**** BEGIN LOGGING AT Thu Sep 04 02:59:57 2008 Sep 04 03:29:11 chrome is going to be part of android.. cool! Sep 04 03:31:02 havent they got a browser already? Sep 04 03:31:13 yeah, based on webkit Sep 04 03:31:29 whats chrome for then Sep 04 03:31:50 you get chrome in desktop/mobile Sep 04 03:31:58 so your apps would work seamless Sep 04 03:32:53 hmpf Sep 04 03:33:14 i though theyll just call androids browser chrome Sep 04 03:33:22 not yet Sep 04 04:03:59 yo muthu jerkface03 Sep 04 04:04:15 still around? Sep 04 04:04:28 trigatch4: yo Sep 04 04:04:31 hey dude Sep 04 04:04:35 you download chrome? Sep 04 04:04:51 waiting on linux Sep 04 04:05:09 Proof that AndroidForums will succeed once the first handsets come out.... Sep 04 04:05:16 http://ChromeSpot.com Sep 04 04:05:41 yeah just registered Sep 04 04:05:57 remembered androidforums suddenly Sep 04 04:10:07 what? Sep 04 04:10:12 just registered for which? Sep 04 04:10:23 your forum Sep 04 04:10:42 Which one? Sep 04 04:10:44 ;) Sep 04 04:10:50 how many you have? Sep 04 04:11:02 Thats a secret Sep 04 04:11:08 hehe Sep 04 04:11:13 then let me keep my secret Sep 04 04:11:21 What secret do you have? Sep 04 04:11:37 registered for one of your forums Sep 04 04:15:41 hahahah Sep 04 04:15:58 google chrome is sick and i already cant wait for the android version Sep 04 04:16:17 i think they'll be able to do a lot of cool things with syncing your phone browser to your computer browser, bluetooth and all that jazz Sep 04 04:16:50 chrome + gears Sep 04 04:16:54 will do all the magic Sep 04 04:20:18 muthu can you tell me if this is spam or real? http://www.androidforums.com/showthread.php?t=254 Sep 04 04:22:19 did you extract the rar? Sep 04 04:22:39 no Sep 04 04:22:41 me = scared Sep 04 04:24:02 i was wondering whether or not it was possible, based on what it claims Sep 04 04:24:15 should be Sep 04 04:24:25 0.9 ships with a few tones Sep 04 04:24:35 may b he's providing more Sep 04 04:26:37 no... i think he is just providing a link to download only the tones Sep 04 04:30:55 this is amazing Sep 04 04:30:59 how many people have asked you if you have tried chrome 'yet'? Sep 04 04:31:08 so far 5 people have asked me today ...and it came out yesterday Sep 04 04:32:09 its all the rage Sep 04 04:32:12 its only in beta Sep 04 04:32:19 i wonder what will happen when/if google promotes it Sep 04 04:32:32 apparently a techcrunch article said it already had 2% global market share Sep 04 04:33:28 heh yeah Sep 04 04:33:29 tried it Sep 04 04:33:32 used it for 3 minutes Sep 04 04:33:33 closed it Sep 04 04:33:54 i already have enough trouble syncinc 4 browsers Sep 04 04:34:05 FF/Opera in win32 and linux Sep 04 04:34:25 theres an article on slashdot that says.. according to clicky, google chrome has 3% of the web browser market Sep 04 04:34:30 rofl Sep 04 04:34:45 yeah Sep 04 04:34:47 insane lollers Sep 04 04:34:48 after day 2 of a beta for windows only Sep 04 04:34:56 that is insane Sep 04 04:35:27 i'm keeping firefox Sep 04 04:35:37 i just switched to swiftfox 3.0.2pre Sep 04 04:35:39 uninstalled all other browsers.. in anticipation of chrome Sep 04 04:35:58 probably a lot of people trying chrome, but 3% is a bit high Sep 04 04:35:58 because the ubuntu packaged 3.0.1 firefox.. likes to crash randomly Sep 04 04:36:01 thats 1% greater than 5 minutes ago...at this rate the market will be conquered by Chrome in 8 hours Sep 04 04:36:12 hehe Sep 04 04:36:14 also that might have been many tech-sites Sep 04 04:37:24 i know quite a few noobs who have tried it .. its almost as if google has tapped into this subconcious need for new web browser Sep 04 04:38:02 how do you manufacture excitement like that Sep 04 04:39:14 you can, if you have a lot of money in the bank Sep 04 04:45:35 yeah but they're not giving the money to media or marketing people Sep 04 04:45:46 they dont have to Sep 04 04:45:49 its because people despise IE Sep 04 04:45:57 and Google is somewhat of the "it" thing Sep 04 04:46:07 the average person doesn't even know what FireFox is... Sep 04 04:46:18 they think its some hackerish linux version of a web browser Sep 04 04:46:34 have you ever had a conversation in like... a grocery store and said "ubuntu" to somebody? Sep 04 04:46:42 try it.. they'll look at you like you havae 23523 heads Sep 04 04:48:06 Chrome is going to be awesome once it comes out of BETA... until then I can see why people would keep FF... I use Chrome for regular browsing and FF for web dev work Sep 04 04:48:35 haha.. ubuntu Sep 04 04:49:01 in ApiDemos, there's a servicebinding and a servicecontroller example Sep 04 04:49:04 muthu.. did you unzip that ringtone folder? Sep 04 04:49:10 it is all mp3 to me and works fine Sep 04 04:49:20 trigatch4: looks like a valid link Sep 04 04:49:24 but i am wondering if someone can confirm this is what is pre-packaged on android? Sep 04 04:49:51 there are 32 ringtones Sep 04 04:49:57 give the forum users a chance to rate etc., Sep 04 04:50:04 so spams can be voted down Sep 04 04:50:27 is there a preference between binding and controller? Sep 04 04:51:04 if i need a service, do i follow the remoteservicebinding or the remoteservicecontroller example? Sep 04 04:51:16 jasta: the android guru Sep 04 04:51:24 has he yawned yet? Sep 04 04:51:26 lol Sep 04 04:52:03 looks like the binding also registers the callbacks Sep 04 04:55:16 hmm, the controller calls start and the binding calls bind Sep 04 04:55:20 that's the only diff i see Sep 04 04:56:18 oh ok Sep 04 04:56:26 bind does not call onstart Sep 04 04:57:54 a few design decisions in android is ambiguous Sep 04 05:35:09 muthu: depends on what you're looking for, you could follow some code in five-client if you want. Sep 04 05:35:22 specifically, i have a ServiceActivity class which you might find useful. Sep 04 05:35:38 its for critically binding an activity to a service. Sep 04 05:38:05 muthu: and yes, this part of Android is in my opinion quite ambiguous. the ApiDemos are not practical at all Sep 04 05:38:42 it took a lot of experimentation to figure out the proper way to do this. see activity/SourceList.java in five-client for some hints as to how this can be done well. Sep 04 05:42:01 that activity also demonstrates my StatefulListView widget, which offers a very efficient way to attach UI state to a list backed by a cursor Sep 04 06:26:12 you know what would be a great Android app i just thought of... Sep 04 06:26:15 a stargazing aid. Sep 04 06:26:38 a program to map the star landscape for your location and given time, or other arbitrary settings. Sep 04 06:26:45 and to show you what's neat to look at, etc. Sep 04 06:26:55 when a good time would be to see upcoming events like eclipses, meteor showers, etc. Sep 04 06:27:22 that would be a really great tool. Sep 04 06:27:28 actually you can make this even better Sep 04 06:27:45 like stellarium for linux, but more of a guide feature added to it Sep 04 06:27:50 if you have a device that supports compass, GPS and orientation sensors, you can find what you're looking at :) Sep 04 06:28:29 the basic premise is not so difficult (it is a well solved problem with lots of public data). the task would simply be miniaturizing it all some fare bit. Sep 04 06:28:50 and it would be a very practical application for a mobile device. Sep 04 06:28:55 i would absolutely love something like this. Sep 04 06:31:09 it could even become really popular among lay persons if it was made more like a tour guide of the night sky as well as a general mapping utility Sep 04 06:31:23 i just did a quick google search and the crop of apps like this for existing mobiles are disappointingly basic. Sep 04 06:35:26 romainguy: i looked at buying a new, modern telescope a while back and they have really gotten incredibly advanced. Sep 04 06:35:49 romainguy: you can orient them with a simple calibration exercise and they will pick what's interesting to look at and automatically move the telescope to look at it Sep 04 06:36:22 so you get a menu on the telescope and you could pick, say, Mercury, wait a moment, and look through the telescope at Mercury. Sep 04 06:36:25 incredible. Sep 04 06:36:27 whats the name of that app to stitch photos together? that+your phone camera is your sextant + System.Time...voila GPS calibration utility/error calculator. Sep 04 12:02:31 1 big service, or many small service - is there a recommended practice here? Sep 04 12:03:53 1 big service will be easier to handle Sep 04 12:04:17 right, but smaller services might be more flexible Sep 04 12:05:20 maybe you can change to smaller services later if you need to. Sep 04 12:05:36 good idea Sep 04 12:20:24 start with one big service, but define multiple interfaces. Sep 04 12:20:42 so there wont be much work later when splitting it up Sep 04 12:21:14 in the apidemos, there's primary and secondary Sep 04 12:22:29 two interfaces exposed via a single interface Sep 04 12:22:36 not sure if this is a good design Sep 04 12:23:48 oh Sep 04 12:24:01 its a single service with 2 service connections Sep 04 12:25:57 got it Sep 04 12:26:53 the example demonstrates how you can nicely group your related services Sep 04 12:37:00 is there a plan for using languages other than java for android progrmming? Sep 04 12:38:45 there are two -- near term, one could write other front-ends that would generate .dex output for different languages. long term, supporting different runtime environments is doable, partially due to the platform security model being based around linux processes and process boundaries, not the VM itself Sep 04 12:40:28 this is a great move, to pull all kinds of developers into android Sep 04 12:48:00 I mainly use objective caml, c, c++ which seems like I might need to wait a long time before these languages are supported. "pull in all kinds of developers into android" - I'm loooking for some sort of time table Sep 04 14:04:18 any example for AliasActivity? Sep 04 14:19:03 did something change with Notification? I'm creating a notification like: Sep 04 14:19:03 Notification notification = new Notification(R.drawable.icon, "testing 123", System.currentTimeMillis()); Sep 04 14:19:19 nm.notify(NOTIFY_ID, notification); Sep 04 14:19:37 and getting "09-04 10:11:18.251: ERROR/AndroidRuntime(193): java.lang.IllegalArgumentException: contentView required: pkg=com.tunewiki.lyricplayer.android id=39340944 notification=Notification(vibrate=null,sound=null,defaults=0x0)" Sep 04 14:20:34 setLatestInfo ? Sep 04 14:20:53 // Set the info for the views that show in the notification panel. Sep 04 14:20:53 notification.setLatestEventInfo(this, getText(R.string.remote_service_label), Sep 04 14:20:53 text, contentIntent); Sep 04 14:21:41 ok...guess that's not optional anymore huh Sep 04 14:21:50 looks like it Sep 04 14:22:21 thanks Sep 04 14:22:41 how you been zhobbs? Sep 04 14:22:47 are you back? Sep 04 14:22:51 yep Sep 04 14:22:57 huh Sep 04 14:22:57 doing well Sep 04 14:23:07 getting everything ported to 0.9 Sep 04 14:23:08 was waiting for you in chennai ;) Sep 04 14:23:12 ha Sep 04 14:37:33 hmmm, not sure why we need PendingIntent in addition to Intent Sep 04 14:38:03 just use PI Sep 04 14:38:41 http://www.ariwriter.com/2008/09/top-260-successful-and-outstanding.html Sep 04 14:38:59 33-34. Michael Martin … Google Android Phones Latest News + gPhone Information MTV Real World Road Rules News Sep 04 14:39:14 wow! a google blog and an mtv blog Sep 04 14:39:37 zhobbs: Are you asking what purpose PendingIntent serves in general? Sep 04 14:40:13 tomgibara: yeah Sep 04 14:40:49 You use it when you want another process to invoke an intent with the privileges of its creator Sep 04 14:41:06 (there may be other uses too) Sep 04 14:42:42 I just never thought about privileges and Intents...seemed like anyone can call any Intent and it would work Sep 04 14:44:26 hmmm, IntentFilters have authority fields now Sep 04 14:58:26 is there a plan for using languages other than java for android progrmming? Sep 04 15:01:04 Proteus: Not outside of the Dalvik VM, no. Sep 04 15:01:32 unfortunate Sep 04 15:02:12 secure Sep 04 15:02:19 not necessarily Sep 04 15:02:55 Proteus, you could always code raw dalvik ... Sep 04 15:03:52 gambler, I'm considering porting the ocaml-java project to get objective caml bytecode to run on dalvik Sep 04 15:04:34 Proteus: Yes, nothing is necessarily secure - I think one of the main reason for Dalvik is keeping memory usage down Sep 04 15:04:39 what does it target now - jvm? Sep 04 15:04:53 yeah, ocaml-java targets the jvm Sep 04 15:06:39 I imagine ocaml would be a fit for mobile apps - much closer to the performance of C, but properly typesafe Sep 04 15:09:55 tomgibara, that's my idea. the language is designed so that so many of the things the jvm/dalvik security has to check for just aren't issues because they're caught at compile time. Furthermore, I like the assurances I can get from the ocaml compiler/debugger that my code is robust and will perform just as I intend it to (ideally). Finally, it allows for much higher level programming when it's useful but is far less verbose than java. Sep 04 15:10:36 It just seems like a good fit for a mobile app and, obviously, my skills are biased towards ocaml Sep 04 15:12:15 java has more extensive libraries. combining the strengths of both is attractive Sep 04 15:12:29 The guarantee that the OCaml compiler offers are not that much better than Java's (but better than C's, that's true) Sep 04 15:12:43 similar to how you can link ocaml code directly to c Sep 04 15:13:49 I think the guarantee is superior, but I'm more interested in higher levels of abstraction and less verbosity. Sep 04 15:14:18 Less verbosity is not always a good thing either... (see Haskell :)) Sep 04 15:14:24 haha Sep 04 15:14:48 ocaml is a very practical language Sep 04 15:14:54 haskell is a religion Sep 04 15:15:01 Fair enough Sep 04 15:16:47 ocaml is a very good language, it allows a higher degree of abstraction than Java and performs well Sep 04 15:17:00 The lack of unicode in the core language winds me up though Sep 04 15:17:03 what I mean is that haskell _enforces_ a way of thinking about coding and about solving problems whereas ocaml is totally mulitiparadigm. You can write ocaml code that doesn't look that different from any other imperitive language, or you can write elaborate functional code. It's about staying as close to the problem as possible Sep 04 15:17:22 unicode? there's a fine library for that Sep 04 15:17:25 camomile Sep 04 15:17:47 I know, I've used it, but it's not the same as having it in the language Sep 04 15:18:33 I was pleasantly surprised at how well ocaml behaved on the coding challenge (http://beust.com/weblog/archives/000493.html) Sep 04 15:19:01 I guess it's arguable. But the syntax is completely customizable with caml4p Sep 04 15:19:50 wouldnt ocaml with android just look like java or would you alter the syntax? ...to me android is mostly just a fancy gui toolkit Sep 04 15:20:54 hrmm Sep 04 15:21:11 I'm not sure what you mean by 'just look like java' Sep 04 15:23:30 well android is a set of classes with java semantics. what is the advantage of doing it in ocaml? Sep 04 15:25:17 if you mean the way it would interface with the rest of the system then I suppose one would want it to be as easy as possible to go between java codded *stuff* and ocaml coded *stuff*. However, I'd mostly just want to write certain algorithimsm with better clarity or performance in one language or the other. Obviously java is the primary language, so at first I'd just want to be able to call ocaml functions/modules/classes in specific ce Sep 04 15:25:17 rcumstanes, with the overall framework being written in java. Sep 04 15:26:41 there's no reason the android system and gui calls couldn't be just as easily written in OO ocaml as they are currently. Sep 04 15:27:24 I never learned ocaml but someone told me it is similiar to lisp...is that right? Sep 04 15:27:33 noooooo Sep 04 15:28:01 lisp is a totally different beast Sep 04 15:28:26 ah...well ill have to try it. Im pretty rusty on anything non imperative these days Sep 04 15:29:00 http://www.cs.caltech.edu/courses/cs134/cs134b/book.pdf Sep 04 15:29:05 this is a good introduction Sep 04 15:30:22 the syntax can look a rather scary at first but once you get what's going on it's quite fluid and elegant Sep 04 15:31:04 for questions, there's the #ocaml channel and two great mailing lists Sep 04 15:32:19 Ive added it to my reading list... Sep 04 15:36:34 if you're curious about security and ocaml you might want to check out the Coq (pronounced like you think, but it's just the word for rooster in french, and it was written in france) theorom assistant. ocaml code can be formally verified with _relatve_ ease because of its design: ocaml is based on the mathematics of the typed lambda calculus. Java, of course, isn't based on any formal system. Sep 04 15:37:23 I studied there (INRIA) Sep 04 15:37:34 hey, cool Sep 04 15:38:26 And yes, the people who named Coq probably didn't master English very well :) Sep 04 15:38:40 hahahaaha Sep 04 15:39:01 Better than naming it clit Sep 04 15:39:07 well, the rooster is a national symbol France. Sep 04 15:39:14 yup Sep 04 15:39:37 it's interesting we pronounce the word the same, spell it differently, but it referrs to the same animal Sep 04 15:44:42 since android is going to be using java for a while though, I have a few questions. First, I've got the Core Java books - is this a good place to start? Second, I want to do real time sound synthesys in java, on android platforms - where can I find more out about doing this in java and are the current and near-future android phones going to be able to handle the computation? Third, doing 2d/3d graphics and animation with openglES from ja Sep 04 15:44:42 va - are there tutorials on the web that would help me do this with the restraints of an embededd system? I've done a good deal of 3d modeling and even some game programming, but I don't know how to account for the severe processor restrictions Sep 04 15:48:40 basically, I'm looking for help getting started. Sep 04 15:49:08 Proteus: I hear the Thinking in Java book is a good starting point for Java but I haven't looked at beginner's book in a long time Sep 04 15:49:27 you can dive into android and then slowly learn the java concepts as required Sep 04 15:50:00 Proteus, You said before you code C++ right? Sep 04 15:50:03 Proteus: http://javapassion.com/ Sep 04 15:50:16 Ah if you already know C++ then learning Java will be trivial Sep 04 15:50:19 Cedric2, I'm not exactly a _beginner_ I don't need to be hand held through what a loop does, I'm just a beginner with java. Sep 04 15:51:04 java is just C++ with fewer features and you dont have to worry about memory mgmt. Sep 04 15:51:09 as much... Sep 04 15:51:34 Yes if you already know C++, you probably don't even need to get a book to learn Java Sep 04 15:52:30 I used to do c++ in class, but that's about it. I'm trying to relearn it while I work with Ogre3D - so, I think I really do need some help learning java Sep 04 15:53:42 I'm also really concerned with getting some pretty math heavy code to run without hiccups on an actual phone Sep 04 15:58:05 incidentally, I've read about aspectJ and I'd like to try it out in an serious program. Google isn't using the usual JVM, am I going to run into problems with incompatable tools/libs? Sep 04 15:58:29 yes... Sep 04 15:58:54 is there a list of stuff to avoid? Sep 04 15:59:00 Yes, you will have a very hard time adapting something like AspectJ to dex Sep 04 15:59:15 Let me rephrase that: it might not be hard, but it's a lot of work :) Sep 04 15:59:32 huh. too bad. Sep 04 15:59:50 i looked into this recently for a little translator I have written using antlr. I wanted to generated aspect J Sep 04 16:00:09 the problem is that it generated JVM bytecode instead of java source Sep 04 16:00:09 The first step would be to implement a dex bytecode weaver Sep 04 16:00:47 you can decompile it back to java source but its really messy, its a feature of the aspect bench compiler (abc) Sep 04 16:01:12 yikes. Sep 04 16:01:30 or you can try to get the runtime dep running on android and change your workflow to suit it Sep 04 16:01:59 Proteus: at any rate, this kind of project doesn't strike me as a good Java starter project :) Sep 04 16:02:10 i think the original version of aspectJ (like v1.0) was able to compile to source. I so far havent tried it yet. Sep 04 16:02:10 haha, yeah. Sep 04 16:02:34 also the aspectJ CVS repository is very poorly maintained in my opinion. Sep 04 16:02:35 There are probably simple weavers out there that would work fine Sep 04 16:03:05 If you end up playing with it, let me know because I really like the AspectJ approach, although I dont love the implementation. Sep 04 16:03:39 will do Sep 04 16:03:53 but like Cedric2 said, it's really not a starter project Sep 04 16:04:45 yah probably not Sep 04 16:04:54 out of curiosity, what sort of stuff are you guys developing for android? Sep 04 16:05:14 I work *on* Android Sep 04 16:05:27 aah Sep 04 16:05:37 cool Sep 04 16:06:01 Im just working on hello world 0.1 Sep 04 16:06:50 Proteus: http://code.google.com/android/adc_gallery/ Sep 04 16:08:49 muthu, when did they post that page? Sep 04 16:09:24 tethridge: its been like a week Sep 04 16:10:20 muthu, that gives a pretty good overview, thanks. Sep 04 16:10:34 welcome Sep 04 16:10:55 I'm surprised that tunewiki isn't available for the iphone Sep 04 16:13:34 it seems like gocart and compare everywhere are the same concept Sep 04 16:13:58 similar, yes Sep 04 16:15:31 Cedric2, perhaps you'd be best to ask about what I can expect in terms of performance restrictions in the near-mid term future? from what I've heard, the first crop of phones aren't too powerful and use the most primitive form of opengles (I think). Perhaps we could do a priviate channel and I can tell you exactly what I want to do and you can tell me if it's flat out impossible? Sep 04 16:16:57 You're right about OpenGL ES but I'm afraid I can't comment more about the hardware part (mostly because I don't know much, to be honest) Sep 04 16:18:03 meh, optimizing even for embedded is a dying art. qualcomm are shipping 1ghz mobile phone processors now Sep 04 16:18:24 gambler: you say that as if it's a bad thing... Sep 04 16:18:45 I still have scars from the early J2ME days where we put everything in a single class to save bytes and made sure all the classes were in the default package Sep 04 16:18:51 just the facts ma'am Sep 04 16:19:34 yes...I wrote a useless gym and diet tracking J2ME app. Even that was rather sluggish. I hope someone writes a better one for android. Sep 04 16:19:56 Proteus: the first phones will ship with OpenGL ES 1.0 and some of 1.1 Sep 04 16:20:22 I posted a message to the android-security-discuss group more than two days ago and it's still not appeared - is anyone actively moderating that group? Sep 04 16:21:05 tomgibara: can you try to resend it? Sep 04 16:21:05 romainguy__, thanks. Sep 04 16:21:20 Cedric2, I think we are on the cusp of a personal computing revolution and I cant wait to throw away my notebook computer Sep 04 16:21:41 gambler: I suggest you keep it around for a while longer, there is something to be said for mice, full size keyboards and big screens Sep 04 16:21:51 Proteus: as for the power aspect of your question, I have played with phones that are considered powerful, and it doesn't give you much Sep 04 16:22:39 Cedric2, in my vision we just cruise up to the robust physical terminals sprinkled everywhere and we get that secure VNC into our phone Sep 04 16:22:41 romainguy, that's disheartening Sep 04 16:22:46 romainguy_: I'll have to do it later, I'm off out now - it was just seeking clarification on what security mechanisms are in place when the phone is answered from a locked state. Sep 04 16:22:51 Proteus: it depends on what you want to do Sep 04 16:23:05 Proteus: remember the early videos of Android we published, we showed Quake 1 running Sep 04 16:23:14 but you won't get much more than that Sep 04 16:23:50 morning Sep 04 16:24:14 hi San Sep 04 16:24:28 aaahhhh, I know - I'd really like to get some feedback on how feasable my projects are with android and the current crop of phones without throwing away my ideas to another app guy (I'm a dirt poor student looking to sell apps to get some kind of an income) Sep 04 16:24:31 yo SanMehat Sep 04 16:24:31 slacking today? :)) Sep 04 16:24:35 gambler: how you doing ? Sep 04 16:24:47 romainguy__: no i'm sick.. got a nasty cold so i'm working on design docs from home :) Sep 04 16:25:07 what's your idea? Sep 04 16:25:35 you work for goole? Sep 04 16:25:44 google Sep 04 16:25:47 ya Sep 04 16:26:37 romainguy__: prv Sep 04 16:27:22 design docs? cool Sep 04 16:27:22 that said today I'm working on documentation Sep 04 16:27:22 Proteus: yes I work on Android Sep 04 16:27:40 would either/both of you give me a min in a private channel? Sep 04 16:28:06 can't you ask your question here? Sep 04 16:28:39 I...guess Sep 04 16:28:55 needing to pay for tuition makes you paranoid Sep 04 16:29:03 sorry Sep 04 16:37:04 Yeah, whatever it is you can ask here. Sep 04 16:37:35 I have three apps planned to take me from novice andrioid programmer to quite advanced. One of my main interests is music and music composition. My overarching goal is to take mobile devices and enable them many of the functions that only musicians working on a desktop would have and many that could only be done with a mobile device. The first is really just to get me going, I doubt anyone would buy it, though people went nuts over that i Sep 04 16:37:35 Brue app. Whatever. Basically I just want to build something like a theramin in java, tied in nicely with the andrioid UI. Tilt sensors and accelerometors manipulate the functions of a relatively simple software synthesizer in real time. Give the user the ability to save their composition or send it via 3g/wifi to their computer (I was told the storage abilites of the early phones are ppretty meagre Sep 04 16:41:42 whoa, I've never seen a theremin before, that's crazy Sep 04 16:43:38 Proteus: that app should be feasible Sep 04 16:49:50 My second app is more ambitious. I'm also a zennist and I had the idea of creating a set of minigames whose purpose isn't to stimulate or challenge the user, but to relax. My target market for this mainly commuters at the end of the day trying to wind down and I suppose it might appeal to anyone needing to chill. This app takes the first little synthesizer and adds visuals and input. Mini sound-visual enviroments would include perhaps a z Sep 04 16:49:51 en rock garden where the placement of the big stones and their spatial relationship sets up the foundation of the sonic environment while raking through the sand - at different speeds, rotations, etc- would add futher layers to the ambient sound scape. Kept interesting via something like markov chains to improvize ineresting flourishes and maybe more (I've read papers on using MC for improvizing complex jazz compositions, so I know it's f Sep 04 16:49:51 easable). This app would also include other interactive ambient soundscapes. For instance drawing sound with colors and tilts and picking different brushes to add different types of sound to the mix. A tool that played meditation bells with configurable times would be easy to add with the sound engine already there I have more ideas, but you get the picture. Sep 04 16:50:42 Proteus: er... wow :) Sep 04 16:51:21 The procedural synthesys is quite a bit more complex with this one Sep 04 16:51:40 sound synthesys, I mean Sep 04 16:52:01 this is where I start wondering whether the phone could handle it Sep 04 16:52:47 that I have no idea Sep 04 16:55:19 app2 has simultaneous graphics (possibly animated), multiple sound sources being mixed, and the real-time generation and manipulation of multiple sound sources in real time Sep 04 16:57:05 proteus: reminds me of electroplankton for the ds, great game! Sep 04 16:57:10 how is all this sounding so far? Sep 04 16:57:17 never heard of it. Sep 04 16:57:26 u should check it out Sep 04 16:57:32 It's sounding very ambitious but fun Sep 04 16:57:33 would love to see more games like that Sep 04 16:58:09 http://www.youtube.com/watch?v=d3v6npP8OZk Sep 04 16:59:35 well, I'd be prototyping all the sound synthesis algorithms in Max/MSP or PD first, which cuts the amount of time required by an enormous amount Sep 04 17:01:10 http://www.ddj.com/mobile/210300551 Sep 04 17:01:43 krosaen: proteus: reminds me of electroplankton for the ds, great game! << I was about to say the same :)) Sep 04 17:19:17 my third app is really ambitious, but I think it's actually revolutionary. I expand on what I've learned, code I've written, from the first two to build a kind of ultra-minimalist digital audio workstation, composition, and the featue that counts: collaborative composition and performance. The phone mic is used as a normal microphone, you can record whatever and then use the touch interface to chop up your samples, whatever. I also add t Sep 04 17:19:17 he sort of functionality you'd see in max or pd - you can take your samples and run them through vocoders, noise, other distortions and then eventually organize your creations into 'instrument' or 'sample' objects. These objects contain all relevant data and can be shared between other phones or to a server somewhere for storage. This app also includes synthesizers - which can be modifed by the user - and a simple sequencer. The main inte Sep 04 17:19:18 rface is where you actually work on compositions. The note interface intuitively works with the touch pad by using a melodic thesauraus - you tap a sqeunce in a melody and rings appear around where you tapped showing increasing complexity in outer rings and sesnible transformations selected by direction, each selection is played to get a feel for how it sounds. It would be a pain to compose whole songs on the tiny screen, the focus is on Sep 04 17:19:23 melodies, rythmic patterns, melodic segues, riffs, etc. Finally, you can either compose alone or transfer the 'instrument', 'loop', 'sample', and/or 'whatever' objects between the phones of a bunch of friends and using a stripped down interface inspired by desktop software designed for live performance, you can hold a jam session with your friends. Latency shouldn't be a big issue over a wifi connection, I don't know about 3g, but the go Sep 04 17:19:28 al is to allow people to compose and perform wherever they are in the world.Sort of a modern drum circle. Sep 04 17:22:02 this one doesn't absolutely require real-time audio synthesis until the android drum circle thing Sep 04 17:22:56 Cedric2, romainguy_, thoughts? Sep 04 17:24:12 is this too crazy? I'm hoping that app2 would give me enough money to maybe hire someone to help, but who knows. Sep 04 17:32:51 does anyone have comments? Sep 04 17:33:07 sounds like a big plan Sep 04 17:34:03 it basically requires hiring at least one or two people to help Sep 04 17:34:31 sounds cool, sort of like pro-tools or garage band on the go? Sep 04 17:34:41 sort of Sep 04 17:35:08 but with the addition of real time collaboration/performance Sep 04 17:37:08 realtime collaboration would be tough i'm guessing; getting each participant in sync to experience the same rythm as well as what others are playing is a hard problem Sep 04 17:43:16 if you can get the latency low enough it's not an issue. There are a lot of ways you could do it. First of all, you don't need split second precision if you set interface up so that what people see is a building queue of icons for their sounds getting layered and then once they feel far enough ahead they go 'live', but they're still not receiveing data which is super time critical, just updates to the queue and the current live output, w Sep 04 17:43:16 hich doesn't need to be in perfect sync Sep 04 17:44:39 I'm thinking of a very high level interface for the collaborative part of it. Sep 04 17:46:28 if you have a delay, active collaboration might be difficult Sep 04 17:46:28 remember, they've worked together (maybe) building their sonic library and at least share the same data, so they're not manipulating much low level stuff by the time they go into jam session mode Sep 04 17:47:59 most electronica musicians, when they're improvising, have a decent head start Sep 04 17:48:22 but it depends on how you want to go about it Sep 04 17:48:52 if you can link the android phones through their wifi connections, latency isn't much of an issue Sep 04 17:49:20 collaboration also doesn't have to be simultaneous Sep 04 17:50:19 you work on something for a few hours, then shoot what you've got via 3g to your partner on the other side of the planet, he tweaks, deletes, adds things, then shoots it back, etc. Sep 04 17:51:52 you collect the sounds of cars and cafes and street-corner preachers, he collects the sounds of forest birds Sep 04 17:52:36 you slice and dice and maybe come up with something interesting. only your studio is with you all the time and the process is much more fluid Sep 04 17:54:23 its a neat idea Sep 04 17:54:36 on the other side I'd set up a web back end for people to meet up and collaborate and share their work Sep 04 17:58:33 one problem would the quality of headphone jacks and microphones Sep 04 17:59:43 assuming it has a headhpone jack. I'm not sure I've seen an android phone that had one - I hope I'm wrong Sep 04 18:01:49 Proteus: good luck Sep 04 18:02:00 ha. thanks Sep 04 18:21:04 new app called Photostream in apps-for-android today: http://bit.ly/android-photostream Sep 04 18:59:37 hello Sep 04 19:25:06 so, java.beans doesn't seem to be implemented...is there any reason why I wouldn't be able to implement this myself? I want to make sure before I start :) Sep 04 19:25:26 you want to add support for java beans? Sep 04 19:26:23 yes Sep 04 19:26:25 i thought beans were one of those 'oh god don't ever use this crap its so 2%@% slow and bloated'.. at least thats what i'd heard.. Sep 04 19:26:41 I want to implement it because a library I'm using supports it Sep 04 19:26:52 supports it or needs it? Sep 04 19:27:01 needs it I mean Sep 04 19:27:06 then use a different library? Sep 04 19:27:17 do you have an alternative to the smack library? Sep 04 19:27:31 don't you try to smack me Sep 04 19:27:36 with the same license Sep 04 19:27:39 heh Sep 04 19:27:39 * SanMehat grumbles and goes back to his feverish sleep Sep 04 19:27:45 <-- sick Sep 04 19:27:51 that stinks Sep 04 19:28:25 herriojr: smack worked in m5 didn't it? Sep 04 19:34:25 yeah, it did Sep 04 19:34:43 but the library uses java.beans which isn't supported in 0.9 Sep 04 19:34:56 java.beans was in m5? Sep 04 19:35:08 maybe I grabbed the wrong source? Sep 04 19:35:36 i would be very surprised if m5 supported beans.. Sep 04 19:35:42 yeah same here Sep 04 19:35:47 it must have been, because it is dying on PropertyDescriptor descriptor = new PropertyDescriptor(name, objectClass); Sep 04 19:36:04 and the same library jar worked in m5 Sep 04 19:36:16 * heat jumps up and down on top of san's office Sep 04 19:36:54 go ahead, i'm lying on the couch Sep 04 19:36:57 :P Sep 04 19:37:08 your domain does not stretch to the place above my livingroom! Sep 04 19:37:45 so I guess I should try and clean all bean references out of the smack source code Sep 04 19:38:28 hehe Sep 04 19:38:30 damn it! Sep 04 19:38:44 or use a different IM lib? Sep 04 19:39:00 down with the beans.. Sep 04 19:39:27 I'll just remove it if I can...there really are only about 3 lines with references to it Sep 04 19:40:05 I just don't know exactly what PropertyDescriptor is :) Sep 04 19:40:40 wait Sep 04 19:41:16 ok Sep 04 19:41:24 oh i dont know what it does either Sep 04 19:41:27 heh Sep 04 19:43:38 "A PropertyDescriptor describes one property that a Java Bean exports via a pair of accessor methods." Sep 04 19:43:49 Clear as mud. Sep 04 19:44:03 i dont like beans.. they make me fart :( Sep 04 19:44:13 fadden: it's the old school way of providing extra information beyond reflection Sep 04 19:44:21 we would do that with annotations or XML in the 21st century Sep 04 19:44:39 but Graham wrote this spec almost 10 years ago now Sep 04 19:45:26 time flies :( Sep 04 19:45:35 heat: oh *thats* why a lot of java programmers say 'java beans smells' :) Sep 04 19:46:10 Like we say in French, "It's beginning to smell of pine" (a reference to the wood used to make coffins) Sep 04 19:46:18 ;) Sep 04 19:50:24 haha Sep 04 19:54:04 hi, will the first android phone support wifi like the nookia n95? Sep 04 19:54:57 no, but it will support wifi Sep 04 19:55:07 not wifi *like* the n95 Sep 04 19:55:41 but you seem like a consumer? Sep 04 19:56:43 yeh i'm just asking, if i wasn't clear; i'm talking about connecting to my router Sep 04 19:57:16 f00f-: Sep 04 19:57:33 yeah then it should work, but no details yet Sep 04 19:57:50 first device wont be that technically great i think Sep 04 19:57:58 lots of limitations due to android api's not stabilizing Sep 04 19:58:33 romainguy_: Photostream is cool, thanks Sep 04 19:59:41 zhobbs: glad you like it Sep 04 19:59:49 btw how do i turn on smooth scrolling on chrome? Sep 04 19:59:52 yeah, will study it a bit Sep 04 20:01:05 and what's bigger, the iphone or the htc? Sep 04 20:03:25 hey wheres jasta? its been somewhat quiet in here Sep 04 20:09:11 so sanmehat, are you an advocate? what so you work on? Sep 04 20:09:30 f00f-: I work on the kernel / system bits Sep 04 20:09:53 SanMehat: Jasta went to ireland today. or is going. Sep 04 20:09:56 awesome, i hope we get a good firewall in there Sep 04 20:10:07 Dougie187: :) Sep 04 20:11:04 I would have expected him to turn off IRC, but i guess he might want to see how much people talk about him? Sep 04 20:11:09 or he hasn't left yet. Sep 04 20:11:12 probably Sep 04 20:11:14 jasta smells Sep 04 20:11:26 maybe he uses java beans Sep 04 20:11:30 lol Sep 04 20:11:41 I doubt it. Sep 04 20:11:46 he'll probably be ssh'ing into his irc from irland Sep 04 20:12:15 I would think his lady friend would get irritated by that. Sep 04 20:12:22 plus i think he intends on being drunk most of the time. Sep 04 20:16:17 herriojr: try adding android:name="android.awt" /> Sep 04 20:16:24 to your manifest Sep 04 20:24:55 anyone get my questions? Sep 04 20:54:52 hello ? Sep 04 20:55:19 hi Sep 04 20:55:41 Im comparing phones Sep 04 20:55:51 the blackberry does have wireless ? Sep 04 20:56:08 I haven't seen a blackberry with wifi, no Sep 04 20:56:40 because it says everywhere wireless access to telco companies Sep 04 20:56:50 They access the phone's wireless network Sep 04 20:56:56 for internet, which is more expensive. Sep 04 20:56:56 but I think that doesn't really mean access to my home network Sep 04 20:56:57 xD Sep 04 20:57:37 ok thanks Sep 04 21:23:44 romainguy_: can you create a shortcut on home from inside an application? Sep 04 21:23:57 yes Sep 04 21:24:14 but the preferred way is to do it the way Photostream does Sep 04 21:24:30 right, will have that option also Sep 04 21:24:55 installing a shortcut directly from an application has two drawbacks: Sep 04 21:25:00 1. it requires another permission Sep 04 21:25:12 2. the shortcut will be placed in the first available slot on Home Sep 04 21:26:25 yeah...was just thinking that if you're watching a video or something would be cool to be able to add shortcut to that video from the menu Sep 04 21:26:51 you can dl that Sep 04 21:26:53 do that Sep 04 21:26:59 it's what Gears does in the Browser Sep 04 21:27:24 ok Sep 04 21:27:46 so you need the permission com.android.launcher.permission.INSTALL_SHORTCUT Sep 04 21:28:25 the Intent to send Home must have the action com.android.launcher.action.INSTALL_SHORTCUT Sep 04 21:28:38 and contain at least the following two extras: Sep 04 21:28:51 Intent.EXTRA_SHORTCUT_INTENT: an Intent to start when the shortcut is launched Sep 04 21:29:01 Intent.EXTRA_SHORTCUT_NAME: the name of the shortcut Sep 04 21:29:18 ah and a third one Sep 04 21:29:25 Intent.EXTRA_SHORTCUT_ICON Sep 04 21:29:45 or Intent.EXTRA_SHORTCUT_ICON_RESOURCE Sep 04 21:29:48 the first one is a Bitmap Sep 04 21:30:06 the second one is a Intent.ShortcutIconResource Sep 04 21:30:12 are resource id's unique to the whole system? Sep 04 21:30:23 yes Sep 04 21:30:30 but it will be converted to a resource name Sep 04 21:30:40 ok Sep 04 21:30:40 for instance: com.myapp.foo:drawable/my_icon Sep 04 21:30:51 that means you should not change the name of the resource :)) Sep 04 21:30:55 how are sendDataMessage's received ? Sep 04 21:31:24 romainguy_: thanks for the help Sep 04 21:31:33 anytime Sep 04 21:31:36 damn, not that easy to use chrome and firefox at once ^^ Sep 04 21:32:25 romainguy_ you know how DataMessages (SMS) ar received? Sep 04 21:32:33 I have no idea Sep 04 21:32:40 k, thx Sep 04 21:33:00 so packing my data to a String should work too... Sep 04 21:33:29 why not use the network +/- Sep 04 21:33:35 eg. a data connection Sep 04 21:33:43 err, i mean, the web Sep 04 21:38:13 foof- I'm writing an SMS-Compressor Sep 04 21:38:18 f00f-: Sep 04 21:42:45 btw how do i turn on smooth scrolling on chrome? and what's bigger, the iphone or the htc? Sep 04 21:43:06 mohbana: chrome isn't on android? Sep 04 21:44:00 mohbana: androidguys.com has a lot of speculative info on the htc phone Sep 04 21:45:27 androidguys has a lot of 'sources' :-D Sep 04 21:45:47 they've generated a lot of traffic I bet though Sep 04 21:48:06 probably Sep 04 21:50:35 romainguy_: I want to rotate an ImageView using a RotateAnimation from its center; by default it rotates from the top-left corner of the view. What's the best way of doing that? Sep 04 21:51:30 from XML or code? Sep 04 21:51:38 tomgibara: android:pivotX="50%" android:pivotY="50%" Sep 04 21:51:42 I'm not fussy Sep 04 21:51:53 form XML, what zhobbs said Sep 04 21:52:05 from code, use public RotateAnimation(float fromDegrees, float toDegrees, int pivotXType, float pivotXValue, Sep 04 21:52:05 int pivotYType, float pivotYValue) Sep 04 21:52:12 with RELATIVE_TO_SELF for the types Sep 04 21:52:38 thanks both of you - I was being dumb and just scanning the class for setters Sep 04 21:59:11 romainguy_: I have another question, but it might be a bit premature because I haven't investigated closely... Sep 04 21:59:20 I'm putting together a simple little game which uses a surface that is continuously updated in a worker thread. The FPS counter shows a healthy 15 fps, but the screen updates are clearly not as frequent. Sep 04 21:59:36 is this your fps counter? Sep 04 21:59:40 yes Sep 04 21:59:45 So my question is: Does the emulator throttle actual updates to the screen? Sep 04 22:00:01 I doubt it Sep 04 22:00:10 you can check the real number of updates easily Sep 04 22:00:25 run this: Sep 04 22:00:27 adb shell service call SurfaceFlinger 1013 ; sleep 10 ; adb shell service call SurfaceFlinger 1013 Sep 04 22:00:32 it will give you two hex numbers Sep 04 22:00:38 subtract them and divide the result by 10 Sep 04 22:00:46 that will be the number of fps as counted by SurfaceFlinger Sep 04 22:00:53 it doesn't get more real than that :)) Sep 04 22:01:08 great, I'll try that out Sep 04 22:06:39 well that cames to 15 Sep 04 22:06:41 *came Sep 04 22:08:48 :) Sep 04 22:08:51 so your eyes are lying :) Sep 04 22:09:01 15 is not much Sep 04 22:09:16 must be :) perhaps my brain's running too fast Sep 04 22:09:19 Or the timing is such that one frame has two updates, and the next has 0 updates. Sep 04 22:09:38 fadden: but SurfaceFlinger indicates 15 fps as well Sep 04 22:09:49 romainguy_: When does SurfaceFlinger fling? When you do an unlockCanvasAndPost ? or continuously? Sep 04 22:10:12 when we do the unlockCanvasetc. Sep 04 22:10:26 and it waits on the vsync Sep 04 22:13:44 In all seriousness, I'm definitely not seeing 15ps on the screen. I can see a simple and slow linear animation clearly judder - I'd estimate 5fps max. I'm still on WinXP for now - I'm wondering if there's something happening between the emulator and the desktop. Sep 04 22:14:22 my SMSCompressor gives already 260 instead of 160 chars :) Sep 04 22:14:26 I'm going to ignore it for now and polish up my code, I'll take a closer look when I've looked for any bugs. Sep 04 22:14:31 plusminus_: what does it do? Sep 04 22:14:52 for now just huffman encoding Sep 04 22:15:30 maybe some word replacement later, if that does not lead to accidential unreadability Sep 04 22:15:31 romainguy_: Thanks for the SurfaceFlinger tip. Sep 04 22:15:46 plusminus_: static or adaptive? Sep 04 22:15:57 plusminus_: what is an smscompressor used for? Sep 04 22:17:22 plusmins_: You can't stop the compressed SMS from appearing in their compressed state within the messages application though, or can you? Sep 04 22:19:52 fadden: static because all clients need exactly the same 'trees' Sep 04 22:20:05 tomgibara: No I can't, system does not allow that Sep 04 22:20:23 the compressed message probably appear as coplete nonsense in the systemapp Sep 04 22:20:59 You're assuming English then? Sep 04 22:21:45 I have not tested 'special-char'-languages for now, but those are also just characters. Sep 04 22:21:57 which would receive their own dictionary then Sep 04 22:22:41 plusminus: You could have a fixed set of 'trees', one for each language, and then prepend a bit sequence indicating the tree used Sep 04 22:23:02 tomgibara: exactly what I'm doing :) Sep 04 22:23:23 I could say great mind think alike but I won't ;) Sep 04 22:23:28 Any transformation that makes some sequences smaller will make others larger. Since you're not dealing with arbitrary inputs, though, it shouldn't be an issue. Sep 04 22:23:43 You should be able to run the input text through every possible dictionary and send it in the one that works the best. Sep 04 22:23:52 That's what I would propose too Sep 04 22:24:18 fadden: yep, its a bit cpu-intense but works smooth, even in the emulator Sep 04 22:24:24 The really interesting question is this: Given a corpus, can you produce a partitioning that provides an optimal set of trees (in the obvious sense) Sep 04 22:24:40 Another idea would be to use splay-tree compression, which adapts quickly; may work well given the short max length of SMS. Sep 04 22:24:59 http://www.picpaste.de/smstretcher.png Sep 04 22:25:21 http://www.picpaste.de/thumbs/smstretcher.1220567104.png Sep 04 22:26:10 sorry for explicit language ;) Sep 04 22:26:25 The neat trick would be to display the number of bytes remaining, and continuously update it as the person types characters in. Sep 04 22:27:05 the "253/257" updates every new character (as I said cpu-intensive, but smooth) Sep 04 22:27:18 its like "used/total" Sep 04 22:27:19 Combining best compression table and updates takes even more processing Sep 04 22:27:47 Shouldn't be an issue for a typical human typist. Sep 04 22:28:05 But the phone battery? Sep 04 22:28:11 what's bigger, the iphone or the htc? Sep 04 22:28:41 Do you generate the output or just compute the length? I would think most of the expense in a Java static Huffman implementation is in the bit-shuffling. Sep 04 22:28:51 All you need is a 256-entry table with the bit lengths, and just sum them up. Sep 04 22:29:06 is google chrome going to be linux compatible? Sep 04 22:29:45 I am computing the 'result' (=encoding) on every new character Sep 04 22:31:13 Perhaps compute the result at each new character, and evaluate "best tree" in the background when the user pauses their typing. Sep 04 22:31:53 yeah, could be a possibility. Sep 04 22:32:12 200 chars should be 200 int array lookups (and 199 additions). You could further optimize by just adding the new char width alone; works for all keys except "delete". Sep 04 22:32:30 Only compute the full encoded form at the end. Sep 04 22:33:34 fadden: Tricky from a UI perspective, since on calculating the result, you many need to inform the user they have typed too much - unpleasant. Sep 04 22:33:49 scrub that comment Sep 04 22:33:57 :-) Sep 04 22:34:08 I forgot it's a static encoding Sep 04 22:34:44 Static Huffman encoding is actually a lot faster than static Huffman decoding. Let me know if you need 65816 assembly code for these. :-) Sep 04 22:35:00 :D probably not Sep 04 22:35:50 I have an efficient java implementation too if you want one :) Sep 04 22:36:36 tomgibara: thanks I'l come back if mine turns out to be too slow Sep 04 22:41:29 trying to get smack to work is a pain in the ass Sep 04 22:41:38 it looks like it might take a bit longer than I thought Sep 04 22:47:57 herriojr: did you find android.awt.jar? Sep 04 22:48:49 no, what's in it? Sep 04 22:49:27 PropertyDescriptor, among other things Sep 04 22:49:31 < heat> herriojr: try adding oh, android.awt is an outside library Sep 04 22:50:27 when adding a uses-library, what should the parent node be? Sep 04 22:50:34 application or the manifest? Sep 04 22:51:32 Looks like it's under "application" in the sample code. Sep 04 22:53:40 omg that works Sep 04 22:54:18 I wish they would document such things Sep 04 22:54:19 ... Sep 04 23:01:19 also - why would it be in the java.awt library and not java.beans? Sep 04 23:01:38 sorry, I'm not usually a java programmer, so it may be a dumb question Sep 04 23:18:12 is Sending SMS between emulator proven to work ? Sep 04 23:19:40 that would be awesome Sep 04 23:20:35 xD it works... Sep 04 23:20:56 your emulator has a phone number? Sep 04 23:21:00 guess what: "mEtTargetPhoneNumber.toString();" is not correct xD Sep 04 23:21:11 use the PORT Sep 04 23:21:15 (like 5556) Sep 04 23:43:38 what's bigger, the iphone or the htc? Sep 04 23:45:04 we'll know in november i guess Sep 04 23:46:54 got a quick question regarding the spinner in 0.9 Sep 04 23:47:19 did they take out the onItemSelectedListener? Sep 04 23:47:46 is there anyway I can take actions based on the selected item in spinner? Sep 04 23:49:16 given how it works now, it doesn't really make sense Sep 04 23:49:32 u mean the spinner? Sep 04 23:49:35 yes Sep 04 23:49:39 wat do u mean Sep 04 23:54:09 can you elaborate a little more, not quite understand what you mean Sep 04 23:56:54 herriojr: its not documented to the reason that its not suported, and not encouraged Sep 04 23:59:28 Hello I'm following the book, Android Essentials, however is out of date! Sep 04 23:59:48 hello ? is there anybody ? Sep 05 00:00:00 no Sep 05 00:00:09 hey! Sep 05 00:00:44 Which good book about android, recomends me ? Sep 05 00:00:50 recomendation Sep 05 00:01:00 Advise me ? Sep 05 00:02:09 npelly, humm ??? Sep 05 00:02:45 heat: so if they plan on not supporting it, then I need to find another way of doing it since it will probably be completely removed Sep 05 00:03:11 sad of me :S Sep 05 00:05:03 yxz97: I'd just use the android documentation Sep 05 00:06:41 herriojr, yea, probably its the better way.. Sep 05 00:07:03 but sometimes is good follow a "Good Book" to get a more general overview about the technology. Sep 05 00:07:12 herriojr, the bug picture Sep 05 00:07:16 herriojr, the big picture Sep 05 00:07:29 I've found that most books don't give a *big* picture or really give you anything useful Sep 05 00:07:39 at least that's how it is for me Sep 05 00:08:21 where are you from ? Sep 05 00:09:05 I'd go to plusminus's site to learn stuff Sep 05 00:09:38 http://www.anddev.org/ Sep 05 00:09:51 well, I don't know if it is really his site, but he posts a lot of tutorials there Sep 05 00:10:45 and it will save you money ;) Sep 05 00:10:51 herriojr, http://www.plusmin.us/ Sep 05 00:10:58 is like mnusic, ??? what ? Sep 05 00:11:35 go to the link I gave you Sep 05 00:11:42 not the one you searched for ;) Sep 05 00:11:47 yea I'm in Thank you heat Sep 05 00:11:49 yea I'm in Thank you herriojr Sep 05 00:11:54 sorry heat Sep 05 00:12:37 hey heat do u know why the onItemSelectListener is taken out for Spinner? Sep 05 00:14:48 chouman82: I told you, because it doesn't make sense with the new spinner Sep 05 00:15:19 instead you should use setOnItemClickListener() Sep 05 00:15:41 the OnItemSelectedListener would fire only if the user moves the highlight around while the choice popup is up Sep 05 00:22:27 Hey - it's true that you can use Java 6 features in your android code base, right? Will the dalvik translator take care of it? Sep 05 00:23:42 fpotter: what features are you talking about? Sep 05 00:23:51 you can only use the APIs that are bundled with the SDK Sep 05 00:24:09 the only language change brought by Java 6 was the ability to put @Override on interface methods Sep 05 00:25:05 Sep 05 00:25:05 A spinner does not support item click events. Sep 05 00:25:27 http://code.google.com/android/reference/android/widget/Spinner.html Sep 05 00:27:05 er sorry about that :)) Sep 05 00:30:03 romainguy_: thx for the info. i actually wasn't sure what the language differences were from 5 to 6. Sep 05 00:30:57 fpotter: we develop Android using Java 5 btw Sep 05 00:31:04 I mean Sep 05 00:31:13 javac 1.5 from the standard Java SE 5 distribution Sep 05 00:31:44 romainguy_: i'm on a mac, and apple only released Java 6 for 64-bit macs which mine is not, so I've been having a hell of time getting this thing called 'soylatte' (an openjdk build for mac os x) up and running so i could use java 6. but you're telling me there's no need, which is great. thx. Sep 05 00:32:17 many of us work with Java SE 5 on Tiger Sep 05 00:32:23 so we can't use Java 6 either Sep 05 00:38:14 oh awesome yay! full psd included in the photostream sample ^.^ Sep 05 00:38:22 oh wait, im on linux :/ doh Sep 05 00:38:58 can't Gimp open it? Sep 05 00:39:45 umdk1d3: my application ShutterSpeed even comes with the full Illustrator file for the icon & logo :p Sep 05 00:41:05 gimp tries but fails :( Sep 05 00:41:22 :(( Sep 05 00:41:40 somethnig about the layers Sep 05 00:41:48 let me try to export it again Sep 05 00:41:48 but i found the model layer in there Sep 05 00:42:03 its like all the layers are the right shapes, but they are pure white Sep 05 00:42:42 weird, I did save the file in compatibility mode Sep 05 00:43:06 http://umdk1d3.mine.nu/dump/layers.png <-- thats what gimp is giving me :/ Sep 05 00:43:23 * umdk1d3 checks gimp version Sep 05 00:44:05 using gimp 2.4.2 right now, which is the newest in the gentoo portage Sep 05 00:44:18 *newest stable Sep 05 00:44:36 its not too big of a deal, because there is that model layer in there Sep 05 00:44:40 exactly what i was looking for Sep 05 00:44:51 i can do logo stuff in inkscape now =D Sep 05 00:49:10 inkscape roxx! Sep 05 00:49:20 umdk1d3: why do you want to open the psd file? Sep 05 00:49:33 it doesn't contain much :)) Sep 05 00:52:46 I can crush the emulator by sending an sms with that inverted "!" in it (probably all such special chars) Sep 05 00:53:17 (through my app) Sep 05 00:53:23 I'll test this further... Sep 05 00:56:41 yes, also crashes the eulator doing that in the Messaging-APplication Sep 05 01:00:57 http://code.google.com/p/android/issues/detail?id=856 Sep 05 02:10:42 umdk1d3: btw, gimp 2.4.2 is pretty old Sep 05 02:10:49 (current is 2.4.7) Sep 05 02:13:46 like i said, 2.4.2 is current stable in portage Sep 05 02:19:41 umdk1d3: you could build something yourself... Sep 05 02:50:37 hey guys, do you also think the SMS sent via "mSMSManager.sendTextMessage(...)" should be automatically stored to the outbox ? Sep 05 02:51:49 nope, but an option for it would be cool Sep 05 02:52:11 or that **** ENDING LOGGING AT Fri Sep 05 02:59:56 2008