playing MP3s only with the screen off should give >12 hours of life. Jul 22 13:00:38 ScaredyCat: That's basically suspend-RAM Jul 22 13:00:44 ahh k Jul 22 13:02:03 * daxxar tries "make update-makefile && make update setup openmoko-devel-image" Jul 22 13:02:07 http://wiki.openmoko.org/wiki/Kernel#Power_Management Jul 22 13:02:53 Yeah - you can turn on/off all the other stuff at will pretty much Jul 22 13:03:51 looks good :) Jul 22 13:04:52 I like the idea of suspend2 keep mode Jul 22 13:06:48 hello Jul 22 13:06:57 Hello Jul 22 13:07:27 What does NOTE: Retreived remote revisions: [] mean during make clean-package-foo? Is it fetching something from an RCS? Jul 22 13:07:28 heelo Jul 22 13:07:50 daxxar dont know whats going on Jul 22 13:07:57 but its strange Jul 22 13:08:01 They're seemingly random, sometimes '0', sometimes a two-three digit number, and sometimes what seems to be a hash Jul 22 13:08:08 "NOTE: Retrieved remote revisions: ['cc3023b9f95d7ac959a764471a65001062aecf41', '0', '2375'] Jul 22 13:08:10 and I never used mokomakefile Jul 22 13:08:20 ah thats from monotone Jul 22 13:08:28 I guess Jul 22 13:09:32 Okay. :-) Jul 22 13:09:55 Now I've cleaned quilt, goin to retry openmoko-devel-image Jul 22 13:10:44 going* Jul 22 13:15:15 Could someone who's experienced with MokoMakefile look at this buildlog and help me debug the problem? http://rafb.net/p/Xiswcn55.html Jul 22 13:18:30 Gonna try without -j3 Jul 22 13:18:52 yes Jul 22 13:19:56 But then it'll probably take forever, ;-) Jul 22 13:19:58 * daxxar takes a shower Jul 22 13:21:57 hm no Jul 22 13:28:55 * * OM Bug 645 has been created by dando(AT)gmx.de Jul 22 13:28:56 * * +/- function does not inverse the result. Jul 22 13:28:57 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=645 Jul 22 13:29:29 3 available? Jul 22 13:29:31 oops Jul 22 13:31:26 'lo Jul 22 13:32:37 daxxar: what type of machine are you building on? Jul 22 13:34:33 Strange, it worked. Jul 22 13:34:37 Intel dualcore Jul 22 13:34:53 T2400 @ 1.83Ghz Jul 22 13:34:55 GHz* Jul 22 13:35:02 NOTE: package quilt-native-0.45: completed Jul 22 13:35:23 So without -j it worked. (I've tried that earlier, but then I forgot my globally set MAKEFLAGS, I guess it heeds that too) Jul 22 13:35:48 | utimens.h:2: error: conflicting types for 'futimens' Jul 22 13:35:48 | /usr/include/gentoo-multilib/amd64/sys/stat.h:370: error: previous declaration of 'futimens' was here Jul 22 13:36:02 I need to fix for glibc-2.6? Jul 22 13:36:06 * ewon sees the word "gentoo" Jul 22 13:36:08 ewanm89_laptop: there is a patch in the OE bugtracker for coreutils Jul 22 13:36:09 ewanm89_laptop--; Jul 22 13:36:14 ewanm89_laptop: it is applied in org.oe.dev Jul 22 13:36:27 Both PARALLEL_MAKE (in local.conf) and MAKEFLAGS (in global env) was set to "-j 3" Jul 22 13:36:49 zecke: And one in openmoko bugzilla (this one didn't work). Jul 22 13:37:07 But the wiki says it should parallelize nicely on multicore systems. Strange. Jul 22 13:37:07 ewanm89_laptop: well, this is why it is the openmoko bugtracker? :) Jul 22 13:37:24 daxxar: don't set MAKEFLAGS :) Jul 22 13:37:36 Do you think it conflicts? Jul 22 13:37:47 I did MAKEFLAGS="" make openmoko-devel-image Jul 22 13:37:51 And that is currently running Jul 22 13:37:55 parallel_make? Jul 22 13:38:10 zecke: The patch didn't work and it's reported as fixed? Jul 22 13:38:31 fixed, as in oe.dev Jul 22 13:38:57 ewanm89_laptop: no idea, I don't check the openmoko bugtracker for OE bugs Jul 22 13:39:10 ScaredyCat: Yes, it's a local.conf setting to use -j-flags. Jul 22 13:39:16 oh, in locl# Jul 22 13:39:18 ewanm89_laptop: I assume it is fixed in org.oe.dev Jul 22 13:39:19 local Jul 22 13:39:34 I removed the MAKEFLAGS and I'm trying only PARALLEL_MAKE now Jul 22 13:41:27 I'll post on the wiki if it works nicely now - after my email verifies. Jul 22 13:43:43 (script) wiki RecentChanges: [[P1_Owners]] [[Wish_List_-_Hardware]] Jul 22 13:47:41 ph5, SpeedEvil - JFI: http://www.flickr.com/photos/elrond/869237803/ Any idea, what this value from the chip might be? Jul 22 13:48:07 oe patch worked :) Jul 22 13:48:23 No it didn't. Jul 22 13:50:12 :( Jul 22 13:55:10 Elrond, possibly - some feedback on phase sync or bit sync. Jul 22 13:55:23 Dmitry_Platonov - Hi!! Jul 22 13:55:34 Hi, Elrond! Jul 22 13:55:40 Dmitry_Platonov - You know, that this value is one from the two you noticed? Jul 22 13:55:57 Dmitry_Platonov - The other one is the max of the CMD23 packet. Jul 22 13:55:58 Yes. Jul 22 13:56:26 hello Dmitry_Platonov Jul 22 13:57:18 It worked when I got the patch to apply. Jul 22 13:57:45 Elrond: nothing trivially obvious pops out of my head. Jul 22 13:58:06 SpeedEvil - Yeah, same here. Jul 22 13:58:12 What's the peak - 2000? Jul 22 13:58:26 Dmitry_Platonov, SpeedEvil - Notice that his is averaged over all channels, all states of those channels. Jul 22 13:58:42 SpeedEvil - You mean the max value? 2043. Jul 22 13:58:46 hi Elrond, Dmitry_Platonov Jul 22 13:59:21 This feels close to 2*PRN-bit-pattern-length. ;o) Jul 22 13:59:23 Elrond: this is from stracing gpgll, or talking to it yourseld? Jul 22 13:59:44 err Jul 22 13:59:50 SpeedEvil - From ph5's cold2.log and longtrace.dat, gllin. Jul 22 13:59:54 err, not bit/phase. something else. Only known it goes down with time. 8) Jul 22 13:59:55 the gps bin, whatever it's called. Jul 22 14:00:02 hi pH5! Jul 22 14:00:48 The spike _could_ nicely indicate locking in software. Jul 22 14:00:53 Elrond: which channel is this? Jul 22 14:00:57 Dmitry_Platonov - Well, when not in sync, it is just noise between 0 and 2000. ;o) (I tried this morning. My trying got me to the idea, that it is only between 0 and 2040.) Jul 22 14:01:05 pH5 - All channels. ;o) Jul 22 14:01:32 Elrond, it's certainly some king of feedback! Jul 22 14:01:41 ho, I think I have an idea. Jul 22 14:01:42 2048 - would nicely correlate to I and Q bins for 1024 code positions. Jul 22 14:02:07 datasheet says massive parallel engine or something like this. Jul 22 14:02:10 yeah Jul 22 14:02:25 It doesn't say how many. I was assuming 2048 Jul 22 14:02:26 Yep. Jul 22 14:02:41 i think it's number of correlator shown in cmd 23 Jul 22 14:03:18 Dmitry_Platonov - Huh? If I redo the same measurement over and over again, this value is noise between 0 and 2043. Jul 22 14:03:36 So how can that be number of correlations? Jul 22 14:03:38 Elrond: plot it as time/x Jul 22 14:03:41 hm. let me explain - my english is not so good. Jul 22 14:03:48 so that if it's a moving peak, you can spot it Jul 22 14:03:51 SpeedEvil - For one channel? Jul 22 14:03:59 yes Jul 22 14:04:12 SpeedEvil - I did this morning. For a non-locked SV. Jul 22 14:04:26 SpeedEvil - hoping around like mad. Looks like noise to me. Jul 22 14:04:37 Plotting it against time? Jul 22 14:04:44 there are 2043 instances of PRN for each channel. with own counters Jul 22 14:04:44 Yes. Jul 22 14:05:15 Dmitry_Platonov - correlators? Jul 22 14:05:51 and post-processor finds max, put window around max in CMD 23 and it's index in lw REG 11 Jul 22 14:05:54 It could also be the carrier phase correlator Jul 22 14:06:06 which tells you about exact bit timing. Jul 22 14:06:57 Dmitry_Platonov - Ahh, that might make sense. Jul 22 14:07:26 If we're not locked, the max will be anywhere... if we're locked, it will be somewhere. Jul 22 14:07:53 Elrond: can you take the packet that gllin sends to the GPS, copy it, and before the satellite moves much, resend it. Jul 22 14:08:03 This may get you into a sort-of-locked state Jul 22 14:08:04 does not looks like phase. i think PLL is set up when SEND FL6 CMD05 / RECV FL6 CMD9d takes place with wery short timings. Jul 22 14:08:04 Okay, I'll extract channel 01 from cold2.log and plot it over "time" Jul 22 14:08:26 * Dmitry_Platonov agrees with elrond Jul 22 14:09:05 (while neo is recreating noise curve ;o) ) Jul 22 14:09:07 That actually may not tell you much. Jul 22 14:09:25 If it's locked in software - then... hmm Jul 22 14:09:51 You'd see minor movements as the software PLL oscillates about the set-point Jul 22 14:10:18 datasheet shows hardware PLL. Jul 22 14:10:42 IIRC, there is only one hardware PLLL though. Jul 22 14:10:50 Isn't that simply the local clock Jul 22 14:11:54 I guessed that you sent the hardware instructions on how to setup the correlators to listen to a satellite at a specified time, and frequency offset from the local clock Jul 22 14:12:05 http://phillryu.com/2007/07/21/the-power-of-good-ui-design/ Jul 22 14:13:09 So you can say 'listen for PRN3 at a doppler of -1Km/s starting at time 0.38834532s Jul 22 14:14:01 And if you've got all the numbers right, then the correlator peak will stay in the place you were expecting, as you're varying the doppler and time settings so it does. Jul 22 14:15:28 Hm, no IR on GTA01, right? Jul 22 14:15:40 Where 'starting at' is 'time if correlator peak was at 0' Jul 22 14:15:45 daxxar: nor 02 Jul 22 14:16:49 http://www.flickr.com/photos/elrond/869364425/ Jul 22 14:17:48 it's known as carrier/code lock. but it's not all. Jul 22 14:18:11 daxxar: bluetooth is better Jul 22 14:18:14 we need bit sync too. Jul 22 14:18:25 and phase too. Jul 22 14:18:39 daxxar: after bluetooth started being used more, phones stopped coming with IR Jul 22 14:18:58 daxxar: and I'd guess palm PDAs only come with IR for backwards compability reasons Jul 22 14:19:31 A while since I saw a laptop with IR. Jul 22 14:20:10 what happens with CMD10 before #48 in that last plot where the value freezes? Jul 22 14:20:22 cesarb: Hehe, true. ;-) Jul 22 14:20:35 eh CMD00 Jul 22 14:20:39 cesarb: IR is nice for using your device as an universal remote, but I can't think of anything else I'd want to use it for. Jul 22 14:20:52 ewanm89_laptop: mine has IR and i regular use it. It's slower than wlan, but much harder to intercept Jul 22 14:21:12 daxxar: universal, as long as none of your remotes use radio... Jul 22 14:21:24 cesarb: You can't have everything. ;-) Jul 22 14:21:25 * cesarb wonders how long until remotes start using bluetooth too Jul 22 14:21:33 I use IR for my media-center at home Jul 22 14:21:37 do you count wiimotes? Jul 22 14:21:40 (which runs Linux and custom software) Jul 22 14:21:53 Because IR receivers and remotes are cheap and easy to find. ;-) Jul 22 14:21:58 ben_goodger: that's no remote, that's a thinly disguised joystick Jul 22 14:22:15 daxxar: just plug a USB BT dongle on it ;-) Jul 22 14:22:35 daxxar: I didn't say they aren't I said they don't build them into the newer laptops. Jul 22 14:22:39 cesarb: Then I need to find / write BT software for my cellphone or similar, and BT remotes are more expensive. ;-) Jul 22 14:22:46 ewanm89_laptop: What? o.0 Jul 22 14:22:48 daxxar: or, with GTA02, use TCP (with neo's wifi) Jul 22 14:22:54 Hehe, yeah, that'll be nice. <34 Jul 22 14:22:57 <3* Jul 22 14:23:11 pH5 - I don't know... But I guess, it's the lock... It's channel 01 from cold2.log, so you can count yourself. ;o) Jul 22 14:23:11 daxxar: you can keep your IR remote, and use BT/TCP as an alternate way of controlling it... Jul 22 14:23:18 Elrond: there are only four values on PRN27 before that register hits the 0x37, the rest is from fruitless searching on other PRNs Jul 22 14:23:20 I think, someone must start working on coldstart.c (or .py) ^_^ Jul 22 14:23:26 I'd want to write the application for J2ME, so that my flatmates can use their non-freed phones ;-) Jul 22 14:24:01 daxxar we will have midpath faster in openembbed than moko can send sms Jul 22 14:24:05 pH5 - Ahh. :) Jul 22 14:24:15 daxxar: if their non-freed phones are like mine, you don't have AFAIK a way to get to BT (it doesn't have IR) from the J2ME VM Jul 22 14:24:24 and it's already at 0x39 the first time after this ominous 'range switch' takes place where your reg6 calculation doesn't hold anymore. Jul 22 14:24:35 BTW, how much cpu time eats gllin? Jul 22 14:24:52 Dmitry_Platonov - About 10%. Jul 22 14:25:24 woglinde: "midpath"? Jul 22 14:25:48 Dmitry_Platonov - Starting gllin for you. ;) Jul 22 14:25:52 cesarb: I *think* at least the phone one of them has allows for it, I think he wrote some bt-application, but I might be mistaken. Jul 22 14:25:56 woglinde: A j2me VM? Jul 22 14:25:59 daxxar right Jul 22 14:26:15 its cacao with cldc1.1 and midp1.0 Jul 22 14:26:16 (I'll google in a bit, just need to finish this mail;) Jul 22 14:26:37 PID USER STATUS RSS PPID %CPU %MEM COMMAND Jul 22 14:26:38 10% - is that in steady state? Jul 22 14:26:38 1067 root S < 6872 1048 8.6 5.4 Xfbdev Jul 22 14:26:44 1184 root S 2928 1 3.5 2.3 gllin Jul 22 14:26:44 That seems a bit high. Jul 22 14:26:48 oh Jul 22 14:27:11 That is in search-mode, because it can't get any locks (despite being close to my window ;o) ) Jul 22 14:27:18 cesarb: isn't there a JSR that enables BT access from J2ME? Jul 22 14:27:44 Elrond: this register ends up at 0x37 for most channels. Jul 22 14:27:46 drath yes Jul 22 14:27:53 drath_: yes. And, AFAIK, the Razr's (very weak) J2ME VM doesn't have it. Jul 22 14:28:00 ah, ok Jul 22 14:28:21 (which is AGAIN the reason I want a freed phone: if it doesn't work, I CAN FIX IT) Jul 22 14:28:21 in the back of my head i seemed to remember using a BT java app on my nokia 6130 Jul 22 14:28:41 ch1, ch2, ch3, ch6, ch7 and ch8 Jul 22 14:28:44 in cold2.log Jul 22 14:29:01 superwaba's graphics primitives are... primitive Jul 22 14:29:16 pH5 - See http://www.flickr.com/photos/elrond/869237801/ ;o) Jul 22 14:29:16 I can only draw ellipses that have the minor and major on the same plane as the x and y axes Jul 22 14:29:33 Elrond: right :) Jul 22 14:29:37 drath/cesarb/daxxar one .bb I only have to create a last .bb for midpath than jvm is in oe Jul 22 14:29:55 woglinde: Cool :-) Jul 22 14:29:58 And the new JavaFX for Mobile, isn't a good choice for SuperWaba ?? Jul 22 14:30:34 Vndmtrx javafx is something like c# Jul 22 14:30:48 aeh no visbual basic# Jul 22 14:30:49 ewon: that's not a limitation if you can apply a transformation matrix to it Jul 22 14:30:49 or so Jul 22 14:31:05 woglinde, humm Jul 22 14:31:18 ewon: Don't all elipses that have major and semi-major axes not on the plane when projected onto the plane have identical elipses that are defined by major and semi-major axes on the plane? Jul 22 14:31:33 ewon: (of course, I have no idea if you can use transformation matrices in superwaba) Jul 22 14:31:40 cesarb: it appears not Jul 22 14:31:48 SpeedEvil: not that I've seen, but my maths sucks. Jul 22 14:32:05 I just want to draw a bounding ellipse around some arrow impact points Jul 22 14:33:13 What's the biggest package built by default? glibc? Jul 22 14:33:20 ewon: Do you mean 'does not support major and semi-major axes in line with the plane'? Jul 22 14:33:26 daxxar: X I'd imagine Jul 22 14:33:32 ewon: Do you mean 'does not support major and semi-major axes in line with the plane axes'? Jul 22 14:33:35 Ah, yeah. Jul 22 14:33:41 i'm more interested in meaning of REG 06 of CMD 00 Jul 22 14:33:44 ewon: Do you mean 'does not support major and semi-major axes not in line with the plane axes'? Jul 22 14:33:45 rather Jul 22 14:34:23 daxxar biggest or most time consuming? Jul 22 14:34:57 Dmitry_Platonov, ph5, SpeedEvil - One channel only, time series for the "base search command": http://www.flickr.com/photos/elrond/869433877/ Jul 22 14:35:01 busybox is sure taking a while. Jul 22 14:35:14 woglinde: most time consuming Jul 22 14:35:19 Elrond: is this in a place where the GPS can get a fix? Jul 22 14:35:25 Or is this form the log Jul 22 14:35:42 SpeedEvil: yes. Jul 22 14:35:52 SpeedEvil - This is from indoors, so no chance for lock/fix. Jul 22 14:36:50 Ah Jul 22 14:37:14 SpeedEvil: X is built once. glibc is built three times. Jul 22 14:37:21 Oh. Jul 22 14:37:25 SpeedEvil - I really need to get a long usb cable soon, so I can sit at my desktop while neo is outside. Jul 22 14:37:53 SpeedEvil: and, if you are playing with EABI like me, you'd have to build it SIX times (uboot doesn't like EABI, so you also need to compile it with the old ABI, just for uboot's sake) Jul 22 14:38:07 Elrond: USB is only any good upto 5 metres I think. Jul 22 14:38:27 add a hub, and it goes up 5m/hub Jul 22 14:38:45 * daxxar is watching the youtube-video-ads about neo1973 Jul 22 14:38:48 SpeedEvil: there's a limit to the number of hubs Jul 22 14:38:49 Hilarious :-P Jul 22 14:38:52 SpeedEvil: If hub is powered. Jul 22 14:39:01 Or if hub is not powered. Jul 22 14:39:36 You can in theory string 4 hubs and a low-powered device in series. Jul 22 14:39:38 I suppose you could use the power off the usb. Jul 22 14:39:40 non-powered Jul 22 14:39:53 ewanm89_laptop - My desktop is 2m away from the balcony door. The 1m usb cable, that came with the neo lets it sit about 80cm to the window. Jul 22 14:40:15 move your desktop closer to the window ;) Jul 22 14:40:37 I'll just take my laptop outside, and sit out there. Jul 22 14:40:39 Or move the window closer to the desktop. Jul 22 14:41:12 SpeedEvil - *lol* Jul 22 14:41:34 ewanm89_laptop - laptops are quite unconfortable to me. Jul 22 14:41:35 take desktop out. enjoy the weather Jul 22 14:42:01 Dmitry_Platonov: Network cables become a problem. Jul 22 14:42:04 Elrond: REG6 = int( 0x122fc9b1 + REG5 * 124.0/38265.0 ) during the second phase of the lock? Jul 22 14:43:49 pH5 - That comes very close... Jul 22 14:44:00 I have: Jul 22 14:44:01 # m = 307.97542 +/- 0.00582 (0.00189%) Jul 22 14:44:01 # BASE2 = 305121712.5821 +- 0.0088 (2.888e-09%) Jul 22 14:44:14 While you: Jul 22 14:44:26 m = 308.588 Jul 22 14:44:42 BASE2 = 305121713 Jul 22 14:44:44 I didn't do a proper regression Jul 22 14:45:14 though I guess you can fix BASE2, given we know the value for 0 offset in REG5 Jul 22 14:45:44 pH5 - Probably. Jul 22 14:46:09 pH5 - I didn't look much into reg6-range2 for a while. Jul 22 14:47:51 what is SAT_CMD00_REG5_BASE? Jul 22 14:48:20 Dmitry_Platonov - The "0" offset of that register. Jul 22 14:57:53 Oh well, I guess, we can certainly be much, much better than gllin. I have put the neo on the balcony, tail -f gps.nmea in gtkterm2. It can see 40% of the sky directly. No lock. Jul 22 14:57:57 s/lock/fix Jul 22 14:58:48 Anyway, I'm getting some cake. Jul 22 15:00:22 stefan_schmidt: ping Jul 22 15:00:37 alphaone: ping Jul 22 15:00:54 Shoragan: ping Jul 22 15:01:15 lazy germans waking up at 5 pm Jul 22 15:01:33 that's gotta be the lonelyest game of ping pong ever... Jul 22 15:05:19 mickeyl: I got a design question as well Jul 22 15:06:34 zecke: i'm all ears Jul 22 15:07:07 mickeyl: already? I thought I would need to draw a number and sit down :) Jul 22 15:07:27 mickeyl: WebView is the Mac API for WebKit (http://developer.apple.com/documentation/Cocoa/Reference/WebKit/Classes/WebView_Class/index.html) Jul 22 15:07:48 mickeyl: it is using Delegates to implement Policy,Tracking of Progress,UI Changes Jul 22 15:08:20 mickeyl: mapping that to GObject/Gtk+ is the question. The easy ones are simple notifications like "Title Changed" and it is just a simple signal Jul 22 15:08:24 Elrond, pH5 if REG5/REG6 seems to be coupled, can it be that one is increment for carrier generator, and other for code generator? Jul 22 15:09:00 mickeyl: more interesting are policies (e.g. download a file, open a new window, follow the link). Which is async and the first approach would be to force subclassing but that is not optimal Jul 22 15:09:58 mickeyl: three options I see are: signal+default handler (forcing people to use g_signal_stop_emission_by_name), force subclassing, use delegates as well Jul 22 15:10:59 zecke - Default handler can also be overriden by returning "True" instead of "False". That's the way x11 events are handled in Gtk. Jul 22 15:11:02 zecke: in that case my gut feeling says signal+default handler Jul 22 15:11:42 Elrond: I would have a signal with return type != G_TYPE_NONE how would that work? Jul 22 15:15:09 WishList:BuiltinScriptingLanguages is funny Jul 22 15:15:14 (But if we're all lucky, a clear-minded concensus will prevail in everyone's understanding that Java just doesn't belong on an embedded device.) Jul 22 15:15:22 ... Jul 22 15:15:26 It's not an embedded device. Jul 22 15:15:28 # If we're really lucky, people will realize that Java Virtual Machine belongs on embedded devices and has great performance there. Jul 22 15:15:30 It's a desktop. Jul 22 15:16:04 I'm not arguing the fact, but that page contradicts itself - since it's based on people's opinions. ;P Jul 22 15:16:11 It's a pentium 100 desktop that nobodies told it that it's a phone. Jul 22 15:16:39 Java - as a scripting language - is insane though. Jul 22 15:19:15 java is an insane language anyway. Jul 22 15:19:27 Give me ruby anyday. Jul 22 15:20:15 Perl Jul 22 15:20:21 awk! Jul 22 15:20:25 sh Jul 22 15:20:58 Aw, heck, let's just code everything in assembly. Jul 22 15:21:08 amen Jul 22 15:21:12 Can be more readable than perl. Jul 22 15:21:20 I only write in Java when writing superwaba apps. And I only write for Superwaba because it lets me write cross-PDA gui apps relatively fast. Jul 22 15:21:30 with the right macros asm is best Jul 22 15:21:34 Though practically it has fundamental problems. Jul 22 15:21:39 Java is today's version of COBOL Jul 22 15:21:43 heh Jul 22 15:21:51 * mickeyl applauds mwester Jul 22 15:21:58 Basically that modern CPUs with multiple execution units are hard for sane humans to write optimal code for. Jul 22 15:22:06 cobol & java & vb - languages for the mongolian hoards Jul 22 15:22:34 writchie: what language whould you consider optimal ? Jul 22 15:23:00 depends on application Jul 22 15:23:08 c++ c or asm Jul 22 15:23:26 all unsuitable for scripting Jul 22 15:23:38 IMO, that's why one has multiple languages -- depends on the problem. SQL is a fine language -- for database queries. Jul 22 15:23:45 ruby or php for scripting :) Jul 22 15:24:19 vb is even worse than java. Jul 22 15:24:23 We just need to agree on a fine language for scripting the device -- what has the best bindings to the UI, fits the device, and fits the type of developer and application being written Jul 22 15:24:32 haskell rules the world! Jul 22 15:24:41 zecke - Huh, I would need to check gtk sources, but signals can have return values. Jul 22 15:24:42 what about python Jul 22 15:24:46 No it doesn't ruby does. Jul 22 15:25:02 eugh, python isn't freeform. Jul 22 15:25:06 mwester: And is small. Jul 22 15:25:21 Javascript! Jul 22 15:25:28 +XUL+XBL Jul 22 15:25:33 emacscript Jul 22 15:25:36 * ScaredyCat lol's Jul 22 15:25:37 Millions of firefox users can't be wrong. Jul 22 15:25:51 why can't they be wrong? Jul 22 15:26:01 SpeedEvil: Accept I need a whole different javascript for IE. Jul 22 15:26:05 millions of vb user can't be wrong? Jul 22 15:26:31 i mean vb pseudo programmers Jul 22 15:26:34 writchie: one could say .... millions of windows user's can't be wrogn .... Jul 22 15:26:34 :P Jul 22 15:26:38 The damn well are, idiots, think they can code, I'll show them real code. Jul 22 15:26:39 we use python for the gps test thing on the device. works fine for me Jul 22 15:27:06 python has too many wasted bytes in whitespace for an embedded device Jul 22 15:27:13 (yes, I'm trolling) Jul 22 15:27:20 Qwell: I agree. Jul 22 15:27:32 not for embedded with arm9 and gb of storage Jul 22 15:27:40 I hate being forced to indent. Jul 22 15:27:44 Qwell - So we want sendmail-cf for scripting, great! ;-) Jul 22 15:27:53 Qwell: so we should use something broken like perl? Jul 22 15:27:55 yay m4 Jul 22 15:27:57 with an m4 front-end Jul 22 15:28:05 m4? Jul 22 15:28:08 Wimp. Jul 22 15:28:17 No m4. sendmailCF, not sendmail.mc Jul 22 15:28:18 gnu macro processor Jul 22 15:28:19 ouch. Jul 22 15:28:39 'Use voice recognition to unlock the phone. "Hi. My name is ... My voice is my passport. Verify me."' - Isn't that the string from Uplink? :-P Jul 22 15:28:41 yay, sendmail.cf \o/ Jul 22 15:29:10 I wrote a sendmail.cf once. Jul 22 15:29:15 daxxar - If "uplink" is that movie, where they break into the bank at the beginning: Yes. Jul 22 15:29:38 (to mail all messages of the form 0000.*NO_UCE to one user) Jul 22 15:29:49 messages to Jul 22 15:30:00 Actually, I was referring to the PC game. Guess they stole it from somewhere else. ;-) Jul 22 15:30:03 It was an idea to post each usenet message with a unique ID. Jul 22 15:30:12 (by Introversion) Jul 22 15:30:14 This was just after the green card spam. Jul 22 15:30:26 So did all of that result in some idea of what 'scripting' language should be used? I assume it's a toss-up between Ruby and Python, they're they flavours of the month... Jul 22 15:30:30 uplink is a great game, was just playing it :) Jul 22 15:30:36 This eventually lead to me recieving 7% of the entire incoming email for my ISP one day. Jul 22 15:30:52 ewanm89_laptop: does this game work under linux/debian? Jul 22 15:31:00 noon: Of course. Jul 22 15:31:09 jgm: It's ideally gotta be in under a few meg. Jul 22 15:31:19 wasn't that quote from sneakers ? Jul 22 15:31:30 noon: introversion.co.uk, grab the demo, and darwinaia and defcon, all 3 are great. Jul 22 15:31:34 As you really want the base OS to fit on the internal RAM - which is 64M Jul 22 15:32:08 How much space does the existing install take up i.e. how much is there to play with? Jul 22 15:32:39 I might load them on to my neo, not sure on what to do with a few of the defcon and darwina controls (yet). Jul 22 15:33:10 daxxar - Sneakers: http://www.imdb.com/title/tt0105435/ Jul 22 15:34:30 Aha. :-) Jul 22 15:34:31 Elrond: in Sneakers the plot is all about a decrypting box. nothing with voice auth. Jul 22 15:34:50 noon - voice auth to get into cosmo's building. Jul 22 15:35:15 noon - The women needs to catch the relevant words from that one engineer. Jul 22 15:37:08 "hi my name is werner brandes. my voice is my passport. verify me" Jul 22 15:37:23 didn't even have to look that up... Jul 22 15:37:34 getting him to say passport proves near impossible tho Jul 22 15:38:00 Elrond: i only remember the infared scanner Jul 22 15:38:18 * ewon considers a GPIO-fired thermite charge Jul 22 15:38:25 "secure!" Jul 22 15:38:31 *ppppffffssssssssssst* Jul 22 15:38:59 good luck boarding a plane then Jul 22 15:39:04 heh :) Jul 22 15:39:43 ewon: you could plug a usb-rocket launcher onto the neo, no need to waste a GPIO :-) Jul 22 15:39:59 nice :) Jul 22 15:40:07 or the usb red button to launch the thermite charge Jul 22 15:40:12 noon, but then you'd need to get external power for it Jul 22 15:40:51 http://www.reghardware.co.uk/2006/07/18/usb_nuclear_war_button_box/ Jul 22 15:41:25 i wonder if you can interface with the switches Jul 22 15:42:20 !!!BIG READ BUTTON!!! Jul 22 15:42:50 buttons are not HID devices alas, the thing just uses the USB for power Jul 22 15:42:53 it's a pity Jul 22 15:43:04 I looked into it a while back :) Jul 22 15:43:23 (script) wiki RecentChanges: [[Java]] [[Basic_End-user/fr]] [[P1_Owners]] [[OpenMoko_Local_Groups:_Birmingham]] [[Category:OpenMoko_Local_Groups:_Norway]] [[OpenMoko_Local_Groups:_Trondheim]] Jul 22 15:44:14 ewon: I'm sure we could make a USB HID version. Jul 22 15:44:30 anyone here in Toronto, Canada? Jul 22 15:45:40 mhh might need to build our own then Jul 22 15:45:46 take apart a usb mouse or so ;) Jul 22 15:46:17 gets you the hid part easily Jul 22 15:46:26 What... that thing is only a hub + taking power from usb? No usb-device on its own. And htey expect to get 36Eur for it? Jul 22 15:46:47 oh sure why not Jul 22 15:46:51 it looks killer ;) Jul 22 15:46:54 (pun intended) Jul 22 15:46:55 Elrond: a hub is a usb device :-) Jul 22 15:47:23 take cheap hub, add case and mouse ;) Jul 22 15:47:33 They could have made it 39Eur and use usbhid, job done. Jul 22 15:47:33 thats why you can't put 127 mouse onto your computer... Jul 22 15:48:13 heck use a usb numpad Jul 22 15:48:22 16 buttons ;) Jul 22 15:48:44 i think i must build one of those ;) Jul 22 15:49:21 buz: cypress sells nice hid usb chips. the one i had my fingers on had 24 I/O Jul 22 15:49:43 i'm totally incapable of soldering Jul 22 15:50:02 philips and stc make arm7 based usb devices for a couple bucks Jul 22 15:50:09 BTW: Can usbhid be abused for serial emulation? (yes, cdc_serial is better anyway) I think, I read somewhere, tht some devices abuse usbhid for this. Jul 22 15:50:54 could be used for a crude serial but why bother Jul 22 15:51:00 10? for a usb numpad Jul 22 15:51:01 usbhid is report based Jul 22 15:51:12 including the usb hub ;) Jul 22 15:51:59 maybe add a webcam for good measure Jul 22 15:52:05 for iris auth Jul 22 15:52:09 *G* Jul 22 15:52:23 Elrond: usbhid is slow. Jul 22 15:52:32 THAT would actually be a good use for a cam on neo Jul 22 15:52:53 Does iris auth work with those tiny webcams? Jul 22 15:52:57 I doubt it. Jul 22 15:53:21 probably not Jul 22 15:53:32 you most likely need a cam that has some lighting of its own Jul 22 15:53:39 more of a scanner than a cam Jul 22 15:53:55 buz: you need a kewl blue or green laser for this to work Jul 22 15:54:16 plus point if it throws a line on the whole room Jul 22 15:54:19 Which brings us to "everything is better with lasers"(from the wiki). ;o) Jul 22 15:55:44 mickeyl, pong Jul 22 15:55:59 * Shoragan was away diving Jul 22 15:56:16 diving? nice. Jul 22 15:56:31 Shoragan: when will you see stefan and daniel next time? Jul 22 15:57:13 i don't know when daniel will return from the UK, probably later today Jul 22 15:57:18 stefan perhaps monday Jul 22 15:57:21 Which reminds me... where's g_serial in the default kernel builds on buildhost? Jul 22 15:57:52 Shoragan: ok. please tell them to give me a call. we need to schedule some work. August will be a tough month Jul 22 15:58:00 Elrond: yes - with the caveat that you'd need a close focus lens. Jul 22 15:58:05 ok Jul 22 15:58:09 thanks Jul 22 15:58:31 SpeedEvil - Yeah, and most webcams don't have that, I'd guess. Jul 22 15:58:32 i've started to port our patches for 2.6.22 Jul 22 15:58:41 oh cool Jul 22 15:58:50 2.6.22 breaks a lot (again) on arm as I've heard Jul 22 15:58:54 Shoragan - Ahh, nice. :-) Jul 22 15:59:00 is harald already working on it? Jul 22 15:59:25 Elrond: The iris is around 1.5cm across - the idea is that you treat the iris as a 2D barcode basically. Jul 22 15:59:44 Elrond: and most actually can - if they have manually variable focus. Jul 22 16:00:07 Shoragan: i hardly doubt that Jul 22 16:00:14 mokomakefilefails: http://pastebin.com/m1b717e72 any idea? Jul 22 16:00:46 mdt: this means you are using a too recent version of ubuntu Jul 22 16:00:54 cesarb, i don't Jul 22 16:00:57 mdt: it has been discussed several times around here already Jul 22 16:01:04 oups, sry Jul 22 16:01:22 mdt: it's an incompatibility with the most recent glibc Jul 22 16:01:47 mdt: There is a patch in openembedded that works. Jul 22 16:01:49 * cesarb doesn't know the current best fix, sorry Jul 22 16:01:52 it is not mentioned in the wiki, or did i miss that? Jul 22 16:02:15 mickeyl, how should i handle the asoc patch? some of it seems to have been merged in .22 Jul 22 16:02:23 mickeyl - BTW: Did you get to look at #631? Jul 22 16:03:55 ewanm89_laptop, can u tell me a url where to find that? Jul 22 16:05:15 http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/coreutils/coreutils-5.3.0 Jul 22 16:06:01 Elrond: not yet, no Jul 22 16:06:25 atm. i'm unable to look at any bugs. hopefully in a couple of weeks -- after the p1 snapshot release Jul 22 16:06:27 Put it in the same place in local tree and overwrite the coreutils-5.3.0 bb file with the one from http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/coreutils Jul 22 16:07:07 mickeyl: you are the one who is managing the distribution, right? Jul 22 16:07:26 mickeyl: I'm trying to make things compile with EABI, and will post a guide on the wiki after I make it work Jul 22 16:07:58 mickeyl - It's a patch only. ;o) Jul 22 16:08:24 cesarb: cool. Jul 22 16:09:32 mickeyl, i'll skip the alsa stuff for now... if the rest works, alsa should be easier Jul 22 16:09:45 Shoragan: ya Jul 22 16:10:31 ewanm89_laptop, wouldn't it be helpful to put that info into the wiki? so newbies like me don't have to ask here? Jul 22 16:11:06 mdt: it took until yesterday IIRC for it to be put on the wiki the FAQ that you need gcc 3.4 or 3.3 installed to compile qemu Jul 22 16:11:23 mdt: so, expect it to take a few more weeks until it gets put on the wiki page ;-) Jul 22 16:11:28 mdt: I only found the exact details this morning, when the patch in the openmoko bugzilla didn't work. Jul 22 16:12:12 ... replacing the .bb alone does not work, it needs the actual patch also. where can i find that? Jul 22 16:13:17 mdt: In the first link I gave. Jul 22 16:13:38 http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/coreutils/coreutils-5.3.0 Jul 22 16:14:45 mdt: you could add that info to the wiki yourself. ;) Jul 22 16:14:56 Kensan, can i? Jul 22 16:15:37 ScaredyCat: http://richtlijn.be/~larstiq/neo/owners.txt\ Jul 22 16:15:58 mdt: every registered user can edit the wiki. Jul 22 16:17:18 mokobot: glibc Jul 22 16:17:18 Grab futimens.patch from http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/coreutils/coreutils-5.3.0 and replace the coreutils_5.3.0.bb file with the one from http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/coreutils/ Jul 22 16:17:59 nice LarstiQ .. but what happened to the last one Jul 22 16:18:55 * * OM Bug 646 has been created by cesarb(AT)cesarb.net Jul 22 16:18:56 * * Bashism on linux-gta01.inc Jul 22 16:18:58 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=646 Jul 22 16:19:49 ScaredyCat: some people insist on not following conventions, not filling in order ids, etc Jul 22 16:19:58 :) Jul 22 16:20:03 ScaredyCat: they get sorted at the end ;) Jul 22 16:20:13 where they belong ;) Jul 22 16:20:36 anyway, the script to produce the output is next to it Jul 22 16:23:06 :) Jul 22 16:23:55 * * OM Bug 647 has been created by cesarb(AT)cesarb.net Jul 22 16:23:56 * * linux-gta01.inc uses wrong objcopy Jul 22 16:23:57 * * http://bugzilla.openmoko.org/cgi-bin/bugzilla/show_bug.cgi?id=647 Jul 22 16:28:00 okay, thanks alot, it works, i wrote something in the wiki about that... Jul 22 16:28:45 mokobot: glibc | mtd Jul 22 16:28:52 mokobot: glibc Jul 22 16:28:53 Grab futimens.patch from http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/coreutils/coreutils-5.3.0 and replace the coreutils_5.3.0.bb file with the one from http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/coreutils/ Jul 22 16:29:24 mdt: I put it there too. Jul 22 17:02:20 hi. any idea what raster graphics libraries (e.g. GD, Magick, Cairo) will be available on openmoko? Jul 22 17:02:44 anything you want Jul 22 17:02:50 5 DOLLAH Jul 22 17:02:59 * ewon has a full metal jacket flashback Jul 22 17:03:06 bye Jul 22 17:03:24 cairo would be nice, but it's quite intensive on the cpu. Jul 22 17:04:30 ~lart ewon Jul 22 17:04:30 * apt readies the nuke launcher and fires some rounds at ewon Jul 22 17:04:30 * mokobot frags ewon with his BFG9000 Jul 22 17:04:39 ewanm89_laptop: every raster lib is Jul 22 17:04:48 ewanm89_laptop: at least cairo is fixed-point inside Jul 22 17:05:02 I thought of that after I hit enter. Jul 22 17:05:11 anyone know what the difference between us and eu wifi is? number of channels? Jul 22 17:05:27 buz: none Jul 22 17:05:36 buz: 802.11g is 802.11g Jul 22 17:05:46 koen: There are differences. Jul 22 17:05:49 yeah but not all jurisdictionsa llow all channels Jul 22 17:06:09 I'm not sure on the exact details. Jul 22 17:06:18 isn't us usually configured for 6 and up Jul 22 17:06:27 eu 5 and down? Jul 22 17:06:45 US is channel 1 to 11 and EU 1 to 14or something like that Jul 22 17:06:47 i have linksys eu equipment, that goes from 1 to 11 Jul 22 17:07:01 ithink i could patch it to go to 14 Jul 22 17:07:15 you lucky EU bucko's Jul 22 17:07:19 anywhoo back to potter Jul 22 17:07:29 I should know them, I plan to make some ham equipment on that band, um must check. (yes it's a amateur radio band too). Jul 22 17:08:39 CoreDump|afk - *ping* Jul 22 17:09:03 pong Jul 22 17:09:28 CoreDump|afk - I somehow remember, that you live near FFM? Jul 22 17:09:39 well, cologne Jul 22 17:09:55 Huh, okay, I was quite wrong. ;o) Jul 22 17:09:56 ewanm89_laptop: the GTA02 has a GPU Jul 22 17:09:57 I read the backlog ;) Jul 22 17:10:12 CoreDump|afk - Okay. ;-) Jul 22 17:10:36 Elrond: do they ship neos to Rivendell? Jul 22 17:10:38 loufoque: Yes, I suppose it'll be feasible then. Jul 22 17:10:54 loufoque - They did already. ;o) Jul 22 17:11:22 Elrond: prove it with nifty pictures Jul 22 17:11:51 Pictures can lie. Jul 22 17:12:14 loufoque - http://www.flickr.com/photos/elrond/sets/72157600292781326/ Jul 22 17:12:37 Elrond: but.... it's not orange! Jul 22 17:12:43 heh, and it seems they even have GPS coverage in middle earth. Jul 22 17:13:19 pH5 - It's very bad. ;o) Jul 22 17:14:09 loufoque - Why should it be orange? Jul 22 17:15:07 because it's orange on the website Jul 22 17:15:28 loufoque - When you order it, you can choose the color. Jul 22 17:15:42 what happened to the green neos? Jul 22 17:15:49 green? Huh? Jul 22 17:15:55 Vegar: if there were green neo's i'd have bought one already Jul 22 17:15:55 I never saw a green one Jul 22 17:16:20 http://www.linuxdevices.com/news/NS2986976174.html Jul 22 17:16:25 loufoque - Hint: I have GTA01Bv03 ;o) Jul 22 17:16:28 green and red Jul 22 17:17:08 ahh Vegar those are just mock ups Jul 22 17:17:20 now that they exist they've got a diferent color scheme Jul 22 17:17:32 loufoque: There are 2 colour schemes available atm, orange and white, or black and silver. Jul 22 17:17:33 ah, just like I thought Jul 22 17:18:06 ok, so how many people in here have a neo Jul 22 17:18:14 The green and white does look quite good. Jul 22 17:18:24 About a dozen or so in channel. Jul 22 17:18:30 Of various releases. Jul 22 17:18:47 if they had the green and white ones, i'd own it, i'm a sucker for anything green Jul 22 17:18:55 lol Vegar you're a jerk Jul 22 17:19:08 what did I donow? Jul 22 17:19:10 do now* Jul 22 17:19:10 in the article Sean says " FIC has the capacity to ship 100,000 of the phones per month, Moss-Pultz said, explaining its low cost. " Jul 22 17:19:22 and we're F*ing waiting for a copule thousand Jul 22 17:19:31 hehe Jul 22 17:19:33 They can ship 100000 to one location. Jul 22 17:19:47 Shipping 100000 to 98000 locations is rather different. Jul 22 17:19:51 And something they've never done. Jul 22 17:19:55 it doesn't say that, and to say ship XXX per month implies out of factory Jul 22 17:20:17 but i doubt demand has reached 10K yet Jul 22 17:20:25 not even 5 K probably Jul 22 17:20:31 It doesn't say that. Jul 22 17:20:33 and that's terribly generous Jul 22 17:20:40 But FIC has never dealt with customers before. Jul 22 17:20:46 Only buisnesses. Jul 22 17:20:47 consumers. Jul 22 17:20:55 businesses are generally the customer Jul 22 17:21:05 who order 100k units to this or that warehouse Jul 22 17:21:09 Not end users - rather Jul 22 17:21:37 they need themselves a distributer Jul 22 17:21:43 maybe go on a field trip to dell... see how they managed to do it without imploding Jul 22 17:21:54 dell employ packing monkeys Jul 22 17:22:06 I know this as there was a big Dell factory beside my university Jul 22 17:22:43 ewon: how do they raise them? Jul 22 17:22:53 ewon: what race of monkeys do they use? Jul 22 17:23:00 loufoque: students, local unskilled labour Jul 22 17:23:03 so at some point in the future, when demand is likely to justify it, fic/openmoko.com needs to get packing monkeys too. Jul 22 17:23:18 I read the current situation as vastly overthinking the shipping thing. Jul 22 17:23:23 ckuethe: or ship them a palette at a time to a distributer Jul 22 17:23:32 outsourced packing monkeys. Jul 22 17:23:33 Speed did you see Harold's rant? Jul 22 17:23:40 ewon: or even a pallet :-) Jul 22 17:23:42 Trying to setup a fully featured cheap shipping solution at the start. Jul 22 17:23:43 I can't wait till openmoko runs on the iphone and Prada Jul 22 17:23:46 cjb_ie: indeed Jul 22 17:23:54 And then not testing it long enough beforehand. Jul 22 17:24:09 outsource the online shop, outsource the shipping, problem solved. Jul 22 17:24:16 ewon: ...unless you're talking a unique colour for each moko...? Jul 22 17:24:23 buying dev kits in general seems to suck Jul 22 17:24:27 cjb_ie: I hate you. Jul 22 17:25:12 the last few gps dev kits i bought were all shipped from the companies' headquarters, and packed by some manager / senior secretary Jul 22 17:25:34 ckuethe: FIC is chinese, they don't need to outsource anything Jul 22 17:25:37 ckuethe: microchip is the only company i've found who get dev kits right Jul 22 17:25:53 They need to outsource to someone with a clue on shipping. Jul 22 17:25:57 and mostly they do that by treating them as regular line items for sale by distributors Jul 22 17:26:03 if you're desperately looking for rationalization, be glad that you're helping to debug an opensource shipping process :P Jul 22 17:26:18 lol Jul 22 17:26:21 cjb: most companies seem to have gone that route nowadays. Jul 22 17:26:32 OpenShip? ;o) Jul 22 17:26:34 ...and making them surprisingly cheap (at a rough guess, maybe 50% of what you'd pay for a similar board without a 'dev kit' label on it) Jul 22 17:26:45 ckuethe: lol Jul 22 17:27:21 it's just a no-brainer, even for the stingiest of bosses, to buy the dev kit rather than having a developer (me) work in the dark with nothing to test code on etc Jul 22 17:27:39 just remember that they ship a package/packet to test, then they have to recompile their shipping procedure... Jul 22 17:27:40 erm. not that i wish to sound bitter or anything... Jul 22 17:27:47 console devkits are a different matter Jul 22 17:27:57 the one sitting on my desk costs roughly 5 times the cost of my car Jul 22 17:28:28 the GC devkit was reasonable Jul 22 17:28:32 ewon: what is it, just a console with a debug board in it? Jul 22 17:28:51 and maybe a PC hanging off the debug board. Jul 22 17:28:51 ewon what console? Jul 22 17:28:52 cjb_ie: yeah, larger hard disk, more NICs etc Jul 22 17:29:03 BryceLeo_Away: E_NOTALLOWED Jul 22 17:29:09 BryceLeo_Away: if he tells you he has to kill you ;-) Jul 22 17:29:24 lol Jul 22 17:29:25 uhhuh Jul 22 17:29:53 * ewon wonders why the espresso tastes worse on the weekends Jul 22 17:30:01 so does the mean you're one of those behind the "Phantom game console" Jul 22 17:30:06 hah, no Jul 22 17:30:19 psp? Jul 22 17:30:29 I work for a company called Demonware Jul 22 17:30:37 omg he's given away information! he's working on a real console that actually exists... maybe... nless he's doing the plausible deniability thing. Jul 22 17:30:49 hahahahaha Jul 22 17:31:01 i was going to ask him various consoles, and guess the one that he took the longest to deny Jul 22 17:31:11 heh Jul 22 17:31:19 ckuethe - *lol* Jul 22 17:31:30 timing attack. ;o) Jul 22 17:31:43 those are the best conversations... the one where the informant says "no comment", after a significant pause Jul 22 17:31:54 hehe Jul 22 17:32:02 is it this? Jul 22 17:32:02 no comment. Jul 22 17:32:08 Maybe you can figure it from the games: http://en.wikipedia.org/wiki/DemonWare Jul 22 17:32:10 you fail to realise that ewan has a sleep(random) before answering every question to defeat such attacks. Jul 22 17:32:14 lol oh come on I be he's working on the DreamCast2 Jul 22 17:32:14 is it that? Jul 22 17:32:14 no comment. Jul 22 17:32:34 is it $something_else? Jul 22 17:32:34 *awkward pause* Jul 22 17:32:34 no comment Jul 22 17:32:52 is it $something_else? Jul 22 17:32:58 ckuethe: possibly with a slightly more exasperated tone Jul 22 17:33:01 cjb_ie - I hope he uses physical random number generator. :) Jul 22 17:33:05 /mode +b CoreDump|afk Jul 22 17:33:10 he may have a random sleep, but his random is an LCPRNG Jul 22 17:33:18 /kick CoreDump|afk No! Jul 22 17:33:22 and this assumes he's actually trying to not disclose Jul 22 17:33:45 huh, I didn't realise we had a Wikipedia page :) Jul 22 17:34:08 if the DK is 5x the cost of a car i was figuring it would be a M$ product Jul 22 17:34:37 ckuethe: well, you'll note he didn't specify what the car cost. Jul 22 17:34:41 xb360, or the as-yet-unannounced new hotness Jul 22 17:35:12 ckuethe: depends on the car :) Jul 22 17:35:32 Nx cost of car, or Nx cost of house... great shorthand for the cost of microsoft products Jul 22 17:35:37 ewon: i'll give you a fiver for it. Jul 22 17:35:38 and don't get me started on veritas... Jul 22 17:35:51 cjb_ie: no. Jul 22 17:36:06 ewon: tenner? Jul 22 17:36:27 er, no :) Jul 22 17:36:36 ewon: three quid Jul 22 17:36:39 woo, bad-tasting espresso kicked in, back to work Jul 22 17:36:48 :( Jul 22 17:37:31 ya know after all this ewon is gonna say "i was really just yankin ya leg) Jul 22 17:37:57 sure, but this is more fun than mowing the lawn or cleaning the cat's litterbox Jul 22 17:38:05 lol Jul 22 17:38:16 ckuethe: ugh, why did you have to remind me Jul 22 17:38:34 * cjb_ie has three of the adorable little smelly bastards. Jul 22 17:38:42 only 1. Jul 22 17:39:03 I have a grand total of 0 to worry about. Jul 22 17:39:14 it only took a couple months but training him to realize that being the fuzzy hungry alarm doesn't work Jul 22 17:39:23 Same as the number of neos ordered :( Jul 22 17:39:44 ewanm89_laptop: i didn't order one either if it's any consolation Jul 22 17:39:55 i'll be gagging for the 02 though Jul 22 17:39:56 i've got a chinchilla and a bunny Jul 22 17:41:17 cjb_ie: But I want to order one, but I physically can't :( Jul 22 17:41:51 i would've liked to, but i just don't have the time to devote to care and feeding Jul 22 17:42:27 anyway. no doubt there'll be a glut of them on ebay soon enough when people who buy them realise what "not consumer ready" actually means Jul 22 17:42:46 IMO, a buyback scheme would make sense. Jul 22 17:42:56 Or maybe just ebay would work OK. Jul 22 17:43:03 cjb_ie: lol, but I can't order from ebay either. Jul 22 17:43:03 (script) wiki RecentChanges: [[Wish_List_-_Hardware_-_Atmospheric]] [[Getting_Started_with_your_Neo1973/fr]] [[OpenMoko_Local_Groups:_Bucharest]] [[Talk:Java]] [[Main_Page/fr]] [[Ambient_light_sensor]] [[Sound_Field]] [[OpenMoko_compatible_cellphone_providers/fr]] and other changes Jul 22 17:43:13 ewanm89_laptop: Why not? Jul 22 17:43:16 ewanm89_laptop: right, but you probably know someone who can. Jul 22 17:44:03 SpeedEvil: the ebay thing might work out nicely for actual developers - idiot consumer types get to buy the phone, pay an import charges etc, and then go "oh crap" and sell it Jul 22 17:44:07 parents wouldn't use there credit card on ebay. Jul 22 17:44:26 ewanm89_laptop: 95% of sellers will accept cash in an envelope. Jul 22 17:44:28 Good afternoon gentlemen Jul 22 17:46:50 Anyone know of any tracking number emails? Jul 22 17:46:50 calamou1: See http://wiki.openmoko.org/wiki/FAQ (and also SH1_FAQ, Neo1973, Developer_preview and Neo1973_Hardware#GTA01Bv4 pages) (Hopefully this link answers your question. This autotrigger for new nick and question during first 15 minutes done only once.) Jul 22 18:05:40 It's still compiling (and using an insane amount of disk space), but I have a list of what I did to compile openmoko for EABI at [[User:CesarB/Compiling for EABI]] Jul 22 18:06:44 hm? Jul 22 18:07:04 using angstroem and be fine Jul 22 18:07:34 or do you mean the mokoapss themself Jul 22 18:08:28 Could someone with a unmodified openmoko build setup tell me how many directories it creates on build/tmp/work/arm*linux? Jul 22 18:08:38 * cesarb wants to have a rough idea how many packages are left Jul 22 18:09:15 woglinde: I'm compiling everything under eabi, except for the uboot (which doesn't seem to accept eabi) Jul 22 18:09:38 woglinde: that page is if someone else wants to also do it Jul 22 18:09:50 hm yes but that is in openembbed for serval months Jul 22 18:10:17 woglinde: I'm using the old openembedded openmoko uses for now Jul 22 18:10:25 woglinde: it's just a test to see how it goes ;-) Jul 22 18:11:02 Dude. I finally got glibc-intermediate to build. Jul 22 18:11:35 Aria: were you the one that posted the bugzilla patch? Jul 22 18:12:05 Nope. Jul 22 18:12:13 ahh Jul 22 18:12:20 Still learning the ropes of the OE build system too much to know how to properly patch things yet. Jul 22 18:12:25 I'll get there though. Jul 22 18:14:11 yeah that whole quilt thing is really odd Jul 22 18:14:22 quilt+mtn+svn, oy. Jul 22 18:14:44 I'm used to my distro's build system, which keeps build system and package definitions very unentangled. Jul 22 18:15:41 Aria: whats your distro? Jul 22 18:15:49 PLD Jul 22 18:15:56 RPM-based. Jul 22 18:17:43 intersting Jul 22 18:18:15 The only entanglement the two have is that rpmbuild provides a set of macros, which the distro HAS customized Jul 22 18:18:48 But the entire emphasis is on taking the author's upstream source, annotating with patches, and building. Keeping things so separate makes it really pleasant to push things upstream Jul 22 18:19:16 I fucking hate rpmbuild. Jul 22 18:19:25 whose rpmbuild? Jul 22 18:19:32 rpm.org or redhat? Jul 22 18:21:28 Aria: that's not too bad a setup Jul 22 18:22:15 No, it works well. Jul 22 18:22:29 Keeping cleverness to a minimum, not metaprogramming makes things really unentangled. Jul 22 18:22:41 A bit of cut-and-paste coding. But that's really not a bad thing. Jul 22 18:23:08 yeah but most cut and paste can just be done with a couple patches Jul 22 18:23:33 hey sagacis! Jul 22 18:24:41 Yes, a full openmoko build. Jul 22 18:25:58 Yeah. We've run massive sed scripts on our rpm spec files a bundle of times. Jul 22 18:26:38 And they work. Now and then, we find somethng not up to date -- that's what build tests are for! -- and it gets fixed. But get a PLD spec, and aside from a few macros that don't translate across distros, you get a damn good overview of how to go from the author's source to working. Jul 22 18:34:22 are the credit card charges happening in any particular order? I'm #1970 and haven't seen anything... Jul 22 18:34:26 OK, this is so weird: I'm looking at a phone case, and it says: "Includes hibernization magnet to help maximize battery life." Jul 22 18:34:35 what the crap is that? Jul 22 18:34:57 do some phones have some kind of magnetic leaf switch that makes them go into powersave mode? Jul 22 18:35:11 certainly seen it in laptops Jul 22 18:35:21 magnet in lid, reed switch in body of laptop Jul 22 18:35:32 rather than having a prong that's pushed down when the lid closes Jul 22 18:37:03 I had a slider phone with something of that nature in it Jul 22 18:38:41 It's not an embedded device. Jul 22 18:38:42 It's a desktop. Jul 22 18:38:42 It's a pentium 100 desktop that nobodies told it that it's a phone. Jul 22 18:38:46 hehe, I like this one. :-) Jul 22 18:40:02 HEhe Jul 22 18:45:25 orospakr, many (all?) modern Blackberry devices use this technique as well; the holster they ship includes the magnet. Can lock-on-holster too if paranoid. Jul 22 18:48:22 idarwin: that explains it. The reason why I'm asking is because my carrier is having a little "appreciation" promotion, and they're giving away cases. but there's only two options, with a list of compatible dimensions. Jul 22 18:48:28 thus far, I've just been comparing dimensons. Jul 22 18:48:34 s/dimensons/dimensions/ Jul 22 18:48:34 orospakr meant: thus far, I've just been comparing dimensions. Jul 22 18:49:01 naturally, I want to use it with my Neo. :) Jul 22 18:49:14 pockets are a cruel environment to technology :P Jul 22 18:49:51 http://tinyurl.com/2yarvj Jul 22 18:58:24 i'm not sure how they are figuring out when to charge cc Jul 22 18:58:33 i was charged yesterday Jul 22 18:59:09 but a friend of mine (who ordered before me) has not yet been charged Jul 22 18:59:35 jaebird: both are the same kind of card (visa/mastercard)? Jul 22 18:59:43 there was only 1hr difference to the YES_I_DO email Jul 22 18:59:50 jaebird: because they might be looking at only one of them for now Jul 22 19:00:05 maybe, mine was visa Jul 22 19:00:24 anybody use debit card? Jul 22 19:01:05 mine was normal visa Jul 22 19:01:34 it just seems strange that there wouldn't be a logical progression Jul 22 19:01:51 what if the run out of devices before the earlier orders are completed Jul 22 19:01:55 that doesn't seem fair Jul 22 19:03:15 do you and your friend live in the same area? Jul 22 19:03:27 yep...down the road Jul 22 19:03:48 sagacis..on this chan actually Jul 22 19:04:01 also on the P1 wiki page Jul 22 19:04:15 it just seems a little bazaar to me Jul 22 19:04:45 maybe they took all the orders and threw them down the stairs... Jul 22 19:04:57 then charged based on the step they landed on :) Jul 22 19:05:56 Hallelujah! Jul 22 19:06:14 jaebird: maybe your friend's bank doesn't do transactions during weekends? Jul 22 19:06:17 jaebird: what was your # ? Jul 22 19:06:30 so he'll be charged tomorrow Jul 22 19:06:47 1932 Jul 22 19:06:51 mine is 1952 and not charged yet Jul 22 19:07:40 the P1 page has #2233 already charged Jul 22 19:07:50 yep Jul 22 19:09:54 that is why i'm wondering how they are processing. Jul 22 19:16:59 anyone got their phone yet? Jul 22 19:17:51 edistar: joking? Jul 22 19:22:57 how about an application to configure your wifi device depending on where you are. (GPS) Hotspot @ home or Hotspot @ work ..... Jul 22 19:24:50 jaebird: sry ;) Jul 22 19:25:27 edistar: actually P0 devels have their phones :) Jul 22 19:25:58 this is kinda like pushing rope Jul 22 19:30:20 grobian_: tis on the wiki Jul 22 19:38:58 ok Jul 22 19:42:43 (script) wiki RecentChanges: [[External_Feeds]] [[Wish_List_-_Hardware_-_Atmospheric]] [[Talk:Java]] [[Java]] [[OpenMoko_Local_Groups:_Bucharest]] [[Wishlist:BuiltInScriptingLanguage]] [[User:CesarB/Compiling_for_EABI]] [[User_talk:WikiSysop]] [[User:Johan]] [[Help_talk:Contents]] and other changes Jul 22 19:59:19 lol...! Jul 22 19:59:32 so my bank turned off my credit card when FIC charged me Jul 22 19:59:45 they rejected the payment? Jul 22 19:59:54 way to go FIC... you are on some horrible black list Jul 22 20:00:09 no, my bank allowed it to pre-auth... but locked my account after the pre-auth Jul 22 20:00:26 (aka ... they would have not really paid the request for funds) Jul 22 20:00:41 i tired to buy a latte today and my card was rejected! Jul 22 20:00:45 wtg, FIC ;-) Jul 22 20:00:46 lol Jul 22 20:01:21 luckly i had cash, and all i had to do was call fraud protection to authorize the charge and my card is active again Jul 22 20:01:28 you tried to buy a latte on your credit card? Jul 22 20:01:30 Was that Citibank, but chance? Jul 22 20:01:37 mwester, no Jul 22 20:01:40 ScaredyCat, sure Jul 22 20:01:53 ScaredyCat, it's a nice way to track expenses Jul 22 20:02:06 Citibank left my wife and two daughters stranded in Spain, because they used the credit card to purchase fuel before returning the rental car. Jul 22 20:02:18 fair enough... but don't they usually have a minimum charge Jul 22 20:02:28 ScaredyCat, some places, yes Jul 22 20:02:40 MBNA left me stranded in cuba, because they promised me my card would work there. Jul 22 20:02:46 ScaredyCat, i'm such a regular.. it's no big deal Jul 22 20:02:52 mwester: I presume you don't bank with them now Jul 22 20:03:04 you sure have lousy banks Jul 22 20:03:09 Nope. Switched to Discover. Jul 22 20:03:11 cuba for two weeks on $100 would have been interesting, but i managed to get some money wired to me through the embassy. Jul 22 20:03:34 (the cheapest, nastiest, illegal accom starts at around $15/night) Jul 22 20:03:35 guaqua, it's more there is *sooooo much* fraud... there are a *lot* of automatic systems that just turn off the cards Jul 22 20:03:38 guaqua: yes - I agree! They miss the whole POINT of having a credit card -- it's so that I don't have to carry local CASH ! Jul 22 20:04:01 in some countries the supermarkets don't take credit cards, only debit cards Jul 22 20:04:11 which is a pita Jul 22 20:04:45 .nl they don't take cc in supermarkets Jul 22 20:04:52 i was just really surprised, with FIC being such a large company... that a bank would think they are fraudulent creditors Jul 22 20:04:56 ScaredyCat: I've since found that disabling credit cards at the worst possible time is a very common experience with Citibank -- apparently they are very aggresive about it, and not at all apologetic about leaving you without an operable credit card. Jul 22 20:05:03 ScaredyCat: yeah, that caught me out Jul 22 20:05:11 which was interesting the first time I loaded up 2 trollys of shopping Jul 22 20:05:41 mwester: crappy bank.. ( like them all I guess) Jul 22 20:06:52 well, the issue is once the bank authorizes a debit or credit, they are responsible for paying it; a lot of banks offer zero end consumer liability .... so to control costs, they have all these automated systems to ensure profits stay high Jul 22 20:07:27 works in a different way in finland Jul 22 20:07:47 I know visa etc are starting with this 3d-auth shite soon... Jul 22 20:07:59 most consumers are *stupid* (for lack of a better word) to not hold their credit information with private confidence because they are at ease with zero-fraud liability Jul 22 20:08:04 daMaestro: FIC might be a large company, but they are using some strange scheme with a taiwan bank Jul 22 20:08:05 there's one central authority which provides the connections and takes care of issues that raise with fraud Jul 22 20:08:17 cesarb, yes.. very true Jul 22 20:08:24 daMaestro: which is why the value is being charged in a uncommon currency Jul 22 20:08:25 cesarb, that is why my card was deactivated Jul 22 20:08:40 ouch Jul 22 20:08:45 they should really just have let a distributor handle the whole thing tbh Jul 22 20:09:42 * ScaredyCat wont point out that they are still charging before shipping... Jul 22 20:09:44 farnell, digikey or similar Jul 22 20:09:58 farnell's pretty much worldwide isn't it? Jul 22 20:10:08 yeah... the damn email said i was going to save on shipping.. but it was still $20 Jul 22 20:10:17 so i saved *nothing* Jul 22 20:10:28 ScaredyCat, i hope so Jul 22 20:10:35 ? Jul 22 20:10:45 ScaredyCat, it should be PIN + signature everywhere Jul 22 20:10:58 that' no protection at all Jul 22 20:11:19 for a start most stores in the US don;t check signatures anyway.. Jul 22 20:11:20 ScaredyCat, or we should have a separate smart card or auth system via our phone.. i don't know Jul 22 20:11:36 well you can't... i just scribble... and it's my card! Jul 22 20:11:37 lol Jul 22 20:11:38 and there have been a number of issues with 'hacked' pin machnes Jul 22 20:11:39 ScaredyCat: like the good old zug story Jul 22 20:11:49 (re checking signatures) Jul 22 20:11:53 yeah cjb_ie Jul 22 20:11:58 the whale one ;) Jul 22 20:11:59 a list of passwords for authenticating with the bank is probably the best way Jul 22 20:12:16 ScaredyCat: oh... i don't remember a whale Jul 22 20:12:18 sadly... no US consumer will *ever* remember a list Jul 22 20:12:22 sms is enough Jul 22 20:12:30 they can barely deal with a 4 digit pin Jul 22 20:12:34 ynezz, now that is scary Jul 22 20:12:49 here in ireland, i get a 'code card' to use with my online banking - essentially it's a one time pad Jul 22 20:12:52 telco + banking = scary conglomerate Jul 22 20:13:04 but secure Jul 22 20:13:10 basically, they're too cheap to use time-based/random keys Jul 22 20:13:11 not really. Jul 22 20:13:37 http://www.zug.com/daily/journal/graphics/011705_cc_aqua.gif Jul 22 20:13:39 if they can hijack your phone number, they can do whatever else Jul 22 20:13:46 cjb_ie: Jul 22 20:14:14 lol Jul 22 20:14:59 :) Jul 22 20:15:08 ScaredyCat: hehehe Jul 22 20:15:55 he signed Shamu, at new england Aquarium .. it's just genius ;) Jul 22 20:17:29 who is Shamu? Jul 22 20:17:36 just to get the point... Jul 22 20:17:38 a whale at sea world Jul 22 20:24:12 Hallelujah! Jul 22 20:25:33 A list of onetime keys agreed by you and the bank? Jul 22 20:26:45 In fact, the bank can't even see them, you sit down they are generated than encrypted on a singly terminal, all temperary files lost. Jul 22 20:29:21 right, and then the twat user puts the list in their lwallet Jul 22 20:29:53 ScaredyCat: Then it's his stupid fault, like writting down your pin. Jul 22 20:30:10 the use is always the weakest point... Jul 22 20:30:15 people are dumb Jul 22 20:30:27 ScaredyCat: you can also have another password Jul 22 20:30:54 so someone stealing the pad couldn't access the service by only using the pad Jul 22 20:31:04 they'd also need the special password Jul 22 20:31:05 which they write on the list paper? :) Jul 22 20:31:07 And how can you get over the CKI problem? Jul 22 20:31:15 well that's just darwinism Jul 22 20:31:41 you can't fight natural stupidity, it fights itself Jul 22 20:32:25 yes, but a cc company *will* refund if the user bitches ... which means we have to pay for those losses. Jul 22 20:33:21 ScaredyCat: Not here in the UK they don't. Jul 22 20:33:35 they do ewanm89_laptop... Jul 22 20:33:50 as online retailers it's our biggest problem Jul 22 20:34:52 Not if it's the users stupid fault, in fact (since chip & pin) every possible fraud is blamed on the user, there fault or not. Jul 22 20:35:14 chip & pin is *not* safe Jul 22 20:35:28 Done properly, it could be. Jul 22 20:35:39 The way it's typically operated - nope. Jul 22 20:35:50 ScaredyCat: I know that, the banks know that, the courts know that, it still doesn't work. Jul 22 20:35:52 and no, lots of places do the old 'Cardholder not present' which doesn;t require a pin Jul 22 20:36:44 it protects the banks. that's about it. Jul 22 20:37:03 Any info on actual shipping? It's nice that they take ppl's money, but what about actually packaging and sending stuff? Jul 22 20:37:07 Yeah, when they blame everything on the user. Jul 22 20:37:10 it gives them plausible deniability "customer must have disclosed the pin" Jul 22 20:39:05 Also, I know bank managers who say opensource software is not safe, and can't give a reason, other than someone will be able to find the flaws (the can with propietary software). Jul 22 20:39:18 s/(the/(they/ Jul 22 20:39:18 ewanm89_laptop meant: Also, I know bank managers who say opensource software is not safe, and can't give a reason, other than someone will be able to find the flaws (they can with propietary software). Jul 22 20:39:26 security through obscurity, yay Jul 22 20:40:06 I know one bank in the UK where it's whole online banking is done on IIS. Jul 22 20:40:17 That's secure. ;) Jul 22 20:40:31 The most exploited webserver in the world. Jul 22 20:40:34 LyosNorezel: Ok - try altering the input number Jul 22 20:40:54 there's nothing inherently wrong with that, so long as they keep it patched etc, and don't do anything obviously stupid like installing FPE Jul 22 20:41:19 Now the reason I haven't got a student account yet is the fact that this is the bank with a branch on campus. Jul 22 20:41:38 cjb_ie: Would you trust it? Jul 22 20:42:18 i would trust the bank. they have a strong business incenhtive to cover up for any technological infelicities Jul 22 20:42:20 *shrug* it's up to them to provide a secure ervice... Jul 22 20:42:25 cjb_ie: And keep the system airwalled. Jul 22 20:42:26 I don't trust anything where it can take upto a month to release a patch for a major disclosed bug. Jul 22 20:42:37 Airwall++ Jul 22 20:42:44 SpeedEvil: airwall? new one on me. Jul 22 20:42:56 Like a firewall, but with an airgap. Jul 22 20:42:56 galexande: i wouldn't trust them to do more then fudge it and bill us for their shortcomings Jul 22 20:43:03 And no network connection. Jul 22 20:43:07 ewan, yeah but you aren't trusting the bank to not be compromised, you are trusting the bank to keep track of your money even if they are. Jul 22 20:43:45 i mean, i would boycott a bank because they have a tendency to bill customers for shortcomings (in fact, have already done so), but that's an orthogonal issue to using iis Jul 22 20:44:15 galexande: They wouldn't manage that, unfortunatly none of the possible banks use apache. Jul 22 20:44:23 On a UNIX server. Jul 22 20:44:42 jpmorgan don't do student accounts :( Jul 22 20:44:51 jp morgan is awful Jul 22 20:45:00 you can pretty much tell the qualithy of a bankby the number of customers. Jul 22 20:45:30 if it issignificantly more than 100,000, they are probably bad a daily banking business and are evil on the side. even if they use apache :) Jul 22 20:45:33 jpmorgan target a particular customer. Jul 22 20:46:15 night guys. Jul 22 20:46:32 night ew Jul 22 20:46:58 cjb_ie: Don't you know where tab is? Jul 22 20:49:06 ewanm89: yes, but sometimes i'm lazy Jul 22 20:50:41 and mokobot may not be around for some of tommorrow Jul 22 20:50:58 bye Jul 22 20:53:10 Hm. A lot of the icons are white documents with a red X inside them, when using MokoMakefile & Xephyr. Did I miss a step in the howto? Jul 22 20:53:32 Also, the document is a bit .. contradictory. Jul 22 20:53:40 One part of it lists the command "gdk-pixbuf-query-loaders > /etc/gtk-2.0/gdk-pixbuf.loaders Jul 22 20:53:54 " as one that should be run inside chroot, but the "automated script" does it outside the chroot. Jul 22 21:03:41 so, has anyone gotten a shipping tracking number yet? Jul 22 21:18:54 UPS doesnt pick up shipments on weekends, therefore they cannot ship and they cannot issue tracking numbers Jul 22 21:21:01 Sublime, ok.. so we can assume the shipment is being prepared to go out tomorrow? Jul 22 21:22:28 lol, ive learned that you cannot assume anything with FIC Jul 22 21:22:42 im assuming, as of now, im never getting my phone Jul 22 21:22:55 im 1843 and my card hasnt been charged Jul 22 21:23:23 apparently they're charging them in random order. Jul 22 21:23:44 so it seems you can't assume anything Jul 22 21:23:48 Sublime: I'm in the same boat. I'm trying patience and doing non-phone stuff Jul 22 21:24:32 yeah, i'd like to be patient and supportive of fic but its getting to be really hard Jul 22 21:24:45 they need to communicate to us what they are doing Jul 22 21:24:47 Have faith! ;) Jul 22 21:24:53 i feel for you guys... Jul 22 21:25:06 although, as far as i know, none have shipped yet! Jul 22 21:25:19 so we may all be in the same boat Jul 22 21:25:20 I'm happy, since they took my money. Jul 22 21:25:28 Sublime: lots of people surrounding you have been charged though Jul 22 21:25:36 Anyone here running MokoMakefile w/ Xephyr? Jul 22 21:25:57 Sublime: could be all sorts of reasons, I wouldn't worry at this point Jul 22 21:25:59 one dude from mesa was charged, thats about it as far as i can tell Jul 22 21:26:09 (easy for me to say with 43xxish Jul 22 21:26:20 Sublime: http://richtlijn.be/~larstiq/neo/owners.txt Jul 22 21:26:21 im not worrying, im just a little ticked off Jul 22 21:26:25 I'm getting a lot of ** (openmoko-dialer:7819): WARNING **: Cannot load stock icon from theme : moko-call-answer Jul 22 21:27:11 Sublime: so 35, 39, 40, 49 are confirmed in your area Jul 22 21:27:44 oh i thought you meant in my area as in, southwest US Jul 22 21:28:09 Sublime: ah no, rt ticket area :) Jul 22 21:28:22 hmm, I might extend the script to group based on area Jul 22 21:28:36 but I'm afraid that field is even more freeform Jul 22 21:38:27 LarstiQ, whats the mean, the description is kinda cryptic Jul 22 21:41:05 the=that Jul 22 21:41:10 owners.txt i mean Jul 22 21:42:40 I'm #1844 and not charged, either... Jul 22 21:42:43 (script) wiki RecentChanges: [[OpenMoko_Local_Groups:_San_Diego]] [[P1_Owners]] [[Talk:Wishlist:BuiltInScriptingLanguage]] [[Press_Coverage/fr]] [[Variometer]] [[Wish_List_-_Hardware_-_Atmospheric]] [[Main_Page/fr]] [[Widget_Inheritance_Graph/fr]] [[External_Feeds]] and other changes Jul 22 21:43:07 whats up 1844, im 1843 Jul 22 21:43:14 heh Jul 22 21:43:49 anyone know what the fields mean in --> 2019 30 1x ENODATE Jul 22 21:44:05 obviously 1st is ticket and 1x is qty Jul 22 21:44:28 look on the bright side, you guys are all getting to bond with your order number neighbours :-) Jul 22 21:44:30 entry on wiki P1_OWners page Jul 22 21:44:41 ah Jul 22 21:45:41 how about the ENODATE Jul 22 21:46:19 id rather prefer bonding with my phone than some dude i dont know that live 234234 miles awy Jul 22 21:46:23 no offense Jul 22 21:46:44 heh Jul 22 21:47:18 ah Jul 22 21:47:29 i get it Jul 22 21:47:42 * Psi_ edits p1_owners Jul 22 21:48:32 Sublime: none taken :) Jul 22 21:58:09 Er. Hi. Any activity here? Jul 22 22:00:18 Apparently not. Well, throwing out a question: Anything known about presence or absence of a JVM? Jul 22 22:01:15 odin neo will have midp support faster than sms sending Jul 22 22:01:35 woglinde: is that a challenge? Jul 22 22:01:38 I am now just preparing the last .bb for openembedded Jul 22 22:01:53 zecke a little one, but I was lazy the whole day Jul 22 22:02:19 midp? Jul 22 22:02:24 yes Jul 22 22:02:31 cacao-cldc with midpath Jul 22 22:03:17 * Odin- can't say he's familiar with Java lingo; explain please? :/ Jul 22 22:03:23 What is MIDP? Jul 22 22:03:42 MIDP is a java standard Jul 22 22:03:51 for mobile devices, mostly Jul 22 22:04:01 It defines a set of features that a JVM must support to be qualified for "MIDP", and it's aimed at embedded/mobile devices Jul 22 22:04:07 There's e.g. MIDP 2.0 Jul 22 22:05:02 I see. Jul 22 22:08:27 Hm. What do I need to do to run the qemu image? It fails with "No init found." (I used MokoMakefile & make qemu) Jul 22 22:09:16 daxxar: did you go through http://wiki.openmoko.org/wiki/MokoMakefile#QEMU? Jul 22 22:10:31 Yes, I installed the packages as it wanted, and it found my gcc-3.4 Jul 22 22:10:56 It starts the bootloader and starts the kernel image, but seems that the filesystem is missing the init binary (?) Jul 22 22:10:56 daxxar: did you "make openmoko-devel-image" ? Jul 22 22:11:36 Doesn't it download a prebuilt image when you run make qemu? Jul 22 22:11:45 I haven't run openmoko-devel-image after I switched away from x86 Jul 22 22:13:16 daxxar: did you try to do it "by hand"/not using Mokomakefile? (http://wiki.openmoko.org/wiki/OpenMoko_under_QEMU) Jul 22 22:13:49 Nope Jul 22 22:13:51 I can try that instead Jul 22 22:17:11 daxxar: do you have lynx and wget installed? Jul 22 22:17:23 Yep Jul 22 22:18:01 Doing the "manual" method now Jul 22 22:18:05 Psi_: just a filler for 'no date supplied' Jul 22 22:19:41 daxxar: you can get the images directly from http://buildhost.openmoko.org/tmp/deploy/images/ if download.sh somehow does not succeed. Jul 22 22:20:07 It seems to be working fine Jul 22 22:20:20 Fetching rootfs atm Jul 22 22:25:48 Seems to work OK now Jul 22 22:35:07 it seems, that group orders don't have any CC charge Jul 22 22:36:32 why? Jul 22 22:36:51 who knows :) Jul 22 22:49:20 hey guys Jul 22 22:49:26 and gals) Jul 22 22:49:43 he alphaone Jul 22 22:51:11 hi woglinde Jul 22 22:52:41 alphaone - Hi! Jul 22 22:53:52 alphaone - I think, mickey wanted to be contacted by you, or somesuch. Jul 22 22:53:59 Mailing lists should be closed during conferences. Jul 22 22:54:18 Elrond: Thanks, I gave him a call already. Jul 22 22:54:49 alphaone - Ahh. Jul 22 22:55:22 alphaone - Do you know any openmoko people near FFM except mickey, abraxa_ and me? Jul 22 22:55:49 Elrond: No, sorry. Jul 22 22:56:13 Berlin is not near FFM, is it? :-) Jul 22 22:56:47 no it isnt Jul 22 22:56:54 Not strictly. ;-) But there would be an ice ;o) Jul 22 22:57:18 woglinde - From where are you? Jul 22 22:57:25 berlin Jul 22 22:57:30 Ahh Jul 22 23:04:07 Anyway, good night people. :-) Jul 22 23:10:38 good night Jul 22 23:25:36 godd nite Jul 22 23:30:17 Odin-: in short, MIDP is typically what comes to mind when you talk about mobile phone games Jul 22 23:30:46 granted, it'd be a little weird to be able to play games but not to SMS :) Jul 22 23:31:06 oh, and without a d-pad, most games will be really _interesting_ to play Jul 22 23:36:29 hi everyone Jul 22 23:43:03 * cesarb is getting the vibe that the thing that will most annoy people on the neo is the lack of working SMS Jul 22 23:43:13 perhaps even more than the lack of a working dialer would... Jul 22 23:43:43 (script) wiki RecentChanges: [[Press_Coverage]] [[Talk:MokoMakefile]] [[User:Michaelshiloh]] [[MokoMakefile]] [[User:Markb]] [[P1_Owners]] Jul 22 23:47:18 Why is a simple SMS -> line of text in a flat file, and echo sms |smssender hard? Jul 22 23:47:25 Or difficult. Jul 22 23:47:39 terminal and on-screen keyboard? Jul 22 23:47:47 :) Jul 22 23:47:48 SpeedEvil: the bizarre GSM AT comandset, AFAIK Jul 22 23:48:05 Oh - and I suppose you've gotta integrate it into gsmd Jul 22 23:48:08 SpeedEvil: which seems to be what is tripping people Jul 22 23:48:13 voice communications is so last century...text is the new comm tool :) Jul 22 23:48:28 SpeedEvil: (in particular, it seems to be doing something unexpected IIRC) Jul 22 23:48:31 or a speech to text processor for SMS :) Jul 22 23:48:34 openmoko ! ftw ! openmoko awesome ! openmoko ! Jul 22 23:48:36 http://smstools3.kekekasvi.com/ Jul 22 23:48:40 sms sending tool Jul 22 23:49:05 jaebird: both have their strong points Jul 22 23:49:26 jaebird: SMS is more silent and sutble, while voice makes for faster communication and frees your hands Jul 22 23:50:27 jaebird: and voice can be cheaper than SMS... at R$0,30 each SMS, people who can do a 15-second phone call and talk about everything that's needed (we often do that) spend less Jul 22 23:50:46 so we need a voice to sms gateway Jul 22 23:51:09 * cesarb wants SMS for one reason: the "popup" SMS that comes with *544#, which tells you how many credits you have left (also *545# and *546#) Jul 22 23:52:57 i've been reading bugzilla! Jul 22 23:53:41 Many people can't do short calls at all. Jul 22 23:54:03 that is why i ignore, and let them go to voicemail Jul 22 23:54:03 And putting the phone down on them is rude. Jul 22 23:54:06 Hence SMS Jul 22 23:54:57 it's not rude... 'sorry dude, reception cut out and then my battery died' Jul 22 23:55:18 my policy is that you can call me anytime you want, if i'm busy i won't answer Jul 22 23:56:00 some people are addicted to answering when the phone rings Jul 22 23:56:14 and then say "uh...i'm busy right now" :) Jul 23 00:01:55 jaebird: speaking of answering when it rings Jul 23 00:02:03 still no ship notice... Jul 23 00:02:04 jaebird: one of my pet annoyances with the Razr Jul 23 00:02:13 i want my moko! gimme! Jul 23 00:02:17 * juri_ starts mugging people. Jul 23 00:02:17 jaebird: if you put it on vibrate then ring, it vibrates twice and then rings... Jul 23 00:02:28 jaebird: EXACTLY at the point when you are ALMOST opening the screen Jul 23 00:02:41 jaebird: couldn't it vibrate just once more? Jul 23 00:02:59 now that the first phones are in the US, we may need to switch timezones. working time monday isn't for another 15 hours or so Jul 23 00:05:52 cesarb: Set a ringtone that has a few seconds of silence? ;-) Jul 23 00:05:59 cesarb: (prepended to it) Jul 23 00:06:10 cesarb: i don't know...i have an E2 ROKR and i just hit the red button and it ignores instantly and goes to voicemail Jul 23 00:06:44 daxxar: good idea, I'd try it if I weren't already getting a new phone :D Jul 23 00:07:03 moto's flip phones are notorious for not allowing an ignore...the side button just silences Jul 23 00:07:51 cesarb: :-) Jul 23 00:09:02 I wonder how long it'll take before SMS & power-management is in a usable state. (those are, tbh, my only requirements for dumping this crap WM5.0 mobile) Jul 23 00:09:30 But the SMS-part is more than just a frontend missing? Jul 23 00:09:46 (problem with gsmd or whatever handles it?) Jul 23 00:09:56 daxxar: the SMS part will probably have an army of geeks over it soon, so it'll end up working quite fast (I hope) Jul 23 00:10:21 daxxar: the problem is mostly with gsmd AFAIK, which IIRC gets confused by the modem's replies or something like that Jul 23 00:10:46 daxxar: but the power management part is harder, because it's not just coding (and it could even need kernel-level coding, which can be scary) Jul 23 00:10:56 * cesarb wonders if Intel's powertop works on ARM Jul 23 00:10:56 i don't think people should underestimate the amount of work left before openmoko is useable Jul 23 00:11:22 i'm reading bugzilla (unless it is out of date!) there are lots of outstanding bugs that are really important :) Jul 23 00:11:57 Hm, okay. Jul 23 00:12:01 Well, good night. :-) Jul 23 00:38:17 counter Jul 23 00:38:17 (last update 2007-07-21 06:20) http://londerings.novalis.org/wlog/index.php?title=Neo1973_counter : Order https://direct.openmoko.com/ ; Shipping prob. 75% in 2 days 23:21:37 (2.973±2.5 days) (1769;249) Jul 23 00:59:51 jaebird: It has been suggested - not-completely-in-jest that 'October' may in fact be a codename, not the actual month :/ Jul 23 01:27:08 SpeedEvil: heh Jul 23 01:30:13 SpeedEvil: hah Jul 23 01:30:27 yeah, we're going to be lucky if we see GTA02 before the end of year holiday Jul 23 01:30:41 SpeedEvil: much like YES_I_DO has an specific meaning around here Jul 23 01:31:37 daMaestro: I do not think it's impossible, we don't know how far ahead they are with it (they are quite tight-lipped (sp?) about that... don't they always report after the fact) Jul 23 01:31:46 i'll be lucky to see gta01 before hanukah Jul 23 01:31:47 DO_I?_YES Jul 23 01:31:50 daMaestro: however, I don't know if the software will be ready by then... Jul 23 01:43:03 (script) wiki RecentChanges: [[Current_events]] Jul 23 01:46:03 i wouldn't mind just having gta02 hardware by october. i imagine for a device like this, the software will never be "ready" right? Jul 23 01:46:14 it'll always be improved and modified. Jul 23 01:52:43 xL0giC: we can have a definition of "ready" which means it will sometime Jul 23 01:53:03 xL0giC: just make a list of needed features (dialing, sms, etc), and as soon as they work correctly, it's ready Jul 23 01:53:29 xL0giC: ready as in "a non-programmer can pick it up and use it fine" Jul 23 01:53:39 i see.. Jul 23 01:53:54 and what about releasing the hardware with the unfinished software? is that a possibility? Jul 23 01:54:23 well it's happening for GTA01 ... Jul 23 01:54:46 xL0giC: bad idea, due to the negative publicity it would generate (unlike the "hackers-only you know what you doing right? say YES_I_DO" GTA01 release) Jul 23 01:55:41 although there are undoubtably people ordering now who just want their hands on a nice shiny phone, even though they're not hackers Jul 23 01:57:32 so what exactly does openmoko need to get gta02 w/software released in october? a miracle? more developers? Jul 23 01:57:44 More active developers. Jul 23 01:57:44 both? :) Jul 23 01:57:53 Maybe even funding of some of them. Jul 23 01:58:25 of course all this assumes the october date won't slip, which is a pretty poor assumption given history ... Jul 23 01:58:59 rwhitby: hm, good to see you. I have one more bashism patch, and a general "better error checking" patch for mokomakefile Jul 23 01:59:18 cesarb: pastebin please Jul 23 02:00:45 do patches need to get discussed on the mailing list, or is leaving them on bugzilla generally enough? Jul 23 02:01:21 rwhitby: http://www.pastebin.ca/630284 is the bashism one Jul 23 02:03:03 rwhitby: http://www.pastebin.ca/630288 is the mokomakefile one, which changes a lot of ; to && so that for instance a missing directory will not run things like "quilt pop" in the wrong directory Jul 23 02:03:27 rwhitby: the bashism one goes into patches/openembedded- as usual Jul 23 02:05:05 rwhitby: I made these while playing with EABI, see [[User:CesarB/Compiling for EABI]] if you are interested in that Jul 23 02:05:30 mooo. Jul 23 02:05:45 rwhitby: thanks for makemokofile - you rock. Jul 23 02:06:06 openmoko: 03rwhitby * r101 10mokomakefile/trunk/patches/openembedded-e2dbb52fe39df7ef786b6068f6178f29508dfded/ (fix-bashism-kernel-headers.patch series): Fix bashism in kernel headers Jul 23 02:06:11 or mokomakefile, really. Jul 23 02:06:24 openmoko: 03rwhitby * r102 10mokomakefile/trunk/Makefile: Robustness fixes - thx to cesarb Jul 23 02:06:30 cesarb: thx, done. Jul 23 02:06:36 ferric: thx Jul 23 02:08:09 sure Jul 23 02:08:22 * ferric notes that #24xx seems to have received a shipping email Jul 23 02:14:08 jesus christ Jul 23 02:14:18 ok this has nothing to do with moko or fic or anything Jul 23 02:14:28 you guys are the closest vent i have right now Jul 23 02:14:53 but i jsut spent the last hour doing the wrong god damn diff eq problem for homework Jul 23 02:14:55 i hate life Jul 23 02:15:07 i need a smoke Jul 23 02:15:15 Was it a similar dif eq to the one you were supposed to do? Jul 23 02:15:36 about the same Jul 23 02:15:53 Could always try submitting it - noting your error. Jul 23 02:15:55 sublime: you at UofA? Jul 23 02:16:03 yeah Jul 23 02:16:42 my prof already hates me Jul 23 02:16:51 hes looking for every excuse possible to fail me Jul 23 02:17:03 afk Jul 23 02:17:29 Anyone know if works has been done on power consumption for the open moko project? Jul 23 02:17:43 would approaches like intel's powertop make sense in this context? Jul 23 02:17:45 No. Jul 23 02:17:49 Sort-of. Jul 23 02:18:00 But the hardware drivers are not there. Jul 23 02:18:13 And you need hardware to work on those. Jul 23 02:18:13 SpeedEvil: powertop would be useful... to find excess bogus wakeups at least Jul 23 02:18:22 True. Jul 23 02:18:44 Even just slowing down the CPU will help lots. Jul 23 02:19:09 * cesarb suddenly has a burst of inspiration Jul 23 02:19:14 But that's not implemented yet Jul 23 02:19:22 it's just a regular linux kernel so I should be able to compile it to include the timmers needed to catch bogus wakeups Jul 23 02:19:32 cesarb: i hope that burst didn't use a lot of power... Jul 23 02:19:34 or would my effort be better spent elsewhere? Jul 23 02:19:49 Would it be useful for the programs to be able to signal their "intent" (as in, "I'm finished for now" or "I'm gonna need the CPU")? Jul 23 02:20:10 So that, for instance, we could power down the CPU without having to wait a little bit on higher power mode "just in case"? Jul 23 02:20:13 Not especially. Jul 23 02:20:27 AIUI - speedup or slowdown should be really quite fast. Jul 23 02:20:49 The hard part is deciding about deeper powersaving modes. Jul 23 02:20:59 should just be a matter of switching clocks, which should be doable within one or two slow clock cycles Jul 23 02:21:15 AIUI - not quite. Jul 23 02:21:21 The PLL isn't nearly that fast. Jul 23 02:21:28 Few hundred clocks, but still... Jul 23 02:21:38 ah, of course, one doesn't want to generate the fast clock when one isn't using it. Jul 23 02:22:00 * cesarb still thinks the thing which will save the most power is simply turning down the backlight Jul 23 02:22:00 If you want the screen, I think you have to be on fast clock. Jul 23 02:22:17 It's about a half CPU, and a half backlight. Jul 23 02:22:19 SpeedEvil: why does that make me think of the ZX81? Jul 23 02:22:23 Speaking very roughly. Jul 23 02:22:26 :) Jul 23 02:22:41 though of course the ZX81 was the opposite Jul 23 02:22:51 Turning off the screen will make memory bound tasks faster. Jul 23 02:22:59 ummm, my first love... Jul 23 02:23:00 As it frees memory bus bandwidth. Jul 23 02:23:02 you set it to fast mode for serious numbercrunching, and screen refresh didn't get serviced Jul 23 02:23:15 For low values of serious :) Jul 23 02:23:38 when you really needed every clock cycle you could get out of that 3.5MHz Z80 Jul 23 02:24:52 let's be kind... for 1981 values of serious :p Jul 23 02:25:56 the moko will probably run faster in powersaving mode Jul 23 02:26:15 Well - cycle per instruction, 12Mhz, 32 bit. Jul 23 02:26:18 Not even close. Jul 23 02:26:43 it would be nice if firefox-lite becomes.. Jul 23 02:27:00 wow, the Z80 took a minimum of four clocks per instruction, and that was for a NOP Jul 23 02:27:01 Lite. Jul 23 02:27:13 12 IIRC for a typical one. Jul 23 02:27:22 Involving a memory read Jul 23 02:27:54 yes lite, they are asking the team to create a stripped down version of ff Jul 23 02:28:21 http://slashdot.org/article.pl?sid=07/07/22/0711249&from=rss Jul 23 02:28:28 FF doesn't need stripped down. Jul 23 02:28:44 It needs some decent allocation monitoring tools so the leaks can be found. Jul 23 02:28:55 And so that devs can get an idea where the memory is leaking. Jul 23 02:29:06 well it might run faster and on less hardware if it didnt have fav's, tabs, addons Jul 23 02:29:11 Currently, you write an extension, and you have no idea where it's leaking. Jul 23 02:29:12 SpeedEvil: the leaks aren't the worse... the bad ones have already been fixed... Jul 23 02:29:19 SpeedEvil: the problem is the memory caching Jul 23 02:29:26 I mean leaks in all senses. Jul 23 02:29:36 SpeedEvil: as in, it caches all the images. uncompressed. in memory. twice (once on X and once on FF) Jul 23 02:29:44 are you guys talking about the software for the phone? Jul 23 02:29:47 Actually - no. Jul 23 02:30:00 It stores them only in the server AIUI. Jul 23 02:30:07 SpeedEvil: not to mention the page structure, bfcache, and lots of other memory-hogging "optimizations" Jul 23 02:30:52 The fundamental problem is that if I add code there is no easy way of telling how much memory I'm making it use :( Jul 23 02:31:01 Or how much extra CPU Jul 23 02:31:08 This does not lead to tight code. Jul 23 02:31:38 SpeedEvil: there are a few impossible to fix leaks in gecko Jul 23 02:31:42 Especially if to understand what it might be doing 'under the hood' when writing an extension might take years of study. Jul 23 02:31:50 thanks to refcounting and gc interacting poorly Jul 23 02:31:59 can i build some of the sys with only 6.2gb free?? Jul 23 02:32:07 at least some apps Jul 23 02:32:21 it is also *really* easy to accidentally set up a cyclic reference between an xpcom object and some js so that neither is ever gced Jul 23 02:32:27 is the card slot a Mirco SDHC ? Jul 23 02:32:34 yes, cucumber. Jul 23 02:32:49 What's the largest card you can actually buy these days? Jul 23 02:32:54 unknown_lamer: however. I'm using FF ATM on 128M and it's working OK for me. Jul 23 02:33:04 mrcucumber: 4G - but only 2G has been tested Jul 23 02:33:05 I heard that an 8gig was announced but what is actually for sale? Jul 23 02:33:06 SpeedEvil: yeah, but I doubt you're doing much else Jul 23 02:33:10 8G is sampling Jul 23 02:33:10 so ff works on moko? Jul 23 02:33:20 unknown_lamer: 12 shells, ... Jul 23 02:33:27 SpeedEvil: running conkerorxr on raw xulrunner gets me down to 40M vs 65M as a startup heap size ... Jul 23 02:33:39 BlackBsd: yes - however it's _not_ going to be the optimal solution. Jul 23 02:33:48 I'd like to have memory to run lots of apps instead of having to quit everything for the browser Jul 23 02:33:55 but a lite might work?? Jul 23 02:34:05 just looking into the future Jul 23 02:34:10 2752 ff 15 0 174m 56m 9732 S 5.5 45.8 433:17.94 firefox-bin Jul 23 02:34:14 Maybe. Jul 23 02:34:22 khtml is nice Jul 23 02:34:35 you mean webkit Jul 23 02:34:43 * unknown_lamer remembers back when he had 16M of RAM and konqueror finally let him browse the web with something nicer than dillo Jul 23 02:35:11 * SpeedEvil tries to remember what he was using on 16M. Jul 23 02:35:20 Netscape 3 or 4? Jul 23 02:35:48 non-free Jul 23 02:36:05 and eww motif :-) Jul 23 02:36:26 I don't know there was a free graphical browser back then. Jul 23 02:36:50 I switched over in 99 Jul 23 02:36:53 mozilla Jul 23 02:36:56 so there was early moz Jul 23 02:36:57 sea-monkey Jul 23 02:37:00 M11ish? Jul 23 02:37:10 IIRC mozilla was horrible compared to NS Jul 23 02:37:22 It's been a while though. Jul 23 02:37:28 yes but it was supposed to be used as a reference platform Jul 23 02:37:34 This was a 486/75 lappy. Jul 23 02:37:46 ns was made to be for end user use **** ENDING LOGGING AT Mon Jul 23 02:59:56 2007