**** BEGIN LOGGING AT Mon Jul 14 02:59:56 2008 Jul 14 03:06:24 whattup Jul 14 03:15:19 lot of gphone news Jul 14 03:24:12 there is? Jul 14 03:24:28 yeah, supposed rumors Jul 14 03:24:37 that google is doing gphone actually Jul 14 03:24:40 rumors aren't news :) Jul 14 03:24:49 they are Jul 14 03:24:53 haha Jul 14 03:26:02 jasta: got a iphone? Jul 14 03:26:07 no Jul 14 03:26:16 are you planning to get the 3g one? Jul 14 03:26:20 no Jul 14 03:26:29 what you have now? Jul 14 03:29:01 moto q9h Jul 14 03:29:09 cool Jul 14 03:29:24 looks like iphone2 has activation issues Jul 14 03:36:15 that doesnt really affect me Jul 14 03:40:52 my original moto q is a piece of shit Jul 14 05:16:07 iphone, gphone? Jul 14 05:16:18 both in news Jul 14 05:16:25 they should combine forces and put out a "i am a g" phone Jul 14 05:16:32 for the original gangstas Jul 14 05:16:35 hehe Jul 14 07:32:04 morning not-iphone dudez Jul 14 10:34:40 michaelnovakjr: you there ? Jul 14 11:40:36 hi evryone Jul 14 11:40:59 hu Jul 14 11:41:34 i got iphone classic is it possible to install android on it ? Jul 14 11:43:10 guten morgen cutmasta Jul 14 11:43:50 ist es möglich zu installieren Androiden auf iPhone? Jul 14 11:45:38 hallo bist du da? Jul 14 11:46:19 einefhs, no its not possible to install android on iphone Jul 14 11:46:22 afaik Jul 14 11:46:36 why ? Jul 14 11:47:20 see here the video ?http://code.google.com/android/ Jul 14 11:47:45 ? Jul 14 11:48:28 serge use iphone Jul 14 11:48:34 with android Jul 14 11:49:24 where? Jul 14 11:49:56 when u play the video u will see the iphone Jul 14 11:50:06 its not the iphone Jul 14 11:50:09 wwhere it's writen android Jul 14 11:50:22 r u sure ? Jul 14 11:50:48 ok Jul 14 11:51:17 allright is it possible to make dual boot on it ? Jul 14 11:53:09 no news is bad news Jul 14 11:54:59 hallo ist da jemand hier? Jul 14 12:26:34 is it possible to install android on the iphone ? Jul 14 12:26:58 einefhs, no its not possible to install android on iphone Jul 14 12:26:58 afaik Jul 14 12:27:28 what of the " no its not possible to install android on iphone" do you have not understood? Jul 14 12:27:52 see the video Jul 14 12:28:15 i m 100 sure it's an iphone the guys use it Jul 14 12:28:25 to present the andraoid features Jul 14 12:31:18 einefhs: no Jul 14 12:31:53 watch the video, it's not an iphone Jul 14 12:34:41 see the black phone Jul 14 12:34:47 it's an iphone Jul 14 12:42:24 hi muthu Jul 14 12:42:30 hi Jul 14 12:42:31 how r u ? Jul 14 12:42:35 good Jul 14 12:42:50 what's happening with you? Jul 14 12:42:53 last time u said me that it' s posible to make an dual bookt Jul 14 12:42:56 last time u said me that it' s posible to make an dual boot Jul 14 12:43:11 with android an iphone application Jul 14 12:43:28 how to make it possible ? Jul 14 12:43:38 you have to wait for android source Jul 14 12:44:04 its not opensource yet Jul 14 12:44:30 ok Jul 14 12:44:53 on ur iphone have u got dual boot with linux N Jul 14 12:44:55 ?? Jul 14 12:47:18 ok Jul 14 12:47:21 no pb Jul 14 12:48:00 no not yet Jul 14 12:48:49 ok thx Jul 14 12:48:50 bye Jul 14 13:50:14 yawn Jul 14 13:51:49 howdy Jul 14 13:54:43 i just saw pictures of the treo 800w Jul 14 14:22:35 hey jasta Jul 14 14:25:16 michaelnovakjr__ Jul 14 14:25:20 you there ? Jul 14 14:25:26 hye Jul 14 14:25:35 what's up Jul 14 14:25:43 ahh great. :) I got a new idea for an application for you. :-) Jul 14 14:25:52 oh yea? what's that? Jul 14 14:27:04 At the moment I'm setting up some NAS system with freeNAS installed on it. On to that server I installed SqueezeCenter (formally known als Slimserver) which is a music serving server supporting all music formats and is licensend under the GPL. Jul 14 14:27:29 cool cool Jul 14 14:27:45 From this server I will stream to my Squeezeboxes in my house. ( www.slimdevices.com ) Jul 14 14:28:10 ince Jul 14 14:28:23 The server is capabable of serving different music streams to every device or it is also possible to synchronise the streams Jul 14 14:28:50 that sounds useful Jul 14 14:28:54 Now comes my idea :). It would be great to develop a remote app for the server. Jul 14 14:29:12 There are remotes available from Logitech Jul 14 14:29:25 but they are pretty expensive Jul 14 14:29:45 Another one has written an remote for the iPhone controlling the SlimServer Jul 14 14:30:07 interesting Jul 14 14:30:14 http://penguinlovesmusic.de/?page_id=7 Jul 14 14:30:41 screens of the iPhone app (thats more an webapp caused by the restrictions of the iPhone v1) Jul 14 14:31:29 I'm just figuring out how all the stuff works together but it seems to be really useful and not restricted by file formt restrictions like other music systems in that area. Jul 14 14:32:06 i see Jul 14 14:34:52 I also had an idea to use the GPS of the mobile phone in order to switch the music from one room to the next while walking through the house Jul 14 14:35:07 But I think GPS is not as precise as it would be needed Jul 14 14:35:09 in this case Jul 14 14:35:14 agreed Jul 14 14:35:17 that would be tough Jul 14 14:35:52 is there a gps to-do list for android? Jul 14 14:36:06 or better put (location based to-do list) ? Jul 14 14:36:25 I dont know havent seen one Jul 14 14:40:10 do you have any experience with free nas ? Jul 14 14:40:33 just trying to suspend my nas at the moment :) and it refuses to work :) Jul 14 14:41:07 i have some experience with it Jul 14 14:42:00 What would you do in order to suspend it? My aim is to WOL it from my Macbook here. Jul 14 14:42:20 i'd have to look at it Jul 14 14:42:58 oh I dont want to keep you from work Jul 14 14:43:27 its ok :) i can look into it in a bit though Jul 14 15:11:32 Great WOL is not available for FreeNAS :( Jul 14 15:11:37 I love BSD Jul 14 15:11:50 http://sourceforge.net/forum/message.php?msg_id=5088647 Jul 14 15:16:10 how do I detect that an activity is being cancelled (via back button) ? Jul 14 15:26:38 i am rusty on my android Jul 14 15:26:50 need to program, even if its useless Jul 14 15:59:03 zhobbs_: one approach would be to start the activity as a subactivity Jul 14 16:09:33 michaelnovakjr: just wanted to say that I'm running now FreeNAS 0.7 and WOL is not working :( again Jul 14 16:30:51 good morning ;) Jul 14 16:36:44 yawn Jul 14 16:36:46 im so exhausted Jul 14 16:36:48 and sunburnt :) Jul 14 16:36:58 heh Jul 14 16:37:12 I got a little sun yesterday, but I had sunscreen so its not too bad. Jul 14 16:37:31 that and I went out really early before it was too hot Jul 14 16:58:28 the register is going to call me this afternoon to discuss the lack of SDK updates Jul 14 17:14:14 what is the register? Jul 14 17:15:10 michaelnovakjr: http://www.theregister.co.uk/ Jul 14 17:15:28 zhobbs: you have an NDA SDK right? Jul 14 17:15:33 yeah Jul 14 17:16:06 and google doesn't care that its NDA users are talking about sdk issues? Jul 14 17:16:12 but I won't talk about that...just talking about public sdk Jul 14 17:16:31 I know what you mean though, I was thinking maybe I should just lay low until after adc is over Jul 14 17:17:02 just to make sure you don't get disqualified or even more frightening... sued Jul 14 17:17:12 yeah Jul 14 17:17:34 I don't have much to gain...a link to helloandroid.com is the best outcome Jul 14 17:18:19 i think the community is wasting too much time crying over the sdk Jul 14 17:18:43 what else should they be doing? Jul 14 17:19:02 don't talk about jasta like that ;) Jul 14 17:19:09 keeping sharp on those android skills Jul 14 17:19:31 zhobbs: its not like you had to reinvent the wheel with your update I assume? Jul 14 17:20:26 * zhobbs_ can't comment Jul 14 17:20:56 I'll tell you this...they got rid of activities and intents Jul 14 17:21:28 lol, no they didn't. Jul 14 17:21:32 ssshhh Jul 14 17:21:44 haha, kidding of course Jul 14 17:22:04 what a coincidence Jul 14 17:22:21 ssshh, dans here Jul 14 17:22:22 hehe yeah, morrildl can you comment on the rumor that they got rid of activities and intents Jul 14 17:22:26 lol Jul 14 17:24:35 haha Jul 14 17:25:05 lol Jul 14 17:26:39 jasta: yes. Activities, Intents, Services, ContentProviders, and IntentReceivers are all gone in the next SDK. We are replacing them with something we are calling "COM Objects" Jul 14 17:26:48 hehe Jul 14 17:26:52 * yakischloba /parts Jul 14 17:27:25 hm, COM objects Jul 14 17:27:28 i dig it Jul 14 17:27:39 FYI, i think the documentation should more explicitly call out that all these components relate through the OpenBinder, or more generally an efficient IPC system that underlies the entire system. Jul 14 17:28:02 i think that people are left to assume that intents and content providers are "magic" as it is now. Jul 14 17:28:24 they arent? :( Jul 14 17:28:26 magic is easier to understand! Jul 14 17:28:30 jasta: Hmm, interesting. We intentionally do NOT do that, b/c we don't want n00bz thinking they "need" to understand Binder to write code Jul 14 17:28:42 (BTW, Android's Binder implementation is not OpenBinder; it's a subset. Just FYI.) Jul 14 17:28:50 morrildl: it's easier for me at least to see that they are all just endpoints on the same system. Jul 14 17:29:09 especially to realize when to use a service and when to broadcast an intent. Jul 14 17:29:17 jasta: Granted, we need at a minimum to make sure it's possible to "elevate" your understanding easily once you do get the basics Jul 14 17:29:22 I don't mind the black box Jul 14 17:29:32 a lot of people are forcing service dynamics through broadcast intents because they don't realize they are just different manifestations of the same thing Jul 14 17:29:52 jasta: example? Jul 14 17:30:24 well f00f i believe does this in his current project ;) Jul 14 17:30:50 i remember fighting with him early on about this Jul 14 17:31:43 beside that though, i think it's useful to understand the relationship so that you can also visualize the threading model when interacting with services, activities, and content providers. Jul 14 17:32:04 more clarification is certainly needed, but it's useful to understand the IPC nature of the design to infer the model being used Jul 14 17:32:05 jasta: yeah we have the stuff I put together for I/O that needs to make it into hte docs proper Jul 14 17:32:12 ...hopefully with more detail Jul 14 17:33:40 still, i wouldn't write the documentation for the lowest common denominator. let tutorial sites do that ;) Jul 14 17:35:29 I like that android is so easy to learn and program in...one of it's big strengths Jul 14 17:37:27 it also has a lot of gotchas, especially when you need to write non-trivial programs Jul 14 17:39:12 whenever you are dealing with IPC, you need to think carefully about your design. Android's IPC system has a lot of niceties, but it doesn't avoid the complexity of arranging a program which avoids deadlocks, race conditions, etc. Jul 14 17:40:55 im scared now. Jul 14 17:41:15 yakischloba: it's just jasta :) Jul 14 17:41:35 ok. I'll ignore him and continue to do things the wrong way :) Jul 14 17:42:20 btw please do keep writing docs for the lowest common denominator, i'm having a hard enough time understanding them as it is. Jul 14 17:42:34 says he who deals largely in the single-threaded space ;) Jul 14 17:42:54 ;) Jul 14 17:42:57 [to romain] Jul 14 17:43:02 jasta: not exactly Jul 14 17:44:12 ipc all seems to work fine for me...with very little thinking Jul 14 17:46:12 oh dear Jul 14 17:46:25 if jasta thinks it's primarly single-threaded, then yes we do need to update docs :) Jul 14 17:47:25 morrildl: i don't think the system is single-threaded, but that the UI toolkit, as presented to the user, is inherently not multi-threaded. Jul 14 17:47:51 it depends Jul 14 17:48:00 it is at a single view hierarchy level Jul 14 17:48:11 which is true. the only difference that i see which adds complexity there is that the drawing mechanism seems to be much unlike traditional systems. Jul 14 17:48:17 but if you have two hierarchies (say, an activity and a dialog) then you have one thread per hierarchy Jul 14 17:48:31 and there's another tiny detail: the window manager is in the system process Jul 14 17:48:49 the drawing system unlike tradition systems? how so? Jul 14 17:49:28 romainguy_: just as you described, in that you cooperatively draw multiple view hierarchies. Jul 14 17:49:42 they don't cooperate Jul 14 17:50:02 well they seem to. i thought that was how the surface flinger worked. Jul 14 17:50:30 a view hierarchy draws in a surface Jul 14 17:50:36 they don't know anything else Jul 14 17:50:45 surfaceflinger's job is to compose the surfaces together Jul 14 17:51:06 the layering of the design is confusing to me, though that is what i meant by cooperatively. Jul 14 17:51:42 it's really no different from a desktop OS Jul 14 17:51:56 the only difference is that you can have several UI threads in an application in Android Jul 14 17:52:34 it's just that for some reason the developer seem to think about the compositing engine and the window manager Jul 14 17:52:50 whereas on a traditional desktop, we never think about it :) Jul 14 18:24:00 hello? Jul 14 18:25:33 romainguy_: well thanks for the explanation :) Jul 14 18:25:42 RyeGye24: hi Jul 14 18:26:40 hey Jul 14 18:26:50 can we help you? Jul 14 18:26:55 ok Jul 14 18:27:11 i have the m5 version of the SDK and Im usng eclipse Jul 14 18:27:22 but it keeps telling me that I'm using m3 Jul 14 18:28:15 follow the instructions to install the m5 version of the Eclipse plugin. It requires a special step. Jul 14 18:28:21 kk Jul 14 18:28:23 thx Jul 14 18:28:41 And also make sure you point the Eclipse plugin to the right location, specifically the M5 installation. Jul 14 18:28:52 That is an explicit step that you must perform (and is covered in the installation instructions) Jul 14 18:29:24 thats probably where i screwed up Jul 14 19:01:55 anybody know the ubuntu package name that contains the binary 'id3v2'? Jul 14 19:02:33 morrildl: did you try sudo apt-get install id3v2? Jul 14 19:02:49 * morrildl sighs Jul 14 19:02:56 I have now :) Jul 14 19:03:11 * morrildl tried apt-get install id3 Jul 14 19:03:16 ah Jul 14 19:03:40 you should also be able to see the package name while its installing for future reference too Jul 14 19:08:30 morrildl: use apt-cache search Jul 14 19:08:41 or dpkg -S $(which id3v2) if you already have it installed Jul 14 19:10:53 apt-cache search is handy Jul 14 19:40:42 whew. Jul 14 20:24:07 http://groups.google.com/group/android-discuss/msg/fa17b0e011a3b8db Jul 14 20:24:13 honest post by JBQ Jul 14 20:27:50 anyone want to talk to the register ( http://www.theregister.co.uk/ ), I bailed on the interview and they want to talk to some developers Jul 14 20:27:58 so he wants some referals Jul 14 20:28:20 sure Jul 14 20:28:22 Oh well take jasta :) Jul 14 20:28:42 I'm no real developer so I cant help you. Jul 14 20:28:54 davidw: what's your email address? ( you can /msg it to me if you want) Jul 14 20:29:07 davidw@dedasys.com Jul 14 20:31:48 ok, sent your info Jul 14 20:34:09 zhobbs_: i would like to ;0 Jul 14 20:34:52 ok sure Jul 14 20:35:08 what's your email? Jul 14 20:35:16 jasta@devtcg.org Jul 14 20:35:50 need someone to keep it real Jul 14 20:37:32 ok, sent him your info Jul 14 20:38:08 that was probably a wise move, by the way. Jul 14 20:38:12 to excuse yourself from that interview Jul 14 20:38:23 definitely Jul 14 20:38:29 without legal council on the NDA, the best course of action is to keep quiet about all of it :) Jul 14 20:40:13 yeah Jul 14 20:40:27 zhobbs, damn, that guy was fast Jul 14 20:40:31 hehe Jul 14 20:40:39 he seemed anxious Jul 14 20:41:18 well, attack pieces probably sell well Jul 14 20:41:56 i don't see how it benefits developers to participate in attack pieces Jul 14 20:41:57 well I sent him 2 people who have been openly critical of the situation, so he should have some ammunition Jul 14 20:42:54 michaelnovakjr: all the SDK bitching has been internal so far...maybe if it gets rolling in the press then google will be forced to act quicker....maybe not though Jul 14 20:43:17 i don't think negative press would have that impact Jul 14 20:43:47 Could anyone tell me what is so difficult about saying there will be a new SDK or not? Even if there are hundreds of companies involved I dont get the point why they cant say that. (reading JBQs post) Jul 14 20:44:33 I can understand that JBQ cant answer that but I dont get the management decisicion :/ Jul 14 20:45:19 Just bs. Who fucking knows. Management make about as much sense as women. Jul 14 20:45:40 Some people just like to have secrets Jul 14 20:45:43 I don't even care when the damn thing comes out, I just want to know what the heck is going on. It's very clear that the Google guys know they're taking a huge hit by screwing around like this Jul 14 20:46:06 huge hit? Jul 14 20:46:08 so I'm worried what rock and hard place they're dealing with, and how that will affect the whole 'open source' thing Jul 14 20:46:22 michaelnovakjr, they're getting raked over the coals for this business Jul 14 20:46:26 Thats the point. Even saying senseless things about the developement would keep the community "feel" involved Jul 14 20:46:33 how? Jul 14 20:46:52 there isn't any thing available yet, implications cannot be determined until the product is made available Jul 14 20:47:17 michaelnovakjr, I mean the SDK/communication issue Jul 14 20:47:57 honestly, my position on the whole issue is that Google had the opportunity to air their dirty laundry a long time ago Jul 14 20:48:02 morrildl, romainguy, jbq, etc... are all smart guys with knowledge of the way open source works, and know that it's costing Google credibility Jul 14 20:48:05 failing that, there is no reason to protect them. Jul 14 20:48:06 there is no communication issue Jul 14 20:48:38 and... this is Google we're talking about... not some little tiny shop that lives in fear of Microsoft... Jul 14 20:48:46 michaelnovakjr, s/issue// Jul 14 20:48:50 Hmm not? I havent seen a blog post about anything in the last months Jul 14 20:48:57 that said, i don't intend to "attack" Google. being openly critical and wildly spewing accusations are not the same. Jul 14 20:49:04 so far all the bad press about android has been the latter. Jul 14 20:49:28 jasta agreed Jul 14 20:49:41 the press for android is a joke Jul 14 20:49:42 jasta, I think I might say that they did manage something badly to get this far and get 'stuck' Jul 14 20:49:44 and as i said, if they wanted to ward off this type of reporting, then perhaps it's high time to SAY something ;) Jul 14 20:49:52 it was pretty smooth sailing at the beginning Jul 14 20:50:03 costing credibility or not, people will still use it in the end. Has anyone completely bailed as a result of this yet? Jul 14 20:50:22 yakischloba: that can't be determined Jul 14 20:50:30 how can you bail on something you don't use? Jul 14 20:50:31 do note that our only respond from Google has been uttered by Dan, Jason, or Romain. A group of people hardly representative of the Google, the OHA, or the project at large. Jul 14 20:50:41 yakischloba: probably not, but there are projects that aren't be developed as a result Jul 14 20:50:49 response* Jul 14 20:50:58 -the* Jul 14 20:51:00 i suck :) Jul 14 20:51:09 There is development that is being kept from going on right now, but where else will people turn? Jul 14 20:51:25 Is everyone jumping ship to develop for iPhone? Jul 14 20:51:30 doubt it Jul 14 20:52:07 how can you say google isn't open and they suck and then go program for the iphone? Jul 14 20:52:15 that is very hypocritical Jul 14 20:53:11 exactly. So as lame as I think the situation is (obviously I don't think its as lame as the rest of you have invested a ton of time do,) I'm still holding out. Jul 14 20:53:32 this definitely isn't a long term problem... Jul 14 20:53:44 I think we just wish we could make sense of it Jul 14 20:53:45 i just think people are blowing the whole thing out of proportion Jul 14 20:54:21 I'd sure be upset if I'd sunk 6 months of development into it too. Jul 14 20:54:27 until there's evidence from public statement that the development process won't be open, its all speculation Jul 14 20:54:59 * davidw isn't really interested in the iphone Jul 14 20:55:01 michaelnovakjr: I don't think anyone is saying it won't be open Jul 14 20:55:23 there are some people based on the fact the latest sdk is under NDA Jul 14 20:55:25 michaelnovakjr, in the long term, it's probably not a big deal, no Jul 14 20:55:31 let's not do this :) Jul 14 20:55:31 however, it stinks Jul 14 20:55:39 yes Jul 14 20:55:45 The problem is that the expectations grew up in the first months. There was a lot of communication, the challenge and so on. Now where its getting more quiet around the whole thing. (perhaps just for reasons of getting out the first hardware) Jul 14 20:55:54 but i don't think griping about it to the press is going to do anything but piss people off Jul 14 20:56:49 michaelnovakjr: i am not convinced that The Powers That Be appreciate the gravity of this most recent decision. Griping in the press is quite likely to be the only way to slap them in the face. Jul 14 20:56:51 From my point of view patience at least for the next 2-3 months has to be the goal. Jul 14 20:57:01 I'm pretty clear with what I want to do with the press: talk about Hecl, and use android as bait:-) Jul 14 20:57:20 what the heck is hecl Jul 14 20:57:23 lol Jul 14 20:57:28 www.hecl.org Jul 14 20:57:29 hehe Jul 14 20:57:34 davidw's android ;) Jul 14 20:57:59 its what they've been working on this whole time under NDA Jul 14 20:58:02 its the latest sdk release Jul 14 20:58:04 in disguise Jul 14 20:58:08 ahhahaha Jul 14 20:58:11 what's the need for another scripting language? Jul 14 20:59:30 it runs on Java ME Jul 14 20:59:51 what for? Jul 14 21:00:01 to write programs with, obviously Jul 14 21:00:13 isn't their a java api? Jul 14 21:00:16 there* Jul 14 21:00:17 Are the people under NDA originally Google employees that developed top-50 apps on their own time, or recruited because of their great apps or both? Jul 14 21:00:31 yeah, but lots of people don't like/have trouble with Java Jul 14 21:00:48 does hecl do multi threaded programming? Jul 14 21:00:54 also, as you may have noticed, despite Java being quite popular elsewhere, people also like things like PHP, Python, Ruby etc... Jul 14 21:00:58 yep Jul 14 21:01:12 so why not port one of those existing languages over? Jul 14 21:01:18 yakischloba: the NDA was extended to the ADC winners, who are all outside developers. Jul 14 21:01:32 because they're too big to run in a really limited space Jul 14 21:01:53 so what does hecl sacrifice in order to run on mobile hardware? Jul 14 21:01:55 I think things like Lua and Forth are the only serious candidates Jul 14 21:02:32 davidw, moores law will make your niche nicheless Jul 14 21:02:37 imo Jul 14 21:03:48 jasta: So are those people helping develop the SDK, or simply continuing to develop their applications using the new one? Jul 14 21:03:51 does hecl lack functionality that python has or php has? or god forbid i mention ruby? Jul 14 21:04:00 yakischloba: the latter Jul 14 21:04:06 michaelnovakjr, certainly Jul 14 21:04:26 so then what's the plus side of using it? Jul 14 21:05:23 michaelnovakjr, I would guess because it runs in resource constrained environments Jul 14 21:05:59 for the record, i want to see someone port perl and python natively to android. that is, compile everything to dex :) Jul 14 21:06:15 or to the java bytecode, but i suspect straight to dex would be simpler. Jul 14 21:06:19 now that's a project Jul 14 21:06:29 i don't want to have to learn another language Jul 14 21:06:43 i'd rather use something that exists already Jul 14 21:07:00 it shouldnt be too hard....a CIL->dex convertor is probably the easiest way Jul 14 21:07:15 although it would be most efficient, i think, to simply build perl and python interps natively for Android Jul 14 21:07:23 and then just glue the Android parts. Jul 14 21:07:37 jasta: jython compiles python to Java bytecode Jul 14 21:08:25 romainguy_: Not really, I looked into it already. The newer version is expected to compile directly to Java bytecode, but current versions actually compile to Java source code, and in particular the code generated is ill-equipped to then build for Android Jul 14 21:08:40 the whole thing would still need lots of work to make it to Android well Jul 14 21:09:04 jasta: yep Jul 14 21:09:11 that was exactly my assessment a few months ago, too Jul 14 21:09:37 so instead of wasting resources on something new why not work on python then? Jul 14 21:12:01 also, it doesn't really build Java source code that is a good translation of Python code. It mostly generates stubs to parse the Python that you wrote. In other words, it would run really slow on Android :) Jul 14 21:12:50 ... back ... talked to that reporter Jul 14 21:13:01 in it's current form, it's a weak design, full of bad assumptions. Jul 14 21:13:01 hopefully he doesn't misquote me, but I got my bit in about Hecl:-) Jul 14 21:13:21 michaelnovakjr, Python wasn't designed for embedded environments Jul 14 21:13:31 for instance it assumes that you have things like files present Jul 14 21:13:42 it has regexp's built in, which take huge amounts of space Jul 14 21:14:04 python's a nice language, for sure, but embedded is often quite constrained Jul 14 21:14:13 davidw: Java, and thus Android, also include regular expression engines. Jul 14 21:14:35 even if it is a crappy one :) Jul 14 21:14:36 gambler, you're probably right about moore's law, long term, but as Keynes said, "long term, we're all dead" :-) Jul 14 21:14:44 if i had to learn hecl or java to program for mobile i'd go with java because you can take the knowledge elsewhere Jul 14 21:14:47 davidw: regexp are not built in the language, they are in the library Jul 14 21:14:50 and what happens is that stuff doesn't die, it just gets recycled Jul 14 21:15:03 lots of really old processors are still out there Jul 14 21:15:05 michaelnovakjr: and Java doesn't look as weird as hecl ;-)) Jul 14 21:15:14 romainguy: true :) Jul 14 21:17:17 michaelnovakjr, well obviously you aren't really who I'm targeting: 1) people who struggle with something like Java and want something simpler and easier, and 2) people who don't struggle to learn languages, and can pick up something like Hecl and use it to do quick prototypes or scripting engines for their own apps Jul 14 21:17:28 I'll leave the sort of mediocre middle ground to Java Jul 14 21:18:15 that is all fine and dandy, but i can't use it anywhere else Jul 14 21:18:27 java i can use embedded, desktop, server .... Jul 14 21:18:59 plus hecl looks a bit frightening compared to java Jul 14 21:19:31 morrildl: one thing i hope that Google ultimately promotes is natively building certain libraries and projects on Android. Providing a flexible build system for those sorts of projects would go a long way. Jul 14 21:19:35 you can use Hecl anywhere you use Java, but it's probably best for phones Jul 14 21:19:49 Good toolchain that can be easily adapted to other projects, I mean. Jul 14 21:19:52 on the desktop/server I'd use something like Ruby or Python or whatever Jul 14 21:20:21 that's my point though Jul 14 21:20:36 btw i absolutely hate ruby Jul 14 21:20:49 haha Jul 14 21:20:57 don't let us forget Jul 14 21:21:17 michaelnovakjr, my own opinion is that programmers who are frightened by new languages aren't very good programmers because unless you use something like Lisp, Fortran or Cobol, languages seem to come and go Jul 14 21:21:51 not to say they have to learn or love Hecl Jul 14 21:21:51 davidw: unfortunately, most programmers identify themselves using a particular language Jul 14 21:21:52 i am not frightened by new languages.... i am frightened by hecl Jul 14 21:21:53 but still... Jul 14 21:21:58 as in, "I am a Java programmer." Jul 14 21:22:18 jasta, well, there are definite advantages to getting good at a few... I usually don't use too many at one time Jul 14 21:22:22 but it's fun to play with new ones Jul 14 21:22:23 i use lots of languages.... perl, python, C/C++, java objective-c .... Jul 14 21:22:35 php Jul 14 21:22:44 c# Jul 14 21:22:55 add a few more 'mind bending' ones to that mix...those are all the same thing, in some ways. Jul 14 21:23:08 try doing Lisp, Forth and Erlang, too, just for fun Jul 14 21:23:21 I have done Lisp work before Jul 14 21:24:26 my point was only that its useful in a small context Jul 14 21:24:37 you've said yourself you wouldn't use it on a desktop/server Jul 14 21:24:42 in terms of syntax and stuff though, since languages are made for people, and people have tastes... de gustibus non est disputandum, as they say Jul 14 21:24:54 so if i am looking to get the most from my knowledge i would pick java thats all Jul 14 21:25:46 i'll be back Jul 14 21:25:50 enjoy! Jul 14 21:27:12 * gambler is looking forward to the (near) day when I can sit down at a X-terminal and VNC/rdesktop/xdmcp into my phone's 1Ghz desktop Jul 14 21:28:07 i doubt they will be called phones then Jul 14 21:29:04 pernonal internet communications device Jul 14 21:29:07 before seeing 1 Ghz CPUs in phones, I'd rather see fast memory buses and good 2D and 3D chips Jul 14 21:29:33 jasta: yeah that's in the plan Jul 14 21:29:40 i'd rather see more battery efficient radios ;) Jul 14 21:29:40 jasta: once source drops, we won't be able to STOP it after all Jul 14 21:29:46 when we say "it's not supported" we literally mean that :) Jul 14 21:30:00 morrildl: I just meant that I hope your build system is not incredibly particular and hard to generalize. Jul 14 21:30:08 eventually there will probably be a sort of "NDK" companion to the SDK that you can use for that Jul 14 21:30:21 jasta: you mean the build system for the core OS? Jul 14 21:30:59 jasta: that will not generally "care" about apps Jul 14 21:31:09 morrildl: Yes. I imagine that new libraries and apps will be introduced to the platform by either adapting your toolchain. Jul 14 21:31:12 err Jul 14 21:31:18 by either adapting your toolchain, or simply recreating it. Jul 14 21:31:24 but there will be some kind of way to get native code onto devices as apps, yes Jul 14 21:31:31 jasta: why adapt or recreate it? Jul 14 21:31:48 the toolchain used to build native apps will be separate from the core Android build tree, and contained in the "NDK" thing Jul 14 21:31:56 (Native development kit) Jul 14 21:31:59 romainguy_: Or of course we could just use it directly, that's what I'm referring to. I hope that Google considers this. Jul 14 21:32:10 that's what I am referring to Jul 14 21:32:20 maybe we should just name it "Android native toolchain" instead of NDK Jul 14 21:32:25 :) Jul 14 21:32:26 our build system builds a bunch of native libs Jul 14 21:32:31 it's very easy to add one Jul 14 21:32:39 note that this is for building .so files to load into an app an access via JNI Jul 14 21:32:46 just like it's very easy to add an app in fact Jul 14 21:32:56 security? Jul 14 21:32:56 full-native apps will not be particularly easy to do, b/c there is no native API for the core framework stuff Jul 14 21:32:59 romainguy_: But specifically what I'm referring to is it easy to add one to it that is in a different tree? Jul 14 21:33:10 morrildl, no, give it a name like JSR 322 to really confuse people Jul 14 21:33:12 Because that is the flexibility that will be the most useful. Jul 14 21:33:16 ah, that I don't know Jul 14 21:33:39 morrildl: yeah, good luck to write a UI with native code :) Jul 14 21:33:47 For example, Linux kernel modules can be built in a separate tree, though they must be aware of the main tree. That type of build system is best for projects like this. Jul 14 21:34:11 jasta: as long as you can get by with .so-via-JNI, then yes that will be possible Jul 14 21:34:30 ...modulo the usual "you build your .so for ARM9 and so it won't run on ARM7" type stuff Jul 14 21:34:34 Yes, that's what I'm referring to. For example, to efficiently port Perl this is what you'd want to do. Jul 14 21:34:44 yeah, that'll be possible Jul 14 21:34:51 note again that it would be command-line-only perl Jul 14 21:34:52 Build libperl, glue it, and then wrap around that glue. Jul 14 21:35:07 morrildl: Why couldn't you build libperl this way? Jul 14 21:35:10 although you could easily build something like wx that has a .so to bridge Perl through to Java Jul 14 21:35:18 you could Jul 14 21:35:25 but you'd need to implement the glue Jul 14 21:35:32 the perl binary is just a small program layered on top of libperl. Jul 14 21:35:33 you could build libperl, but not write GUIs *in perl* Jul 14 21:35:47 morrildl: Sure you could, Perl can dynamically typeset and glue ;0 Jul 14 21:35:54 you'd need a bridge or glue .so and a bunch of stubs in Java Jul 14 21:36:05 jasta: okay I think we are in violent agreement Jul 14 21:36:05 hehe, oh you silly Java programmers! :) Jul 14 21:36:13 can't you do reflection through the .so's? Jul 14 21:36:21 davidw: with enough work, yes :) Jul 14 21:36:36 I don't think I'd fancy having a look at the resulting code... Jul 14 21:36:52 Perl would be the easiest to fit for this reason, morrildl. Jul 14 21:37:01 but ultimately if you write a Perl program that wants to put a button on the screen, you're going to have to go through Dalvik bytecode somewhere that calls addView(new Button(...)) Jul 14 21:37:29 that bridge can be handled via the .so and an appropriate top-level Activity Jul 14 21:37:35 but it will be some work Jul 14 21:37:38 how's dalvik end up putting it on the screen? Jul 14 21:37:49 davidw: through other JNI calls, bascially Jul 14 21:38:09 this will be very exciting once released :) Jul 14 21:38:22 ok, but something interacts with the hardware... right? through syscalls or ioctls or something? Jul 14 21:38:57 davidw: yeah, native code does Jul 14 21:39:02 the expensive stuff is all native Jul 14 21:39:15 so theoretically you could ape what it's doing Jul 14 21:39:21 it's accessed from Dalvik via JNI. One of Dalvik's design goals is to make JNI not be horribly expensive Jul 14 21:39:44 davidw: you COULD, but (and this is getting outside my area of direct expertise) I'm told that it would probably be easier to just do it in Dalvik Jul 14 21:39:51 I'm sure it would Jul 14 21:40:22 the problem is that all of the high level UI code is in Java Jul 14 21:40:29 I thought that junk about Perl made it clear we were in the realm of 'could' though;-) Jul 14 21:40:30 native contains "only" the drawing primitives Jul 14 21:40:33 and surfaceflinger Jul 14 21:40:57 so yes, you could write something that interacts correctly with surface flinger but it would not interact with the window manager, the task manager, teh activities, etc. Jul 14 21:41:04 davidw: actually, no, i was thinking very practically. Jul 14 21:41:29 I think Perl is a good candidate for a real alternative to Java on Android. It is sufficiently flexible to make this work properly. Jul 14 21:41:54 I'd bet more on stuff running on top of Java Jul 14 21:42:04 would...did Jul 14 21:42:10 The Dalvik layer is the best target, not Java. Jul 14 21:42:20 well, yeah Jul 14 21:42:29 Dalvik could even be modified to make dynamic languages fit into it better. Jul 14 21:43:12 But that is speculation, since we have yet to see how it works, and how the bytecode looks. Jul 14 21:43:15 that's actually kind of a problem for things like ruby and python compared to Hecl - they're much more advanced in integrating with the JVM, meaning that they're more tied to it Jul 14 21:43:48 the disadvantage of Python and Ruby is that they're robust? Jul 14 21:43:56 isn't that more a disadvantage of Hecl? :) Jul 14 21:45:33 not that i care about all this personally. i think Java is an appropriate fit, and am not interested in using another language. Jul 14 21:45:34 for many things yes, for porting to Android, it was an advantage Jul 14 21:46:43 jasta: dalvik is designed with multiple languages in mind Jul 14 21:46:50 I would actually doubt that anyone on this channel would be in the target market for Hecl Jul 14 21:47:00 in terms of using it as their only language Jul 14 21:47:19 davidw: well I'd much rather use Hecl than Perl, and I don't even need to see Hecl syntax to know that ;) Jul 14 21:47:20 some people might want a small interpreter to use to dynamically update applications, or script extensions Jul 14 21:47:57 morrildl: I'd rather use Whitespace than Perl :) Jul 14 21:48:19 Scala already works on Android, or at least did at one time Jul 14 21:48:44 Jython would work, and if indeed they are working on a bytecode-"native" compiler option for jython, that would be perfect Jul 14 21:49:16 what about php?!?!??!?!!? Jul 14 21:49:21 except if they rely on runtime bytecode generation Jul 14 21:50:00 romainguy, IIRC, they might have some things like that... Jul 14 21:50:10 but I don't recall very clearly Jul 14 21:50:39 duey, ew. Jul 14 21:50:51 morrildl: hey now, Perl is my cherished language ;) Jul 14 21:51:24 lol Jul 14 21:51:35 Perl is a nice language to cherish in the sense of hanging it above the fireplace and not actually ever using it for anything;-) Jul 14 21:51:36 * duey notes excessive use of ? and ! Jul 14 21:52:08 I got my start as a programmer using Perl but ended up getting sick of it Jul 14 22:00:08 you guys just sent an email to the wrong people Jul 14 22:00:15 fyi Jul 14 22:00:37 duey: ? Jul 14 22:00:41 with regard to an adc build Jul 14 22:00:57 * morrildl isn't following that Jul 14 22:01:03 can you be more specific? Jul 14 22:01:08 We're pleased to announce that SDK build 84853 is now available on your private download site. This will be the last build released for ADC Round 2 and is the build that you will need to submit your final application under. Jul 14 22:01:15 okay Jul 14 22:01:21 this went to which list? Jul 14 22:01:28 adc-entrants@google.com Jul 14 22:01:39 and it should have gone where? Jul 14 22:01:54 I assume to the adc finalists Jul 14 22:01:56 ah Jul 14 22:01:57 because im not one of them Jul 14 22:02:00 okay, thanks Jul 14 22:02:06 sorry for the inconvenience. I'll notify David Jul 14 22:02:25 hehe Jul 14 22:04:41 lol! Jul 14 22:05:57 yeah, I just got that one too Jul 14 22:06:36 this probably isn't the most fortuitous time for that to go out;-) Jul 14 22:07:18 I would have to agree with that :) Jul 14 22:09:11 lol Jul 14 22:09:17 I didn't get it! Jul 14 22:11:16 AHHA Jul 14 22:11:17 you idiots! Jul 14 22:11:21 i just got that too Jul 14 22:11:38 its not that bad of a mistake Jul 14 22:11:38 with all the tension and speculation about this, lol. Jul 14 22:12:01 it even addresses us as ADC Entrants Jul 14 22:12:16 duey: are you kidding? don't you think there's gonna be about 1700 people responding "what? where do i download it?" Jul 14 22:12:37 my guess is that calling people idiots isn't in the "how to make friends and influence people" book;-) Jul 14 22:13:14 davidw: we've been called/accused of so many things by now that it doesn't really matter :) Jul 14 22:13:53 davidw: what does it say about forming private invite-only clubs? :) Jul 14 22:15:11 davidw: you talk to that reporter? Jul 14 22:15:15 zhobbs, yeah Jul 14 22:15:23 hopefully he won't try and misquote me to work his angle Jul 14 22:15:42 The best way to avoid being misquoted is to say very little. Jul 14 22:15:42 did it seem like his angle was to bash google? Jul 14 22:15:53 jasta: did you talk to him? Jul 14 22:16:37 zhobbs, probably Jul 14 22:16:53 if I were a reporter trying to get eyeballs, that's what I'd do Jul 14 22:17:10 zhobbs: nope, he didn't send an e-mail Jul 14 22:18:25 hehe, i am so tickled that after saying nothing for months, they accidentally send out an e-mail like that ;) Jul 14 22:19:50 it'll be interesting to see if they FINALLY decide to say something now that they've made this embarrassing mistake ;) Jul 14 22:26:02 hehe, dsm, clicked send too soon! :) Jul 14 22:26:40 the efficiency of auto-complete Jul 14 22:26:43 :) Jul 14 22:26:46 :) Jul 14 22:32:03 hmm? Jul 14 22:32:35 oh, dsm must be this David fellow ;) Jul 14 22:32:42 that really made my day Jul 14 22:56:32 ok. Jul 14 22:56:33 so to do it. Jul 14 22:56:43 hi Dougie187 Jul 14 22:56:47 hows it going? Jul 14 22:57:33 decent Jul 14 22:57:37 nice. Jul 14 22:57:44 you getting a lot of stuff done with your project? Jul 14 22:57:48 Nope ;) Jul 14 22:58:06 lol Jul 14 22:58:07 I've been a little busy with other stuff since my interest finally sparked Jul 14 22:58:27 I got a lot of reading done though and I've kinda started Jul 14 22:58:41 well thats pretty cool. Jul 14 22:58:50 im pumped for my laptop. Jul 14 22:59:10 When I jump in over my head like this, I sometimes have to step back and give myself 'sink-in' periods so I'm not forging ahead blindly Jul 14 22:59:33 hey Dougie187. Jul 14 22:59:37 hows it going jasta? Jul 14 22:59:47 btw, your way behind now. Jul 14 22:59:49 lol Jul 14 23:04:46 Dougie187: lol, i know. Jul 14 23:05:00 im at like 58 or something now. Jul 14 23:05:03 and im in act 4. Jul 14 23:06:18 i'm in act 4 too. Jul 14 23:06:19 nm Jul 14 23:12:35 yakischloba, what are you working on? Jul 14 23:12:48 davidw: not much yet ;) Jul 15 01:15:08 howdy Jul 15 01:18:17 hey michaelnovakjr Jul 15 01:18:21 hows it going? Jul 15 01:20:07 michaelnovakjr_ you around? Jul 15 01:23:34 yep Jul 15 01:23:41 did you get my messages this weekend? Jul 15 01:23:42 what's up? Jul 15 01:23:50 heh Jul 15 01:23:53 nope... just saw you ask if i was around Jul 15 01:32:50 <|Phocion|> hey all Jul 15 01:33:03 hello Jul 15 01:34:07 howdy Jul 15 01:36:48 hey so michaelnovakjr_ I did a logcat when I could recreate Android hanging at startup Jul 15 01:37:21 got the output? Jul 15 01:37:26 "W/SurfaceFlinger (517): executeScheduleBroadcasts() skipped, contention on the client. We'll try again later...." Jul 15 01:37:34 just keeps saying that Jul 15 01:38:46 what does it say right before that? Jul 15 01:39:46 nothing that i can see that strikes me as importanty Jul 15 01:40:03 do you have the whole output? Jul 15 01:40:14 "Granting new permission com.google.android.maps.permission......." Jul 15 01:40:27 yeah 1 sec Jul 15 01:41:21 ah here we go Jul 15 01:41:23 http://markmail.org/message/anybxyarwitlofsr Jul 15 01:41:28 looks like similar problem Jul 15 01:42:58 interesting Jul 15 01:44:06 yeah thats pretty much the same thing im seeing Jul 15 01:44:10 endless "We'll try again later..." Jul 15 01:44:24 and its hit or miss Jul 15 01:44:24 have to attempted the adb kill-server? Jul 15 01:44:28 yeah Jul 15 01:44:42 work? Jul 15 01:44:57 nope Jul 15 01:45:23 od Jul 15 01:46:24 anyone play around with gps on android? **** ENDING LOGGING AT Tue Jul 15 02:59:56 2008