**** BEGIN LOGGING AT Mon Apr 02 02:59:56 2007 Apr 02 03:08:34 how can i get sha256sum? Apr 02 03:09:48 you host distribution should have it in one of it's packages. Apr 02 03:10:43 i run makefile, it prompt no this file Apr 02 03:11:51 Please install following missing utilities: sha256sum Apr 02 03:11:59 http://www.openembedded.org/wiki/OEandYourDistro should have been updated by the OE core team to identify which package is required on each distro to get that. Apr 02 03:12:18 If not, please research that for your distro and update that wiki page. Apr 02 03:12:33 OK, thanks Apr 02 03:32:28 (script) wiki RecentChanges: [[Introduction/ee]] [[Main_Page/ee]] [[UI_Improvements]] [[User:Barnasc]] [[User_talk:Barnasc]] [[Wishlist:Mesh_Networking]] [[Wish_List]] Apr 02 03:41:50 openmoko: 03rwhitby * r64 10mokomakefile/trunk/patches/openmoko-HEAD/series: Removed openmoko-mirrors (already done upstream) and silence-retrieved-versions (henryk is going to commit that upstream) patches in an attempt to stop the bitbake cache rebuild on update Apr 02 04:29:28 (script) wiki RecentChanges: [[Main_Page/ee]] [[Introduction/ee]] Apr 02 04:38:25 counter Apr 02 04:38:25 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 15:40:47 (5.653 +-5.7 days) (718,140) Apr 02 04:39:07 are we on time for phase zero phone ? Apr 02 04:40:16 those are gone ages ago Apr 02 05:44:56 Mornin' all. Apr 02 05:52:44 Morning :) Apr 02 05:52:51 Hummmm.. Is the price for neo1973 in Europe going to be just ~260 euros? Or is there going to be some custom currency conversion from 350 USD -> XXX EUR? Apr 02 05:53:40 I haven't heard anything besides $350, so I hope it will be a fair exchange into euros Apr 02 05:55:01 Kewl.. ^^ I first feared it was going to be some 350 euros before checking the exchange rate and that sounded a bit expensive. Apr 02 05:57:21 good morning Apr 02 06:26:28 (script) openmoko-community: Sean Moss-Pultz An Update Apr 02 06:29:12 counter Apr 02 06:29:12 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 14:45:24 (5.615 +-5.6 days) (719,140) Apr 02 06:35:28 (script) openmoko-announce: Sean Moss-Pultz [openmoko-announce] An Update Apr 02 06:39:01 * koen looks at http://people.openmoko.org/sean/pictures/gta01b_v4_back.png Apr 02 06:39:26 http://people.openmoko.org/sean/pictures/gta01b_v4_front.png Apr 02 06:44:53 darn delays. =/ Apr 02 06:45:01 * Rakshasa got ants on hos pants. Apr 02 06:45:08 s/hos/his/ Apr 02 06:51:24 * rwhitby wonders if P0.5 phones will be sent out like P0 phone ;-) Apr 02 06:51:47 (if there is a hardware problem that requires a respin) Apr 02 06:54:21 Must be an april fool's joke... Aren't the images named the wrong way round - Front is back and vice-versa. XP Apr 02 06:54:38 Rakshasa: no, they are the correct way around. Apr 02 06:55:08 oh, no, you're right - they are the wrong way around. Apr 02 06:55:15 lol Apr 02 06:55:43 I guess it depends whether you're taking a "packaged phone" point of view or a "PCB" point of view. Apr 02 06:55:52 the names are correct from a PCB point of view. Apr 02 06:59:04 True, although a bit fuzzy, it being printed on both sides... Apr 02 07:04:31 mickeyl: ping - can you remove the 'oa' stuff from site.conf? Harald said in the bugzilla entry that it shouldn't be there, but he didn't remove it. Apr 02 07:09:45 openmoko: 03mickey * r1612 10/trunk/oe/conf/site.conf: Apr 02 07:09:45 openmoko: oe/conf: remove OpenAlchemy out of public repository Apr 02 07:09:45 openmoko: NOTE: reverted this locally on buildhost Apr 02 07:10:44 openmoko: 03mickey * r1613 10/trunk/oe/conf/site.conf: oe/conf: remove unused stuff from site.conf Apr 02 07:11:43 rwhitby: sure thing, sorry. Apr 02 07:16:42 * Psi_ reads the latest announcement Apr 02 07:16:51 hmm.. counter already had a bit 'extra' time at upper limit ;-) Apr 02 07:17:28 mickeyl: thx Apr 02 07:19:04 personally i feel that if they said "screw it, well just merge phase 1 and phase 1+ together" i would be fine with it Apr 02 07:21:28 (script) wiki RecentChanges: [[Neo1973]] [[Buying_Interest_List]] Apr 02 07:21:31 *we'll Apr 02 07:22:09 * aloril wants it as soon as possible (when its ready and stable) Apr 02 07:22:51 yes, we all agree, i think, that we are happy to wait until they fix the bugs Apr 02 07:24:13 i think we all agree that we have no other choice :p Apr 02 07:24:22 hehe Apr 02 07:25:01 yeah, but we aren't upset about the delays, we all want a stable phone :) Apr 02 07:26:56 morning Apr 02 07:27:08 TRIsoft, checked your email? Apr 02 07:27:35 Psi_: Not yet, just booting the company :-) Apr 02 07:32:02 openmoko: 03zhiyong_sun * r1614 10/trunk/src/target/OM-2007/panel-plugins/openmoko-panel-gsm/src/ (moko-gsm-conn.c moko-gsm-conn.h openmoko-panel-gsm.c): put gsm gprs signal status image into GtkHbox, which use to realize dynamic image show. Apr 02 07:38:51 counter Apr 02 07:38:51 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 14:10:34 (5.591 +-5.6 days) (720,140) Apr 02 07:39:10 Guess that one needs to be updated a bit.. Apr 02 07:44:28 <_ASK_> does someone know what the overall goal of the project is? Apr 02 07:44:35 <_ASK_> what a 1.0 openmoko will have? Apr 02 07:46:57 the goal is a complete mobile phone software stack.... Apr 02 07:47:04 including a comprehensive set of apps Apr 02 07:47:12 <_ASK_> but the specification sof such stack are... Apr 02 07:47:22 In the wiki? Apr 02 07:47:55 well, we are not interested in GSM specs Apr 02 07:48:03 as long as we have access to AT commands Apr 02 07:48:07 so don't misinterpret that Apr 02 07:48:30 <_ASK_> no, i mean Apr 02 07:48:40 <_ASK_> is there a plan to sync information with a desktop pc? Apr 02 07:48:52 i think so Apr 02 07:48:57 synching is an important field Apr 02 07:49:00 <_ASK_> things like that Apr 02 07:50:05 openmoko: 03zhiyong_sun * r1615 10/trunk/src/target/OM-2007/panel-plugins/openmoko-panel-gsm/src/ (moko-gsm-conn.c openmoko-panel-gsm.c): whitespace changes Apr 02 07:50:19 <_ASK_> but is there something written? Apr 02 07:52:07 text in the wiki Apr 02 07:52:08 code in svn Apr 02 07:52:26 it's not a speciifcations project Apr 02 07:52:29 it's more about coding :) Apr 02 07:53:25 <_ASK_> please... that's the biggest mistake in open source; to code just to see what happens... Apr 02 07:54:07 <_ASK_> open sorce != chaos. Apr 02 07:55:14 well Apr 02 07:55:21 please do a bit more research about the OpenMoko project Apr 02 07:55:36 <_ASK_> i'm asking, because i don't see much in the wiki Apr 02 07:55:37 you will learn how the project runs then Apr 02 07:55:50 see openmoko.org Apr 02 07:56:00 see openmoko.com Apr 02 07:58:39 hi mickeyl Apr 02 07:58:48 openmoko: 03zhiyong_sun * r1616 10/trunk/src/target/OM-2007/panel-plugins/openmoko-panel-gsm/src/openmoko-panel-gsm.c: gtk warning fix Apr 02 07:59:34 cheers XorA, hi xkr47 Apr 02 07:59:47 <_ASK_> i really feel sick sometimes; when people forgets the past and repeat the same mistakes. Apr 02 08:00:32 _ASK_: do you really think that attitude is the best way to influence people running this project? Apr 02 08:01:08 * XorA is scared now Apr 02 08:01:10 <_ASK_> not the best, bur certainly the most clear Apr 02 08:01:36 Perhaps a better way would be to create skeleton wiki pages with the specifications you are looking for, and help the developers to agree on and fill in and comply with such specifications? Apr 02 08:01:54 <_ASK_> i'm doing exactly that Apr 02 08:02:01 <_ASK_> i just don't complain, i work Apr 02 08:02:23 excellent. glad to see you're part of the solution. Apr 02 08:03:21 can you point me to the pages? I'm interested in seeing what level of specifications you're looking at ... Apr 02 08:03:52 (or just give me your WikiName and I can search for myself) Apr 02 08:04:00 <_ASK_> this project is very important because a) it's the first time handheld devices get truly open and b) because working in a new platform is an opportunity to build things right from the ground up Apr 02 08:04:37 Yep, that's a statement that I doubt anyone would argue with. Apr 02 08:06:07 <_ASK_> so my point is very simple: just give it an structure, a set of clearly defined standards that people will need to follow; i know it's limiting, but without standarization software never becomes a finished product Apr 02 08:07:20 I'm not sure anyone would disagree with that either. The point is which standards are the most important to do first, given the limit developer resources? Apr 02 08:07:56 <_ASK_> well, i was wondering if a proposal for the GUI is already done Apr 02 08:08:08 <_ASK_> it's something I would like to work in Apr 02 08:08:30 it's in the wiki Apr 02 08:08:35 _ASK_: there seems to be plenty of GUI design pages on the wiki, in great detail, including widget operations and layout. Apr 02 08:08:40 how many more times do we have to repeat that!?!?!? Apr 02 08:08:45 <_ASK_> where? Apr 02 08:08:46 wki Apr 02 08:08:47 wiki Apr 02 08:08:48 wiki Apr 02 08:08:50 wiki Apr 02 08:08:56 it has a search field Apr 02 08:09:05 <_ASK_> oh, not at widget level Apr 02 08:09:09 <_ASK_> but more on a system level Apr 02 08:09:53 _ASK_: you're looking for something higher level than http://wiki.openmoko.org/wiki/Look_%26_Feel I guess Apr 02 08:10:41 _ASK_: have you read the archives of the mailing lists where dbus and corresponding system-level things have been discussed? Perhaps you can summarise that to a wiki page if that's the level of standardisation you are looking for? Apr 02 08:11:21 <_ASK_> oh, I really think we have very different things in mind Apr 02 08:11:49 <_ASK_> i'll set an example Apr 02 08:12:01 ok, a pointer to that skeleton page you said you'd already started working on would be good at this point then ... Apr 02 08:12:45 <_ASK_> palm os has a "launcher", which is something like the window manager, a command bar, a preferences panel Apr 02 08:13:04 <_ASK_> an application can either be part of the launcher, or can be hidden from it and reside on the preferences panel Apr 02 08:13:10 ("system" means something to everyone at every level of the design hierachy. the guy who designed the GSM chip designed a "system", as did everyone else at each level above him. Apr 02 08:13:44 <_ASK_> it can also be configured in such a way it can be launcher from the command bar Apr 02 08:14:03 <_ASK_> those details might sound simple Apr 02 08:14:17 <_ASK_> but a handheld is a different concept compared to a desktop computer Apr 02 08:14:27 those sound like the top panel area and footer area, plus some conventions for use? Apr 02 08:15:01 <_ASK_> yes, it's just a matter to reach and agreement and write it down Apr 02 08:15:28 <_ASK_> but i feel the project largely lacks an end user "vote" in the election Apr 02 08:16:33 _ASK_: when you start the conversation with a confrontational "you guys are doing it all wrong" approach, you're bound to get the type of response that koen gave you. if you started with that example, and asked where you could help write down the current decisions and move forward from there, you're likely to get a better response ... Apr 02 08:16:57 <_ASK_> well, sorry, i don't wish to be arrogant Apr 02 08:17:46 "when people forgets the past and repeat the same mistakes" can be taken that way - since IRC text lacks the other 80% of communication subtlety. Apr 02 08:18:01 mickeyl: is there any Dbus concept or we yet have to develop one Apr 02 08:18:16 <_ASK_> i just want to call attention to important things, because there are a lot of developers, and i've seen lots of examples for years, who focuse too much on technical stuff forgoting about more general things Apr 02 08:18:28 (script) wiki RecentChanges: [[Main_Page/ee]] [[Buying_Interest_List]] [[Introduction/ee]] Apr 02 08:19:15 <_ASK_> just yesterday i remember someone on the channel saying "i don't want to align with any rules, i just want to code" Apr 02 08:19:24 _ASK_: here's the thing about developers - you need to gain their trust/respect first, and then influence them with reasoned discussions. Apr 02 08:19:43 Magon: dunno what you mean with concept, however we want to supply anything where it makes sense with a dbus interface to make things scriptable and remote controllable. the most interesting part of the dbus interface is the forthcoming neod (neo daemon). i expect to start with working on the API next week Apr 02 08:20:20 _ASK_: you also need to be able to distinguish between the people who are leading this project (like mickeyl, who give very rational responses), and other people on IRC who give cowboy developer responses like the one you just quoted. Apr 02 08:20:43 <_ASK_> ok Apr 02 08:21:19 (no offence to the person you quoted intended - coding for coding's sake has it's place too, but not in a commercial development like openmoko) Apr 02 08:21:32 i wanted to start writting footer, but first I need to know how it is supposed to work so I need to know dbus interface Apr 02 08:21:53 ah, right Apr 02 08:22:00 there has been some discussion about that last weekend Apr 02 08:22:04 i have been looking for dbus spec and dbus in code yeaterday and spec are good, and there is nearly no dbus in code Apr 02 08:22:22 yes, we don't make use of dbus yet Apr 02 08:22:26 except via eds Apr 02 08:22:47 what we expect from the footer API will soon be documented by alphaone Apr 02 08:22:51 (hopefully) Apr 02 08:23:04 also, I need to go over the GSoC proposals Apr 02 08:23:12 this is on my agenda for this week as well Apr 02 08:23:26 (contains also a footer application, i guess that's yours?) Apr 02 08:23:41 I want to help do build the dbus interface Apr 02 08:23:44 yes that is mine Apr 02 08:24:40 but there is not yet anything to build on, and building from scratch is not wise idea without discusion Apr 02 08:25:11 _ASK_: my suggestion would be to create a wiki page with examples of the types of standards you would like to see put in place, with examples perhaps from PalmOS (cause it seems you're familiar with that), and then read the archives of the mailing list to see what decisions have been made at that level already, and then open up the other questions for discussion on the mailing lists (pointing to the wiki page you've created). Apr 02 08:29:22 <_ASK_> well, now that you're suggesting things Apr 02 08:29:37 <_ASK_> what's the easiest way to create dummy apps? Apr 02 08:29:44 <_ASK_> like Visualbasic on windows Apr 02 08:29:50 <_ASK_> just to demonstrate concepts Apr 02 08:30:11 pygtk+? dunno - I'm not a gui expert. Apr 02 08:35:04 openmoko: 03mickey * r1617 10/trunk/oe/packages/openmoko-base/openmoko-libs_svn.bb: oe/openmoko-base: grab latest openmoko-libs from OE upstream Apr 02 08:40:49 i am very surprised by the today's mail of Sean Moss: Citation: My original plan was to announce that Harald convinced us all Apr 02 08:40:49 to turn OpenMoko into a closed source distribution --ClosedMofo. Apr 02 08:41:02 What does that means? no openmoko any more? Apr 02 08:41:29 Georg_Miersch: yesterday was the first of april Apr 02 08:42:03 hahaha Apr 02 08:42:05 :-) So is this also a joke: "we're accepting orders?" Apr 02 08:42:18 where is this annoncement? Apr 02 08:42:35 on announce lists of yesterday Apr 02 08:42:36 Announce mail Apr 02 08:42:38 Georg_Miersch: it was supposed not to be Apr 02 08:42:40 * mickeyl suddenly notices he's not on the announce-list Apr 02 08:42:50 on openmoko-something..discussion ? Apr 02 08:43:00 or announce... Apr 02 08:43:01 it's not so much of a joke as it is a description of a joke Apr 02 08:43:33 mickeyl: he sent it to the community list too, so you should get it anyway :P Apr 02 08:43:43 right Apr 02 08:43:43 * pH5 isn't on announce either. I don't have to be, thanks to IRC :) Apr 02 08:43:48 hehe Apr 02 08:43:52 Well phase 1 phone seems onway. Sean Moss included two hardware pictures in his mail in announce mail. very cool. Apr 02 08:43:53 wow Apr 02 08:43:56 nice unshielded phone pictures Apr 02 08:44:34 I notice more samsung hardware components than before, true? Apr 02 08:44:36 aloril's bot is our announce list ... Apr 02 08:45:11 Georg_Miersch: shouldn Apr 02 08:45:19 t be. probably just reseating on the PCB Apr 02 08:45:23 i.e. moving Apr 02 08:45:34 we still have no SMP phone Apr 02 08:45:38 heh Apr 02 08:45:43 say that the reason for the instability is the hardware, and the problem is discovered now - how long does it take to create a new hardware revision? Apr 02 08:45:49 SMP? Apr 02 08:45:54 symmetrical multi processing Apr 02 08:46:09 but there is basic phone capability Apr 02 08:46:19 mickeyl: you do have AMP though Apr 02 08:46:22 hehe Apr 02 08:46:29 XorA, yeah, but hardly controllable ;) Apr 02 08:46:36 or at least... very limited Apr 02 08:46:37 Does SMP means you can phone and game at the same moment? like multitasking? Apr 02 08:46:49 no no, SMP would mean putting more CPUs into the phone Apr 02 08:46:51 it's nuts Apr 02 08:46:51 good morning Apr 02 08:46:56 (at this point of time) Apr 02 08:47:00 cheers florian Apr 02 08:47:03 already got 2 cpu's in the phone Apr 02 08:47:11 ya, but AP/BP doesn't really count Apr 02 08:47:20 puny arm7tdmi Apr 02 08:47:45 AP/BP that is not AC/DC :-)? What is this processor for? Apr 02 08:48:12 * ozamosi get tired of trying to understand all the acronyms, and goes to watch curling instead Apr 02 08:48:24 AC/DC does that mean we all have to get into our schoolboy outfits? Apr 02 08:48:49 I am looking AP/BP on wiki... Apr 02 08:49:15 not found on the wiki. what the heck is ap? Apr 02 08:49:24 AP = Application Processor Apr 02 08:49:28 BP = Baseband Processor Apr 02 08:49:39 AP runs our shiny apps Apr 02 08:49:44 BP runs the GSM stack Apr 02 08:50:03 Thanks! And there are not a third cpu? Apr 02 08:50:19 not at this point of time Apr 02 08:50:33 perhaps some GP later Apr 02 08:50:38 * mickeyl grins Apr 02 08:50:57 :-) Apr 02 08:51:14 the bluetooth chip? Apr 02 08:51:17 GP? What happened to CP, DP, EP and FP? :) Apr 02 08:51:27 :D Apr 02 08:52:17 mickeyl: is neod supposed to be point controlling all dbus communication? Apr 02 08:53:44 not controlling everything (dbus daemon does that actually), but probably having the largest dbus API Apr 02 08:53:53 the bt is likely arm7 as well Apr 02 08:54:20 Well, I have another issue. Sean Moss wrote yesterday he's accepting orders though there is some delay. So WHERE can you place an order? Apr 02 08:54:47 I mean if it's not a 1st april joke... Apr 02 08:55:21 Georg_Miersch, he was describing what he would've liked to have done as an April Fool's joke Apr 02 08:55:53 Okey, thanx for clarification... Apr 02 08:57:34 So basically Sean Moss gave us the message that there will be more weeks delay because of some hardware issues still not resolved. It's better to have a good hardware than a flawed one. Apr 02 08:58:26 best April fool would have been "openmoko sold to microsoft" Apr 02 08:59:13 :-) or the hardware for free.. Apr 02 08:59:25 I was considering sending an email with some fabricated u-boot patches which implemented a new GTAPR01 device, where 'P' was for performance (400MHz) and 'R' was for radio (wifi). Apr 02 08:59:42 hehehehe Apr 02 08:59:47 rwhitby: :-D Apr 02 09:00:00 gsmd patches for umts Apr 02 09:00:01 but I took my son bowling instead :-) Apr 02 09:00:19 a boot over WiFi patch Apr 02 09:00:41 rwhitby, the joke's on him then ;) Apr 02 09:00:56 gsmd umts patches wouldn't be that much of a joke, now that gsmd runs on htc universal Apr 02 09:00:59 nah, he loved it (he's 6). got a strike and a spare. Apr 02 09:02:03 actually last weekend on the openmoko devel meeting we talked about april jokes as well Apr 02 09:02:10 "OpenMoko bought by MontaVista" Apr 02 09:02:19 "OpenMoko bankrupt" Apr 02 09:02:20 and stuff Apr 02 09:02:22 :D Apr 02 09:02:34 but we rather worked instead of formulating something Apr 02 09:02:35 "openmoko will be based on qt after all" Apr 02 09:02:36 :D Apr 02 09:02:40 hehe Apr 02 09:02:42 right Apr 02 09:04:31 "windows 95 source code released to public by microsoft -- Openmoko will be based on it." Apr 02 09:05:30 Psi_: OMG, what a morning... Checked all the mails. What do i have to look for ? Apr 02 09:05:31 scary :) Apr 02 09:05:46 I think the best april fools jokes are those that are believable, and you can't quite work out whether it's a joke or not. Pity Slashdot has no idea about it. Google's attempts were too far-fetched as well. Apr 02 09:06:19 * XorA has missed the internet cleaner for the last two years, "users" always seemed to fall for that one Apr 02 09:06:37 rwhitby: google always do two way too far-fetched, where one of them is actually true Apr 02 09:07:12 openmoko: 03alex_tang * r1618 10/trunk/src/target/OM-2007/applications/openmoko-messages/src/ (9 files): (log message trimmed) Apr 02 09:07:12 openmoko: Do some clean work Apr 02 09:07:12 openmoko: 1.free used strings Apr 02 09:07:12 openmoko: 2.remove comments and functions which will not be used any more Apr 02 09:07:12 openmoko: 3.clean some definitions Apr 02 09:07:14 openmoko: 4.review detail-area and rewrite member definition Apr 02 09:07:17 so which of the google paper or toilet broadband is the true one this year? Apr 02 09:07:18 openmoko: 5.move page enum from main.h to detail-area.h Apr 02 09:07:47 "we will be doing webmail. the standard 5 meg was not enough, so we'll give you 1 GB each instead" Apr 02 09:08:02 Well... I doubt the toilet broadband is true :) Apr 02 09:09:05 joke: you now can use the neo to switch traffic light phases to green.. Apr 02 09:09:11 ozamosi, Yeah, I thought Gmail was a joke too... )-: Apr 02 09:12:19 Didn't they put up that taxi tracking gmap system another year on 1st of april? Apr 02 09:16:28 (script) wiki RecentChanges: [[Buying_Interest_List]] [[User:Msquared]] [[Main_Page/ee]] Apr 02 09:18:51 If you look the picture of gta01b_v4 and ..v3 there is a big layout chance! Apr 02 09:20:45 Georg_Miersch: its a result of pressing 'reroute' button after making some changes Apr 02 09:22:43 the processor seems to be the same S3C2410... C0649 Apr 02 09:22:57 Georg_Miersch: gta01bv5 will get faster Apr 02 09:23:16 yeah! that is phase 1.5? Apr 02 09:23:45 bv5? yes Apr 02 09:24:05 openmoko: 03rwhitby * r65 10mokomakefile/trunk/patches/openmoko-HEAD/ (7 files): Remove all un-needed patches Apr 02 09:24:35 I think so they are same. I was referring to the wiki announcement that there will be a phase 1.5 device in June, with a faster cpu and possibly a wifi functionality Apr 02 09:25:49 openmoko may well build with no patches right now - need to rebuild from scratch to confirm. Apr 02 09:27:18 how realistic is the neo 1973 mockup (debian packages) to the real device? Differences? Apr 02 09:27:45 Georg_Miersch: ? Apr 02 09:29:30 I thought you can install openmoko in ubuntu and also run a neo device mockup on the pc screen using the mouse to simulate a touchscreen? Apr 02 09:29:50 ah.. you mean xoo with neo1973 skin? Apr 02 09:29:50 using the mouse to imitiate a touch screen? Apr 02 09:30:04 yes xoo Apr 02 09:30:15 * hrw has real neo1973 so no need for xoo ;D Apr 02 09:30:45 you bastard... Im still waiting... :-) Apr 02 09:33:36 hrw: can the touchscreen detect multi touch? two fingers touching the screen at the same time? Apr 02 09:33:41 yeah... I almost had a match in my hand, ready to burn my current phone once and for all, before I realized there would be more delays :( Apr 02 09:34:41 * XorA has three Neos Apr 02 09:34:41 nooo better collect them and plaster your living room like stuffed wildlife animal heads Apr 02 09:35:16 XorA - I have pity with you, only 3 --- how is this possible? Apr 02 09:35:49 although first one is very non working as a phone :-D Apr 02 09:35:55 Georg_Miersch: yesterday I would tell you that it can detect all 5 fingers Apr 02 09:36:08 :-) Apr 02 09:36:09 Georg_Miersch: today I will point you to wiki and faq Apr 02 09:36:39 Im reading hardware details on the second screen... Apr 02 09:36:49 in openwiki Apr 02 09:42:30 Yes neo 1973 has a touchscreen capable of multitouch. citation from openwiki: The Neo1973 has a screen that is capable of multitouch, only the driver is missing. If multitouch is available, then pressure sensitivity may be possible: The harder you press the finger, the bigger the area your finger covers. Apr 02 09:42:40 Pinging Psi_ again Apr 02 09:46:56 I will plan my application for Neo 1973 using multitouch even if the driver is missing. Well it seems that general system-level multitouch behavior still has to be established. Like pressing thumb in the lower left corner and using another finger to move a map image = grab function. if the thumb is not on the screen, a finger tap will activate a map zoom. Apr 02 09:51:46 hey Apr 02 09:51:59 just got openmoko working on a HTC Magician Apr 02 09:52:11 hey i386, congrats Apr 02 09:52:26 I wonder why people insist on multitouch when screen has 2.8" only Apr 02 09:52:26 http://wiki.xda-developers.com/index.php?pagename=HTC_Magician Apr 02 09:52:36 pH5: ty Apr 02 09:52:42 hrw: apple PR monster Apr 02 09:52:45 Im just having a little bit of trouble though Apr 02 09:53:35 I assume the screen on the neo is much larger than the htc magician Apr 02 09:53:45 is there anyway to rescale the ui? Apr 02 09:53:57 i386: neo has 2.8" vga. magician has qvga iirc? Apr 02 09:54:36 i386: there a qvga theme in the svn Apr 02 09:54:46 right Apr 02 09:54:46 i386: but unfortunately a lot of the apps still hardcode ui sizes Apr 02 09:55:03 I have working patches for the guitar app and half of the dialer. Apr 02 09:55:05 is that going to change? Apr 02 09:55:13 i386: sure! Apr 02 09:55:14 oh rock! Apr 02 09:55:15 multitouch makes sense!!! the iphone is designed with multitouch, you can access application and function faster Apr 02 09:55:44 pH5: right Apr 02 09:55:48 Georg_Miersch: show me *working* iphone then I will maybe understand Apr 02 09:56:03 To tell you the truth ive hacked a recent openmoko root image Apr 02 09:56:19 with the right kernel and driver loading info for the htc magician Apr 02 09:56:20 Georg_Miersch: proper gesture support will be much more useful. I usually only have the thumb on the touchscreen. Apr 02 09:56:23 multitouch; one finger is working as "shift" the other can select another functions. Apr 02 09:56:45 is there anyway to get pimped out on powerpc linux for openmoko development? Apr 02 09:57:12 /ignore is powerfull command Apr 02 09:57:22 I am pretty sure you won't miss multitouch. I am not against singletouch. I advocate the using of both possibilities. Apr 02 09:58:07 apple are experts at the vulcan nerve pinch Apr 02 09:58:25 multitouch watch the amazing movie! http://cs.nyu.edu/~jhan/ftirtouch/ Apr 02 09:58:46 better Apr 02 09:58:52 hrw: you were right Apr 02 09:58:56 :) Apr 02 09:59:33 pH5: could I get a look at those patches? Apr 02 10:00:33 multitouch: to enlage a thumbnail image put two fingers on screen and move the fingers in opposite directions (one finger to the left other finger to the right), then you will zoom in the picture accordingly. Apr 02 10:01:13 i386: sure, pretty much WIP: http://en.pastebin.ca/420682 Apr 02 10:01:24 Georg_Miersch: i think the people here are familiar with that Apr 02 10:04:32 about the half yes - the other cannot imagine multitouch possibilities. Apr 02 10:05:17 sorry what was that url again? Apr 02 10:05:26 http://en.pastebin.ca/420682 Apr 02 10:05:31 ta Apr 02 10:05:41 Just had my Xserver crapout :/ Apr 02 10:06:17 Georg_Miersch: yes they can, they just don't see a reason to put their time and effort on the issue, since it's not a priority one Apr 02 10:12:28 (script) wiki RecentChanges: [[Calculator]] [[Buying_Interest_List]] Apr 02 10:14:02 where can I find it in svn? Apr 02 10:14:51 (the gvga theme) Apr 02 10:16:07 anyone can tell me url of download sha256sum Apr 02 10:17:08 yes. but anyway for my application I already plan and design multitouch capability for the future while providing singletouch capability. This will save me some time to implement multitouch user input, if a driver is ready. Apr 02 10:17:11 ping_tu: newer coreutils have it Apr 02 10:17:12 ping_tu: for the slugosle distro, you should look in ipkg.nslu2-linux.org - does that help? or was there some other distro you were interested it? Apr 02 10:20:50 pH5: do you perhaps have a binary of that fix? Apr 02 10:22:02 i386: trunk/src/target/OM-2007/artwork/themes/openmoko-standard-qvga Apr 02 10:22:06 i386: only on the phone Apr 02 10:22:15 pH5: ahh bugger Apr 02 10:22:30 pH5: dont have my dev enviroment setup here :P Apr 02 10:24:38 pH5: thanks anyway :) Apr 02 10:25:18 i386: np, let me know if you find time to do further qvga-ification :) Apr 02 10:26:10 ping_tu it's in slackwares coreutils ftp.slackware.com in the source dirs Apr 02 10:26:20 pH5: hey Apr 02 10:26:27 I just got the themes now Apr 02 10:26:29 and its working Apr 02 10:26:36 s/working/booting/ Apr 02 10:26:36 i386 meant: and its booting Apr 02 10:27:05 so ill let you know when its all done... Apr 02 10:27:14 i386: cool Apr 02 10:29:41 pH5: right Apr 02 10:29:45 that looks much better Apr 02 10:29:46 although Apr 02 10:29:58 the keyboard pixmap looks like its missing from the top left Apr 02 10:30:11 the desktop items are rather large Apr 02 10:30:29 re kb pixmap, It's the same on vga. don't get me started about mb-panel2 Apr 02 10:32:13 pH5: mb-panel2? Apr 02 10:33:33 matchbox-panel-2, the top panel app. Apr 02 10:34:30 pH5: without the patches I suppose I cant dial anything? Apr 02 10:34:58 well, the 'dial' button is off-screen. you could navigate there with the d-pad, I guess :) Apr 02 10:35:32 * rwhitby hopes the next neo device after 1973 will have a d-pad ... Apr 02 10:35:47 (as an alternative to the touchscreen only device) Apr 02 10:36:03 pH5: no dpad on the magician :) Apr 02 10:36:29 * rwhitby dreams of openmoko running on the Treo650 ... Apr 02 10:37:10 * SpeedEvil is going to try to fit a thumbstick to his neo. Apr 02 10:37:12 (except for that speed thing, of course - maybe a turbocharged Treo650v12) Apr 02 10:37:35 i386: what magician are you talking about? Apr 02 10:37:38 http://www.flickr.com/photos/7337648@N08/415587809/ Apr 02 10:40:20 <_ASK_> oh i love the old dell keyboard Apr 02 10:40:35 <_ASK_> newer models have a filmsy one... Apr 02 10:41:55 pH5: HTC Magician Apr 02 10:42:04 pH5: sold as O2 XDA Mini Apr 02 10:42:12 so you have direction keys below the screen, right. Apr 02 10:42:18 yeah I do Apr 02 10:42:20 ok Apr 02 10:48:27 <_ASK_> nite Apr 02 10:54:13 _ASK_: night - I look forward to seeing your specifications wiki page ... Apr 02 11:01:00 openmoko: 03alex_tang * r1619 10/trunk/src/target/OM-2007/applications/openmoko-messages/src/ (sms-contact-window.c sms-contact-window.h): Apr 02 11:01:00 openmoko: Change contact-window parent from MokoWindow to GtkWindow Apr 02 11:01:00 openmoko: For reason that MokoWindow will automatically add window to MokoApplication, so there would be a warning 'window added twice'. Apr 02 11:08:28 (script) wiki RecentChanges: [[Buying_Interest_List]] Apr 02 11:23:46 counter Apr 02 11:23:46 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 12:18:07 (5.513 +-5.5 days) (721,140) Apr 02 11:46:09 tag Apr 02 11:48:18 morsche Apr 02 11:56:12 so, have people already analyzed all the changes in the new to be released hardware? Apr 02 11:56:24 based on the couple pictures, I mean ;) Apr 02 11:57:30 what pictures? Apr 02 11:57:43 http://lists.openmoko.org/pipermail/announce/2007-April/000011.html Apr 02 11:58:15 Hadaka since i can go over to harald and just touch the pcb... ;) Apr 02 11:58:46 roh: heh, well, so, what's different from phase0? :) Apr 02 11:58:46 Oh - right. Apr 02 11:59:12 It's naked. Apr 02 12:00:09 Hadaka mostly minor changes like pins on flatcables or single gpio reorderings Apr 02 12:00:14 Well... Apr 02 12:00:20 Debug connector has moved. Apr 02 12:00:27 err Apr 02 12:00:31 no, never mind. Apr 02 12:01:10 the parallel bus on the debug connector is gone (did more harm than use) and a i2c is added there Apr 02 12:03:30 What are the pads next to the vibrator? Apr 02 12:03:53 headers for the remote-control dongle for the japanese market Apr 02 12:03:56 j/k Apr 02 12:05:09 roh: okay, thanks Apr 02 12:05:28 (script) wiki RecentChanges: [[UI_Improvements]] Apr 02 12:05:45 are there sensors in the phone? temperature? something else? Apr 02 12:06:22 Hadaka not that i know of.. and if they would be unpractical for external use Apr 02 12:09:25 It was stated somewhere earlier that SPI would be brought out to solderpads. Did that end up happening? Apr 02 12:09:49 SpeedEvil i2c Apr 02 12:10:32 Sigh - I always get those confused. Apr 02 12:12:41 roh: thanks Apr 02 12:19:45 * SpeedEvil crosses fingers for "subtle software bug that will be found tomorrow". Apr 02 12:47:22 bahahhh Apr 02 12:47:30 ppc crash in monotone! Apr 02 12:52:05 openmoko: 03dodji * r1620 10/trunk/src/target/OM-2007/openmoko-libs/ (3 files in 2 dirs): Apr 02 12:52:05 openmoko: Went forward with the mokojournal API mockup Apr 02 12:52:05 openmoko: * openmoko-libs/libmokojournal/moko-journal.[c|h]: went forward Apr 02 12:52:05 openmoko: with the API proposal Apr 02 12:54:55 openmoko: 03laforge * r1621 10/trunk/ (2 files in 2 dirs): Apr 02 12:54:55 openmoko: * switch vibrator pin into GPIO OUTPUT mode by default Apr 02 12:54:55 openmoko: ** this prevents vibrator from being default-on at boot time Apr 02 12:54:55 openmoko: ** kernerl driver needs to switch from GPiO output to TOUT3 function mode Apr 02 13:04:04 hmmh, am I the only one who thinks that by-nature extensible stuff, such as the 'journal' for calls and stuff, should not be coded in C - or atleast there should be a framework for memory management and datatypes, such as APR? Apr 02 13:04:39 Hadaka: we all should write in BASIC Apr 02 13:04:57 hrw: noo! FORTH! Apr 02 13:05:16 robtaylor: dont be stupid, COBOL Apr 02 13:05:35 awk. Apr 02 13:06:16 a friend of mine once wrote a java optimiser in awk, when he was illegally fired from company I was only man who could decode it :-) Apr 02 13:06:37 XorA: that's truly sick Apr 02 13:07:28 Awk is very nice for what it's good at. Apr 02 13:07:47 I've even done realtime DSP in it. Apr 02 13:07:57 the only awk i know is '{print $2}' Apr 02 13:08:10 awk '$0=$2' is faster. Apr 02 13:08:13 greets Apr 02 13:08:18 from where i can be able to get the source code of openmoko Apr 02 13:08:19 perhaps I'm just not thinking in phone terms, but if adding, for example, jabber messages to journal requires changing libopenmoko C-API and coding C, that sounds very unextensible Apr 02 13:08:44 its amusing how much of a speed inscrease you can gain from java with a few regexps Apr 02 13:09:02 Hadaka: i think you're confusing code and data Apr 02 13:09:20 java.util.regex++: Apr 02 13:09:23 robtaylor: am I? Apr 02 13:09:26 s/\:/\;/ Apr 02 13:09:43 Hadaka: well, if the api sucks, then the api needs fixing Apr 02 13:09:49 ewon: nah, just fixing some fundamental mistakes in the Java VM Apr 02 13:09:55 Hadaka: thats completely orthogonal to which language to us Apr 02 13:10:01 *e Apr 02 13:10:10 is the openmoko gpu / cpu quick enough to decode a stream of jpegs in realish time? Apr 02 13:10:10 ewon: See http://wiki.openmoko.org/wiki/FAQ (Hopefully this link answers your question. This autotrigger for new nick and question during first 15 minutes done only once.) Apr 02 13:10:18 nice :) Apr 02 13:10:41 weeble? Apr 02 13:10:52 robtaylor: well yeah, I'm bundling those two things only because C-code requires recompilation and API changes are nasty when trying to keep binary compatibility in a library Apr 02 13:11:35 api will always be hard Apr 02 13:11:55 dealing with abi changes is a pretty well understood technique ;) Apr 02 13:12:09 robtaylor: the fact that C is the language is of no importance, except if using C means manual memory management Apr 02 13:12:32 it would be nice to have a phone I can write C for Apr 02 13:12:33 Hadaka: ok, you try writing something that runs fast on Neo hardware using another language.. Apr 02 13:12:45 everytime I have to write something in j2me I die a little bit inside Apr 02 13:12:54 haha Apr 02 13:13:05 robtaylor: yeah, abi changes can be handled - but it's really was a pain when developing subversion for example Apr 02 13:13:11 It's a relatively slow processor. You have to be _reasonably_ efficient. Apr 02 13:13:22 currently have a 6230i, and pretty much the only option is j2me Apr 02 13:13:22 well yeah, perhaps I'm just not thinking in phone terms then Apr 02 13:13:24 how does j2me compare to j2se? Apr 02 13:13:35 Especially when in future cases we may be able to use the display in slow mode (12Mhz) Apr 02 13:13:44 it's like j2se's retarded cousin that touches people inappropriately. Apr 02 13:14:04 ewon: :D Apr 02 13:14:05 Hadaka: feel free to code in any language you want, you don't need to use libopenmoko Apr 02 13:14:24 robtaylor: yeah, ofcourse, I just thought it was a bit odd Apr 02 13:14:25 I write C for a living, having to use j2me to get use out of my phone causes me physical pain Apr 02 13:14:28 its soemthing very specifically for a phone platform Apr 02 13:14:44 Hadaka: you thought using C on a slow, memory constrained embedded platform was odd? Apr 02 13:14:51 interesting Apr 02 13:15:12 robtaylor: I've used python programs on N770 which use GTK and all - they are not unusable at all Apr 02 13:15:25 robtaylor: and python is about the slowest language one could imagine running on that Apr 02 13:15:29 Hadaka: N770 is about 4 times faster than Neo, iirc Apr 02 13:15:49 Hadaka: and import python-glib still takes 30 seconds on a N770 Apr 02 13:16:05 that's bad Apr 02 13:16:09 robtaylor: is it? I thought they were somewhat comparable? I tried asking around for that information earlier Apr 02 13:16:11 lads, why is the micro-sd slot under the battery? is this not what killed the n-gage? Apr 02 13:16:25 ewon: cheapness. Apr 02 13:16:38 It's more expensive to move it off the combined slot. Apr 02 13:16:38 Hadaka: no, the samsung processor is really quite slow for its clock speed Apr 02 13:16:44 I see Apr 02 13:16:54 I'd sacrifice speed for battery life any day Apr 02 13:16:59 In future, other devices may be better. Apr 02 13:17:13 robtaylor: ah, okay, well then it is getting a bit tricky Apr 02 13:17:15 Hadaka: starting udev taks 44s on the neo hardware, about 4 n N770 Apr 02 13:17:16 and have for example SD or microSD slots. Apr 02 13:17:43 Hadaka: i'm not really sure whats so slow though, I don't have one myself Apr 02 13:17:45 Hadaka: i'm all for python use too Apr 02 13:17:58 robtaylor: we suspect the OE mount script to be guilty Apr 02 13:18:05 robtaylor: I just tried python on N770 - 2-3 seconds for start up, 4-5 second for import glib - and not sure if this xterm lags or not Apr 02 13:18:12 Hadaka: also mono's looking quite good on N770/N800 too Apr 02 13:18:35 Hadaka: hmm, thats quite interesting, is that python2.5? Apr 02 13:18:37 mono on openmoko would be really nice Apr 02 13:18:43 robtaylor: 2.4.2 Apr 02 13:19:34 robtaylor: I'm not sure what you mean by importing glib - do you mean starting the main loop as well? I just did 'import glib' Apr 02 13:20:02 Hadaka: no i mean import glib, but maybe armeabi has helped with the speed there Apr 02 13:20:51 Hadaka: and perhaps InDt have applied their glib linking speedups (breaks abi, but doenst really matter on this platform) Apr 02 13:21:01 Hadaka: anyhow, its good to hear its a lot faster than it was Apr 02 13:21:02 robtaylor: I don't know enough of the subject anyway - but the point is that I can start a python program on my N770 that uses GTK - and it starts under 10 seconds and is usable Apr 02 13:21:41 Hadaka: how much memory does something that just imports and runs the mainlooop take? Apr 02 13:21:47 but yeah, if the processor one quarter of the speed, then that is a problem Apr 02 13:21:55 (i guess you don't have exmaps installed?) Apr 02 13:23:02 * guaqua is pondering about python vm Apr 02 13:23:20 robtaylor: I don't know how this is possible - but 2.8M resident Apr 02 13:23:22 * guaqua and feeling really bad in the stomach thinking about garbage collection Apr 02 13:23:43 robtaylor: if exmaps is a separate installation on top of the default firmware, then I probably don't have it installed Apr 02 13:24:11 Hadaka: try http://floopily.org/~robtaylor/memstats-busybox.py Apr 02 13:24:38 | grep python should do it Apr 02 13:25:41 hrm, something iffy with the wlan, will take a moment Apr 02 13:30:17 I presume if you guys go for a wifi chip in $NEXT_VERSION, it will support such useful things as monitor mode, and maybe an external antenna connection? Apr 02 13:31:23 ewon we will see. since we want competely opensourced drivers and i know no hardware which cannot do that i hope so ;) Apr 02 13:31:59 excellent :) Apr 02 13:32:16 i'd love to see the mesh option :) Apr 02 13:32:28 robtaylor: os.getlogin raises no such file on this installation, I'll try hardcoding that Apr 02 13:32:42 (no idea what might be wrong, this device is up for reinstallation soon anyway) Apr 02 13:32:42 I'm looking forward to a phone I can do useful things with :) Apr 02 13:32:58 ewon: useful? what's that? ;) Apr 02 13:33:06 guaqua i'd love to see any mesh protocol that is not complete and utter crap.. but besides that there is still hope Apr 02 13:33:17 guaqua: something I can write apps for without selling my soul to j2me Apr 02 13:33:32 Hadaka: ah, yeah, i think i've seen that issue too Apr 02 13:36:16 robtaylor: heap 380KB, stack 64 Apr 02 13:37:35 sha256sum is missing in the current build Apr 02 13:37:39 counter Apr 02 13:37:39 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 11:11:10 (5.466 +-5.5 days) (722,140) Apr 02 13:37:51 i think that counter should list something else Apr 02 13:39:20 ewon: hang on... I sold my soul when I started working with j2me? wtf? I never got a dollar for it! Apr 02 13:39:46 Stephmw: welcome to the dark side Apr 02 13:40:14 gayness Apr 02 13:40:32 misfire Apr 02 13:41:54 Hadaka: hmm, its not too bad. i'd guess the same thing in c would probably be about 16,16 Apr 02 13:42:40 * zecke gets annoyed by SHA Apr 02 13:43:08 why? Apr 02 13:44:12 ewon: what's your objection btw.. Sun's phoneME? Would you use a GNU Classpath-based J2ME implementation? Apr 02 13:45:06 Stephmw: I wouldn't use any j2me implementation at all... Apr 02 13:45:14 I'd use a C crosscompiler, like God intended. Apr 02 13:45:29 hehehe Apr 02 13:45:39 ewon: I see Apr 02 13:46:34 java on something like the openmoko is, IMO, wrong. Apr 02 13:46:40 we _know_ the hardware. Apr 02 13:46:42 use C. Apr 02 13:47:04 well, you guys know the hardware, I'm sure I'll poke it with a stick when the consumer version comes out Apr 02 13:47:29 So you don't want to have access to existing mobile phone games? Apr 02 13:47:37 Java - for end-user apps isn't silly. Apr 02 13:47:41 for my uses, not really Apr 02 13:47:57 ewon: ok, now that makes more sense Apr 02 13:48:41 If I want to bum a few instructions off something in C, I can, with the java VM I'm stuck with whatever the VM implementer thought was sane at the time Apr 02 13:49:05 I'm not saying there should not be a java VM for the openmoko, I'm stating that it shouldn't be the defacto choice for new userspace development. Apr 02 13:49:30 that said, I've been here around 10 mins, so feel free to ignore. Apr 02 13:49:53 I think no language which has manual memory management and is trivially crashable by typos should be the defacto choice for new userspace development Apr 02 13:50:43 depends how you handle userspace crashes Apr 02 13:50:45 sure... let's put that Core2Duo in the phone, to be able to run all those shiny java apps ;) Apr 02 13:51:02 and with limited hardware, manual memory management is possibly a necessity Apr 02 13:51:20 drath: there are recently developed batteries that may work for that. Apr 02 13:51:24 but, we'll how everything develops - it's ofcourse nice if the basic functionality of the phone is as fast and hand tuned as possible Apr 02 13:51:34 drath: they can completely discharge in around 2 minutes. Apr 02 13:51:40 hehehehehe Apr 02 13:51:53 but, have to jet now, ta ta Apr 02 13:51:59 Wave. Apr 02 13:52:07 guess you could also start a car engine with one of those batteries then :) Apr 02 13:58:16 openmoko: 03mickey * r1622 10/trunk/src/target/OM-2007/examples/openmoko-panel-demo/ (configure.ac src/Makefile.am): openmoko-panel-demo: fix dependencies, closes #351 Apr 02 14:00:28 (script) wiki RecentChanges: [[UI_Improvements]] Apr 02 14:09:43 drath: you're obviously referring to desktop Java, ie J2SE. Apr 02 14:10:03 drath: J2ME is more performant than that Apr 02 14:10:16 drath: as well as being quite happy within a 2mb heap Apr 02 14:10:39 it's still a VM, doing its own garbage collection / hand-holding / ass-wiping Apr 02 14:10:57 which is exactly what you want for untrusted code Apr 02 14:11:45 i think it would be a nice to have Apr 02 14:12:04 I'm unsure if chroot might not be a better option. Apr 02 14:12:26 Or similar. Apr 02 14:12:27 there are a lot of devs who only have experience from java - and they would be a nice addition Apr 02 14:13:25 SpeedEvil: that's a valid point - I'm still uneasy with the ability of a native process to potentially block emergency calls Apr 02 14:13:36 SpeedEvil: something trivial to avoid in Java Apr 02 14:13:42 SpeedEvil: at least, at the vm level Apr 02 14:14:01 heh, always thought one of the points of having free software is that i can actually /trust/ the code Apr 02 14:14:24 Umm - how can a native process block emergency calls? Apr 02 14:14:32 If gsmd says no. Apr 02 14:14:48 it can kill gsmd Apr 02 14:14:49 by blocking user interaction Apr 02 14:14:54 fuser -k /dev/ttySAC0 Apr 02 14:15:12 surely this could be implemented like the magic sysreq key in the kernel? get-through-at-all-costs kind of thing? Apr 02 14:15:12 It can't kill gsmd if it's running as an untrusted process in a chroot. Apr 02 14:15:33 no, but it can startve other processes: excessive IO, etc Apr 02 14:15:38 *starve Apr 02 14:15:49 just open an unclosable system modal dialog Apr 02 14:15:51 so have the gsm handling as a kernel module? Apr 02 14:15:54 There are existing solutions for that though. Apr 02 14:15:54 full screen Apr 02 14:16:18 SpeedEvil: I'm happy to cross that hurdle when we come to it though Apr 02 14:16:18 The trivial option would be not to allow untrusted applications fullscreen. Apr 02 14:16:33 keep the header/footer always. Apr 02 14:34:40 counter? Apr 02 14:35:08 bahh .. Apr 02 14:39:28 cydork: Apr 02 14:39:29 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 10:40:17 (5.445 +-5.4 days) Apr 02 14:39:41 aloril: ahh thanks. :) Apr 02 14:40:19 cydork: it didn't like extra space at end (and it gives shorter version without '?') Apr 02 14:40:55 aloril: oh ok.. didn't knew that :) Apr 02 14:44:40 is there already handwriting recognition in openmoko (or will there be in near future)? Apr 02 14:46:08 depends if there is an opensource handwriting recog solution - maybe something ripped from the 770? Apr 02 14:46:40 there are Apr 02 14:46:50 openmoko: 03mickey * r1623 10/trunk/src/target/OM-2007/applications/openmoko-appmanager/openmoko-appmanager.pro: openmoko-appmanager: repair qmake building Apr 02 14:51:37 did someone mention 'open source' and 770 in the same sentence? Apr 02 14:52:02 I did Apr 02 14:52:06 *hangs head in shame* Apr 02 14:52:28 openmoko: 03mickey * r1624 10/trunk/src/target/OM-2007/inputmethods/: add inputmethods directory, this is where the openmoko-inputmethods should reside Apr 02 14:54:23 mickeyl: you noticed openmoko-keyboard_svn.bb in OE? ;) Apr 02 14:54:32 sure thing :) Apr 02 14:54:48 took a look aet? Apr 02 14:54:49 yet Apr 02 14:55:01 not yet Apr 02 14:55:02 om-keyboard is a good start, far from what we'd like, but not bad imo Apr 02 14:55:18 I'd like to have it as a patch to matchbox-keybaord Apr 02 14:55:43 openmoko wont have a zillion chinese code slaves in the future to maintain all forks Apr 02 14:56:07 well Apr 02 14:56:07 ya Apr 02 14:56:15 i told that Apr 02 14:56:18 "do it as a patch" Apr 02 14:56:30 I know how stuff like that goes Apr 02 14:56:40 some engineers keep ignoring me Apr 02 14:56:43 we will change that Apr 02 14:56:54 you don't want to see the 'utils' repo at the university Apr 02 14:56:59 hehe Apr 02 14:57:28 (script) wiki RecentChanges: [[Translation]] Apr 02 14:57:29 s//UTwente /g Apr 02 14:57:44 and hacks to keep it working with gcc 2.95.3..... Apr 02 14:57:54 openmoko: 03www-data * r1625 10/trunk/src/target/OM-2007/ (2 files in 2 dirs): move openmoko-keyboard to inputmethods directory. NOTE: need adjusting OE recipes Apr 02 15:01:30 mickeyl: FIC isn't sponsering GUADEC? Apr 02 15:01:42 no idea Apr 02 15:01:52 that's business issues Apr 02 15:03:27 mickeyl: could you ask the ipkg engineers to split up the is-processing patch? Apr 02 15:03:36 i can try... Apr 02 15:05:14 openmoko: 03mickey * r1626 10/trunk/oe/ (3 files in 2 dirs): oe: catch up w/ renaming openmoko-keyboard location in SVN Apr 02 15:06:03 * koen tests the diffsplit script Apr 02 15:10:23 oeeeh... where does the oe get "sha256sum" from? Apr 02 15:10:39 I hate SHA :) Apr 02 15:10:58 roh: coreutils of debian and ubuntu feature sha256sum Apr 02 15:11:04 it complains about missing it after i removed build/tmp since it was grown too much Apr 02 15:11:13 zecke i havent removed packages Apr 02 15:12:17 roh: a couple of options: a) wait b) remove conf/checksums.ini and remove the checksums c) alias sha256sum="shasum -a256 -p" and install the perl module d) ... Apr 02 15:12:49 ok, diffsplit works as advertised, but is not what I want Apr 02 15:13:20 koen: use mc to edit patches Apr 02 15:14:11 zecke: that, or I send you some frikandellen ;) Apr 02 15:15:36 mickeyl: the best example of stubborn engineers was when the uni website framework needed an update and one engineer did everything in ASP .NET while the framework was in php Apr 02 15:15:55 (and ASP .NET doesn't work to well on apache 1.3.x) Apr 02 15:18:22 yeah, back in college we had the token db-driven-website project... 90% of the class did it in ASP with an Access backend... 2 teams (including the one I was on) did it in PHP / MySQL - the ASP server used to die twice a day due to the sheer volume of atrocious code running on it Apr 02 15:48:39 hi! Apr 02 15:48:47 in the Wiki, there is Estonian translation... Apr 02 15:48:57 but the language code is mistakenly "ee" Apr 02 15:49:05 it should be "et" instead... Apr 02 15:49:24 enemy-territory? Apr 02 15:49:50 well, was for 50 years.. in 91 we declared our independence Apr 02 15:49:50 openmoko: 03mickey * r1627 10/trunk/oe/packages/openmoko-base/ (openmoko-session_svn.bb openmoko-theme-standard-qvga_svn.bb): oe: actually package the proper theme Apr 02 15:50:24 plaes: ISO language codes does not match other tables Apr 02 15:50:43 plaes: .uk is United Kingdom or Ukraine? Apr 02 15:51:24 united Kingdon is en_GB, ukraine is uk or ukr Apr 02 15:54:51 pH5: your gsmd now has vendor and machine plugins? Apr 02 15:57:06 koen: yes, tested on magician and neo, goxboxlive will test universal. Apr 02 15:57:19 I'll wait what LaF0rge has to say about that patch. Apr 02 15:58:23 * koen hopes the ezx dudes will find some time to work on openezx Apr 02 15:58:55 people got the bluetooth working on ax80 and a1200 Apr 02 16:00:38 mickeyl: isn't that change to openmoko.bbclass superfluous? Apr 02 16:00:51 is it? Apr 02 16:00:53 + elif section == "inputmethods": return "inputmethods" Apr 02 16:00:53 let me check again Apr 02 16:00:54 else: return section Apr 02 16:00:59 uah Apr 02 16:01:06 ~lart me Apr 02 16:01:06 * apt runs at mickeyl with an origami Swiss Army knife, and inflicts a nasty paper cut Apr 02 16:01:22 that's what i get for doing too many things in parallel Apr 02 16:01:27 That's why I didn't change it :) Apr 02 16:01:31 right Apr 02 16:01:46 OTOH, it marks it as a valid section Apr 02 16:02:00 note to self: don't make conference calls while coding Apr 02 16:02:19 yeah, i want to change it to a dictionary anway soon Apr 02 16:02:26 mickeyl: http://sourceforge.net/mailarchive/forum.php?thread_name=460F8B38.50704%40iki.fi&forum_name=gtk-webcore-devel Apr 02 16:02:32 (in case you didn't see that yet) Apr 02 16:15:43 koen: I didn't! Thanks Apr 02 16:16:52 does openmoko support scandinavian characters (æøåäö)? Apr 02 16:17:27 koen: so gtk-minibrowser can be better now? Apr 02 16:17:38 Vegar: on screen and on keyboard Apr 02 16:17:49 cool Apr 02 16:20:45 hm, I'd like to see a picture of the HTC Universe and the Neo1973 Apr 02 16:33:30 Vegar: the locale is utf-8 Apr 02 16:37:02 hrw: gpe-mini-browser you mean? Apr 02 16:37:07 yes Apr 02 16:37:51 Yep, gtk-webcore seems to work a lot better. I ironed out some last issues with Kimmo after that commit and updated the svn stuff in OE Apr 02 16:41:56 I also added a patch that silences the gtk-webcore output so that we save some cpu cycles. I'll try it out on the openmoko soon. Apr 02 16:42:41 ~lart mb-panel2 in openmoko Apr 02 16:42:42 * apt send killer squirrels to attack mb-panel2 in openmoko Apr 02 16:43:19 ~lart in process plugins Apr 02 16:43:19 * apt beats in process plugins into protomatter with the andromeda galaxy Apr 02 16:43:32 ~lart non executable plugins Apr 02 16:43:32 * apt pours gasoline all over non executable plugins, ignites the fire, and then enjoys some toasty marshmallows with the glorious blaze Apr 02 16:43:52 ~lart panels that don't allow runtime loading plugins Apr 02 16:43:53 * apt pries panels that don't allow runtime loading plugins's back open with a screwdriver and flashes a new bootldr to panels that don't allow runtime loading plugins Apr 02 16:45:39 openmoko: 03mickey * r1628 10/trunk/src/target/OM-2007/inputmethods/ (inputmethods.pro openmoko-keyboard/openmoko-keyboard.pro): add qmake build support for openmoko-keyboard Apr 02 16:46:18 hmm, koen seems to be full of rage :-) Apr 02 16:46:37 I do not remember when last time I had mb-panel2 running on neo Apr 02 16:47:26 .dev should have it sorted out right now Apr 02 16:48:48 koen: I just upgraded Apr 02 16:51:28 (script) wiki RecentChanges: [[User:Phlash]] [[Buying_Interest_List]] [[Introduction/ee]] Apr 02 17:08:28 counter Apr 02 17:08:28 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 09:25:46 (5.393 +-5.4 days) (723,141) Apr 02 17:11:19 good evening all Apr 02 17:12:32 I get this error when I run make openmoko-devel-image http://paste.ubuntu-nl.org/13542/ Apr 02 17:13:44 tuxmaniac: yes? check line 11? Apr 02 17:20:04 damn, that was a hard few days. Apr 02 17:20:16 * noidd hits the mailing list to see what he's missed the last 7 days Apr 02 17:26:04 The neo doesn't have hardware problems, they are just waiting for OSM to be usable in taipei! Apr 02 17:28:05 zecke, whats sha256sum coreutils contains only sha1sum Apr 02 17:28:30 coreutils 6.0 has sha256sum Apr 02 17:29:57 koen i have 5.93-5ubuntu4 which doesnt Apr 02 17:30:09 and? Apr 02 17:30:14 I said 6.0 does have it Apr 02 17:30:28 mplayer 4.5 also doesn't have sha256sum Apr 02 17:31:22 My edgy machine also does not have sha256sum Apr 02 17:31:39 koen *sigh* then note it somewhere that anybody can recompile their distro and do not just say '$revision which most people do not use has it' Apr 02 17:31:52 counter Apr 02 17:31:53 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 09:14:03 (5.385 +-5.4 days) (724,141) Apr 02 17:32:05 roh: coreutils-6.0-native is in OE... Apr 02 17:32:17 roh: engage brain before commenting Apr 02 17:32:58 koen then the makefile is broken for not building it? Apr 02 17:33:06 makefile? Apr 02 17:33:15 openmoko: 03laforge * r1629 10/trunk/src/target/kernel/patches/gta01-pcf50606.patch: Apr 02 17:33:15 openmoko: * resolve mutex deadlock in suspend path Apr 02 17:33:15 openmoko: * disable interrupt until we've been scheduled to serve last one Apr 02 17:33:15 openmoko: * make sure we cannot disable IOREG, since that powers PCF50606 I2C interface Apr 02 17:33:15 what's a makefile? Apr 02 17:33:21 * koen doesn't use any makefile Apr 02 17:33:45 which one... just guess.. youre in #openmoko Apr 02 17:33:46 its a perl thing /nod Apr 02 17:33:47 openmoko: 03laforge * r1630 10/trunk/src/target/kernel/patches/gta01-core.patch: * Ask PCF50606 to not disable CORE_1V8. The PMU will do this as final step in hardware Apr 02 17:33:52 Makefile.PL Apr 02 17:34:16 roh: every channel has a makefile? Apr 02 17:37:35 koen: your comments are neither helping nor compliant with any documentation which we have out there. we HAVE the mokomakefile and atm it seems broken. atleast thats my view of it. (reproducable here) Apr 02 17:38:37 koen, should nt the makefile download and get the sha256sum instaleed? Apr 02 17:38:42 I wish people would use their common sense instead of starting each sentence with "the makefile is and I have " Apr 02 17:44:23 hello :) Apr 02 17:44:43 counter Apr 02 17:44:43 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 09:07:38 (5.380 +-5.4 days) (725,141) Apr 02 17:45:33 koen: but the Makefile says I don't think Apr 02 17:46:04 zecke: that'd be it Apr 02 17:46:29 grr.... u-boot-gta01v3-r4_6db7d0af2336c126e4d4b2f248cc23516bdd46a8_0_1621.bin seems not to work Apr 02 17:47:05 guess there is no way to recover without the debug board? Apr 02 17:47:18 * koen hands philippe u-boot-gta01bv3-r3_0_0_1364.bin Apr 02 17:48:22 how's the software developement going? Apr 02 17:48:42 is openmoko coming together yet? Apr 02 17:48:43 openmoko: 03mickey * r1631 10/trunk/src/target/OM-2007/ (8 files in 8 dirs): update .pro file Apr 02 17:49:08 philippe: PLEASE NOT YOU! Apr 02 17:49:15 zecke since i get the same error with and without the makefile... ;) Apr 02 17:49:27 philippe: gta01v3 is not Bv3 :/ Apr 02 17:49:41 philippe: you now got a fancy doorstopper until you find someone w/ a debug board Apr 02 17:49:45 zecke and cannot build coreutil-6.0-native with bb erroring before Apr 02 17:51:00 mickeyl: OH FUCK! Apr 02 17:51:32 ya Apr 02 17:52:01 mickeyl: guess I'll see if I can get the Lauterbach at work fired up Apr 02 17:52:18 ok Apr 02 17:52:22 roh: I just checked in a mini sha256sum which is trustworthy to sum degree Apr 02 17:52:34 roh: RP and me will check that it works with bitbake1.6 and 1.8 Apr 02 17:52:38 philippe: if not, we can get you a debug board. in Braunschweig the guys have one spare Apr 02 17:52:44 philippe: and when you get it working again, flash a angstrom-x11-image to test webcore :) Apr 02 17:54:53 counter Apr 02 17:54:53 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 09:02:33 (5.377 +-5.4 days) (726,141) Apr 02 17:55:32 :) Apr 02 17:56:09 mickeyl: seems I will have some trouble connecting the JTAG... Apr 02 17:56:46 if you have a debug board, it's fairly easy Apr 02 17:57:26 meh, still don't think I'd want openmoko on a FIC Neo1973... Apr 02 17:57:51 still looking for a potential openmoko-device... Apr 02 17:58:11 think the Motorola E690 would work? ROKR E6? anyone? Apr 02 17:58:32 mickeyl: if you can get me one that would be great. Apr 02 17:59:04 in the meantime I om decoding the OOCD link schematics Apr 02 17:59:50 mickeyl: hi, have you heard about the neo1973 emulation under qemu stuff? Apr 02 18:00:26 mickeyl: we would really like to make the code public and i was told to ask you for some place to post it osmewhere in openmoko's repos Apr 02 18:01:33 balrog-kun: /src/host/qemu-neo1973 Apr 02 18:01:47 balrog-kun: please mail LaF0rge for an account Apr 02 18:01:48 bbiab Apr 02 18:02:06 mickeyl: ok, thanks for the suggestion Apr 02 18:02:17 i'm gonna mail him right away Apr 02 18:03:31 mickeyl: btw should I open a bug for the fact that the GPS antenna cable could get damaged during assembly? Mine has two nice marks on it from the cover... Apr 02 18:04:53 havent kept up on the progress of this phone much lately. Is it for sale yet? Apr 02 18:05:03 does anyone have info about what is needed for a device to run moko? like the ROKR E6 Apr 02 18:05:42 mjm counter Apr 02 18:05:58 LuitvD: touchscreen and a recent linux kernel Apr 02 18:06:05 koen *g* Apr 02 18:06:16 mickeyl: poor boy. Do you know if projects.om gets fixed anytime soon? Apr 02 18:06:38 http://www.core77.com/blog/object_culture/cmmn_the_worlds_first_opensource_car__5847.asp Apr 02 18:06:38 woglinde: counter? Apr 02 18:06:52 counter Apr 02 18:06:53 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 08:56:33 (5.373 +-5.4 days) (727,141) Apr 02 18:06:56 koen: so the ROKR E6 should somewhat be a guaranteed moko-device? it's QVGA touchscreen and runs linux from the box Apr 02 18:07:20 woglinde: im not pickin up wht your throwin down Apr 02 18:07:45 mjm type counter in your irc-client Apr 02 18:07:48 mjm: look at what aloril did after woglinde said counter Apr 02 18:07:52 counter Apr 02 18:07:53 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 08:56:03 (5.372 +-5.4 days) (728,141) Apr 02 18:07:54 than you get the status for p1 phase Apr 02 18:08:12 LuitvD: I said 'recent linux kernel' Apr 02 18:08:13 or type /msg aloril counter? to get more details Apr 02 18:08:18 LuitvD: that means 2.6.19 or newer Apr 02 18:08:25 LuitvD: not 2.4.1.montevista Apr 02 18:08:33 koen: hmm, meh, have to find out the kernel on the ROKR :P Apr 02 18:08:54 1.2 !! Apr 02 18:08:55 motorola only ships 2.4 kernels Apr 02 18:08:59 :( Apr 02 18:09:11 sucky montevista Apr 02 18:09:15 MVMD! Apr 02 18:09:41 wtf Apr 02 18:09:43 but it's possible to get a 2.6.19 running, with the right attempts ... right? Apr 02 18:10:02 with the right hacking... Apr 02 18:10:05 LuitvD: ask the people on #openezx Apr 02 18:10:38 counter Apr 02 18:10:38 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 08:54:40 (5.371 +-5.4 days) (729,142) Apr 02 18:10:50 woglinde: ty Apr 02 18:11:10 koen: oeh, thanks for mentioning openezx :D that's what I was looking for Apr 02 18:11:59 LuitvD: http://poh9.blogspot.com/2007/03/openmoko-dialer-ngstrm-magician.html Apr 02 18:13:32 zecke: i'll repush the ones responsible :/ Apr 02 18:13:40 koen: this was the info I was looking for :) http://wiki.openezx.org/Rokr_E6 Apr 02 18:13:53 mickeyl: rate my GSoC proposal :) Apr 02 18:16:59 d/topic Apr 02 18:17:05 hmm Apr 02 18:21:53 LuitvD: you buying an E6? Apr 02 18:25:14 buz: I think I actually will yes Apr 02 18:25:29 but no wifi and qvga? that's sooo 2005 Apr 02 18:25:30 buz: if e28 ain't responding, I'll go for moto Apr 02 18:25:36 heheheh Apr 02 18:25:50 bluetooth and 2MP cam is good enaugh Apr 02 18:26:09 GPRS + 1GB data bundle :) Apr 02 18:26:14 don't need wifi for that Apr 02 18:26:45 i wont support moto Apr 02 18:27:00 all phones i've seen from them were truly crappy quality Apr 02 18:29:03 which ones have you tried? Apr 02 18:29:36 rarzs, v3i, v360 Apr 02 18:29:44 no candybar though Apr 02 18:29:54 Anyone comment on the UK 'Orange World' service? Apr 02 18:31:49 if its anything like the swiss one consider it useless Apr 02 18:32:19 It's unclear if it's a walled garden or not. Apr 02 18:32:25 At least from the webpage. Apr 02 18:32:26 if it's anything like the dutch one consider it useless too Apr 02 18:32:37 its a walled garden Apr 02 18:32:43 supposedly it can send email Apr 02 18:32:47 but thats it Apr 02 18:33:00 I was worried by 'It's like the internet in your pocket' being used often. Apr 02 18:33:10 rather than 'it is' Apr 02 18:33:22 Sigh. Apr 02 18:33:27 in switzerland the page explicitely states different traffic rates for inet and orange word Apr 02 18:33:29 world Apr 02 18:34:03 What I'm annoyed about is that at the moment I have credit rating problems. Apr 02 18:34:32 So possibly may not be able to get a pay-monthly contract, for reasonable priced data. Apr 02 18:34:54 Even though I could actually pay upfront for 12 or 18 months. Apr 02 18:35:29 some of my usually near bankrupt friends always managed to get postpaid contracts Apr 02 18:36:29 around here they usually actively push away from prepaid Apr 02 18:47:28 (script) wiki RecentChanges: [[Wish_List]] Apr 02 18:53:39 openmoko: 03mickey * r1632 10/trunk/src/target/OM-2007/applications/openmoko-calculator/ (18 files in 4 dirs): initial import of openmoko-calculator 0.0.3 courtesy Rodolphe Ortalo Apr 02 18:53:58 openmoko: 03mickey * r1633 10/trunk/src/target/OM-2007/applications/applications.pro: add openmoko-calculator to applications.pro Apr 02 19:04:01 openmoko: 03mickey * r1634 10/trunk/oe/packages/ (2 files in 2 dirs): oe: add recipe for openmoko-calculator Apr 02 19:05:41 openmoko: 03mickey * r1635 10/trunk/oe/packages/ipkg/ (5 files in 2 dirs): oe: reintroduce our ipkg overlay to make the updated patch sitting in files/ effective Apr 02 19:07:33 openmoko: 03mickey * r1636 10/trunk/oe/packages/openmoko-inputmethods/openmoko-keyboard_svn.bb: oe: bump openmoko-keyboard recipe to make overlay active Apr 02 19:28:06 are there sources for libgsmd or some development headers somewhere? Apr 02 19:30:04 philippe: either bitbake libgsmd or look in openmoko svn src/target/gsm Apr 02 19:30:14 ok. just found Apr 02 19:30:18 thanks pH5 :-) Apr 02 19:31:23 * philippe works on deb packages and an automated script to set up an openmoko native development kit Apr 02 19:44:28 (script) wiki RecentChanges: [[Wish_List]] [[UI_Improvements]] [[Applications]] Apr 02 20:05:21 openmoko: 03laforge * r1637 10/trunk/src/target/kernel/patches/ (gta01-core.patch s3c_mci_platform.patch): extend the per-platform set_power() function of s3mci to also be able to send 'power off' events to the per-platform callback function. Apr 02 20:19:25 counter Apr 02 20:19:25 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 07:50:17 (5.327 +-5.3 days) (730,143) Apr 02 20:19:29 yay :) Apr 02 20:21:03 I just hope they find whatever bug there is in the hardware or software Apr 02 20:21:25 likewise I hope it's a bug in the software Apr 02 20:21:54 3 days later and harald pulls a mosquito out of the assembly machine Apr 02 20:22:00 * LarstiQ would really like a phone now, even an imperfect one Apr 02 20:22:01 But, personally, I don't care if it's a month late.. I'll be on vacation almost all of April, then moving to a new apartment May 1:st ;) Apr 02 20:22:03 koen: heh :) Apr 02 20:22:14 * koen is happy with his p0 phone Apr 02 20:22:19 :-P Apr 02 20:22:24 np: Queen - I want it all. Apr 02 20:22:40 koen: I suppose that means you're not willing to transfer it to me? ;) Apr 02 20:22:59 hehe Apr 02 20:24:23 I care if it's a month late. Apr 02 20:24:26 Personally - P0 wouldn't be very interesting to me. P0, with GPS driver, maybe. Apr 02 20:24:41 My student's SoC project likely can't complete if it's a month late Apr 02 20:24:47 :( Apr 02 20:25:10 I mean, I guess we could do reference implementation work, but that's kinda lame :\ Apr 02 20:25:18 dho: I don't think it will be that late even if there's some hardware bug Apr 02 20:25:35 Worst case, it could be a reroll. Apr 02 20:25:48 dho: What SoC project? Apr 02 20:25:51 If for example it's a misroute under a BGA, and can't be wired round. Apr 02 20:25:56 Porting Plan 9 to the Neo1973 Apr 02 20:26:09 SpeedEvil: (I don't think he means System on a Chip) ;) Apr 02 20:26:29 dho: Cool :) Apr 02 20:26:44 CM: I was talking of your comment. Apr 02 20:27:07 SpeedEvil: I know, i refered to last time SoC was mentione here :) Apr 02 20:27:12 Ah. Apr 02 20:27:25 * CM hides in shame after a bad joke Apr 02 20:42:16 openmoko: 03andrew * r1638 10/trunk/src/host/qemu-neo1973/: Neo1973 emulator (Initial import). Apr 02 20:42:28 (script) wiki RecentChanges: [[Wish_List]] [[Wishlist:PilotDB]] [[Wishlist:PersonalWiki]] [[Wishlist:EggTimer]] Apr 02 20:46:38 er Apr 02 20:46:43 ok Apr 02 20:47:05 Including devices? Apr 02 20:50:39 openmoko: 03andrew * r1639 10/trunk/src/host/qemu-neo1973/ (101 files): Populate qemu-neo1973 main directory. Apr 02 20:50:55 er Apr 02 20:51:10 But I'm not seeing that directory? :S Apr 02 20:52:05 Oh, I had some screwed up svn settings Apr 02 20:52:11 dho: https://svn.openmoko.org/trunk/src/host/qemu-neo1973/ Apr 02 20:52:19 hmm, it seems my svn is also screwed Apr 02 20:52:33 it won't add subdirectories for some reason Apr 02 20:55:31 I just blew it away and tried again Apr 02 20:55:47 jesus Apr 02 20:56:26 Which code is the neo code? Apr 02 20:57:28 ? Apr 02 20:57:56 I'm assuming the point of this is to have neo1973 drivers in there :) Was wondering if any of that was already there. Apr 02 20:58:50 SpeedEvil: how long would such a reroll take? Apr 02 20:59:04 Well... Apr 02 20:59:36 Physically if you order a large quantity of boards, and get them assembled, you can probably get them out the door in a couple of weeks absolute minimum. Apr 02 20:59:55 dho: still committing, i made a stupid error and the subdirectories didn't get in yet Apr 02 20:59:56 However, that would be with absolutely no testing other than 'it boots'. Apr 02 21:00:03 balrog-kun: ah ok Apr 02 21:00:15 i'm a noob :( Apr 02 21:00:17 balrog-kun: you're andrew@ then? Apr 02 21:00:26 SpeedEvil: so a month at least then Apr 02 21:00:31 dho: yes, that's me Apr 02 21:00:31 It depends. Apr 02 21:00:57 If the problem with this version can be fixed in software... Apr 02 21:00:59 balrog-kun: good to know :) Apr 02 21:01:16 then they may feel confident enough to order a large batch of V4 Apr 02 21:01:23 this upload will take a while on my cable Apr 02 21:01:42 balrog-kun: So is this inclusive of neo device emulators? Apr 02 21:02:10 there are halfway houses, for example, you can order 500 PCBs, 500 sets of components, and then only stick components on 50 of them, until you've tested them. Apr 02 21:02:10 SpeedEvil: oh sure, I meant in case of a reroll as you described Apr 02 21:02:18 openmoko: 03andrew * r1640 10/trunk/src/host/qemu-neo1973/ (405 files in 24 dirs): Apr 02 21:02:18 openmoko: Import all of the hardware emulation. Apr 02 21:02:18 openmoko: The qemu tree should be buildable and usable from this point. Apr 02 21:02:19 * LarstiQ is really hoping it's a software issue Apr 02 21:02:27 dho: yes, although there is no BT (yet), no GSM and no GPS (for obvious reasons) Apr 02 21:02:54 That cuts costs as you then only have to eat the costs of 50 sets of hardware and 500 PCBs, rather than 500 of both, but if it works, you can make 500 without waiting for the PCB. Apr 02 21:02:59 balrog-kun: Having a serial device that says ``you poked bit abc'' for implementing the other end of the GSM and GPS serial spec should be ok for that though. Apr 02 21:03:06 SpeedEvil: right Apr 02 21:03:14 balrog-kun: What's BT? And is there anything I can do to help out with this? This is extremely useful for me. Apr 02 21:03:22 BlueTooth Apr 02 21:03:25 Ok Apr 02 21:03:48 For GPS emulators, you don't need it. Apr 02 21:03:52 dho: well, you can see what's being output on the serials by the OS and you can type from keyboard a fake response from the "hardware", or you can have a serial console Apr 02 21:03:54 gpsd already has a 'fake' mode. Apr 02 21:04:00 dho: bluetooth Apr 02 21:04:07 Right Apr 02 21:04:27 And the gps on the neo doesn't output serial. Apr 02 21:04:35 Well - not 'normal' serial. Apr 02 21:04:44 hmm right Apr 02 21:04:56 Explain? Apr 02 21:05:04 It talks to a binary plugin which talks to gpsd Apr 02 21:05:11 Oh. Apr 02 21:05:13 the rest of the software talks to gpsd Apr 02 21:05:20 Reversing that should be simple. Apr 02 21:05:23 gpsd is a standard bit of software. Apr 02 21:05:24 dho: i think everyone is welcome to help adding more hardware, and otherwise i think there's not much to do, i'd be happy to here if it works for anyone, and perhaps a good howto on the wiki would be good to have as qemu has *tons* of options Apr 02 21:05:32 *hear Apr 02 21:05:40 balrog-kun: Well I'll try to get it building on FreeBSD Apr 02 21:05:48 balrog-kun: This will be amazingly useful for the Plan 9 port. Apr 02 21:06:29 dho: it's not guaranteed to build ok under FreeBSD, i think the ports version has some custom patches Apr 02 21:06:34 user-mode-plan9? Apr 02 21:06:43 SpeedEvil: no Apr 02 21:06:46 balrog-kun: It does Apr 02 21:06:55 SpeedEvil: native Apr 02 21:09:44 balrog-kun: What version of qemu is this? Apr 02 21:11:55 dho: some three weeks ago snapshot, i'm waiting for Paul Brook to merge some changes he promissed to merge before i update Apr 02 21:12:22 Ok. Apr 02 21:13:04 I'm going to try to build this now. Apr 02 21:13:21 thanks Apr 02 21:17:40 yay it's going to be a pita! Apr 02 21:20:21 last time i tried building undef FreeBSD, not using ports, it was just a couple of errors Apr 02 21:20:32 it generally worked Apr 02 21:30:03 balrog-kun: So just running ``make'' in here should build neo1973 support? Apr 02 21:30:55 dho: provided that arm-softmmu (or was it arm-system) is in the list of targets chosen by ./configure, yes Apr 02 21:36:09 here we go.... Apr 02 21:39:29 cc -o qemu-system-arm... Apr 02 21:39:33 gmake[1]: Leaving directory `/usr/ports/emulators/qemu-1973/work/qemu-neo1973/arm-softmmu' Apr 02 21:39:36 :) Apr 02 21:41:49 balrog-kun: Is there a firmware image to use with this? Apr 02 21:42:22 (navigating around the snapshots directory, I'm not sure where to find what I'm looking for) Apr 02 21:42:58 dho: well, there is an easy way to get OpenMoko up and running, and i'm writing the HOWTO on the wiki at this moment about how it works Apr 02 21:43:18 also http://people.gta01.hmw-consulting.de/laforge/tmp/rootfs/rootfs.jffs2 is a 404 Apr 02 21:43:43 balrog-kun: Ok, I'll wait for that. Apr 02 21:43:54 dho: basically, in the qemu directory, where you did "make", you run "openmoko/download.sh" and then "openmoko/flash.sh", this is the easiest way as far as i can tell Apr 02 21:44:05 it's very much experimental but it should work Apr 02 21:44:16 and if it doesn't work i will be also happy to hear about it Apr 02 21:49:19 * rwhitby notes 1906 unique IP downloads of mokomakefile Apr 02 21:50:43 BAH Apr 02 21:50:45 it's bash :( Apr 02 21:51:14 oh, right Apr 02 21:51:18 i don't have sh installed Apr 02 21:51:50 dho: do you know which bits are the bash specific ones? Apr 02 21:52:04 anything [[ Apr 02 21:52:54 Why can't that just be if [ ] Apr 02 21:52:56 also the code of these scripts is extremely ugly (unlike the emulator code) Apr 02 21:53:08 it can Apr 02 21:53:53 i consider these scripts a temporary solution basically, but it would be good to get them fixed Apr 02 21:53:59 ok Apr 02 21:55:01 ./splashimg.pl System_boot.png > splash Apr 02 21:55:01 480 x 640 (255) Apr 02 21:55:01 gzip -c < > splash.gz Apr 02 21:55:01 Syntax error: redirection unexpected Apr 02 21:55:01 *** Error code 2 Apr 02 21:55:18 oh, GNU make Apr 02 21:55:31 the scripts should probably use "gmake" Apr 02 21:55:49 Ah I see Apr 02 21:55:57 heh, any build system these days assumes gnu make :-) Apr 02 21:56:04 not ours :( Apr 02 21:56:25 stat: illegal option -- c Apr 02 21:56:30 dho: please tell me it doesn't expect imake though? :) Apr 02 21:56:38 nah, bsdmake of course. Apr 02 21:56:47 Or if we're talking about Plan 9, mk Apr 02 21:56:59 also, ls: uImage-2.6-*-fic-gta01.bin: No such file or directory <-- where should I have gotten that from? Apr 02 21:57:21 dho: download.sh should have downloaded it for you Apr 02 21:57:35 there's an url in "openmoko/env" to some images Apr 02 21:57:41 Oh Apr 02 21:57:51 No, it downloads a bunch of index.htmls Apr 02 21:57:52 ah Apr 02 21:57:57 lynx: invalid option: -listonly Apr 02 21:57:57 ouch Apr 02 21:59:13 How does AAC work on neo? Apr 02 21:59:50 Lynx Version 2.8.5rel.4 (25 Oct 2005) -> I think that should be -hiddenlinks=listonly for that version Apr 02 21:59:54 hmm, does the itms work without itunes? Apr 02 21:59:55 now it's downloading stuff. Apr 02 22:00:56 what's stat -c supposed to do Apr 02 22:01:11 Itunes have recently announced you can now buy (some) stuff DRM free. Apr 02 22:01:31 SpeedEvil: yay, P60 time :) Apr 02 22:01:46 SpeedEvil i think nobody tested... i'll see if i get around to compile some command line player soon, to try out Apr 02 22:02:14 dho: output the filesize iirc Apr 02 22:02:32 SpeedEvil: aye, that is how I interpreted your aac question Apr 02 22:03:12 SpeedEvil: xmms2 supports aac, I haven't tested it but I think it shouldn't have a lot of trouble running on the neo Apr 02 22:04:23 SpeedEvil its only a question if its fast enough.. but it'll eat battery much faster than e.g. mp3 Apr 02 22:05:19 when speaking about battery - what is estimated call time/standby time for P1 ? Apr 02 22:07:57 well that failed miserably Apr 02 22:09:55 Wheeeeeeeeeeeeeee, love those netsplits Apr 02 22:12:26 yeah, that's not working at all, heh. Apr 02 22:15:01 openmoko: 03mickey * r1641 10/trunk/src/target/OM-2007/openmoko-libs/libmokoui/moko-menu-box.c: libmokoui: force application and filter menu to always be of a certain size. closes #130 Apr 02 22:16:05 balrog-kun: maybe you'll have a better idea what's goign on Apr 02 22:16:15 balrog-kun: http://www.databits.net/~dho/typescript.txt Apr 02 22:16:37 * dho goes home Apr 02 22:17:49 dho: almost for sure "echo" is misbehaving Apr 02 22:18:10 other than that all the errors you see except the bad block ones, are part of the process and ar ok Apr 02 22:20:03 dho: it basically means that the emulator works :) Apr 02 22:26:11 openmoko: 03mickey * r1642 10/trunk/src/target/OM-2007/applications/openmoko-calculator/po/POTFILES.in: openmoko-calculator: add missing POTFILES.in. closes #380 and #384 Apr 02 22:38:28 (script) wiki RecentChanges: [[OpenMoko_under_QEMU]] [[FAQ]] Apr 02 23:02:46 counter Apr 02 23:02:46 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 06:28:37 (5.270 +-5.3 days) (731,143) Apr 02 23:09:54 * SpeedEvil contemplates. Apr 02 23:10:09 The LCD, speakers, vibrator, case don't change at all. Apr 02 23:11:00 If the V4 bugs aren't too bad - and they have already made some hardware, it would cost little to send out the hardware now, with V4, with the V5 board to follow. Apr 02 23:11:43 As any existing V4 boards - if they are not bugfixable - will be going in the trash. (if there are lots) Apr 02 23:12:23 harald got his first v4 board today, so give it a few days to check for bugs before they fire up the factory again Apr 02 23:13:32 What are "the bugs"? Apr 02 23:13:49 Ah. Apr 02 23:13:59 the didn't state any details in the announce mail Apr 02 23:14:21 high-rez mostly gpio and irq as well as pmu changes.. small things which need to be sorted out Apr 02 23:14:48 So some may be 'oops''s - like the 64M bug. Apr 02 23:15:02 And some may be actual hardware bugs. Apr 02 23:15:43 Aha. Apr 02 23:15:47 e.g. the data and addresslines on the debug connector are gone. so the device doesn't get instable if running on 266mhz with debugboard attached Apr 02 23:16:10 besides we don't use em and need em. Apr 02 23:16:11 neat Apr 02 23:17:51 What's the leadtime between the trigger being pulled, and boxes shipping to users? Apr 02 23:18:12 Yeah, when can I put one in my lab. :D Apr 02 23:18:21 SpeedEvil dunno.. we'll see ;) Apr 02 23:18:51 I think it's safe to assume that the counter needs put back three weeksish though. Apr 02 23:18:55 :/ Apr 02 23:19:12 ouch Apr 02 23:19:15 who set it that way at all and why? Apr 02 23:19:22 Announcements. Apr 02 23:19:33 Previous announcements. Apr 02 23:20:06 ah... ok.. but we'll know more anyways in a week or so.. Apr 02 23:20:09 As long as the basic RF section works, that's what I want to test. ;) Apr 02 23:20:18 so send me a bad board. ;) Apr 02 23:20:23 I got a BS i can hook it upto now. Apr 02 23:20:53 what do you wanna do with it? Apr 02 23:21:35 roh: Play with it in the lab, test it against all of the equiptment we have. And ultimately - but not now write a custom app. Apr 02 23:21:36 the rf parts (gsm) are unchanged afaik. Apr 02 23:21:45 * balrog-kun would be happy with a bad board too Apr 02 23:22:04 high-rez: you have gsm-test equipment around? Apr 02 23:22:09 roh: yeah Apr 02 23:22:21 i rarely use a mobile phone Apr 02 23:22:25 nice. i'd be glad to get your results Apr 02 23:23:05 ericsson bs's, nokia bs's, nortel bs's, a whole array of bsc's msc's etc etc etc. A bunch of HSDPA/HSUPA gear to - not that it would help with this device. ;) Apr 02 23:28:44 counter Apr 02 23:28:44 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 06:15:38 (5.261 +-5.3 days) (732,143) Apr 02 23:29:09 high-rez: you never know, maybe you'll find the secret HSDPA jumper! :) Apr 02 23:29:10 * don-o wishes planet.openmoko.org had more content Apr 02 23:29:22 * wookieface agrees Apr 02 23:29:30 surely there is interesting stuff going on 5 days from 'launch' Apr 02 23:30:42 yes:p Apr 02 23:31:17 Alas, it looks like >5 days. Apr 02 23:31:47 Probably at least 14, more likely 20-30. Apr 02 23:31:59 Optimistically. Apr 02 23:32:09 SpeedEvil: oh yeah? could you get someone to blog about that? or are you one of the people on the planet? Apr 02 23:32:42 does the on-screen keyboard have a dvorak layout? Apr 02 23:33:27 roh earlier said that there is one board, and it'll need approval before the factory makes more, so that's at least a couple of weeks delay, even if the current problems turn out to be software fixable. Apr 02 23:34:13 Dvorak'd probably be easy to add if it doesn't have it. Apr 02 23:34:28 (script) wiki RecentChanges: [[FAQ]] [[Neo1973]] Apr 02 23:35:07 SpeedEvil doesn't need to be. if the testing is done and no other changes are made the production can happen __very__ fast. i dunno about shipping Apr 02 23:35:42 don't forget that we are moving extemely fast right now. i've never seen this before... developing in hyperspace Apr 02 23:36:08 yeah - if the issues get resolved in the next couple of days, and there is no need for a hardware respin, it _could_ happen in a bit over a week. Apr 02 23:36:58 Assuming all the components are sitting on the shelf somewhere and it's only PCBs that are being waited on. Apr 02 23:37:21 SpeedEvil the parts are already there, yep Apr 02 23:37:47 so no panic. you will get your playthings soon enough Apr 02 23:38:11 Assuming no hardware respin - which would delay a little, for another testing cycle. Apr 02 23:39:00 ATM I'm just doing paper development, which isn't really constrained by lack of hardware. Apr 02 23:41:43 I wanna touch the thing. Apr 02 23:41:45 :) Apr 02 23:42:12 high-rez: indeed you want. It has a special feel to it =) Apr 02 23:43:00 I'm surrounded by devices, but this one has the potential to not suck (read: not symbian or RIM). Apr 02 23:50:47 <_ASK> http://standardos.wikidot.com/introduction Apr 02 23:51:59 roh: can you put some of that juicy info into planet.openmoko? Apr 02 23:52:11 inquiring minds want to know Apr 02 23:55:22 don-o planet is just an aggregator so someone needs to blog Apr 02 23:56:32 roh: i assumed from your authoritatve tone you were a core dev and probably were already one of the aggregated blogs. :) Apr 02 23:59:30 don-o i work for openmoko but i do not blog (and hopefully will not be forced to ;) Apr 03 00:01:00 roh: i see. well there are teaming hordes scrounging for any info you can throw at them. Apr 03 00:01:02 I've put some into [[neo1973]] and [[faq]] entries in the wiki. Apr 03 00:01:09 or at least thats how it feels to me :) Apr 03 00:06:01 SpeedEvil where do you take the 14th from? Apr 03 00:11:17 does anyone know specifically what the problem is they are having with the latest hardware(or software as the case may be) revision? Apr 03 00:13:34 peepsalot no new problems. its just about testing before producing Apr 03 00:15:45 roh: ok. from reading the community update from Sean Moss-Pultz, it sounded like there was some new issues. the exact phrase was "stability issues" Apr 03 00:16:04 i was just curious what sort of issues those might be Apr 03 00:16:25 is this the same FIC http://www.fic.com.tw/, and does that page load for anyone? Apr 03 00:16:51 does not load for me don-o Apr 03 00:18:22 and yes it would appear that should be the correct address Apr 03 00:24:30 peepsalot i dunno exactly what he means. Apr 03 00:25:58 are you part of the core team roh? if so may I ask what role you play? Apr 03 00:26:50 peepsalot i will take care of supporting the devs Apr 03 00:29:53 roh: the devs meaning other paid openmoko developers? Apr 03 00:30:25 don-o yes. as well as unpaid so far possible Apr 03 00:32:22 anyone tried to put their neo1973 in their pocket? Apr 03 00:32:28 (script) wiki RecentChanges: [[Neo1973]] [[Wish_List_-_Hardware]] Apr 03 00:32:58 aevin: yeah. Apr 03 00:33:19 it comes w/ a protective case/bag Apr 03 00:33:44 CoreDump|home: does it "work" or is it highly uncomfy? Apr 03 00:33:54 s/highly/mostly Apr 03 00:34:02 s/highly/mostly/ Apr 03 00:34:02 aevin meant: s/mostly/mostly Apr 03 00:34:12 works for me I guess. Apr 03 00:34:30 Not that I'm carry it around a lot ATM ;) Apr 03 00:36:21 i see. (but that will hopefully change) Apr 03 00:37:03 with P1, yes I hope so =) Apr 03 00:44:22 hmm.. would it be possible to make mailman's archives at lists.openmoko.org show a timestamp for each individual post (while sorting by thread), or should I rather shut up and just use news.gmane.org or similar? Apr 03 00:45:21 or perhaps easier: change the CSS style sheet to make visited and non-visited hyperlinks have different colors? Apr 03 00:53:07 openmoko: 03zhiyong_sun * r1643 10/trunk/src/target/OM-2007/panel-plugins/openmoko-panel-gsm/src/moko-gsm-conn.c: openmoko-panel-gsm: delete gprs signal check testing codes Apr 03 00:53:19 +++http://lists.openmoko.org/style.css \n +a:visited { color: #FF9933; // yes, I am color blind ;-) } , or something. Apr 03 01:23:57 openmoko: 03zhiyong_sun * r1644 10/trunk/src/target/OM-2007/panel-plugins/openmoko-panel-gsm/src/ (moko-gsm-conn.h openmoko-panel-gsm.c): openmoko-panel-gsm: bug fix, initialize gsm and gprs signal with UN_INIT status, instead of UN_CONN avoid to UN_CONN status image show error. Apr 03 02:57:29 counter Apr 03 02:57:29 http://wiki.openmoko.org/wiki/Neo1973 P1: 5 days 04:31:15 (5.188 +-5.2 days) (733,143) **** ENDING LOGGING AT Tue Apr 03 02:59:56 2007