**** BEGIN LOGGING AT Thu Feb 14 02:59:57 2008 Feb 14 03:01:09 I'm glad I'm learning stuff :-) Feb 14 03:01:21 :) Feb 14 03:01:38 I've only done web app development before I worked with Android and I never thought much about mobile phone screens Feb 14 03:01:38 Dan_U: I'm glad you're learning stuff too. Feb 14 03:01:46 I know it's hard when you work on an emulator but you really need to remember you're writing for cell phones :) Feb 14 03:02:15 For as many questions as I answer in google groups, you'd think I'd know this stuff, but there's a ton I don't know Feb 14 03:03:20 Dan_U: There are no hard problems, just problems you have not learnt the answer to yet. Feb 14 03:03:25 so true Feb 14 03:05:48 chomchom: I'm glad you didn't say that this afternoon, when it took me and another guy 5 hours to fix a bug in gridview :) Feb 14 03:06:31 Oh dear, I've yet to approach the looming beast that is my grid Feb 14 03:06:39 was that the one where the screen was turning black or something? Feb 14 03:06:48 Dan_U: this one's been fixed a long time ago Feb 14 03:07:06 this bug was due to performance improvements introduced this week but that are not present in M5 Feb 14 03:07:55 Could you be more specific as I wasn't online and I'm interested. Feb 14 03:08:10 more specific about? Feb 14 03:08:27 Well it was a bug that took like 5hours to fix Feb 14 03:08:50 "romainguy: this bug was due to performance improvements introduced this week but that are not present in M5" Feb 14 03:08:59 was it just grinding to a halt due to performance issues or was it an implementation issue that there was a work around to. Feb 14 03:09:14 neither Feb 14 03:15:51 Ok, on that bomb shell, I'll bid everyone good night. Thanks for the help. Feb 14 03:15:56 romainguy: are you guys allowed to comment on how many submissions have been entered in the challenge or other interesting stats related to the challenge? Feb 14 03:16:11 I have no idea :) Feb 14 03:16:16 ah Feb 14 03:16:22 it'd be fun to know Feb 14 03:16:32 I'm planning on entering at least 4 submissions Feb 14 03:16:43 perhaps 5 or 6 Feb 14 03:16:57 it'd be neat to know how many submissions I'm up against Feb 14 03:16:59 Why not just one, but very good?:)) Feb 14 03:16:59 worldwide competition, everyone and their mum will have a submission. Feb 14 03:17:16 I have one that is very good, and a few others that are not bad Feb 14 03:17:35 I've only got the one. I'm a slow poke. Feb 14 03:17:38 I think the whole thing about it being good depends more on how good the idea is Feb 14 03:17:53 and it's not the easiest thing to do, coming up with great ideas Feb 14 03:18:37 I do have one idea implemented that I think nobody has thought of, although I could be wrong. Feb 14 03:19:45 Dan_U: Ideas are cheap. Everyone and their mum will have one. Tangible, commercially viable proofs of concept and indeed products, are not. Feb 14 03:20:00 you must think the opposite of me :-) Feb 14 03:20:30 to me, developing an application is simple Feb 14 03:20:47 coming up with the idea for it in the first place is hard Feb 14 03:20:54 at least if you intend not to be a copycat Feb 14 03:21:14 You can't deploy an multilingual stable, dependable idea on a phone network. Feb 14 03:21:48 why would you say that? Feb 14 03:22:28 because the idea is only an idea until it is truly implemented Feb 14 03:22:45 ah, I didn't see the word "idea" in your first message Feb 14 03:22:57 you can't really deploy an idea at all Feb 14 03:23:14 no not at all! Feb 14 03:23:26 but making the implementation multilingual, stable, and dependable is possible Feb 14 03:23:28 at least I think so Feb 14 03:24:12 I have complete confidence in your ability to create a good application fast. But no matter how good you are, quality takes testing. Feb 14 03:24:25 true Feb 14 03:24:29 Anyway, nighty night. Feb 14 03:24:35 goodnight Feb 14 03:44:47 romainguy: where are all the icons in the android.jar? Feb 14 03:44:56 i want to reuse a few of them Feb 14 03:47:47 I don't know if they are packed in android.jar Feb 14 03:47:57 since there's no need to Feb 14 03:48:01 where else would they be? Feb 14 03:48:08 in the emulator image Feb 14 03:48:12 or rather, how could i access them? Feb 14 03:48:33 android.R.drawable.* Feb 14 03:49:21 are they gauranteed to be there? Feb 14 03:49:43 only if they are part of the framework Feb 14 03:49:48 apps might have their own Feb 14 03:49:54 oh, that's true. Feb 14 03:50:05 a stock icon set would be really helpful :) Feb 14 03:53:32 argh a bug in the bitmap cache Feb 14 04:01:01 oops Feb 14 04:01:07 there's bug in the new 3D Transition demo ^^ Feb 14 04:01:35 a big one? Feb 14 04:01:57 jasta: no, you might just blow up the heap if you play with it for too long Feb 14 04:03:07 Can you help me understand what the XML might look like for simple_list_item_1? i'm trying to figure out what all that padding is. Feb 14 04:03:16 and how to set the text appearance to textAppearanceLarge Feb 14 04:03:37 it's just a TextView Feb 14 04:04:08 how can that be? when i create an XML layout like: Feb 14 04:04:20 Feb 14 04:04:30 Feb 14 04:04:33 Feb 14 04:04:42 all the padding goes away, and the text size is small. Feb 14 04:04:59 android:id="@android:id/text1" Feb 14 04:04:59 android:layout_width="fill_parent" Feb 14 04:04:59 android:layout_height="?android:attr/listPreferredItemHeight" Feb 14 04:05:00 android:textAppearance="?android:attr/textAppearanceLarge" Feb 14 04:05:00 android:gravity="center_vertical" Feb 14 04:05:02 android:paddingLeft="27dip" Feb 14 04:05:10 ah-ha :) Feb 14 04:05:14 thanks a BUNCH Feb 14 04:05:42 great.. let me get the new one Feb 14 04:21:39 I think theres some new issues with eclipse in this latest release Feb 14 04:22:25 twice now I've had eclipse not want to deploy my application to the emulator and when I go to close eclipse, it doesn't stop the eclipse process Feb 14 04:22:32 so I have to manually kill it Feb 14 04:28:58 romainguy: are you here? Feb 14 04:29:03 yes Feb 14 04:29:16 I think I'm having serious problems... Feb 14 04:29:16 [2008-02-13 20:24:41 - ddm-hello] Now disconnected: [Client pid: 593] Feb 14 04:29:16 [2008-02-13 20:24:41 - ddms] execute 'am start -n localhost.athena/localhost.athena.Main' on 'emulator-tcp-5555' : EOF hit. Read: -1 Feb 14 04:29:16 [2008-02-13 20:24:41 - ddms] execute: returning Feb 14 04:29:16 [2008-02-13 20:24:41 - ddms] execute 'pm -p localhost.athena' on 'emulator-tcp-5555' : EOF hit. Read: -1 Feb 14 04:29:17 [2008-02-13 20:24:41 - ddms] execute: returning Feb 14 04:29:46 I keep seeing this in eclipse Feb 14 04:29:58 and it will just sit there saying it's launching my app Feb 14 04:30:07 any suggestions? Feb 14 04:30:57 none Feb 14 04:30:59 I know restarting eclipse fixes this, but I only got to run my app once this last time before it did this Feb 14 04:31:03 I don't use the plugin Feb 14 04:31:08 ah Feb 14 04:31:32 well, I guess I'll just have to get used to restarting eclipse a lot Feb 14 04:32:23 Should I add that to the bug tracker, or do issues with the plugin go somewhere else? Feb 14 04:32:31 bug tracker :) Feb 14 04:32:38 ok, thanks Feb 14 04:32:49 romainguy: if i wanted to, how would i actually get the value of listPreferredItemHeight? My thought was to call getResources().getDimensionPixelSize(R.attr...), is that not right? Feb 14 04:33:36 you first need to extract the value from the theme Feb 14 04:34:51 is it not simple? Feb 14 04:35:03 ah Feb 14 04:35:09 actually you're calling it the right way Feb 14 04:35:14 but it's asking for a resource id Feb 14 04:35:31 yeah, aren't i giving it one? Feb 14 04:36:10 not with .attr Feb 14 04:36:38 * jasta seems confused Feb 14 04:37:23 really i'm just trying to update some old code that did not utilize XML layouts. maybe i should just convert it to use them. Feb 14 04:40:49 i will just do that, forget it :) Feb 14 04:41:32 ^^ Feb 14 04:41:53 sorry I'm not helping much but I'd like to fix this bitmap caching bug before the last shuttle Feb 14 04:42:07 it's ok, i will work it out myself. Feb 14 04:43:37 zhobbs: are you there ? Feb 14 04:49:13 chadkouse: what's up? Feb 14 04:53:59 hi, eclipse is flagging an error in my manifest.xml, but I can't seem to figure it out: http://pastebin.com/d7387344f Feb 14 04:54:14 I just upgraded to M5 Feb 14 04:55:20 on line 10 of my pasted code, eclipse says 'no resource identifier found for attribute name in package android' Feb 14 04:55:59 is that what the @@ is for? Feb 14 04:56:05 to signify the error Feb 14 04:56:54 i compared it to the code on http://code.google.com/android/devel/bblocks-manifest.html, but it seems ok to me. any ideas? Feb 14 04:57:06 straz: do you know where the @@ came from? Feb 14 04:57:52 I'd guess that error means your eclipse is looking at the incorrect sdk location. Feb 14 04:58:00 as in, an older one Feb 14 04:58:25 yes, the @@ was suggested by pastebin.com to flag the line Feb 14 04:58:31 ah, ok Feb 14 04:58:34 there are no @@'s in my code :) Feb 14 04:58:48 well, double check the location eclipse is looking at for the sdk Feb 14 04:59:16 yes, thanks, I did change the library but it felt like awkward surgery Feb 14 05:00:12 in the eclipse project, under 'referenced libraries', I added m5 and deleted m3. Feb 14 05:00:26 I doubt that would be a good thing to do Feb 14 05:00:44 I do wonder whether there's some other path somewhere to make the change. what's the right way to do this? Feb 14 05:00:56 just specify the correct location for the sdk :-) Feb 14 05:00:58 at least that worked for me Feb 14 05:01:11 like you did when you first installed the sdk Feb 14 05:01:34 straz: right click on project and select Android -> Firx project properties Feb 14 05:01:50 or something like that, and it will move it to the new .jar Feb 14 05:02:06 ok, will try. I was on step 6 of the steps on http://code.google.com/android/intro/upgrading.html but didn't see anything about redirecting eclipse to the new jar. Feb 14 05:03:38 hmm, tried Fix Project Properties, and it appears to have linked back to the old lib. Feb 14 05:04:25 I'm not surprised Feb 14 05:04:35 for example, earlier after changing the lib to what I thought was M5, it asked me to change code in some places. Like EaseInOutInterpolator has become AccellerateDeclerateInterpolator. Feb 14 05:04:36 it's still probably looking at the old sdk location Feb 14 05:04:49 yes, dan, that's what I think too. Feb 14 05:04:56 first, don't change any libs, you don't need to Feb 14 05:04:59 not sure where to change it though Feb 14 05:05:04 ok Feb 14 05:05:07 let me find the location Feb 14 05:05:13 it's where you originally set it in eclipse Feb 14 05:05:18 couldn't find where the path was set. Feb 14 05:05:26 thanks Feb 14 05:05:35 window -> preferences -> android Feb 14 05:05:46 and I'd assume you might need to update any path variables you may have set Feb 14 05:06:14 what I did was to put the sdk in a directory with a generic name "android" and then set the path and eclipse location to that Feb 14 05:06:27 then when a new sdk comes out, I remove everything in that dir and extract the new sdk there Feb 14 05:06:41 aha Feb 14 05:09:11 ok, that updated the library nicely, thank you Feb 14 05:09:48 now chasing some other renamings (like easeinoutinterpolator -> acceleratedecelerateinterpolator) Feb 14 05:10:01 you'll probably be chasing a lot of stuff :-) Feb 14 05:10:58 alright, got my RSS reader building and somewhat updated/working. Feb 14 05:13:18 i guess i'll start updating my ADC project now Feb 14 05:13:18 Sorry for the trouble guys, but it's better if we change APIs now rather than later :) Feb 14 05:13:30 Anyway, I finally fixed this @!# bitmap cache bug Feb 14 05:13:55 it's fine, actually Feb 14 05:13:56 I'm fine with making changes due to the sdk Feb 14 05:14:07 the largest amount of work i did was in updating a program that wasn't even written well to the last sdk Feb 14 05:14:13 at least I have a lot of time between now and end of the challenge submission phase Feb 14 05:14:28 I only wish some other bugs had been fixed Feb 14 05:14:36 like the timezone issue Feb 14 05:14:41 thanks Dan et al, it's working fine now. the line 10 error was changed to a different one on line 7, easily fixed by changing name>android:name Feb 14 05:15:05 you're welcome Feb 14 05:15:09 Dan_U: We also would like to fix more bugs but as you can see, a lot of work went into M5 already :) Feb 14 05:15:15 right Feb 14 05:15:31 but I have an issue where the time I show to the user is all messed up Feb 14 05:15:44 But I suppose I can mention it in my docs for the challenge judges Feb 14 05:16:14 But what is really odd about it is that the first two sdks worked fine Feb 14 05:16:19 I wrote a lot of code to work around bugs in animation in M3, now I have to see what M5 fixed and hopefully rip it all out Feb 14 05:16:53 i, thoughtfully, deferred writing my UI code in favor of the framework :) Feb 14 05:17:19 a lot of fillAfter and animationListener events not being called properly, so I wrote my own Feb 14 05:17:24 so my ADC project is barely affected. but it does have some heavy weight content provider stuff and that totally exploded Feb 14 05:17:51 straz: that amost completely fixed Feb 14 05:18:06 straz: don't rely on onAnimationRepeat() if you set a listener on an AnimationSet Feb 14 05:18:14 I didn't write that case yet Feb 14 05:18:29 oh good. I see you have layout animation, which I'll read about next. I wrote my own too, based on changing absoluteLayouts Feb 14 05:18:52 layout animation might not do what you think it does Feb 14 05:19:00 hmm Feb 14 05:19:00 ok Feb 14 05:19:07 you can see it in action in ApiDemos Feb 14 05:19:15 but it's super easy to use from XML Feb 14 05:19:26 also, for cool animations look at ApiDemos > Animation > 3D Transition Feb 14 05:19:36 it showcases layout animations + custom animation Feb 14 05:20:28 ok thanks romain, I'll look for it Feb 14 05:20:32 :) Feb 14 05:20:41 good night Feb 14 05:20:45 see ya Feb 14 05:31:24 oops Feb 14 05:31:27 I'm almost home Feb 14 05:31:28 ++ Feb 14 05:40:03 ugh, I think I found another bug. Anybody here use a progressdialog in m5 yet? Feb 14 05:40:34 i ported my m3 project which uses them Feb 14 05:40:39 didn't notice anything unusual Feb 14 05:40:53 What happens when you try dismiss()? Feb 14 05:41:15 Dan_U: how many have you found? Feb 14 05:41:22 I'm getting an IllegalArgumentException in android.view.UIThreadUtilities.isUIThread on line 45 Feb 14 05:41:23 i've only found 1 hard crash so far Feb 14 05:41:28 Dan_U: nothing at all unusual. the progress dialog dismisses. Feb 14 05:41:38 I've found some bugs Feb 14 05:41:46 this one is bad and I don't know what I'm doing wrong Feb 14 05:41:58 Dan_U: However, my existing code did not utilize any of the UIThreadUtilities stuff, it just posts mBusy.dismiss(); Feb 14 05:42:05 mine doesn't either Feb 14 05:42:07 the new sdk uses it Feb 14 05:42:16 i know that Feb 14 05:42:22 I have an activity with a method that does this... Feb 14 05:42:23 if (progressDialog != null) { Feb 14 05:42:23 progressDialog.dismiss(); Feb 14 05:42:23 } Feb 14 05:42:25 but i don't invoke the new code path, is all i was saying. Feb 14 05:42:36 I guess I don't understand Feb 14 05:42:53 perhaps you have always been using the progress dialog wrong and itj ust didnt show that until now Feb 14 05:42:54 are you saying your code doesn't use UIThreadUtilities? Feb 14 05:42:59 because my code works exactly as expected. Feb 14 05:42:59 that could be Feb 14 05:43:02 hmm Feb 14 05:43:17 is your dismiss call executed with a Handler? Feb 14 05:43:22 Dan_U: no, i'm saying it does, but not to invoke the new code paths it offers. That is, it doesn't try to dismiss the dialog from the non-UI thread. Feb 14 05:43:29 Yes, through a posted Runnable. Feb 14 05:43:34 my code looks as: Feb 14 05:43:40 mBusy = ProgressDialog.show(...); Feb 14 05:43:52 let me try that Feb 14 05:44:04 Thread t = new Thread() { public void run() { ... mHandler.post(...mBusy.dismiss()...); ... } }; t.start(); Feb 14 05:44:06 I think I've been misunderstanding the UI thread for a long time Feb 14 05:44:14 Dan_U: Seems like you have. Feb 14 05:44:22 I thought that if you had a method in your Activity that it would be executed in the UI thread Feb 14 05:44:30 lol, what!? Feb 14 05:44:31 but apparently that's not the case Feb 14 05:44:39 well, lots of things worked fine doing that Feb 14 05:44:42 at least, they used to Feb 14 05:44:49 another poor assumption on my part Feb 14 05:44:52 that's the thing about concurrency, though. lots of things work fine that aren't. Feb 14 05:45:25 probably because I'm not used to using something like Android. I'm thinking in terms of swing where this wouldn't have happened Feb 14 05:45:31 at least I don't think it would have Feb 14 05:45:56 Swing must have the same paradigm as Android. It is an extremely common practice for UI toolkits to not be thread-safe. Feb 14 05:46:21 ok, but afaik, there is no notion of a Handler in swing Feb 14 05:46:24 at least I never used one Feb 14 05:46:40 but it's been awhile since I used swing, so I might not be remembering correctly Feb 14 05:46:55 the Handler is just a utility laid on top of the threading model in Java. Feb 14 05:47:35 it serves as a simple message passing mechanism, which conceptually exists in any langauge supporting threading. Feb 14 05:48:01 most applications interacting with threads communicate using message queues, i mean. Feb 14 05:48:27 re Feb 14 05:48:57 what does re actually mean? Feb 14 05:49:26 I'm looking up how it's done in swing so I don't make myself look like an idiot :-) Feb 14 05:49:34 I recall it's quite different Feb 14 05:49:37 hmm Feb 14 05:49:46 I don't know off-hand how it is done in Swing, since I have never used it, but I assure you, it is similar. Feb 14 05:49:48 You know, I worked on Swing at Sun Feb 14 05:49:58 Which is why I introduced new UI threading utilities Feb 14 05:50:05 hmm Feb 14 05:50:10 Knowing what damages the single-UI thread paradigm did to many developers Feb 14 05:50:22 heh, you mean locked up apps Feb 14 05:50:29 stuff like that Feb 14 05:50:32 The damage it does is the result of developers ignoring the complexity it introduces. Feb 14 05:50:42 Would you agree, romainguy____? Feb 14 05:50:44 well, that's because too many developers don't understand threads Feb 14 05:50:45 No Feb 14 05:51:10 It was because either developers didn't know about the 1 thread rule Feb 14 05:51:15 or found it too complicated to handle Feb 14 05:51:28 really? Feb 14 05:51:29 Isn't that just restating what I just said? Feb 14 05:51:31 I thought it was simple Feb 14 05:51:34 No jasta Feb 14 05:51:54 You're saying the utilities were hiding this complexity and thus making developers not understand Feb 14 05:52:00 No, I'm not. Feb 14 05:52:03 it was because they did not understand the utilities were introduced Feb 14 05:52:26 and thinks like SwingWorker helped tremendously Feb 14 05:52:31 I'm saying the model of a single non-concurrent UI thread is confusing to developers because they ignore the complexity that it introduces. Feb 14 05:52:37 Nothing to say about the utilities Android just implemented. Feb 14 05:52:45 At some point I was tempted to add something like Visual Basic's DoEvents to Android :) Feb 14 05:52:57 jasta: Ah ok, sorry for the confusion Feb 14 05:52:59 Developers enter with the notion that threading is not relevant, and only learn it to the degree that it impairs their programs in a very immediate sense. Feb 14 05:53:06 (And I was talking about Swing's utilities :) Feb 14 05:53:49 I wish UI toolkits were thread safe but already using one thread is very complex Feb 14 05:54:05 Being thread safe and efficient would be very hard ^^ Feb 14 05:54:53 romainguy____: Yes, they would be. Which is my central argument against the introduction of utilities that claim to permit ignorance when in reality they do not. Feb 14 05:55:11 Developers must understand the threading model that UI toolkits introduce. Failure to do is disasterous. Feb 14 05:55:20 I agree Feb 14 05:55:25 but many don't and won't Feb 14 05:55:31 So, permitting their ignorance in small areas with the usage of leaky utilities helps no one, ultimately. Feb 14 05:55:44 It actually does (unfortunately) Feb 14 05:56:09 I spoke at many conferences and I had to teach about threading to developers who had 5 to 10 years of experience with the toolkit Feb 14 05:56:10 Instead, their programs should explode in a fiery mess. If you're going to make a utility that can determine misuse of the rule, might as well make that utility abort on them :) Feb 14 05:56:41 And when their program explode because of threading issue, they either drop the toolkit or keep blaming it Feb 14 05:57:11 so a lot of developers just don't understand threads at all? I find that amazing. Feb 14 05:57:39 A lot of UI developers don't understand hardly anything at all. Feb 14 05:57:41 * jasta hides Feb 14 05:57:45 haha Feb 14 05:57:54 Dan_U: not threads, the notion of a single UI threads and how to handle it Feb 14 05:58:01 ah, ok Feb 14 05:58:10 That's why, for many people I met, Visual Basic's DoEvents worked Feb 14 05:58:11 I don't at times. Apparently this is one of those times Feb 14 05:58:14 (unfortunately :) Feb 14 05:58:22 Probably due to my assumptions Feb 14 05:58:30 onCreate is not exectued in the UI thread? Feb 14 05:58:35 it is Feb 14 05:58:39 hmm Feb 14 05:58:45 that's quite interesting Feb 14 05:58:46 romainguy____: Perhaps our philosophies just can't agree. I prefer dangerous things to have sharp edges. Maybe that's because I have 12 years experience writing C. Feb 14 05:58:49 Dan_U: you know, many bugs I fix in Android are due to scheduling issues in our UI thread :) Feb 14 05:59:08 I'm glad you are fixing them and not me Feb 14 05:59:09 :-) Feb 14 05:59:22 jasta: and I agree with you, unfortunately when you provide an API to large audiences (millions in the case of Swing), you have to be pragmatic Feb 14 05:59:42 ok, so I have a method I call from onCreate and it attempts to dismiss the progressdialog and that's when I get the exception Feb 14 06:00:00 Dan_U: what is the exception? I missed that part Feb 14 06:00:10 ERROR/AndroidRuntime(2265): Caused by: java.lang.IllegalArgumentException: View android.policy.PhoneWindow$DecorView@402d07f0 must be attached to a window Feb 14 06:00:10 ERROR/AndroidRuntime(2265): at android.view.UIThreadUtilities.isUIThread(UIThreadUtilities.java:45) Feb 14 06:00:10 ERROR/AndroidRuntime(2265): at android.view.UIThreadUtilities.runOnUIThread(UIThreadUtilities.java:64) Feb 14 06:00:10 ERROR/AndroidRuntime(2265): at android.view.UIThreadUtilities.runOnUIThread(UIThreadUtilities.java:102) Feb 14 06:00:10 ERROR/AndroidRuntime(2265): at android.app.ProgressDialog.dismiss(ProgressDialog.java:127) Feb 14 06:00:16 Dan_U: why would onCreate dismiss your progress dialog? That makes no sense. Feb 14 06:00:27 well... Feb 14 06:00:44 I have a method that is called at other times and it must dismiss the progress dialog if it happens to be shown Feb 14 06:00:52 I probably need to do more checks in my code Feb 14 06:00:55 onCreate is called at a point where your activity is not even active. It hasn't been started yet, how could it be busy? Feb 14 06:01:19 true Feb 14 06:01:39 I guess I was trying to just make one method and make it generic no matter when I called it Feb 14 06:02:02 A ProgressDialog is intended in all cases to be displayed and dismissed at some event in the future. The UI thread cannot be asked to block for that amount of time. Feb 14 06:02:09 Dan_U: this exception means you're trying to dismiss a dialog that's not on screen yet Feb 14 06:02:22 romainguy: yes, I agree Feb 14 06:02:42 for whatever reason, I thought I needed to do this. Might have been a long time ago when I didn't understand things Feb 14 06:03:01 onCreate under no circumstance should either display or dismiss a dialog :) Feb 14 06:03:11 agreed Feb 14 06:03:21 neither should onStart. your thinking is busted Dan_U :) Feb 14 06:03:39 although showing a dialog will not break anything really Feb 14 06:03:40 hmm Feb 14 06:03:48 since our dialogs are not truly modal Feb 14 06:04:08 ok, so if I want to do some long running task before showing the activity I have arrived at, why shouldn't I show a progress dialog? Feb 14 06:04:08 romainguy____: I understood the modal dialogs were just a nested message queue that is being run, right? Feb 14 06:04:20 jasta: yes, and they don't exist in Android Feb 14 06:04:29 because we precisely wanted to avoid that case Feb 14 06:04:52 Dan_U: you can, make sure to spawn this task in a Thread Feb 14 06:04:52 Oh, you mean Android's are not that way? Well actually, now that I think for a moment that is the only way it could work. Feb 14 06:05:06 You wouldn't want to inhibit the rest of the events going on. Feb 14 06:05:08 jasta: yes, our dialogs are "modal" in the user-side way Feb 14 06:05:16 jasta: that is they stay on top of the app Feb 14 06:05:20 but they don't block the events Feb 14 06:05:47 Right, that is sensible. Feb 14 06:05:48 I am spawning the task in a thread Feb 14 06:05:52 usually the nested events loop somehow manage to send repaint events and such to the parent loop Feb 14 06:06:01 but nested events loops are nasty :) Feb 14 06:06:08 basically, show the progress dialog and start the thread. After thread ends, dismiss progress dialog Feb 14 06:06:19 Dan_U: that should work Feb 14 06:06:27 Dan_U: More precisely, right before the thread ends, dismiss the dialog. Feb 14 06:06:31 Dan_U: but only if you dismiss the progress dialog after it has started Feb 14 06:06:36 the problem here is interesting Feb 14 06:06:39 right Feb 14 06:06:45 ProgressDialog.show() will post a Runnable on the UI thread Feb 14 06:06:57 so when your code reaches dismiss(), the ProgressDialog is actually not here yet Feb 14 06:07:04 hence what jasta says Feb 14 06:07:14 well, I was trying to handle a case I thought might come up, with the progress dialog being visible and trying to show it again Feb 14 06:07:22 but I guess that case shouldn't happen Feb 14 06:10:45 alright here goes. into my ADC project :) Feb 14 06:10:47 I hope not :) Feb 14 06:11:15 I've learned to handle all cases that I can think up. Maybe I go overboard. Feb 14 06:11:17 :-) Feb 14 06:11:30 when it comes to threading, sometimes it's better not to do that :) Feb 14 06:11:41 for instance, the Android UI toolkit Feb 14 06:11:50 if you use it from another thread, it might work... or not Feb 14 06:11:51 Dan_U: Well, just make sure you understand the model you're working with. Otherwise, you will imagine cases that are logically invalid. Feb 14 06:12:26 jasta: yeah, problem is I keep thinking in terms of other models and don't understand why things are the way they are Feb 14 06:12:31 but I'm learning Feb 14 06:13:25 I'm amazed I am still proficient writing UIs myself, actually. My first large project was heavy in the UI area, and ever since then I've stuck to daemons and networking ;) Feb 14 06:14:00 re Feb 14 06:14:00 But, I was introduced to UI programming with GTK, so I suppose I learned it all the hard way :) Feb 14 06:14:03 re Feb 14 06:14:18 ah GTK ^^ Feb 14 06:14:23 guess it helps indeed :) Feb 14 06:14:44 I wish I had someone to teach me UI programming. I might not be so bad at it if that were the case Feb 14 06:14:53 Hey now. I will defend that GTK is an excellent toolkit given its fundamental constraints (uhh, C) Feb 14 06:15:11 I like GTK Feb 14 06:15:19 Oh, ok. We're cool then :) Feb 14 06:15:28 I just think that object oriented languages are a little easier to work with when it comes to UI Feb 14 06:15:33 but that's my point of view Feb 14 06:15:53 romainguy: I agree, completely. The problem for me was timing more or less. Feb 14 06:15:53 there are only three toolkits I cannot stand: MFC, wxWindows and Tk Feb 14 06:16:19 I like QT Feb 14 06:16:21 (and SWT without JFace because it reminds me of MFC and Win32) Feb 14 06:16:25 yeah QT is nice Feb 14 06:16:40 QT I like, but the problem is that C++ is a miserable example of an OO language. Feb 14 06:16:56 jasta: I've done only PyQT :)) Feb 14 06:17:01 Good ;) Feb 14 06:17:17 but wxWindows in C++ ... @!# Feb 14 06:17:20 I chose GTK to start largely because I preferred C over C++, and I prefered GNOME over KDE at the time. Feb 14 06:17:36 I wish GNOME was ported to QT Feb 14 06:17:39 because I prefer those two :) Feb 14 06:18:17 GLib is damn impressive, I will say. It's amazing how they've whipped C into shape :) Feb 14 06:18:31 ^^ Feb 14 06:18:43 In fact, my android project has a server component written in C with GLib :) Feb 14 06:18:56 nice Feb 14 06:20:01 I was going to write it in Java at first, but I just couldn't help myself. Feb 14 06:20:05 I'm only 24 and I'm turning into a dinosaur ;) Feb 14 06:20:36 jasta:why do you say you are turning into a dinosaur? Feb 14 06:20:39 so that means at 25 I'm still younger than you because I use only Java? :) Feb 14 06:20:42 Actually though, C seemed like the right choice for portability. Feb 14 06:21:01 C is a nice portable lanaguage, if you're careful Feb 14 06:21:33 Dan_U: I was mostly joking, but because I stubbornly choose languages for legacy's sake. Feb 14 06:21:41 you mean C++? Feb 14 06:21:49 romainguy_: Of course, a very nice portable language. If you're very careful ;) Feb 14 06:21:49 or C Feb 14 06:21:56 Dan_U: I mean C. Feb 14 06:22:05 Doesn't make you a dinosaur Feb 14 06:22:20 I graduated school in '04 and the entire 4 years was almost completely C/C++ Feb 14 06:22:26 romainguy_: But like I said, I've written C for a long time. Writing code correctly using it is not a challenge for me ;) Feb 14 06:22:34 makes sense Feb 14 06:22:46 I'm just too lazy to do my malloc and free correctly :) Feb 14 06:22:49 Dan_U: Yeah, but the problem is you didn't learn either language. Feb 14 06:22:52 I've used too many GC'd language Feb 14 06:23:04 jasta:how so? Feb 14 06:23:53 Universities today use C as a whipping tool, with no serious intent to teach students how to be truly proficient with it. Feb 14 06:24:10 yeah, I guess I can see your point Feb 14 06:24:17 well a lot of people start out with C++ and no real C foundation either Feb 14 06:24:21 Nearly every CS student graduating with some C or C++ coursework under their belt could not functionally exist in a UNIX C dev shop. Feb 14 06:25:00 the biggest problem with my education would be not understanding some of the more advanced features of the language Feb 14 06:25:22 That's the thing though. C has no advanced features. C is a very small, very simple language. Feb 14 06:25:45 well, if you went to the school I did, you'd understand what I mean by basic and advanced :-) Feb 14 06:25:52 Everything it offers, however, is dangerous. Feb 14 06:25:58 there's a lot more to C than I learned, and even still I don't know a lot Feb 14 06:26:11 Then you didn't learn the basics either :) Feb 14 06:26:22 so i just upgraded the SDK, everything compiles but i'm unable to call bindService at runtime -- java.lang.NoSuchMethodError: bindService Feb 14 06:26:30 jasta:probably not well enough Feb 14 06:26:49 it seems bindService is still supported according to the docs? Feb 14 06:26:53 I guess what I'm talking about is libraries and such Feb 14 06:27:41 C's size and simplicity are why it is perceived to be large and complex. It is neither, and that causes folks to believe they can navigate through the language without understanding the so-called "advanced" topics. Feb 14 06:27:55 So while you think you know the basics, what you know is squat and couldn't write a functional program to save your life ;) Feb 14 06:28:05 you are correct Feb 14 06:28:05 jasta: C++ did a lot of damage to C's image :) Feb 14 06:28:12 you as a general pronoun, sorry. I am not trying to insult :) Feb 14 06:28:17 lol Feb 14 06:28:24 I don't insult that easily Feb 14 06:28:45 I wouldn't say I couldn't write a functional program. I've written a compiler in C Feb 14 06:28:49 a long time ago Feb 14 06:29:37 romainguy_: Agreed :) Feb 14 06:30:10 jasta: I decided to stop using C++ the day I discovered how to create a pointer to a class member Feb 14 06:31:00 Hehe, C fortunately has no anomalies whatsoever. It's just those pesky compilers... ;) Feb 14 06:32:09 Anyway folks, this has been fun, but I am dying to see my ADC project running on the new SDK :) Feb 14 06:32:13 bbl Feb 14 06:32:18 cya Feb 14 06:32:24 jasta: good luck Feb 14 06:32:36 * romainguy_ is fighting with animations and drawing Feb 14 06:32:51 grr, I really must not understand progress dialogs Feb 14 06:33:32 so I got my progressdialog to display now. I took out my dismiss check that I didn't need Feb 14 06:33:43 but when I do need to dismiss it, I get the same exception Feb 14 06:34:03 are you creating the dialog in onCreate again? Feb 14 06:34:22 yes, basically (in a method called by onCreate) Feb 14 06:34:25 given android's lifecycle, that would be unwise ;) Feb 14 06:34:25 Dan_U: can you show me the code? or a skeleton of your code? (i.e. without the logic, just the methods and classes) Feb 14 06:34:44 progressDialog = ProgressDialog.show(FriendList.this, null, Feb 14 06:34:44 getResources().getString(R.string.loading_progress_label), true, Feb 14 06:34:44 true); Feb 14 06:34:44 Thread thread = new Thread() { Feb 14 06:34:44 public void run() { Feb 14 06:34:45 requestFriendList(); Feb 14 06:34:47 handler.post(updateResults); Feb 14 06:34:49 } Feb 14 06:34:51 }; Feb 14 06:34:53 thread.start(); Feb 14 06:35:01 and the dismiss()? Feb 14 06:35:04 and then updateResults is a runnable that calls dismiss() Feb 14 06:35:07 and a few other things Feb 14 06:35:31 can you show the content of updateResults() and the exception you get? Feb 14 06:35:48 ERROR/AndroidRuntime(2672): java.lang.IllegalArgumentException: View android.policy.PhoneWindow$DecorView@401311e0 must be attached to a window Feb 14 06:35:48 ERROR/AndroidRuntime(2672): at android.view.UIThreadUtilities.isUIThread(UIThreadUtilities.java:45) Feb 14 06:35:48 ERROR/AndroidRuntime(2672): at android.view.UIThreadUtilities.runOnUIThread(UIThreadUtilities.java:64) Feb 14 06:35:48 ERROR/AndroidRuntime(2672): at android.view.UIThreadUtilities.runOnUIThread(UIThreadUtilities.java:102) Feb 14 06:35:48 ERROR/AndroidRuntime(2672): at android.app.ProgressDialog.dismiss(ProgressDialog.java:127) Feb 14 06:36:09 let me comment some stuff out to see if it matters and then I'll paste a copy without all the unimportant code Feb 14 06:36:14 ok Feb 14 06:36:50 Did you guys change id= to android:id= because it wasn't really unique? Feb 14 06:37:21 the explanation was so long and complicated that I forgot it after reading it Feb 14 06:37:27 lol Feb 14 06:37:29 but basically, it's more coherent :) Feb 14 06:37:34 romainguy_: would you know why i'd be getting NoSuchMethodError when calling bindService in an Activity, with the new SDK? Feb 14 06:37:39 (it turns out that XML namespaces are very complicated :) Feb 14 06:38:22 f00f-: I have no idea, what parameters do you pass? Feb 14 06:38:49 romainguy: here's my updateResults: Feb 14 06:38:51 private final Runnable updateResults = new Runnable() { Feb 14 06:38:52 public void run() { Feb 14 06:38:52 progressDialog.dismiss(); Feb 14 06:38:52 } Feb 14 06:38:52 }; Feb 14 06:40:16 romainguy_: bindService(new Intent(MyClass.this, MyService.class), null, mConnection, Context.BIND_AUTO_CREATE); Feb 14 06:40:29 f00f-: this method exists Feb 14 06:40:39 Dan_U: Hmm... Feb 14 06:40:44 Dan_U: let me give it a try Feb 14 06:40:49 ok Feb 14 06:40:55 * romainguy_ is afraid there's a huge bug in dismiss() ^^ Feb 14 06:41:34 hmm Feb 14 06:41:36 f00f-: Do you mean this, not MyClass.this? Feb 14 06:42:17 updating eclipse plugin gives error: Android Editors (0.2.0.200802081635) requires plug-in "org.eclipse.wst.sse.ui". anyone else run into this? Feb 14 06:42:30 jasta: yes even with 'this' it throws the same exception Feb 14 06:43:03 Well then I don't know. That's precisely how I bind my services. Feb 14 06:43:22 actualy Feb 14 06:43:31 bindService() takes only 3 parameters Feb 14 06:43:38 I see you pass 4 Feb 14 06:43:39 Oh, the bastards changed it. Feb 14 06:43:50 I'm noticing that right now in my project too actually Feb 14 06:44:05 the parameter for which you pass null is gone Feb 14 06:44:54 Dan_U: I cannot reproduce your bug, can you try to just call dismiss() instead of putting it in the Runnable you post? Feb 14 06:45:05 I'll try Feb 14 06:46:49 same thing Feb 14 06:46:55 let me try to code you up a simple example Feb 14 06:46:57 how weird Feb 14 06:47:03 I coded a similar version myself Feb 14 06:47:11 can you send it to me? Feb 14 06:47:14 sure Feb 14 06:47:45 you can email it to dan.ulery@gmail.com if it's too long to paste here Feb 14 06:47:53 Dan_U: http://pastebin.com/m1cd66298 Feb 14 06:47:58 or that works Feb 14 06:48:20 and in the Thread, if you remove h.post() and do just progressDialog.dismiss() it works as well Feb 14 06:48:42 but only by magic ;) Feb 14 06:49:00 jasta: I find Handler more magical :) Feb 14 06:49:19 you create a new Handler instance in the air, and boom, you can post stuff on the UI thread :) Feb 14 06:49:27 hmm. you guys took out uri.getPathLeafId() :( Feb 14 06:50:28 romainguy: yes, yours does work for me. I'll try to figure out the difference Feb 14 06:50:28 romainguy_: it is strange that you actually construct the thing, instead of accessing it like a singleton. Feb 14 06:51:02 jasta: not really Feb 14 06:51:04 surely that must be what it is? it's just binding to the message queue on the current thread, right? Feb 14 06:51:13 well... Feb 14 06:51:21 the thing is there is one UI thread... per view hierarchy Feb 14 06:51:34 what? per view hierarchy? Feb 14 06:51:37 yep Feb 14 06:51:46 you can't be serious Feb 14 06:51:52 I am Feb 14 06:52:12 So for every layout we make, we're spawning a new thread? Feb 14 06:52:15 no Feb 14 06:52:18 I said per view hierarchy Feb 14 06:52:27 Well I asked you to clarify :) Feb 14 06:52:29 a view hierarchy is the whole tree of views inside of a window Feb 14 06:52:44 Ahh, then per window you mean :) Feb 14 06:52:54 not necessarily, but yes, if you prefer :) Feb 14 06:52:58 hehe Feb 14 06:53:07 anyway, hence the non-singleton Feb 14 06:53:15 we could use a thread local but... Feb 14 06:54:02 Well, I am not interested in discovering this relationship without code to examine anyway :) Feb 14 06:54:06 I will happily wait hehe Feb 14 06:54:14 romainguy_: quick question, how come when I set the background color on a View object now, it doesn't seem to actually change anymore? Feb 14 06:54:30 was there some background transparency setting added in? Feb 14 06:54:34 jerkface03: what view are you changing the background of? Feb 14 06:54:41 wait wait Feb 14 06:54:42 there it goes Feb 14 06:54:46 ^^ Feb 14 06:54:50 eclipse just needed to be shut down and restarted ;x Feb 14 06:54:55 it messes up on me sometimes Feb 14 06:55:01 * romainguy_ is always afraid the new bitmap caching system could screw something up Feb 14 06:55:01 thanks. Feb 14 06:55:04 sure :) Feb 14 06:56:46 ok, here goes. everything updated except the ContentProvider :\ Feb 14 07:04:25 romainguy_: I think I found my error. Probably another bad assumption I made. I'm checking the docs now Feb 14 07:04:53 hehe Feb 14 07:05:06 progressbar's behave REALLY weird when you give them unusual dimensions. Feb 14 07:05:13 they like spin all over the place Feb 14 07:05:16 instead of rotating in place Feb 14 07:05:25 it almost looks cool Feb 14 07:06:26 yeah, I'm an idiot Feb 14 07:06:36 odd that it even worked correctly before Feb 14 07:08:49 Ok, so I had put code that started my long running thread in both onResume and onCreate. Feb 14 07:09:31 I had originally just put it in onCreate, but found a case where I needed to refresh the UI, so I put it in an overridden onResume Feb 14 07:09:43 and it didn't cause any issues with the onCreate, so I didn't even think about it. Feb 14 07:34:42 alignment is such a mystery to me in this sdk Feb 14 07:37:25 jasta: yeah, don't mess up with circular progress bar's dimensions :) Feb 14 07:37:47 jasta: what alignements are a mystery? Feb 14 07:38:05 romainguy_: I just think I have a very different understanding of what weight, padding, and gravity do. Feb 14 07:38:23 they're pretty straightforward though Feb 14 07:38:41 but I must admit it helps to understand the internals Feb 14 07:39:00 For example, I have a progress bar (horizontal) to the left of a button. I want the progress bar to stretch the window, with the button placed next to it. Feb 14 07:39:15 but i want the progress bar vertically aligned, and with left and right padding. Feb 14 07:39:48 i assume that using a progress bar with height=wrap_content and gravity=center_vertical, with left/right padding and weight=1 would accomplish this. Feb 14 07:40:03 with a button placed to the right of it in a linearlayout, of course. Feb 14 07:40:14 yep Feb 14 07:40:25 and a progress bar width of 0 Feb 14 07:40:29 so far so good Feb 14 07:41:01 except that does nothing of what i want. the left/right padding is ignored, and the progress bar is positioned so that it is aligned to the top of the linearlayout these two widgets are placed in. Feb 14 07:41:24 so it's not gravity you want Feb 14 07:41:26 it's layout_gravity Feb 14 07:41:38 oh god damnit Feb 14 07:41:43 yep Feb 14 07:41:48 yup, sure enough. Feb 14 07:41:51 gravity = applies to the content of the widget Feb 14 07:41:56 layout_gravity = applies to the widget Feb 14 07:42:14 padding still deosn't seem to work, though. Feb 14 07:42:23 that might because of the background drawable Feb 14 07:42:31 try layout_marginLeft and layout_marginRight instead Feb 14 07:42:51 ahh, yes that does work. Feb 14 07:43:02 interesting, well, that clears up quite a lot for me. Feb 14 07:43:32 i adore the new progress bar widget, btw ;) Feb 14 07:45:12 yeah it's nice Feb 14 07:45:26 is the vertical alignment still busted? Feb 14 07:45:38 no, layout_gravity fixed it for sure. Feb 14 07:45:45 cool Feb 14 07:45:48 i guess i just didnt realize there was a difference between gravity and layout_gravity. Feb 14 07:46:01 yeah, everybody gets confused the first time Feb 14 07:46:13 thanks Feb 14 07:46:19 also, when you have vertical alignment issues in LinearLayout, it can be due to the baseline alignment Feb 14 07:46:35 so if you have widgets with text and weird alignements, try to disable baselineAlignement :) Feb 14 07:46:44 hmm, what is that feature? Feb 14 07:47:00 well for instance, imagine you have a TextView next to a EditText Feb 14 07:47:18 since they both have different heights you could normally align them in 3 manners: top, center or bottom Feb 14 07:47:35 but in all these 3 cases it will look weird because the text contained in the two widges will not be aligned Feb 14 07:47:41 baseline alignement prevents that Feb 14 07:47:56 it makes layouts more readable, even though most people will never notice it Feb 14 07:48:01 oh, interesting. Feb 14 07:48:04 yep Feb 14 07:48:22 Swing layout for instance suck because they do not support this by default Feb 14 07:48:29 Android does Feb 14 07:48:42 unfortunately when you do fancy layouts with weird widgets, "strange" things happen Feb 14 07:48:55 that's why you can disable it Feb 14 07:49:08 hmm, i'll look out for that, thanks. Feb 14 07:49:23 (interestingly this feature is one of the features that requires the less code to implement, but one of the most difficult to implement too :) Feb 14 07:52:13 hehe neat, just fixed two other layouts i was struggling with because of that stupid gravity problem :) Feb 14 07:52:17 ^^ Feb 14 07:52:36 yeah, once you understand how it works, it "clicks" and it becomes a lot easier Feb 14 07:52:36 what's a dip, btw? i don't understand what makes it device-independent? Feb 14 07:53:16 the device independent pixels? Feb 14 07:53:22 yeah Feb 14 07:54:40 the dips are not affected by the resolution of the screen Feb 14 07:55:05 but what are they normalized to? Feb 14 07:55:08 96 dpi Feb 14 07:55:14 er Feb 14 07:55:16 yes Feb 14 07:55:20 96 dip == 1 inch Feb 14 07:55:27 ahhh ;) Feb 14 07:55:34 so they represent a measure of physical space. Feb 14 07:56:00 basically they make sure that a big button will stay big on a high-dpi screen Feb 14 07:56:04 beautiful Feb 14 07:56:06 you've been very helpful today :) Feb 14 07:56:10 nio isn't throwing a linkerror Feb 14 07:56:15 everything works! Feb 14 07:56:17 jerkface03: :) Feb 14 07:57:22 romainguy_: how would you say the emulators bytecode execution rate compares to that of the concept devices coming out soon Feb 14 07:57:49 maybe about 50% slower on an actual device? Feb 14 07:58:01 jerkface03: it really depends on what you do Feb 14 07:58:13 the bytecode execution is actually comparable Feb 14 07:58:24 the huge difference I saw was in the drawing code Feb 14 07:58:41 in whose favor? Feb 14 07:58:45 emulator Feb 14 07:58:55 cell phones have little video memory Feb 14 07:58:59 and very slow buses Feb 14 07:59:07 so, how much slower did you estimate it was? 50%? 75%? Feb 14 07:59:23 it varies too much Feb 14 07:59:47 it's a function of what you do, how many pixels you have on screen, what the speed of the bus is, etc. Feb 14 07:59:55 aye, but in a broadview i should expect it to be slower? Feb 14 08:00:09 if you do animations with blending yes Feb 14 08:00:16 kk Feb 14 08:00:17 thx Feb 14 08:00:25 there's a good example in ApiDemos Feb 14 08:00:32 Animation > 3D Transition Feb 14 08:00:41 this is much slower on a physical prototype Feb 14 08:00:48 that thing actually runs slow on my emulator Feb 14 08:00:57 because it's doing full screen blending Feb 14 08:01:29 jasta: I've only used the emulator on Mac OS X and it runs quite fast Feb 14 08:01:38 (both on a Mac Book Pro and a Mac Pro) Feb 14 08:01:51 apparently the Windows version is slower, and I don't know about the Linux one Feb 14 08:04:17 man, this viewbinder thing is great. Feb 14 08:04:54 jasta: I'm glad you like i :) Feb 14 08:04:56 it Feb 14 08:05:06 it took me 10 minutes to implement and I immediately liked it too ^^ Feb 14 08:05:25 it's a great way to do stupid little things like fix-up data out of a cursor Feb 14 08:05:50 yeah, I wish we had thought of that for M3 Feb 14 08:09:13 sort of a weird name, tho. Feb 14 08:09:24 makes you think of a literal binder, as in a notebook, not something which binds. Feb 14 08:10:00 I guess I didn't think about that Feb 14 08:10:16 it made sense to my French mind :) Feb 14 08:10:50 i mean, i knew right away what it was based on the context Feb 14 08:11:02 sure but I understand your point Feb 14 08:13:58 There is currently no hardware out/or announced release dates for that runs Android right? Feb 14 08:14:07 No Feb 14 08:14:17 well, the announced released dates are end of 2008 and beginning of 2009. Feb 14 08:14:41 damn, I just bought a blackberry curve.. I guess I'll prolly get another phone in a year then Feb 14 08:15:55 hmm, on second thought, i should probably extend CursorAdapter anyway. i have the need to inject data not related to the database like status updates and stuff. Feb 14 08:16:21 What kinda apps are you guys writing? Feb 14 08:16:56 i'm writing a media synchronization system which allows you to have access to all your images, videos, and music stored on your PC at any given time (assuming a data connection on your phone) Feb 14 08:17:25 so, meta data is synchronized local, content is on demand (and from cache). Feb 14 08:17:45 jasta: I do that through a PSP through a PS3, it slow and works... sometimes Feb 14 08:18:06 Are there a list of projects in the works? base apps Android includes? Feb 14 08:18:23 romainguy_: Right, but mine will work well. I've already implemented a proof of concept on Windows Mobile. Feb 14 08:18:29 jasta: nice Feb 14 08:18:31 Windows Mobile is just ugly and stupid. I hate developing for it. Feb 14 08:18:46 I'm always amazed by how slow WM can be Feb 14 08:18:51 even on rather powerful hardware Feb 14 08:18:54 me too. it doesn't do anything. Feb 14 08:19:16 romainguy_: Right now I'm doing a test sync of all my music meta data hehe. A refresh from a clean database takes like 10 minutes almost :) Feb 14 08:19:21 (simulating EDGE speeds) Feb 14 08:19:30 how many songs? Feb 14 08:19:34 that said, most of the bottleneck is in the parser. i am using XML currently; haven't switched over to WBXML. Feb 14 08:19:46 romainguy_: I have yet another stupid question, probably due to my inexperience. Want to help? Feb 14 08:19:55 Dan_U: If I can :) Feb 14 08:19:56 romainguy_: 10,483 items synchronized. Feb 14 08:20:20 Ok, so I decided against my clickable textview I mentioned in an earlier chat... Feb 14 08:20:23 jasta: then it would work beautifully with my music library ^^ (< 3,000 items) Feb 14 08:20:29 Dan_U: Awesome! Feb 14 08:20:41 does Android run off X? or some sorta customer framebuffer? Feb 14 08:20:52 romainguy_: Well, that's just the first time synchronize anyway. Updates beyond that are just getting changes from day-to-day, which shouldn't be much more than a couple of hundred for the common case. Feb 14 08:20:55 WesLappy: the latter Feb 14 08:20:57 a couple of hundred would sync in a few seconds. Feb 14 08:21:16 I have a button that I'm setting a background that is basically a black border line a couple pixels wide and the rest is transparent. It's also a nine-patch Feb 14 08:21:16 plus, the meta data synchronization is off-line. Feb 14 08:21:23 or rather, non-interactive. Feb 14 08:21:41 romainguy_: The biggest "if" of the system is the speed at which real data can be downloaded, buffered, and played for the user. Feb 14 08:22:03 I'm doing the same with an imagebutton and it works fine, but when I do this with just a button, it seems to ignore my nine-patch until an event happens Feb 14 08:22:08 For 3G networks, i calculate that the buffering time would be 10 seconds under ideal conditions, 20 practically. Feb 14 08:22:13 jasta: makes sense Feb 14 08:22:29 like if it gets focus, it then shows the padding I've put in the nine-patch Feb 14 08:22:34 Dan_U: when and how do you set the packground? Feb 14 08:22:38 but when first rendered, there is no padding Feb 14 08:22:38 romainguy: how can you test your software with the SDK w/o the hardware? I mean, does it assume some basic input specs or do you have to kinda design a virtual phone to run the software off? Feb 14 08:22:58 WesLappy: we use emulators and we have physical development units Feb 14 08:23:11 romainguy_: well, I extended button, and I set the background in the constructor of my class. Feb 14 08:23:24 Is that a bad thing? Feb 14 08:23:33 not really Feb 14 08:23:35 dev units really expensive? Feb 14 08:24:07 Dan_U: what does it show before an event occurs? Feb 14 08:24:16 or not open to public yet? Feb 14 08:24:47 romainguy_: it renders fine except that the padding is ignored Feb 14 08:25:11 what layout do you put your button in? Feb 14 08:25:23 it's a linearlayout Feb 14 08:26:11 romainguy_: I recently had the idea to extend my system to synchronize any sort of content, very generically. So you could, for example, access your "My Documents" without ever synchronizing them ahead of time. The security implications are staggering, however. Feb 14 08:26:24 you could try to call requestLayout() after setBackround() but it should do it for you Feb 14 08:26:42 jasta: Most definitely, but that could be very useful too Feb 14 08:27:03 romainguy_: Just tried requestLayout. Didn't help Feb 14 08:27:40 Dan_U: stupid question, but you call super(...) first right? Feb 14 08:27:52 yes Feb 14 08:28:19 there's only one thing I can think it would be, but I don't know the specifics Feb 14 08:29:05 I extended the linearlayout that contains these buttons and when this layout is constructed, I programmatically add the buttons Feb 14 08:29:26 the custom layout is added to the content view in the layout xml Feb 14 08:29:38 but I don't know if that would cause any problems Feb 14 08:30:22 hmm you could be interfering with the requestLayout() process indeed Feb 14 08:30:33 alright, I'll try experimenting a bit Feb 14 08:33:38 i should go to bed :\ Feb 14 08:35:33 ok, so if I just add these buttons to a regular layout using xml it works fine. Feb 14 08:36:00 svn commit -m "fuck yeah, updated to m5" Feb 14 08:36:03 hehe :) Feb 14 08:36:06 but the idea for this custom linear layout I have is that I shouldn't be specifying the buttons it includes when I use it in the layout xml Feb 14 08:36:53 Dan_U: Yep, I understand, and I don't know why it did not work Feb 14 08:37:30 ok. I thought I could call requestLayout() after adding all the buttons, but it doesn't help Feb 14 08:37:41 is there a forceLayout() method? Feb 14 08:37:51 or try to call requestLayout() on the buttons themselves Feb 14 08:38:23 no forcelayout. I'll try requestlayout on the buttons Feb 14 08:38:41 ok forceLayout() was in M3 Feb 14 08:39:05 requestlayout on the buttons doesn't help Feb 14 08:39:11 doh! Feb 14 08:39:23 anybody feels like porting this to Android: http://www.bytonic.de/downloads/jake2_jogl11.jnlp ? :) Feb 14 08:39:24 doh? Did I do something wrong again? :-) Feb 14 08:39:31 Dan_U: No, not you :) Feb 14 08:39:56 I guess my requestLayout code path is too agressive Feb 14 08:39:57 dang it Feb 14 08:40:10 so it's actually a bug? Feb 14 08:40:24 I don't know Feb 14 08:40:39 ok Feb 14 08:40:43 Could you make it a small reproducable test case I can run on my machine? Feb 14 08:40:49 That way I could debug and see what's going on Feb 14 08:40:50 will do Feb 14 08:40:55 Thanks, that helps a lot Feb 14 08:41:29 heh, I've been in the position of needing a good test case from a user too, so I understand. Feb 14 08:41:39 I can't tell you some of the oddball bug reports I've seen Feb 14 08:41:46 :) Feb 14 08:42:22 one the poor lady somehow combined the description of 3 bug reports into one. You'd be reading a sentence describing one bug and before the end of the sentence, it'd completely switch to another Feb 14 08:42:33 Haha Feb 14 08:42:45 * jasta shuts down for the night Feb 14 08:42:52 Night jasta Feb 14 08:43:05 night, thanks for all the info tonight Feb 14 08:43:11 You're welcome Feb 14 08:51:24 romainguy_: this might be interesting. I coded an example and it works fine. Feb 14 08:51:39 I'll have to try to track down what I'm doing wrong Feb 14 08:52:02 I'm glad to hear it's your problem ;-) Feb 14 08:52:23 It's always risky to go fix this part of the UI toolkit Feb 14 08:52:43 yeah, I could imagine it might turn into a mess Feb 14 08:53:18 well, not necessarily and that's the problem Feb 14 08:53:37 if you introduce a bug it might be very subtle and surface only in a very specific situation that you never tested Feb 14 08:53:49 (like the bug that took me 5 hours and 2 lines of code to fix in GridView today) Feb 14 08:54:06 that's probably what I've done with my code. Something that I don't see because it's not obvious Feb 14 08:54:25 don't you hate it when it takes all that time and it's only 2 lines? Feb 14 08:54:46 Actually I prefer that than discovering that the fix requires a huge rewrite :) Feb 14 08:54:53 good point Feb 14 08:55:11 I recently rewrote the invalidate() mechanism Feb 14 08:55:20 it was pretty stressful :)) Feb 14 08:55:31 anyway, time to go to bed Feb 14 08:55:35 see ya! Feb 14 08:55:40 ugh. I think I'm glad I don't have your job Feb 14 08:55:55 ok, I'll try to figure this out and if I get a working example, I'll talk to you tomorrow on here or something Feb 14 08:56:04 Well, I love doing that kind of stuff :)) Feb 14 08:56:14 Sure, don't hesitate to ask, I might just not answer :) Feb 14 08:56:22 haha Feb 14 08:56:31 ok, good night Feb 14 08:56:56 man Feb 14 08:57:01 are you still up Feb 14 08:57:19 god damn it, you lucky unemployed duys Feb 14 08:57:28 I'm not lucky Feb 14 08:57:44 I'd like to have a job. I have too much anxiety Feb 14 08:57:49 good night / good morning :) Feb 14 08:57:51 need something to do Feb 14 08:58:00 plus a paycheck would help Feb 14 08:58:32 maybe I should apply for a job at google, but I hear only about 1% of people get hired there Feb 14 08:59:06 chomchom: hey, I am employed! :) Feb 14 08:59:18 Dan_U: last year we received 1 million resumes... Feb 14 08:59:26 wow Feb 14 08:59:28 Dan_U: maybe try somewhere local, unless you live next to a head quarters. Feb 14 08:59:31 I didn't think it was that much Feb 14 08:59:32 haha Feb 14 08:59:47 I live in Orofino, Idaho. Look on a map to see just how far out in the sticks that is :-) Feb 14 08:59:56 Very little opportunity here Feb 14 09:00:13 And seems no places want to hire telecommuters Feb 14 09:00:14 :-( Feb 14 09:01:12 sweet, I was able to reproduce my problem Feb 14 09:01:19 now if I could just figure out what I did wrong Feb 14 09:01:59 Dan_U: mail me your test case, I'll take a look tomorrow Feb 14 09:02:16 I'm not sure if I have your email addr Feb 14 09:02:42 Dan_U: romain.guy@mac.com Feb 14 09:02:48 ok, thanks Feb 14 09:06:11 well, I figured it out for anybody who is wondering Feb 14 09:06:24 although not sure if it's a bug. Seems if I call setPadding, it messes it up Feb 14 09:06:32 ah Feb 14 09:06:43 setBackground() and setPadding() don't like each other Feb 14 09:06:48 I thought you were going to sleep? Feb 14 09:06:49 it's not a bug, but it's an annoying feature Feb 14 09:06:52 hmm Feb 14 09:06:55 and we plan on doing something about it Feb 14 09:07:03 anyway, bed time, for real :)) Feb 14 09:07:03 can you think of a workaround? Feb 14 09:07:13 (I just walked by the other computer and saw your message so... :) Feb 14 09:07:17 haha Feb 14 09:07:29 maybe I just won't set padding Feb 14 09:07:33 I think I can do that Feb 14 09:07:39 you should just let the 9 patch set the padding Feb 14 09:07:52 ok. The padding was left over from my textview implementation Feb 14 09:10:34 romainguy: if you are still here, thanks for the help Feb 14 15:30:00 Hi... Im reading about android (but no work at all) and got few ideas... So, Im wondering: is there a way to know wich view item has been selected? Or know wich application/viewGroup has been selected/opened? Feb 14 15:44:34 is there a new plugin for eclipse as well? Feb 14 16:28:49 mike1o: yeah Feb 14 16:29:11 zhobbs, I had to reinstall it Feb 14 16:29:16 cause it wouldn't update Feb 14 16:29:26 yeah, you have to uninstall/reinstall Feb 14 16:30:37 zhobbs, doin' it right now :) Feb 14 16:31:06 then you gotta update the path to the SDK also Feb 14 16:31:30 zhobbs, did that too restarting eclipse now... Feb 14 16:34:44 zhobbs, here's what happends when I try to open androidmanifest.xml http://img225.imageshack.us/my.php?image=screenshotproblemoccurrak5.png Feb 14 16:35:59 when you installed the plugin did you install both plugins (plugin + editor)? Feb 14 16:36:47 There is a new editor for the android manifest file Feb 14 16:36:48 zhobbs, there is another plugin? Feb 14 16:37:06 yeah, there are two listed when you install it Feb 14 16:37:26 zhobbs, I think we have found the bug Feb 14 16:37:29 eheh Feb 14 16:37:39 :) Feb 14 16:41:49 so ringtones, i'm assuming there is or there will be a contentprovider so you can add them? Feb 14 17:00:11 hi all Feb 14 17:02:23 jasonlee: presumably? Feb 14 17:09:59 has anyone ran into the null pointer exceptions with Bundles? Feb 14 17:10:16 no, but havent done anything there yet Feb 14 17:10:53 I think it's a bug. plusminus posted in the developer group with an issue and mine is similar Feb 14 17:11:06 I cannot make a non indeterminate progress bar for some reason... Feb 14 17:11:45 * jasta wanders off to work Feb 14 17:15:53 * jasonchen is happy now that's he's wrestled Colloquy into displaying the way he wants it to look Feb 14 17:16:59 jasonchen: might I suggest snak? Feb 14 17:21:12 awww Feb 14 17:21:31 Kraln: oh? i hadn't heard of snak Feb 14 17:21:50 just made a toasted cinnamon raisin bagel and dropped it on my dirty carpet Feb 14 17:22:24 jasta, yesterday I dropped my thinkpad on the floor :) Feb 14 17:22:34 jasonchen: it's quite a bit better than colloquy Feb 14 17:23:07 mike1o: so you're saying i should still eat it? Feb 14 17:23:34 depends on which side it fell Feb 14 17:23:50 the buttered side, i'm really upset Feb 14 17:24:01 but i just got an e-mail that we have donuts at work, so forget it. Feb 14 17:24:29 Kraln: cool; i'll try it out later today Feb 14 17:24:39 any more updates to the sdk before the challenge? Feb 14 17:25:21 sushmu: There will be if Google meets their deadlines, but they are being shy on the details in case they don't meet their deadlines :) Feb 14 17:26:03 ok Feb 14 17:26:21 so i guess this is the last one Feb 14 17:26:30 The update may not come before the first round of the challenge is over, however. Feb 14 17:26:43 some people were unhappy with the current gui Feb 14 17:26:53 Yes, I realize that. Feb 14 17:27:06 I have mixed feelings about it myself, but am certain that they are working on it Feb 14 17:27:13 and at the very least, the new UI is an improvement over the previous. Feb 14 17:27:31 The previous UI would have been totally unusable on the very first device to launch using android. Feb 14 17:27:44 Which, as we all know, will be a large screen touch device. Feb 14 17:27:54 right Feb 14 17:28:31 I think a large amount of the complaints are coming from people who aren't generalizing the UI away from their desktop PC monitor. Feb 14 17:28:38 may be Feb 14 17:28:49 So they say "It's so big!" without thinking of the DPI change from monitor to mobile screen. Feb 14 17:29:10 The buttons in the new UI, for example, are not larger than an iPhone. If anything, many of them are still smaller. Feb 14 17:30:00 i am not an mobile expert.. so i haven't really thought about the new ui Feb 14 17:30:15 You don't have to be an expert to have an opinion. Feb 14 17:30:21 ha ha Feb 14 17:30:26 then it might be half baked Feb 14 17:30:30 you fine with that Feb 14 17:30:34 :) Feb 14 17:31:00 Sure, but if you aren't careful you might make yourself look like a retard. Feb 14 17:31:16 You fine with that? :) Feb 14 17:31:30 i told you i'm not an expert :) Feb 14 17:32:01 Let me rephrase: the UI is targetting people, not experts. If people don't like it, it's not good. Feb 14 17:32:30 but as you said, we need to factor the touch thing in Feb 14 17:32:52 clearly the current ui is ugly compared to the last one Feb 14 17:32:58 Yes, you must generalize what you are seeing to a touch screen, high DPI screen. Feb 14 17:33:02 from a pure aesthetic sense Feb 14 17:33:07 correct Feb 14 17:33:13 jasta, is this normal? http://img108.imageshack.us/my.php?image=screenshotandroidemulatbn6.png Feb 14 17:33:17 How would you say clearly? Feb 14 17:33:42 take the mobile factor out.. just compare the ui visually Feb 14 17:34:00 I think the bigger buttons thing is made for lower res phones... Feb 14 17:34:08 sushmu: That's true. Feb 14 17:34:15 or lower screen size Feb 14 17:34:28 Yes, the DPI is really the magic bit here. Feb 14 17:34:48 but in that screenshot I posted we have double buttons Feb 14 17:35:08 yes.. one's like the favorites Feb 14 17:35:11 and its confusing Feb 14 17:35:23 the right hand ones are favorites. it will make more sense when there are more than like 5 regular icons Feb 14 17:35:52 I say Google people know what they're doing... at least more than I do... Feb 14 17:36:01 sushmu: the new UI is growing on me. today i plan to port my VNC project to visualize the DPI shift onto a mobile device. Feb 14 17:36:01 violates DRY Feb 14 17:36:47 we'll get used to it Feb 14 17:36:55 what's DRY? Feb 14 17:36:56 jasta, but I can't see the apps button Feb 14 17:37:09 mike1o: oh, the ApiDemos? Feb 14 17:37:17 DRY = Dont Repeat Yourself Feb 14 17:37:24 i dunno, maybe you borked it. Feb 14 17:37:57 mike1o: -wipe-data? :) Feb 14 17:38:00 gotta run to work for real now guys Feb 14 17:38:40 sushmu: do me a favor, find someone with an iPhone (they're everywhere), and e-mail him a screenshot taken from emulator -skin hvga-p and have him open it up. Feb 14 17:38:46 You'll see what I mean :) Feb 14 17:39:00 ha ha Feb 14 17:39:08 you got an iphone? Feb 14 17:39:14 Everyone does. Feb 14 17:39:21 So I don't need one ;) Feb 14 17:39:41 actually i don't have an iPhone strictly because it launched with no developer SDK. Feb 14 17:39:49 i thought that was such bullshit that i boycotted them :) Feb 14 17:39:50 true.. lame Feb 14 17:40:04 anyway, work Feb 14 17:40:25 cya Feb 14 17:40:33 I want a gphone Feb 14 17:49:16 I've a "perfect geek tool for a museum" Greenphone :) Feb 14 18:11:09 romainguy_: is there a determinate round ProgressBar anymore? Feb 14 18:11:53 nope Feb 14 18:11:57 it's gone Feb 14 18:12:20 romainguy____: dude, why do you have THREE underscores? Feb 14 18:12:22 gotcha Feb 14 18:12:25 or is that 4? Feb 14 18:12:31 :) Feb 14 18:12:43 morrildl: because I use three computers on IRC and they keep adding underscores :) Feb 14 18:14:04 romainguy____: you have hardware there so you can test these ui's with touchscreen? Feb 14 18:14:23 romainguy, there's some poor guy out there who longs to, one day, have just one underscore of his very own - you should be ashamed of wasting them like that! Feb 14 18:15:39 underscores are like pennies Feb 14 18:15:45 they give them to you! Feb 14 18:15:49 What I think is awesome is that romainguy_ and romainguy__ recently both joined, but it's romainguy____ who's actually talking Feb 14 18:16:02 morrildl: not quite :) Feb 14 18:16:04 morrildl: yeah, that threw me off too Feb 14 18:16:34 they are clones! Feb 14 18:16:43 i think we know by now that romainguy and romainguy_ and romainguy__ and romainguy___ are all the same dude Feb 14 18:16:56 michaelnovakjr: Another victory for Science. Feb 14 18:17:06 haha Feb 14 18:17:43 what is the google viewpoint on gmo's? Feb 14 18:17:44 He's spread around the internet though, romainguy__ = gfx@72.14.224.1, romainguy_ = adsl-75-36-210-92.dsl.pltn13.sbcglobal.net Feb 14 18:18:01 Google is Soylent Green! Feb 14 18:18:15 at least he is consistent with his name! and not romainguy and then like turtlewatcher or mr_rogers Feb 14 18:18:22 zhobbs: well, two of the computers are on the same desk Feb 14 18:18:28 just a KVM :) Feb 14 18:18:35 hehe Feb 14 18:18:37 haha Feb 14 18:19:13 lol Feb 14 18:19:28 irc at work lets me be benley_______________________ Feb 14 18:19:56 freenode is clearly deficient! Feb 14 18:24:52 hahaha Feb 14 18:26:02 he added the underscores to confuse iphone spies! Feb 14 18:31:45 live by the _, die by the _ Feb 14 18:31:55 but I'm still here! Feb 14 18:32:34 what, the iphone doesn't have _ ? Feb 14 18:33:32 i_phone :) Feb 14 18:33:32 I don't know hy I'm doing htis, but whatever. Feb 14 18:33:39 no, Jobs thought it 'sucked' and wasn't very slick looking Feb 14 18:34:39 so he went the Java camel-case route instead Feb 14 18:35:49 actually, he's a Lisp fan, so it's all this-and-that Feb 14 18:55:29 how can I set style="?android:attr/progressBarStyleHorizontal" in code? Feb 14 18:55:50 why are you trying to do? Feb 14 18:56:41 well, just making a horizontal slider and it would be nice if I didn't have to provide that in the xml (ie, that would be the default for this custom view) Feb 14 18:57:03 updating a tutorial on my site Feb 14 18:57:11 ah yes Feb 14 18:57:19 didn't you just extend ProgressBar? Feb 14 18:57:22 yeah Feb 14 18:57:26 in that case, just set it horizontal Feb 14 18:58:48 what you mean? Feb 14 19:18:57 previously, you made a progressbar horizontal by setting the background. perhaps they have changed that. Feb 14 19:19:11 yeah Feb 14 19:19:32 it needs that style="?android:attr/progressBarStyleHorizontal" now Feb 14 19:19:55 that "?android:" is new too I guess... Feb 14 19:20:00 nope Feb 14 19:20:04 it's always been there Feb 14 19:20:07 oh ok Feb 14 19:20:43 it just wasn't documented correctly before Feb 14 19:20:48 it still probalby isn't ;) Feb 14 19:20:52 I didn't find any setStyle type of methods for progressBar Feb 14 19:21:15 zhobbs: you can pass that to super Feb 14 19:21:19 zhobbs: your call to super could just use the ProgressBar constructor with defStyle. Feb 14 19:21:33 and pass android.R.attr.progressBarStyleHorizontal as the defStyle. Feb 14 19:21:41 cool Feb 14 19:21:42 exactly Feb 14 19:23:09 works Feb 14 19:24:37 it's some magic voodoo with context.obtainStyledAttributes to collect a bunch of defaults from a predetermined style set. you could inject all those defaults manually, if you knew what they were (which you don't) Feb 14 19:25:28 i suspect, hwoever, that it is still internally controlled with setBackground. Feb 14 19:26:05 part of this style is to set the background Feb 14 19:26:13 FYI, not having source code to the XML content used by android is a real pain. Feb 14 19:27:32 Yep Feb 14 19:36:47 that's why we gotta just decode those binary xml files Feb 14 19:37:59 I agree. I am just not sure that the amount of time it would take to figure out is worth the time it would save :) Feb 14 19:38:40 yeah...I don't have the time either Feb 14 19:38:48 by the time I have the time the source will be out hopefully Feb 14 19:39:06 yeah, not a lot of long term benefit :) Feb 14 19:40:04 otoh, i bet the format is really simple Feb 14 19:40:19 and it seems strings are kept in tact, so it's easy to get your bearing in there. Feb 14 19:40:38 yeah, you can get a pretty good idea just using a hex editor Feb 14 19:41:32 * jasta strokes his beard Feb 14 19:53:13 has anyone ran into a null pointer exception with Bundle putAll? Feb 14 19:57:41 hmm, i wonder if the linux kernel changes at all from m3 to m5. Feb 14 20:01:59 it did, damn :) Feb 14 20:47:01 I take it passing serializables between activities with Bundle.putSerializable doesn't work yet? Feb 14 20:48:51 ERROR/JavaBinder(513): java.lang.RuntimeException: Parcelable encounteredClassNotFoundException reading a Serializable object Feb 14 20:58:56 can we submit more than one project? Feb 14 21:00:12 yes. Feb 14 21:00:46 i suspect that if the scope of your work is so small that you will likely not win. Feb 14 21:00:55 s/ that /, / Feb 14 21:00:56 jasta meant: i suspect, if the scope of your work is so small that you will likely not win. Feb 14 21:01:11 ugh, i hate that bot Feb 14 21:06:17 I feel I made a huge mistake moving my code to the new sdk without making a backup copy Feb 14 21:14:53 Dan_U: you're not using a VCS? Feb 14 21:18:46 nope, I got too lazy :-( Feb 14 21:18:52 and I should know better Feb 14 21:19:15 yeah I lost large parts of a project once that way Feb 14 21:19:22 since then I always have CVS or SVN installed locally Feb 14 21:19:36 I worked for a place that lost pretty much everything for the whole company because their cvs box died Feb 14 21:19:41 apparently it must have taken the disks out Feb 14 21:20:10 but that's probably a freak occurence, although it's a reason to have backups Feb 14 21:20:24 yep :) Feb 14 21:20:40 my machine is backed up 3 times on different hard drives :) Feb 14 21:20:48 nice Feb 14 21:20:57 mine would be, if I had money to buy some new hardware Feb 14 21:21:23 you can probably buy an old machine to do that Feb 14 21:21:30 yeah, true Feb 14 21:21:37 I used to recycle old laptops in cheap backup servers at home Feb 14 21:21:40 yeah, I found an old machine to be my file/print server Feb 14 21:21:46 I do back up to a usb thumb drive occasionally Feb 14 21:21:57 and I have an older copy of everything on my linux box Feb 14 21:22:18 but I'm going to figure out how to make everything work with the new sdk. I'm dedicated to doing that Feb 14 21:22:26 put a web-based bittorrent client on my fileserver...and love it Feb 14 21:22:30 except Bundle seems horribly broken Feb 14 21:22:54 Seem like the geocoding stuff is broken too Feb 14 21:23:14 I haven't gotten to the point I can test geocoding :-( Feb 14 21:23:21 but I assume that will be a big hurdle too Feb 14 21:24:33 hmm, I wonder if I can work around my serialization issue by making a setClassLoader call on the bundle. Feb 14 21:26:08 apparently not Feb 14 21:26:37 zhobbs: my file server is a retired Isilon (isilon.com) developer's node with 1.2tB of storage. Feb 14 21:26:48 nice Feb 14 21:26:58 60% full ;) Feb 14 21:27:09 what do you put on there? :) Feb 14 21:27:14 mine is an old celeron POS with 300 GB Feb 14 21:27:20 and it's pretty much full Feb 14 21:27:44 romainguy___: all the things you'd expect, only lots of them :) Feb 14 21:27:48 my /Users/romainguy fits in a 20 GB DMG :) Feb 14 21:27:55 that's all the data I need ^^ Feb 14 21:28:11 well... except for the 500 GB dedicated to my RAW photos ;-) Feb 14 21:28:32 nice photos on your site btw Feb 14 21:28:39 thanks zhobbs Feb 14 21:28:45 Only very recently I realized that there is a central theme in my geek life. Feb 14 21:28:59 jasta: Hard drives? Feb 14 21:29:11 Every major project I've ever done, Android including, has been to support my insatiable thirst for media ;) Feb 14 21:29:24 that explains the 1.2 TB :) Feb 14 21:29:42 My problem is that since the Intel 486 came on the market, I've only ever used laptops Feb 14 21:29:54 So I quickly learned how to use as little hard drive space as possible Feb 14 21:29:55 :) Feb 14 21:30:03 i have a laptop, but also a workstation at work and at home. Feb 14 21:30:06 And this is a habit I cannot shook off now Feb 14 21:30:13 -shook+shake Feb 14 21:30:40 i actually mount my file server at home while i'm at work, though. Feb 14 21:31:07 I use back to my mac for that Feb 14 21:31:11 works pretty well Feb 14 21:31:28 but i could not live without my workstations. my laptop is great, but i use it like once every couple of weeks i'd say. Feb 14 21:31:51 my work and home environments both have dual 22" widescreens, so staring at my little 15" laptop is frustrating :) Feb 14 21:31:56 I could not live without my desktops Feb 14 21:32:06 yeah, same here, dual 22" on the one box Feb 14 21:32:16 I prefer laptops by far (especially since I spent the past 4 weeks moving to a different country every 6 months) but I have to admit a MacPro would be ncie at home for my pictures Feb 14 21:32:30 jasta: hey I use laptops but when at home or work, they're plugged to a 30" :)) Feb 14 21:36:26 ugh, it is so hard to do my day job anymore. Feb 14 21:36:31 all i want to do is work on Android. Feb 14 21:36:58 all your base are belong to us Feb 14 21:38:10 i don't even typically like UI programming Feb 14 21:38:49 but after years of working with Windows Mobile and PalmOS, I find it satisfying :) Feb 14 21:50:42 oh well, i'll live :) Feb 14 21:51:28 my background is web stuff, and I'm so tired of that anything else seems exciting Feb 14 21:53:48 hmmm, what is a Google Application Specialist? Feb 14 21:53:51 my background, i suppose, is network programming. but i have done many things :) Feb 14 21:54:52 my background is pretty much web stuff, and it is a bit tedious and boring Feb 14 21:55:04 Professionally, I have only worked with Perl, C, and C++. But I don't feel that my professional career accurately reflects where my talents and interests lie. Feb 14 21:55:42 I mean, I like Perl and C, but I feel it is unfair to label :) Feb 14 21:56:17 so where do your talents and interests lie? Feb 14 21:56:48 irc chatting, procrastinating :) Feb 14 21:56:58 haha Feb 14 21:57:16 Yes, zhobbs is right. :) Feb 14 21:57:48 let me rephrase that... Feb 14 21:58:06 More seriously, I would say I am uncommonly adept at designing large systems efficiently, scaling either internally or remotely (as in, a cluster of networked nodes). Feb 14 21:58:06 If someone offered you the job of a lifetime and it didn't include chatting/procrastinating, what would that job be? Feb 14 21:58:43 Dan_U: Truthfully, that job would include an asterisk for job description ;) Feb 14 21:59:13 interesting Feb 14 21:59:41 But I'd take anything to do with critical scalability :) Feb 14 22:00:19 My preference is to apply that to networked systems, however. Feb 14 22:02:37 But, in general, I am a markettable programmer simply because I don't ignore complexity and I read critically. Feb 14 22:07:53 jasta: I wish I was enthusiastic as you Feb 14 22:07:59 because you seem to be Feb 14 22:09:26 I used to be, but got a bit burnt out Feb 14 22:11:18 I'm trying to get out of full time programming...got my electrical engineering degree, want to get to the hardware side Feb 14 22:11:54 probably need to get my masters though Feb 14 22:12:19 that might be fun Feb 14 22:12:31 I think I'd go for a business degree if I went back to school Feb 14 22:12:43 electrical engineering would probably be my second choice Feb 14 22:12:45 yeah, actually I'm more interesting in the business side Feb 14 22:12:56 interested* Feb 14 22:13:21 playing with microcontrollers is really fun though Feb 14 22:16:08 zhobbs: it depends :)) Feb 14 22:18:03 zhobbs: something I really enjoyed was programming games for GameBoy Advance Feb 14 22:18:13 romainguy__: I haven't been forced to do it all day everyday...which probably kills just about anything Feb 14 22:18:22 involved both cool graphics and low-level programming Feb 14 22:18:37 zhobbs: well, I love doing graphics stuff all day long Feb 14 22:18:52 I'm interesting in conrol systems...I want my program to open valves and move servos, etc Feb 14 22:19:04 /s/conrol/control Feb 14 22:21:25 romainguy__: good thing you like it, cause as far as I can tell you do it all day long :) Feb 14 22:24:52 Dan_U: I'm not enthusiastic about it, really. I just identify with being a programmer. That's what I am. Feb 14 22:25:20 Makes it easy to be OK doing it all day long ;) Feb 14 22:25:44 Eventually, though, I've got my eye on a director of software engineering gig ;) Feb 14 22:26:31 jasta: so do I, but I got burnt out with too many long hours and it seems too much like programmers are treated like an unwanted expense versus someone useful Feb 14 22:26:33 * romainguy__ would like to go back to UI design Feb 14 22:26:59 of course, maybe that's due to just the places I've worked for Feb 14 22:27:44 Dan_U: Probably. Feb 14 22:27:56 For example, I am not overworked at all. I think, in fact, I'm underworked :) Feb 14 22:28:18 just ported 4 small tutorials fairly painlessly... Feb 14 22:28:22 I also remember one of my college profs saying that if you wanted a job in software development, don't even think about being a programmer Feb 14 22:28:47 What did he mean by that? Feb 14 22:28:51 Dan, you have to remember that he's in Academia, far removed from the 'real world' Feb 14 22:29:08 that the only jobs in the US would be designing software, not coding it Feb 14 22:29:20 I definitely agree he was far removed from the real world Feb 14 22:29:45 Dan_U: That's just not very likely, honestly. Feb 14 22:29:49 in fact, if he knew any of you were using anything that even remotely resembled Java, he'd have a fit :-) Feb 14 22:30:05 actually, I saw it with my own eyes, so I kinda believe it Feb 14 22:30:06 everyone I know that's outsourced code has ended up bringing it back in-country. Feb 14 22:30:16 not everyone Feb 14 22:30:26 Dan_U: Did you ever consider that he's just an idiot? Feb 14 22:30:28 I said everyone that *I* know. Feb 14 22:30:30 I've worked on projects with outsourced developers and it never really worked Feb 14 22:30:33 oh yes, he is an idioiut Feb 14 22:30:37 *idiot Feb 14 22:30:42 Then who cares what he says about anything? Feb 14 22:30:51 All the projects I worked on used outsourced devs Feb 14 22:31:02 and they generally were people who hadn't even graduated from college Feb 14 22:31:05 some were quite good though Feb 14 22:31:10 but it was real hit and miss Feb 14 22:31:17 I didn't graduate college, nor did I ever attend. Feb 14 22:31:24 wow Feb 14 22:31:40 I also enjoy a debt-free life because of it :) Feb 14 22:32:07 oddly enough, I never went in debt because of college, and it's not because my parents had lots of money either Feb 14 22:32:15 Although I did take a number of maths courses at a community college for fun. Feb 14 22:32:21 jasta: I graduated college and because it was in France I also enjoy a debt-free life :) Feb 14 22:32:25 shoestring budget combined with enough financial aid to pay for tuition and books Feb 14 22:32:53 I don't think I'd be very good at programming if I hadn't went to college Feb 14 22:33:03 romainguy__: I was just being quippy. College wasn't right for me for a number of reasons. Regardless, it is a luxury not a necessity. Especially in this crazy industry. Feb 14 22:33:17 depends where Feb 14 22:33:31 Only culturally. Feb 14 22:33:31 in France it definitely helps a lot to have a degree to get a decent job Feb 14 22:33:41 college is cheap in florida, tuition was $300/semester if you get a 3.0 in highschool Feb 14 22:33:42 Making six figures as a government contractor pretty much requires a college degree. sorry Feb 14 22:34:10 romainguy__: College always helps, but it is one tool among many. Being a rockstar programmer, for example, also helps :) Feb 14 22:34:21 good companies don't hire rockstars Feb 14 22:34:49 have you ever had to work with one, or maintain their code? Feb 14 22:34:56 redefines pita. Feb 14 22:35:00 ^^ Feb 14 22:35:17 Kraln: I believe you Feb 14 22:35:24 Once I had to maintain Keith Richards code. Feb 14 22:35:30 No, I'm kidding. He maintained it all himself. Feb 14 22:35:57 the rockstar programmer I was brought on to 'deal with' also did acid and pcp. Feb 14 22:35:59 I'll do you one better. I had to maintain Keith Moons code. It went ok until my computer blew up :-) Feb 14 22:36:08 have you ever seen code written while on acid? Feb 14 22:36:15 it makes my head hurt. Feb 14 22:36:18 Kraln: I think I have Feb 14 22:36:36 Kraln: Perhaps you aren't really part of this conversation. You seem to be speaking with little relevance. Feb 14 22:36:43 Kraln: I wouldn't get too far on acid Feb 14 22:36:51 imagine if you will, calling a simple method in a class to obtain a value to display on a web page... Feb 14 22:36:52 Insane people doing drugs and not performing their job function is not dependant on a college education. Feb 14 22:37:04 Instead, the developer writes a couple hundred lines of reflection code Feb 14 22:37:14 Dan_U: that's about right Feb 14 22:37:29 jasta: no it's not related to college at all. Feb 14 22:38:14 college isn't really a place to go to learn anything; you can learn what you need really out of books or other freely available stuff. College itself is the 'test', some people won't hire you unless you 'pass' Feb 14 22:38:34 My point is, simply, that college is one tool to acquire knowledge. It is a very good tool, and that is why it costs money. It is not, however, an indication of how well that tool was used. Feb 14 22:38:40 if you want to go your own way, it doesn't really matter, cept the banks won't finance you... Feb 14 22:38:46 I don't know. I found that I learned a lot better having the lectures in college Feb 14 22:40:05 College also gives you access to stuff you perhaps wouldn't normally look at. For instance, I would never have written VHDL or worked with an FPGA if it hadn't been a required college course Feb 14 22:40:42 Like I said, college is a very good tool, and that's why it costs money usually :) Feb 14 22:40:45 Dan_U: there's a lot of 'this is the way that it's done' stuff in college too, that is really helpful for self taught people. Feb 14 22:42:56 Kraln: That's a dangerous attitude, though. Inheriting or copy someone's logic and approach in any particular industry is an easy way to find yourself stagnating in much the same way that person originally did. Feb 14 22:43:13 copying* Feb 14 22:43:41 or can make you learn from their experiences Feb 14 22:44:20 zhobbs: Depends on how you approach the situation. I would say that in order to learn from their experience, you would need to analyze their logic, not copy it. Feb 14 22:44:55 But anyway, this conversation is too engaging. I'm not gonna get home by 5 if we keep this up :) Feb 14 22:45:01 bbl Feb 14 22:58:02 Dan_U: I'm torn. Feb 14 22:58:10 On the one hand: FPGA designers unite! Feb 14 22:58:11 your torn? Feb 14 22:58:17 lol Feb 14 22:58:18 On the other hand: Verilog forever, man!! Feb 14 22:58:32 It's like we're both rappers, but I'm East Coast and you're West Coast Feb 14 22:58:41 haha Feb 14 22:59:03 we actually were recommended to get a verilog book for the class and I think I still have it Feb 14 22:59:08 don't think I even opened it up Feb 14 22:59:26 hehehe Feb 14 22:59:34 that was an awesome class though. We wrote a processor Feb 14 22:59:40 didn't have enough time to put it on the fpga though Feb 14 22:59:41 My objection to VHDL is that you have to write the same code three times Feb 14 22:59:51 heh, doh Feb 14 22:59:57 I don't remember enough about it to argue with you :-) Feb 14 23:00:09 we actually did some simple stuff with the fpga. Stuff like a traffic light Feb 14 23:00:52 ahh, the infamous blinking-LEDs assignment :) Feb 14 23:01:22 the guy who taught us the processor stuff I think was working on using fpgas to morph into different kinds of hardware, like sound cards, etc. Feb 14 23:01:30 unfortunately I didn't take enough interest in it to find out more Feb 14 23:01:37 I was more interested in AI Feb 14 23:02:14 I think FPGA's are interesting in how they work....haven't had a chance to program for them though Feb 14 23:02:37 I hear a lot of people curse vhdl though Feb 14 23:02:40 imagine setting time delays in your code :-) Feb 14 23:02:46 it's a bit different Feb 14 23:03:41 but the xilinx tools we were using had the ability to draw an actual schematic, so you could see the logic gates and connecting wires of your processor. It was really cool Feb 14 23:04:07 Dan_U: ahhh, floorplanner. Feb 14 23:04:17 You bring back extremely painful memories Feb 14 23:04:29 Is that what it's called? Feb 14 23:04:30 I wrote a tool that did bitstream linking Feb 14 23:05:08 I can probably still recite the structure of the logic fabric on the Virtex-II family Feb 14 23:05:12 I'm not sure I could ever adapt to developing hardware though. I much prefer java Feb 14 23:05:24 morrildl: you really are hardcore Feb 14 23:05:26 Dan_U: yes, it's a rather radically different mindset Feb 14 23:05:48 Dan_U: heh, the scary thing is I'm not, really Feb 14 23:05:50 guess that's why you work for google Feb 14 23:06:11 I know enough to have profound sympathy for people who work at AMD, Intel, ARM, and so on Feb 14 23:06:19 same here Feb 14 23:06:37 we only implemented a simple mips processor and it was a pain Feb 14 23:06:44 Dan_U: that's pretty hardcore too Feb 14 23:06:46 even with the schematic in the book Feb 14 23:06:53 yeah, I had a professor just get a job doing semiconductor R&D at IBM... Feb 14 23:07:01 and the book didn't even get it right! Feb 14 23:07:32 actually Feb 14 23:07:44 since you did it on an FPGA, I guess that "core" would actually be "soft" ;) Feb 14 23:07:52 haha Feb 14 23:08:17 that's true...fpga is soft-core by definition Feb 14 23:08:20 zhobbs: IBM designed the cell, which is pretty interesting as processors go Feb 14 23:08:30 that should have been Cell, not cell Feb 14 23:08:34 i.e. the PS3 processor Feb 14 23:08:42 The thing is like SIMD on steroids Feb 14 23:09:02 yeah, wonder if any programmer have figured out how to use it all yet Feb 14 23:09:07 It sits between a traditional processor and an FPGA Feb 14 23:09:14 morrildl: makes me wonder how programmers use it when I compared my games on PS3 and Xbox... :) Feb 14 23:09:15 zhobbs: yes, although they aren't writing games Feb 14 23:09:37 Some of my buddies who are still at GE have been using Cell for high performance computing Feb 14 23:09:54 One tells me that it really IS an incredible processor, everything they say is true Feb 14 23:10:12 but to use it properly you have to redesign your algorithms to take advantage of it Feb 14 23:10:25 yeah, PS3 dominates one of the distributed computing apps... Feb 14 23:10:32 not sure if it's SETI or that cancer one Feb 14 23:10:34 morrildl: I had wanted to do that with some evolutionary computing stuff I was working on, but my friend who used to code for playstations said it wouldn't do near what I needed it to do Feb 14 23:10:41 He doesn't think a C compiler is ever really going to be able to full exploit the Cell Feb 14 23:11:10 romainguy____: heh :) Feb 14 23:11:35 zhobbs: The Folding@Home Feb 14 23:11:41 * morrildl wants a cell phone with a Cell Feb 14 23:11:47 a Cell phone? Feb 14 23:12:11 you'd probably have to carry the battery strapped to your back though Feb 14 23:12:17 cell cell phone Feb 14 23:12:34 cell-squared phone Feb 14 23:17:09 morrildl: Are you guys allowed to comment on how many submissions there have been to the challenge so far? Feb 14 23:25:08 I don't know why people would submit it this early Feb 14 23:25:27 well, maybe they are done and figure they may as well submit Feb 14 23:25:58 I wouldn't expect there to be a lot, but if there is a lot it would be surprising Feb 14 23:35:19 i don't envy the OHA members who have to judge this competition :) Feb 14 23:35:28 jasta: why? Feb 14 23:35:29 they will have to wade through so many shitty little projects :) Feb 14 23:35:36 you think so? Feb 14 23:35:45 i'm almost certain that people will submit things like tip calculators. Feb 14 23:35:54 I think they are going to have to wade through shitty coded projects Feb 14 23:36:31 I don't mean to make an example out of anyone, but I've tried to help some people out in the developers group and they don't even know what a stack trace is Feb 14 23:37:04 my friend wrote a distributed compiler for ps3 Feb 14 23:37:06 or something Feb 14 23:37:21 distributed language that compiled for ps3 Feb 14 23:37:23 Dan_U: nuts, that was me. Feb 14 23:37:29 lol Feb 14 23:37:33 you're kidding right? Feb 14 23:37:35 yes Feb 14 23:37:42 thank god Feb 14 23:37:43 Dan_U: sorry, can't comment on it :( Feb 14 23:37:54 morrildl: no problem Feb 14 23:38:07 can you say if the response in general to android is what you were hoping for? Feb 14 23:38:46 its more than i expected Feb 14 23:39:06 i thought it would be people in this channel Feb 14 23:39:18 and then i get reminded when i check my mailing list mail Feb 14 23:51:55 does anyone know when there will be some info available about challenge II? Feb 15 00:05:08 Dan_U: probably when the phones come out Feb 15 00:06:28 yeah, I suppose Feb 15 00:28:33 does android use a linux kernel? Feb 15 00:31:46 ping Feb 15 00:32:29 yes it does Feb 15 00:33:50 cool Feb 15 00:33:59 and the software is writen in java? Feb 15 00:35:20 ahh, home Feb 15 00:35:27 l33tlinuxh4x0r: Most of it, yes. Feb 15 00:36:33 interesting Feb 15 00:37:35 is there a list of supported hardware? Feb 15 00:37:43 yes, and it contains 0 entries. Feb 15 00:37:48 lol Feb 15 00:37:54 what about the sdk? Feb 15 00:38:08 the SDK is supported on Linux, Mac OS X, and Windows. Feb 15 00:38:10 shouldn't the software that we make run on something Feb 15 00:38:34 l33tlinuxh4x0r: Yes, but it currently only runs on an emulator. Feb 15 00:38:47 l33tlinuxh4x0r: Please read the FAQ on Android's site. I am not interesting in reading it to you. Feb 15 00:39:30 ok Feb 15 00:40:27 do you work for google Feb 15 00:40:52 No. Feb 15 00:40:55 l33tlinuxh4x0r: http://code.google.com/android/documentation.html Feb 15 00:41:00 That page should help you get started Feb 15 00:41:12 romainguy___: Got a tablelayout issue, want to help me out? Feb 15 00:41:21 Dan_U: I can't right now Feb 15 00:41:28 ok, no problem Feb 15 00:41:30 try shrinkColumns and stretchColumns :) Feb 15 00:41:35 l33tlinuxh4x0r: some FAQs are here http://code.google.com/android/kb/index.html Feb 15 00:41:47 romainguy___: doubt that would fix it Feb 15 00:43:11 If anybody is wondering what is wrong, http://pastebin.com/d2c646c25 has the example code/xml Feb 15 00:43:29 * jasta taps his foot as the new Android kernel recompiles Feb 15 00:43:41 gonna fire this up on my VNC viewer and take a look at the new UI on a real handset :) Feb 15 00:43:51 err my VNC server Feb 15 00:48:00 is there an irc channel for the msm7k kernel Feb 15 00:52:41 ohhhh, pretty :) Feb 15 00:54:35 wow seriously, this UI does look nice a handset Feb 15 00:54:56 :)) Feb 15 00:55:11 it actually isn't big looking at all Feb 15 00:55:16 it looks quite sensible Feb 15 00:55:19 told ya Feb 15 00:55:43 told me? you must be confused, i was defending the UI. Feb 15 00:55:52 i just hadn't taken the time to actually load it up yet :) Feb 15 00:56:38 :) Feb 15 00:57:01 i love that i wrote this VNC server hehe Feb 15 00:57:11 even though i know it will be totally pointless once phones come out Feb 15 00:57:19 i'm gonna post some pictures for people to drool over :) Feb 15 00:57:23 hehe Feb 15 00:57:27 this time though i'm gonna try to make them not look so damn fuzzy Feb 15 00:57:47 you had some tips for me last time but i can't remember them. should i try taking the photo outside or something? Feb 15 00:58:01 ah no Feb 15 00:58:06 the trick is to take two pictures Feb 15 00:58:13 one exposed for the phone Feb 15 00:58:17 one exposed for the screen Feb 15 00:58:17 oh right, i remember that. but i don't understand what you mean at all :) Feb 15 00:58:31 well the screen is very bright compared to the device itself Feb 15 00:58:36 i was just borrowing my girlfriends basic digital camera. Feb 15 00:58:44 does it have a manual mode? Feb 15 00:58:52 yeah probably Feb 15 00:59:01 if so, you'll want to use it Feb 15 00:59:53 and just control the exposure? Feb 15 01:00:15 yes Feb 15 01:01:47 hmm Feb 15 01:01:57 I used to have pictures I took that way but I seem to have deleted them Feb 15 01:02:04 aah, I know Feb 15 01:02:43 jasta: http://progx.org/gallery/1081447377.jpg << this is an example of the technique I'm talking about Feb 15 01:03:13 interseting. Feb 15 01:04:09 hehe neat, there's a dropshadow on the windows title now Feb 15 01:04:11 i never noticed that Feb 15 01:04:26 couldn't see it on my monitor, but i can see it on the device :) Feb 15 01:09:46 When I make an invisible tablerow visible, why would the columns in all the rows beneath it appear to shrink in width? Feb 15 01:11:35 romainguy___: i really think you guys went the right route with the menu. looking at it on a phone really drives this home Feb 15 01:12:25 I take that back, apparently it's only the first column Feb 15 01:12:51 jasta: thanks :) Feb 15 01:13:07 Dan_U: TableLayout always makes sure that all the cells in a column have the same width Feb 15 01:13:17 if you introduce new columns then the sizes are recomputed Feb 15 01:13:31 teehee, i searched for "mexican near seattle" in Google Maps here and it came up with "El Gaucho's Restaurant", which is a fancy suit and tie steak house in Seattle. Feb 15 01:13:41 but I'm adding a row, not a column Feb 15 01:14:06 I mean, I could see it widening the column width if my new row had wider content Feb 15 01:14:15 but it is getting narrower Feb 15 01:17:23 hehe maps is unusable on QVGA-L :) Feb 15 01:17:41 (dont lecture me, i understand why) Feb 15 01:18:50 @!# VPN Feb 15 01:18:59 never ever buy an Airport Extreme Base Station if you want to use VPN Feb 15 01:22:14 re Feb 15 01:43:59 I give up. I'm just gonna let my layout look a little weird Feb 15 02:26:40 anyone know why setting the Blur_Behind flag on my activity's window causes the background to go solid black? (without the flag, its transparent, which is what its supposed to be) Feb 15 02:27:09 im using the same code as the fancy translucent demo, but instead of the neat clear blue I get solid black Feb 15 02:30:38 or why setting the text color for any textview in xml doesn't do anything Feb 15 02:31:03 im stumpd Feb 15 02:31:10 *stumped **** ENDING LOGGING AT Fri Feb 15 02:59:56 2008