**** BEGIN LOGGING AT Tue Feb 06 02:59:57 2007 Feb 06 03:25:38 nick count records: 106 2006-12-04; 108 2006-12-07; 109 2006-12-12; 114 2006-12-13; 116 2006-12-20; 134 2007-01-11; 141 2007-01-12; 143 2007-01-13; 157 2007-01-15; 163 2007-01-16 ; 182 2007-01-17; 185 2007-01-19 ; 197 2007-01-20 ; 201 2007-01-21; 205 2007-01-22; 206 2007-01-23; 211 2007-01-25; 215 2007-01-26; 219 2007-02-01; 228 2007-02-05 Feb 06 03:38:29 cool Feb 06 03:38:33 still growing Feb 06 03:38:46 the latest beryl rc1 is sweet on my dull dell Feb 06 03:39:07 with only 8mb video ram intel chipset it plays videos around the cube while rotating Feb 06 03:39:09 i tried beryl back in the day.. decided it just wasted my time Feb 06 03:39:11 that is sweet Feb 06 03:39:17 Agrajag-, its not Feb 06 03:39:27 it's not what? Feb 06 03:39:30 the zoom feature along for people with any eye problems ... Feb 06 03:39:34 that is so nice Feb 06 03:39:49 just press "super" key (aka windows key) and scroll mouse Feb 06 03:40:08 to zoom.. it stays at the mode until you click out with same sequence Feb 06 03:40:29 i find that helps my wife who has issues with reading the screen Feb 06 03:40:43 i dont have eye problems.. when i switch desktops or alt-tab i want things to happen quickly - i found that it just wasted my time by not instantaneously changing Feb 06 03:40:44 and it does not dork up font layouts like in web pages... Feb 06 03:40:53 oh i still have that Feb 06 03:41:10 and ... new beryl has on/off button in tray Feb 06 03:41:25 choosing kde/metacity deco also... Feb 06 03:41:28 if so desired Feb 06 03:41:39 i really want one of these phones though Feb 06 03:41:42 yeah but i didn't find any of the features were actually useful Feb 06 03:41:55 just need to decide on how i can help develop Feb 06 03:42:12 they are great for making windows folks jealous Feb 06 03:42:19 even vista can't compare Feb 06 03:42:20 I had few and little bugs with beryl, but those were enough for me to uninstall it. I didn't use the effects anyway, it's just annoying. Feb 06 03:42:30 its so nice now Feb 06 03:42:45 When I want to open a window, I want that window to be opened. I don't want to see some stupid animation before it opens. Feb 06 03:42:57 much improved ... my intel card plays stuff it would never before i had beryl Feb 06 03:43:05 i agree Feb 06 03:43:18 i tweak the settings to get fast open/close Feb 06 03:43:22 etc... Feb 06 03:44:08 [A Feb 06 05:30:01 counter Feb 06 05:30:01 5 days 23:57:47 (5.998 days) for source for *all* developers and devices for selected developers (2007-02-11); a month, 5 days (33.998 days) for anybody (targeting developers) (preorders: not yet) (2007-03-11); 7 months, 5 days (217.998 days) for mass market (2007-09-11): see topic for more info Feb 06 07:08:54 morning Feb 06 07:09:05 aloril: ahhgr.. Feb 06 07:13:00 hrw: ? Feb 06 07:14:55 aloril: its morning and you remind me that I have to wait 1.5-2 weeks... Feb 06 07:28:39 counter Feb 06 07:28:39 5 days 21:59:08 (5.916 days) for source for *all* developers and devices for selected developers (2007-02-11); a month, 5 days (33.916 days) for anybody (targeting developers) (preorders: not yet) (2007-03-11); 7 months, 5 days (217.916 days) for mass market (2007-09-11): see topic for more info Feb 06 07:32:30 someone here use ubuntu? Feb 06 07:33:31 yes Feb 06 07:36:55 show me /etc/apt/sources.list Feb 06 07:40:40 hrw: about 6 days + shipping (in 6 days it should be middle of Monday at Taiwain, so... there should be no extra time except actual shipping) Feb 06 07:43:28 g'day all Feb 06 07:44:52 http://www.fluff.org/ben/unbuntu.aptsrc Feb 06 07:46:16 thx Feb 06 07:48:47 fluffs: 404 ;( Feb 06 07:50:07 http://www.fluff.org/ben/ubuntu.aptsrc Feb 06 07:50:30 thx Feb 06 08:09:23 * fluffs gets out the fishing rod, and dangles into samsung's r+d to see what he can hook Feb 06 08:22:35 morning Feb 06 09:44:27 is there any binary version of XMPP? seems kinda inefficient to send XML instant messages over GPRS Feb 06 09:45:12 use a ssh tunnel to the jabber server :) Feb 06 09:46:39 no need, jabber can do ssl itself Feb 06 09:46:42 mmh, yeah, the xml stuff is some overhead... wonder if it supports negotiating compression Feb 06 09:46:48 but even compressed XMPP is overhead Feb 06 09:46:53 buz: ssh -C, not ssl Feb 06 09:47:04 i think openssl can negotiate compression no? Feb 06 09:47:21 hmm, maybe, not sure Feb 06 09:47:42 Jabber at least does seem to have a draft spec for negotiating it, if openssl/gnutls/friends don't Feb 06 09:48:09 rfc3749 does seem to spesify TLS compression options Feb 06 09:48:30 yeah TLS has compression Feb 06 09:48:53 but still, a simple binary protocolwould work with less traffic... Feb 06 09:49:22 well, probably Feb 06 09:49:52 i mean what do you really need once a connection is established? a way to get the contact list (perhaps) and one to send/receive messages Feb 06 09:50:08 some sort of UID:message would be plenty for most uses i guess Feb 06 09:50:14 apparently openssl and gnutls have implemented deflate as per rfc 3749: http://www.ietf.org/IESG/Implementations/rfc-3749-implementations.txt Feb 06 09:51:13 can't be bothered to read the detailed report now though Feb 06 09:51:15 meeting soon Feb 06 09:52:21 http://www.agiledelta.com/w3c_binary_xml_proposal.html is that supported by anything? Feb 06 10:01:57 big step is XML -> compression, XML -> binary might be even less efficient, though most efficient is probably XML -> binary -> compression Feb 06 10:02:06 mornin' Feb 06 10:11:38 using a slimmer protocol would probably be biggest step Feb 06 10:12:07 * XorA points at ASN.1 Feb 06 10:12:18 stupd XML numpties Feb 06 10:12:21 mmh, theoretically a good compressor should be about the same with or without binary xml format, but of course in practice there may be differences Feb 06 10:13:25 ~prod lrg Feb 06 10:13:26 * apt zaps lrg with a high voltage cattle prod Feb 06 10:23:56 There is a draft standard about stream compression outside TLS in XMPP Feb 06 10:24:01 http://www.xmpp.org/extensions/xep-0138.html Feb 06 10:24:37 yeah I noticed that Feb 06 10:25:04 but if gnutls and openssl implement tls-level compression, perhaps easier to use that Feb 06 10:26:47 Probably yes. Feb 06 10:27:34 it depends on how big your xmpp messages are Feb 06 10:28:09 I wonder what kind of protocol Agile Messenger uses Feb 06 10:28:46 iirc when I used it the bandwidth usage was really low Feb 06 10:28:59 I think that best would be compress-before-send Feb 06 10:29:54 counter Feb 06 10:29:54 5 days 18:57:52 (5.790 days) for source for *all* developers and devices for selected developers (2007-02-11); a month, 5 days (33.790 days) for anybody (targeting developers) (preorders: not yet) (2007-03-11); 7 months, 5 days (217.790 days) for mass market (2007-09-11): see topic for more info Feb 06 10:29:59 yep. and encrypt, in case the cpu overhead isn't too big Feb 06 10:30:06 * XorA thinks using a real protocol would be better Feb 06 10:31:14 * koen is wondering about what the best mix of moko/g(pe)2/poky is Feb 06 10:31:28 koen: Angstrom you phewl Feb 06 10:31:45 XorA: the mix will be used in angstrom of course :) Feb 06 10:32:07 koen: and how much of poky is in moko Feb 06 10:32:29 XorA: real protocol as in "whatever network the user is in" or something else? Feb 06 10:32:29 hrw: at least that wretched xserver-kdrive-common Feb 06 10:32:49 Installing xserver-xorg (X11R7.1-1.1.0-r0) to root... Feb 06 10:33:14 hrw: if you have ipkg available, you can put those in the feeds Feb 06 10:33:17 (or I could) Feb 06 10:33:39 * koen is still celebrating the dbus fix Feb 06 10:34:31 jannu: as in XML is fscking stupid idea for an on the wire protocol Feb 06 10:35:30 XorA: XML is a stupid idea for most things it's used in Feb 06 10:35:47 jannu: damn right there :-) Feb 06 10:36:33 And where it is used it's usually used "wrong" Feb 06 10:37:25 It's just like ldap was a few years before it Feb 06 10:37:40 and java and .net and .... Feb 06 10:38:29 People like saying "Yes, we send it as an xml-file" even if it's just a big table wrapped in ... Feb 06 10:38:55 * CM has a slight SOAP allergy Feb 06 10:39:17 that's because Sales see it as...: XML? [Yes/No] Feb 06 10:40:00 Exactly, or SOA or Ajax or what ever is in right now Feb 06 10:40:33 And saying that an application is based on J2EE, XML and SOAP pumps up the price by 200% Feb 06 10:40:59 damn, OpenMoko phones just went upto 700$ :-) Feb 06 10:41:05 * Stephmw adds those keywords to his MIDlet Feb 06 10:41:16 Hehe Feb 06 10:41:19 Even though in reality the Java part is just wrappers around 20 year-old fortarn code Feb 06 10:41:35 or cobol Feb 06 10:41:55 jannu: if it ain't broke, don't fix it.. if it's fossilised, hide it Feb 06 10:42:04 and sell it Feb 06 10:42:18 eh? that happened before the dev occured :) Feb 06 10:43:00 no, I mean hide it and get rid of the whole thing, including support Feb 06 10:43:08 especially support Feb 06 10:43:34 outsourcing is fun Feb 06 10:43:35 "Oh our competitor does XYZ? So does our Development version!" Hi random developer, it's Sales here, we've sold our stuff with XYZ... best add it Feb 06 10:45:22 hmm. there must be some really good old cobol-based protocol which we could use for IM Feb 06 10:45:56 IP over SMS Feb 06 10:46:02 can't get better Feb 06 10:46:14 SMS-anything Feb 06 10:54:20 btw, anyone familiar with the vibration hardware on the Neo (or mobiles in general)? Feb 06 10:54:29 jannu: in general Feb 06 10:55:03 what kind of response times do those have? Feb 06 10:55:32 tens or hundreds of ms Feb 06 10:56:48 jannu: they tend to be a motor with a misbalanced weight attached Feb 06 10:57:59 XorA: yep. that's what I was thinking Feb 06 10:59:28 XorA: I was just wondering about using it in providing ui feedback for the touchscreen Feb 06 11:00:00 that'll basically require response times of 10 ms or less Feb 06 11:00:04 jannu: my feeling is it would lag behind any event too much, but I dont actually know what the NEO has fitted Feb 06 11:00:50 phone with a rumble pack though :-) Feb 06 11:01:46 XorA: that's another option Feb 06 11:02:03 it's also a bit hungry on power Feb 06 11:02:12 bass-boost Feb 06 11:03:21 koen: this phone isnt gonna need bass boosting :-) Feb 06 11:03:59 some sort of touch feedback would be nice though Feb 06 11:04:34 XorA: because it already have a bass-reflex unit at the bottom? Feb 06 11:04:36 touchscreen keyboards tend to suck mainly due lack of decent feedback Feb 06 11:04:40 s/unit/channel/ Feb 06 11:04:40 koen meant: XorA: because it already have a bass-reflex channel at the bottom? Feb 06 11:05:28 jannu: audible feedback on my old Ericsson R380 was pretty sufficient when using a stylus for writing Feb 06 11:05:48 it had a loud *CLICK* sound Feb 06 11:05:58 koen: no idea about the case Feb 06 11:06:42 raynet: that's one option. I'm just a bit sceptical on whether audio is fast enough when you type with your thumbs Feb 06 11:07:02 root@h2200:/var/lib/ipkg# uptime Feb 06 11:07:02 11:39:39 up 83 days, 20:45, 2 users, load average: 2.60, 1.81, 1.66 Feb 06 11:07:26 jannu: well, a *click* sound only has to last 20-50ms Feb 06 11:11:06 yes, but the neural response delay is much longer for audio signals than it is for touch Feb 06 11:13:26 true, but i doubt that you will be writing that fast on it Feb 06 11:13:41 plus the connection between physically pressing the touchscreen and getting a physical feedback is obviously a lot easier Feb 06 11:14:09 use a mini cattleprod as a stylus Feb 06 11:14:44 extra feedback also uses more battery, though someone will probably write a nice app that shows how many keypresses there are left before you need to recharge Feb 06 11:14:54 koen: or implant my thumbs with some sort of self-electrocuting device Feb 06 11:16:23 raynet: I would suspect that one could start the vibrating motor with less power Feb 06 11:17:09 raynet: after all, you hardly need the regular vibration strength for button press feedback Feb 06 11:17:27 jannu: with less power than with sound... humm, i doubt. and the motor probably has even longer delay than with audio + neural response time (100-150ms) Feb 06 11:17:47 raynet: yes. that's quite likely Feb 06 11:18:20 If you have the vibrator on low power, it's going to take longer to start. Feb 06 11:18:28 with audio you need to move X gram object for couple mm for couple times. with motor you need to rotate a mass that is 10-100X Feb 06 11:18:33 You pretty much have to blip it on high power. Feb 06 11:19:00 and it is quite possible that the motor is on/off controlled Feb 06 11:19:07 That doesn't matter. Feb 06 11:19:19 You blip it on and off under software control. Feb 06 11:19:26 To do coarse PWM. Feb 06 11:19:37 SpeedEvil: and waste even more battery :) Feb 06 11:19:43 obviously Feb 06 11:20:24 Actaully - generally it uses less power than having it on all the time. Feb 06 11:20:47 SpeedEvil: i wouldn't be sure without actually measuring it first Feb 06 11:21:04 I'm talking of general experience with motors. Feb 06 11:21:15 it's quite likely that to make touch feedback work in a usable way one would need a separate, more lightweight vibrator Feb 06 11:21:29 yes. Feb 06 11:21:35 well, if you have a "special" circuit to control the motor, but emulating the control with ARM cpu isnt (possible) very power efficient Feb 06 11:21:38 probably something piezo-based Feb 06 11:21:53 Stephmw: that might work Feb 06 11:21:57 No, even blipping the motor on and off at 20Hz. Feb 06 11:22:11 Peizio _sucks_ at low frequency. Feb 06 11:22:24 yep. a small piezo-element might work Feb 06 11:22:30 Unfortunately not. Feb 06 11:22:42 the root problem is that motors' windings have shit power efficiency unless the thing is superconducting Feb 06 11:22:44 piezo element glued to the back of the case so you keep your fingers close to it Feb 06 11:23:04 after all, it only needs to shake the phone for something like 20-50 ms Feb 06 11:23:06 small pezio elements only produce low frequencies with reasonable drive. Feb 06 11:23:13 also you could then say the device has 2-way audio system :) Feb 06 11:23:34 You need multiple kilovolts and stacked elements to get low frequency response. Feb 06 11:23:58 why would one need low frequency response? Feb 06 11:24:18 Because a pulse feelable by fingers pretty much needs it. Feb 06 11:24:21 5 pulses at 1000Hz would work if the vibration can be directed to the casing Feb 06 11:24:34 (pulse of eg 10ms) Feb 06 11:24:39 It can't effectively with the powers available. Feb 06 11:24:51 The fingers don't really pick up that sort of frequency well. Feb 06 11:25:09 The casing also takes _lots_ of momentum to shake it. Feb 06 11:25:32 hmm. wonder if it could be directed straight to the touchscreen Feb 06 11:25:58 SpeedEvil: my nokia phone does have nice resonance vibrations with some ringtones so i dont think there is need for that much power if the frequency is selected carefully Feb 06 11:26:19 Yes - but it's going to be very audible. Feb 06 11:26:40 yeah... think of what effect you could have if one hand rotated the cube... and the other vibrated at a freq. related to the rate of rotation.... Feb 06 11:26:43 humm, too bad i am at the office, i could test a piezo at various frequencies at home Feb 06 11:26:43 lol Feb 06 11:27:21 Another problem is that you can only feel a vibration really on the touchscreen if it's touched. Feb 06 11:27:40 Otherwise, you're touching a floating layer that won't really transmit anything. Feb 06 11:27:57 which it tends to be, if you're thumb typing on it Feb 06 11:28:15 I've done the pezio experiment - basically, you need a _lot_ of energy to make it click. Feb 06 11:28:26 provided that the response is really fast that is Feb 06 11:28:37 And it gets worse when you are not touching the pezio directly, but it's in a case. Feb 06 11:28:55 The problem isn't the energy directly, it's that you need a 50V or so of drive. Feb 06 11:29:25 yep. that's a bit of a problem Feb 06 11:29:34 It's actually easier for touching to use a loudspeaker with a little weight on it. Feb 06 11:29:51 Or equivalent. Feb 06 11:30:14 and frequencies around 250Hz or so Feb 06 11:30:49 Yeah. Again - this isn't low power though. Feb 06 11:32:17 Still. One can always hope Feb 06 11:33:20 I went into this when I was designing an mp3 player - not yet built. Feb 06 11:33:50 SpeedEvil: you mean touch screen feedback? Feb 06 11:34:01 No - touch feedback Feb 06 11:34:03 no screen Feb 06 11:34:34 ok. sounds interesting Feb 06 11:35:06 One surprisingly power efficient way was a stepper motor, that wedged the case open and closed by a fraction of a mm. Feb 06 11:35:35 That has other problems though. Feb 06 11:36:23 Another option might be to somehow mount the touch screen on a spring Feb 06 11:36:29 Or somesuch Feb 06 11:36:55 In order to get that press-and-release feel Feb 06 11:37:34 I think you'd need ~2mm travel for that to be effective though Feb 06 11:38:20 how about a tiny electrical discharge? Feb 06 11:38:42 coupled with pressing the layer above the touchscreen you might get away with less than 2 mm Feb 06 11:38:51 enough to feel, but not to shock Feb 06 11:39:25 you'd need a conductive surface somewhere Feb 06 11:39:30 call it, The ElectroTouch feature ;-) Feb 06 11:39:34 or a microgrid Feb 06 11:40:10 actually that might work Feb 06 11:40:54 because with a grid you could direct the discharge exactly where your finger is Feb 06 11:41:11 downside, dunno about efficiency.. and how do you determine when the contact isn't a finger? Feb 06 11:41:29 do you need to? Feb 06 11:41:44 dunno about you, but I stick my phone in my pocket... Feb 06 11:41:49 if the touchscreen isn't locked, just shoot Feb 06 11:42:07 jokes aside, that could get the battery down fast Feb 06 11:42:43 You really don't want to put touchscreen stuff in your pocket with any other stuff. Feb 06 11:42:59 as it'll scratch up horribly. Feb 06 11:43:21 yep. fingers aren't exactly the world's most sensitive devices when it comes to electricity Feb 06 11:43:27 my p910 confirms the scratching :) Feb 06 11:43:36 Thay are quite sensitive actually. Feb 06 11:43:41 in terms of joules. Feb 06 11:43:51 that's why you have more than one pocket Feb 06 11:44:01 The voltage may need to get quite high to put that number of joules into the fingertip. Feb 06 11:44:12 Plus, unless you have a metal case, there is no ground. Feb 06 11:44:31 yes, but in the range of using the shock for feedback on each and every keypress Feb 06 11:44:55 Compared with keeping the display backlight on, it's not really horrible. Feb 06 11:45:10 good point Feb 06 11:45:17 This is why I like hardware buttons. Feb 06 11:45:23 I wonder how good the speakers are, what the lowest hz is Feb 06 11:45:42 For this sort of speaker - it's generally not a good question. Feb 06 11:45:50 The answer depends more on the housing. Feb 06 11:45:57 And the volume. Feb 06 11:46:10 Probably not much power at 30hz anyway for a small speaker Feb 06 11:46:12 Generally, a few hundred Hz to get reasonable output. Feb 06 11:46:17 Essentially zero. Feb 06 11:46:49 To keep the same output pressure with a given size of speaker, is a problem. Feb 06 11:47:15 Because the speaker needs to press in and out on the air by a constant amount. Feb 06 11:47:34 And if it has to keep on doing that for a substantial fraction of a second, it runs out of travel. Feb 06 12:01:34 for the community list: http://www.murrayc.com/blog/permalink/2007/02/06/dealing-with-zealots-in-open-source-communities/ Feb 06 13:14:47 we need http://www.hkepc.com/bbs/itnews.php?tid=738057 in v2 Feb 06 13:18:16 koen: that article is a good one, and any reasonable/compromising sort of person would probably agree with it. But the reason it's so easy to agree with is because it's making one simplistic assumption: that the people on a given project are rational whereas the people expressing counter-opinions are not. Unfortunately, that doesn't apply at all when a given project is surrounded by regressive fanboy groupies ... and sadl Feb 06 13:19:01 Morgreet: you got cut off after 'sadl' Feb 06 13:19:02 koen: in other words, the article deals with only the easiest problem area in forum life. :P Feb 06 13:19:15 k, damn limits: Feb 06 13:19:30 buz lol Feb 06 13:19:33 REPOST TAIL: ... and sadly, that happens a fair bit. Feb 06 13:19:50 together with gf7400go Feb 06 13:25:58 The worst kind of fanboys are the "That's impossible, said so" ones. Hate them, the most regressive type. Feb 06 13:27:07 counter Feb 06 13:27:07 5 days 16:00:38 (5.667 days) for source for *all* developers and devices for selected developers (2007-02-11); a month, 5 days (33.667 days) for anybody (targeting developers) (preorders: not yet) (2007-03-11); 7 months, 5 days (217.667 days) for mass market (2007-09-11): see topic for more info Feb 06 13:27:41 thank god february has only 28 days :D Feb 06 13:27:54 had it 31, I'd have to wait 3 days more :D Feb 06 13:28:02 Hehe Feb 06 13:29:10 Feb[25] = '\0'; Feb 06 13:29:15 ?-) Feb 06 13:29:17 There ya go, shaved another 3 days off. Feb 06 13:29:22 lol Feb 06 13:29:40 I assume you use 0-indexing.. Feb 06 13:29:46 Of course ;p_ Feb 06 13:34:48 counter Feb 06 13:34:49 5 days 15:52:57 (5.662 days) for source for *all* developers and devices for selected developers (2007-02-11); a month, 5 days (33.662 days) for anybody (targeting developers) (preorders: not yet) (2007-03-11); 7 months, 5 days (217.662 days) for mass market (2007-09-11): see topic for more info Feb 06 13:38:07 Hi Feb 06 13:38:30 can u tell me where to buy a phone with openmoko? Feb 06 13:38:50 you can't yet Feb 06 13:39:30 why? Feb 06 13:39:49 counter Feb 06 13:39:49 5 days 15:47:57 (5.658 days) for source for *all* developers and devices for selected developers (2007-02-11); a month, 5 days (33.658 days) for anybody (targeting developers) (preorders: not yet) (2007-03-11); 7 months, 5 days (217.658 days) for mass market (2007-09-11): see topic for more info Feb 06 13:40:30 7 month???!?! Feb 06 13:40:44 too much Feb 06 13:40:48 1 month Feb 06 13:40:59 Yep, unless you want it in a developer-only state, in which case just 5 weeks. Feb 06 13:41:09 2 weeks Feb 06 13:41:21 But for plain users who expect all the goodies working, Sept 11 Feb 06 13:41:49 the phone will ship worldwide even in developer preview, right ? :) Feb 06 13:42:00 Allegedly, yeah Feb 06 13:42:23 and u know if it will have have a wifi connection? Feb 06 13:42:32 Yes we know. It won't. Feb 06 13:42:39 :-) Feb 06 13:42:54 this is a problem Feb 06 13:43:41 Just a matter of priorities: openness, or features? For Neo, it's openness. Feb 06 13:44:07 openness then features :) Feb 06 13:44:20 is it known if v2 will have wifi ? Feb 06 13:44:21 yay open features Feb 06 13:44:36 Neo v2 will probably have wifi, since the OLPC group is working on an open driver. Feb 06 13:44:53 aha Feb 06 13:45:15 ha, but will it do mesh? Feb 06 13:45:47 And what's more, OLPC has exactly the same reqs as phone: very low power, very low cost. Feb 06 13:46:27 hmm i think i've read an article on that ... or it was disscussed here yesterday Feb 06 13:46:29 :) Feb 06 13:46:52 No reason why it shouldn't do mesh. However, I assume that would not be able to share the same RF front-end as cellular. Just guessing. Feb 06 13:46:55 Morgreet: thats a huge leap, from OLPC network card to Neo one :-) Feb 06 13:47:21 XorA: yep. But huge leaps'R'Us :P Feb 06 13:48:38 can u tell me the difference between openmoko and qtopia? Feb 06 13:52:43 qtopia has a good support for developping new software Feb 06 13:52:44 slight text update: Feb 06 13:52:45 counter Feb 06 13:52:45 5 days 15:35:01 (5.649 days) for source for *all* developers and devices for selected developers (2007-02-11); a month, 5 days (33.649 days) for devices for anybody (targeting developers) (preorders: not yet) (2007-03-11); 7 months, 5 days (217.649 days) for mass market (2007-09-11): see topic for more info Feb 06 13:53:20 openmoko too? Feb 06 13:53:40 you can install qt at least I think Feb 06 13:54:11 in Neo? Feb 06 13:55:14 I see that the braindead city of Boston won its suit over the ATHF joke (Slashdot), got $2m in "damages". No surprise though, judges don't remain judges by telling the valiant defenders in the War Against Terror that they're all on crack. Feb 06 13:56:01 It was an out of court settlement I thought Feb 06 13:56:03 luca: OpenMoko is based on openembedded: http://www.openembedded.org/filebrowser/org.openembedded.dev/packages/qt Feb 06 13:57:30 so for developping I also need qt? Feb 06 13:58:31 luca: i'm in the process of porting Mono if you're interesed. i have managed to port portable .net but with some quirks atm Feb 06 13:58:38 Morgreet: i heard that it was an out-of-court settlement Feb 06 14:00:20 sorry for all this question but I must to understand... Feb 06 14:00:42 I know how to develop application with qt Feb 06 14:00:59 so it will be easy to develop for qtopia Feb 06 14:01:05 but for openmoko? Feb 06 14:02:47 luca: main differences, GTK versios QT, LGPL versus GPL, c rather than c++ (but you can use GTKmm), the platform is a cutdown version of gnome, you can make commercial software on it wihtout buying a license. Feb 06 14:03:01 luca: qt not needed, but if you want its available Feb 06 14:03:46 aloril: though he probably would have problems accessing the framework from qt Feb 06 14:04:02 openmoko is based on gtk? Feb 06 14:04:08 luca: yep Feb 06 14:04:32 ok, now I'm understanding... :-) Feb 06 14:04:32 robtaylor: I'm still smiling because dbus finally works Feb 06 14:04:54 koen: heh, i'm still perplexed why that hack was needed ;) Feb 06 14:05:34 Thanks to all, u were very usefull Feb 06 14:05:35 you can image the suckage of having no session-bus Feb 06 14:05:53 I must go, bye Feb 06 14:06:11 koen: yeah, that would suck hard Feb 06 14:06:53 mickeyl: does openmoko have a session bus? any idea why you didn't hit the same problem as GPE? Feb 06 14:07:25 robtaylor: they don't use gpe-dm, but a shell wrapper Feb 06 14:07:25 ahh Feb 06 14:07:39 robtaylor: that, or dbus 0.61 Feb 06 14:07:45 *shudder* Feb 06 14:07:45 it only started with dbus 0.9x Feb 06 14:07:59 interesting Feb 06 14:08:54 robtaylor: we have a session bus and we run into the same problem, but due to more important problems no one had time looking into it Feb 06 14:09:25 mickeyl: ok, well its fixed in OE now Feb 06 14:09:31 excellent. Feb 06 14:09:42 it was a complete weirdo Feb 06 14:10:16 for some reason, SIGCHLD is blocked for anything launched by GDM (i think the blocker is earlier) Feb 06 14:10:31 and that stops waitpid working Feb 06 14:11:28 mickeyl: does openmoko have libglade? Feb 06 14:11:40 yes Feb 06 14:11:41 cool Feb 06 14:11:43 if not, it's only a package install away :) Feb 06 14:11:44 no support for openmoko widgets though Feb 06 14:11:46 (yet) Feb 06 14:12:06 ok, not suprising :) Feb 06 14:12:39 mickeyl: checkout my current project: http://live.gnome.org/High5/PeopleBrowser Feb 06 14:12:50 mickeyl: judging from your comments on image use, the integer pixops patch would speed it up a lot, but sapwood would speed it up even more Feb 06 14:13:00 mickeyl: although I still haven't found docs on sapwood Feb 06 14:13:09 just got it talking to ebook last night, so starting to get excited about it :) Feb 06 14:13:35 as far as I can tell, integer pixops is 100% important as it speeds up a lot more than just the pixmap engine Feb 06 14:13:43 I'm still not sure whether we can use sapwood Feb 06 14:13:47 since it removes a couple of features Feb 06 14:13:47 robtaylor: don't forget to include some geo information Feb 06 14:13:51 and only supports 1 channel alpha Feb 06 14:13:55 which might be a showstopper Feb 06 14:14:06 1 channel or 1 bit? Feb 06 14:14:14 bit :/ Feb 06 14:14:27 right Feb 06 14:14:36 koen: we wont :) Feb 06 14:15:39 koen: oh, that's a nice idea for a dynamic group - people near me :) Feb 06 14:19:50 robtaylor: I'm still thinking about how to send my position back home in a sane, and cost-friendly way Feb 06 14:21:08 koen: what do you mean exactlY? Feb 06 14:21:50 when I'm on the road and I have my gps + {n800 or moko} on I want to send my position to the server back home Feb 06 14:22:06 koen: SMS Feb 06 14:22:17 XorA: naa, data rates are checper than sms Feb 06 14:22:20 sms cost €0.20 a piece Feb 06 14:22:26 gprs is free in .nl Feb 06 14:22:38 but outside .nl sms is cheaper Feb 06 14:22:38 koen: even when roaming? Feb 06 14:22:57 when roaming gprs is *@($*(@# expensive Feb 06 14:23:10 in the neighbourhood of a few pints per mb Feb 06 14:24:04 robtaylor: any screenshots of your people thing? Feb 06 14:24:43 koen: only the mockup for now Feb 06 14:25:17 robtaylor: make it finger friendly on high-dpi screens Feb 06 14:26:38 koen: thats the plan :) Feb 06 14:26:39 counter Feb 06 14:26:40 5 days 15:01:06 (5.626 days) for source for *all* developers and devices for selected developers (2007-02-11); a month, 5 days (33.626 days) for devices for anybody (targeting developers) (preorders: not yet) (2007-03-11); 7 months, 5 days (217.626 days) for mass market (2007-09-11): see topic for more info Feb 06 14:27:15 koen: it should just be a matter of plopping in some different glade Feb 06 14:27:35 robtaylor: I hope it doesn't use libpanelapplet Feb 06 14:27:39 koen: any idea how fast gdkpixbuf scaling is on a 200Mhz arm? Feb 06 14:27:45