**** BEGIN LOGGING AT Sat Dec 29 02:59:56 2007
Dec 29 04:07:30 <jasta> why is taking a photograph of a cell phone so fucking hard?  it always comes out blurry
Dec 29 04:09:57 <absentia> need very bright natural light
Dec 29 04:17:52 <jasta> really?
Dec 29 04:17:54 <jasta> like outside?
Dec 29 04:23:34 <romainguy> jasta: and don't hesitate to use two exposures
Dec 29 04:23:37 <romainguy> one for the cell phone itself
Dec 29 04:23:40 <romainguy> one for the screen
Dec 29 04:29:37 <jasta> two exposures?
Dec 29 04:30:49 <romainguy> two photos
Dec 29 04:31:04 <jasta> and digitally combine them?
Dec 29 04:31:08 <romainguy> yes :)
Dec 29 04:31:14 <jasta> oh, well, that is too fancy for me :)
Dec 29 04:31:25 <jasta> romainguy: check this out anyway...
Dec 29 04:31:49 <romainguy> sure
Dec 29 04:32:14 <jasta> http://android-vnc.googlecode.com/files/CIMG0359.JPG.jpg
Dec 29 04:32:35 <romainguy> nice
Dec 29 04:32:37 <jasta> teehee ;)
Dec 29 04:32:44 <romainguy> did you write the VNC client or the server?
Dec 29 04:32:48 <romainguy> (or both)
Dec 29 04:32:51 <jasta> the server, your guys' VNC server sucks :)
Dec 29 04:33:04 <jasta> the VNC client is .NET VNC Viewer (dotnetvnc.sf.net)
Dec 29 04:33:26 <romainguy> cool
Dec 29 04:33:37 <jasta> actually, i had a few questions about that though
Dec 29 04:34:12 <jasta> what is the deal with the RFB implementation in libsurfaceflinger?  was that just a toy or are there plans to build it into something useful?
Dec 29 04:34:54 <jasta> i tried forever to make it work right but ultimately gave up because it was just implemented way too sloppily, like it was never meant to even exist
Dec 29 04:35:54 <jasta> the only reason i found it was because i got bind: Address already in use when I tried to deploy mine :)
Dec 29 04:37:32 <jasta> hmm?
Dec 29 04:37:47 <jasta> romainguy: i assume you missed all that? :)
Dec 29 04:38:38 <romainguy__> No but I cannot comment on that :)
Dec 29 04:38:53 <jasta> why not?
Dec 29 04:39:21 <jasta> just curious if there was some purpose for the creation of the built-in RFB server.  seems pointless, plus it sucks.
Dec 29 04:39:32 <jasta> it doesn't even handshake correctly (come on!)
Dec 29 04:40:55 <swetland> it was a one-off for a demo ^^
Dec 29 04:41:06 <swetland> needed to display the device screen on a laptop to project
Dec 29 04:41:21 <romainguy__> hey swetland :)
Dec 29 04:41:26 <jasta> swetland: really?
Dec 29 04:41:29 <swetland> yes, really
Dec 29 04:41:43 <swetland> at some point we'll either spiff it up to be more useful or disable/remove it
Dec 29 04:42:09 <jasta> well, in case you're in the area of that code, you might want to change "RFB %03d.%03d" to "RFB %03d.%03d\n". ;)
Dec 29 04:42:39 <swetland> I'll pass that on to the guy who wrote it
Dec 29 04:43:41 <jasta> Well, I didn't really spend much time developing my server, honestly.
Dec 29 04:43:52 <jasta> libvncserver made it all very easy, and it cross compiled without any troubles.
Dec 29 04:45:25 <swetland> sadly it's GPL'd, so that's not a route open to us
Dec 29 04:45:58 <jasta> well, it would be if you didn't taint yourself in the surfaceflinger library, but i suppose that you must for efficiency.
Dec 29 04:46:23 <jasta> i didn't have to worry about efficiency, since my host machine running the emulator is an athlon x2 4400+ :)
Dec 29 04:46:26 <swetland> yeah, integrating it with the compositer is the least-overhead way to do it
Dec 29 04:46:50 <jasta> hehe, i could have gone further, but it was just a hack job.  i just poll a mmap'd framebuffer device.
Dec 29 04:47:04 <jasta> and difference each frame to determine differences.
Dec 29 04:47:07 <swetland> how do you figure out which fb to grab?
Dec 29 04:47:18 <jasta> what do you mean?
Dec 29 04:47:19 <swetland> the fb is doublebuffered and swaps on each update
Dec 29 04:47:36 <swetland> so when you mmap /dev/graphics/fb0 you get two framebuffers one after the other
Dec 29 04:47:38 <jasta> yeah, i noticed that, actually.  i have currently ignored the problem, but did plan to revisit it.
Dec 29 04:47:56 <swetland> the current yoffset from the fbparams indicates the active one
Dec 29 04:48:11 <jasta> i dumped a bunch of debug last night and discovered the double buffering.  my server currently maps offset=0, length=xres*yres, which is not the correct buffer i have found.
Dec 29 04:48:11 <swetland> I think you'd have to extend the fb driver to allow a third party to cleanly observe when it flips
Dec 29 04:48:55 <swetland> (which is the other reason to put it in the surfaceflinger, since the sf always knows the buffer state, etc)
Dec 29 04:49:25 <jasta> yes, of course.  the surfaceflinger is obviously correct, it's just that i do not have access to that area of the code (**ahem**)
Dec 29 04:49:30 <swetland> yet!
Dec 29 04:49:34 <jasta> **ahem**
Dec 29 04:49:47 <jasta> yes, of course.  yet.
Dec 29 04:50:05 <swetland> I fully expect that when we finally do a full source release people will totally pull stuff apart and build things we never thought of.  It'll be fun to watch
Dec 29 04:50:30 <jasta> well, perhaps.
Dec 29 04:50:42 <swetland> almost certainly. lots of fun stuff in the system.
Dec 29 04:50:43 <jasta> the open source community is not to be taken for granted.  sometimes they don't do squat :)
Dec 29 04:51:17 <jasta> it all depends on how everyone "feels" about Android.  i know, it sounds absurd, but it's significant.
Dec 29 04:51:23 <swetland> whether they add/modify stuff, pull bits out and use it for something else, or what, I can't predict
Dec 29 04:51:45 <jasta> not to nit pick all the fine work that has gone into this platform so far, but releasing the source code too late is a great way to rub the entire community the wrong way and end up with very little innovation.
Dec 29 04:52:17 <swetland> community relations is tricky business.  I'm just a simple kernel engineer.
Dec 29 04:52:45 <jasta> i'm a member of said community, and i still agree.
Dec 29 04:53:33 <romainguy__> swetland: Don't use "simple" to describe your work :p
Dec 29 04:54:12 <swetland> you get a bunch of people wanting a bunch of different things. and some who will consider it all vapor until it ships in hardware and some who will consider it all vapor until it ships in source and some who'll never be happy because we didn't use their favorite libfoo or bard.  ^^
Dec 29 04:54:38 <swetland> romain: systems stuff is simple! therein lies the complexity
Dec 29 04:55:22 <romainguy__> swetland: this description reminds me of how I think about UI toolkits
Dec 29 04:55:39 <swetland> simple is a *lot* harder than complex to get right at times
Dec 29 04:55:44 <swetland> I'll grant you that
Dec 29 04:55:50 <jasta> swetland: the lion's share of the work will come from people who think it is vapor until you ship source.  trust me.
Dec 29 04:55:57 <swetland> anyone can solve problems by stacking a few more layers on top of things
Dec 29 04:56:26 <romainguy__> so true
Dec 29 04:56:45 <swetland> jasta: we hope the community will be interested, but we are not expecting the community to finish it for us (though they may well add things we didn't consider)
Dec 29 04:57:24 <jasta> i don't know that i can believe that, actually.
Dec 29 04:57:26 <swetland> when it ships in hardware and source is released it should be a pretty functional system. probably not the end-all be-all of mobile whatevers, but complete enough to be a good experience (if we do our jobs right)
Dec 29 04:57:38 <jasta> it definitely sounds like google is leaning on innovation from outside its doors.
Dec 29 04:57:58 <jasta> whether that's in members of the OHA or the open source community i can't say.  i'd hedge my bets on the open source community, however.
Dec 29 04:58:21 <swetland> of course the system makes use of a lot of great existing open source stuff
Dec 29 04:58:36 <jasta> i mean future innovation, not just what you've built upon currently.
Dec 29 04:58:49 <swetland> true of any platform
Dec 29 04:58:53 <swetland> nobody thinks of everything
Dec 29 04:59:11 <swetland> the "killer apps" as it were rarely originate with the platform developers
Dec 29 04:59:31 <jasta> that's true, of course.
Dec 29 04:59:39 <jasta> of course, Google created the killer app for my phone.
Dec 29 04:59:45 <romainguy__> swetland: who are too busy writing the platform anyway :)
Dec 29 05:00:00 <swetland> jasta: which one?
Dec 29 05:00:04 <jasta> gmm.
Dec 29 05:00:31 <swetland> for my money, it's all about awesome email and IM, but that's just my personal preference (4+ years developing and using hiptop weigh into that strongly)
Dec 29 05:00:40 <jasta> i have it in my sights to create a killer app for android, though.  so i certainly hope you folks do your jobs well :)
Dec 29 05:01:32 <jasta> i was hoping to wait until source code was released to start, but i am losing faith that is a sensible path.
Dec 29 05:01:39 <rwhitby> swetland: for me, it's all about offline-ability of email, web, feeds, etc.  In Australia, mobile data is too expensive for always-on.
Dec 29 05:02:14 <rwhitby> (merry <insert appropriate holiday greeting here> to all)
Dec 29 05:02:29 <swetland> rwhitby: even if always-on is affordable, the networks aren't always there when you need 'em. which was something I loved about email on the hiptop. grabs it when it can, you can compose when offline, will send when it can, etc)
Dec 29 05:03:02 <rwhitby> swetland: I use a treo650, with bluetooth access points (nslu2 with usb dongle) at home and work.  that covers all of my offline reading needs :-)
Dec 29 05:03:17 <rwhitby> (reading and writing).
Dec 29 05:03:37 <rwhitby> if it wasn't for a 30 min each way bus ride each day, there is no way I would keep up with mailing list traffic.
Dec 29 05:03:51 <swetland> I think that'll be one of the nice fallouts of doing this as an open platform -- it becomes a lot easier for people to specialize for niche or vertical applications that you'd never see a big OEM or carrier (focused on volume) go after
Dec 29 05:05:24 <swetland> and regarding concerns about source release, trust me, everyone working on this wants to see the source out there as much (or more) than people watching from the outside.  I've been working on it for 3 years now -- I've obviously been waiting longer ^^
Dec 29 05:05:54 <jasta> swetland: so, would we be wise to assume that source code will not arrive until handsets do?
Dec 29 05:07:03 <swetland> I don't think we have an official statement on that at this point
Dec 29 05:07:18 * swetland pokes around
Dec 29 05:07:21 <romainguy__> if we have, it's on the OHA web site
Dec 29 05:07:27 * benley pokes swetland for no good reason
Dec 29 05:07:42 <jasta> swetland: hmm, well i certainly hope that it comes soon.
Dec 29 05:07:51 <jasta> i am starting to feel disillusioned about the whole thing.
Dec 29 05:08:29 <jasta> it's turning me cynical :)
Dec 29 05:08:43 * swetland was cynical long before he started working on this
Dec 29 05:09:34 <benley> I should work on that. I still possess some of the youthful wide-eyed hope.
Dec 29 05:09:45 <benley> but that can't be healthy to keep around.
Dec 29 05:09:53 <rwhitby> swetland: most official thing I heard was your marketing dude say something in england that source wouldn't be released before handsets, probably in the 2H08.
Dec 29 05:10:09 <rwhitby> but of course you won't be able to confirm or deny that ;-)
Dec 29 05:10:12 <jasta> i had not heard that.  but it would certainly be bad news.
Dec 29 05:10:21 <romainguy__> jasta: how so?
Dec 29 05:10:25 <swetland> indeed (cannot confirm or deny or they shoot me!)
Dec 29 05:10:31 <jasta> rwhitby: i'm hoping if we catch one of them off guard they'll spill the beans ;)
Dec 29 05:10:51 <jasta> romainguy__: how so about what?
Dec 29 05:11:03 <romainguy__> jasta: why would that be bad news?
Dec 29 05:11:20 <rwhitby> jasta: I imagine the Google training for people allowed to hang in here is like that scene from clockwork orange ...
Dec 29 05:11:36 <benley> rwhitby: I'm a googler and nobody's given me specific instructions
Dec 29 05:11:45 <swetland> oh come on! do you see us putting anybody in eyelid clamps? ^^
Dec 29 05:11:52 <benley> (I don't work on android fulltime though)
Dec 29 05:12:19 <swetland> oh, misread, training for *google* people
Dec 29 05:12:47 <rwhitby> swetland: eyes still blurry from the clamps, hey? ;-)
Dec 29 05:13:00 <swetland> me, I just push my luck and hope I've got enough open P1 bugs that they won't fire me for being a smartass on irc
Dec 29 05:13:05 <jasta> romainguy__: Because the platform as it stands now is insufficiently documented to decipher all the magic pieces that Google has already written themselves.  It also does not give us a fair glimpse of the maturity of the code base, since we must trust that there is ever more than we can see.
Dec 29 05:13:19 <romainguy__> jasta: Ok
Dec 29 05:13:23 <swetland> jasta: it's not done yet
Dec 29 05:13:23 <jasta> And that extra measure of faith is unsettling, especially to members of a community who are not accustomed to it.
Dec 29 05:13:49 <jasta> swetland: We know it is not done.  But there are shades of gray, as you know.
Dec 29 05:13:50 <swetland> that kinda being the point of an early-look SDK release
Dec 29 05:13:56 <rwhitby> jasta: the way I see it is that there are companies that have paid lots and lots of money to be part of OHA (either directly or indirectly), and they will want to keep some competitive advantage (i.e. the source) until they release hardware.
Dec 29 05:14:30 <jasta> swetland: Well, keep in mind that you can't maintain the "early look" status for much longer.  Once you officially begin the developer challenge you are encouring a presumption on our side that this is no longer vaporware, and that we ought to care about how it proceeds.
Dec 29 05:14:30 <swetland> all I can say for people who will not be happy until they have the source code is that maybe they should wait until source is released -- if that's the only thing that will make 'em feel better there's not a whole lot else one can do
Dec 29 05:14:41 <jasta> Again, the extra measure of faith is unsettling.  Source should accompany the developer challenge.
Dec 29 05:15:09 <romainguy__> jasta: Why is it so unsettling?
Dec 29 05:15:14 <benley> jasta: if only it were that simple
Dec 29 05:15:16 <romainguy__> (Just curious)
Dec 29 05:15:25 <rwhitby> jasta: would you have preferred that they delay the start of the challenge until after hardware was released?
Dec 29 05:16:14 <jasta> rwhitby: Of course not, but these two pieces are not logically connected.  Source and hardware are not tied so closely together.
Dec 29 05:16:25 <rwhitby> I mean you have to remember you're dealing with billion dollar companies here, not just a bunch of open source aware engineers.
Dec 29 05:16:56 <jasta> romainguy__: Because you're asking for a large commitment of effort to a project that we cannot see, with perceivably no fair explanation as to why.
Dec 29 05:17:03 <swetland> obviously there is a chicken and egg problem regarding development and incomplete platforms.  the whole developer challenge addresses this (partially) by offering cash prizes (no strings attached) as an incentive.
Dec 29 05:17:07 <romainguy__> jasta: I understand
Dec 29 05:17:30 <jasta> romainguy__: Very specifically, because we cannot see the components that hold up our code to determine if our code should be written in a different way or what design elements are sensible.  You make our job harder.
Dec 29 05:17:31 <romainguy__> From personal experience though, I know that timing the release of source code is not very easy
Dec 29 05:17:32 <rwhitby> jasta: the explanation is simple.  there's companies investing millions of dollars involved.
Dec 29 05:17:37 <romainguy__> I experienced it first hand at Sun
Dec 29 05:18:17 <romainguy__> I've seen source code released too early and it led nowhere
Dec 29 05:18:49 <jasta> My girlfriend is literally poking me now.  I think I'd better digress and pay attention to her :)
Dec 29 05:18:58 <romainguy__> Same thing here :)
Dec 29 05:19:01 <romainguy__> Good night guys
Dec 29 05:19:16 <jasta> Night.
Dec 29 05:46:27 <jasta> swetland: hey, something you might be able to help me with.  what is the keycode for the "Back" key (escape in the emulator)?
Dec 29 05:47:10 <jasta> my VNC server injects keyboard input via serio_interrupt, sending the keyboard scancode for the key pressed.  I built dumpkeys on the emulator and it shows keycode 1 as Escape, which is what I expected, but sending that keycode seems to do something entirely different.
Dec 29 05:47:19 <jasta> do you know of any way to figure this out?
Dec 29 05:49:42 <swetland> you mean the linux keycode (like you get from /dev/input/event*)?
Dec 29 05:54:43 <jasta> yes.
Dec 29 05:54:53 <jasta> can you just read /dev/input/eventn to figure it out?
Dec 29 05:55:14 <swetland> run getevent from the shell
Dec 29 05:55:39 <jasta> thanks.
Dec 29 05:55:43 <swetland> it'll chatter about all /dev/input/event* events
Dec 29 06:24:06 <zgold> Good evening!
Dec 29 06:24:17 <benley> guten tag
Dec 29 06:24:52 <zgold> be this the channel for the google os Android or just another funky gang 'o people who like androids?
Dec 29 06:25:01 <benley> the google one.
Dec 29 06:25:10 <zgold> Be you somebody with experience with said OS?
Dec 29 06:25:16 <benley> minimally.
Dec 29 06:25:22 <zgold> know much of services?
Dec 29 06:25:24 <benley> there are other people lurking I think.
Dec 29 06:25:32 <benley> who know Things
Dec 29 06:25:47 <billmeltsner> true, there are others of us lurking.
Dec 29 06:25:48 <billmeltsner> I mean
Dec 29 06:25:49 * billmeltsner lurks
Dec 29 06:26:10 <zgold> I'm writing my first app, trying to launch a service and bindService returns true, however the ServiceConnection's onServiceConnected method never gets called
Dec 29 06:30:05 <zgold> Hm I takes it all the lurkers of this area be off making merry holiday
Dec 29 06:45:52 <zgold> whoot was just an AndroidManifest issue (I think)
Dec 29 06:46:24 <zgold> Yay for missing an s in a plural ><
Dec 29 07:06:44 <zgold> So, before I knew about AIDLs i was half way through implementing it myself.. sigh
Dec 29 07:41:51 <zgold> What do I do if I find an incompatibility in a java library with android (that android supports, java.net)
Dec 29 08:15:55 <jasta> zgold: report it to google via the android-developers forum.
Dec 29 09:00:17 <jasta> hmm, i thought i was able to create res/ folders like layout-land-320x240 and stuff.  apparently not.,
Dec 29 12:32:15 <cutmasta> morning
Dec 29 13:00:43 <tulcod> is there a multihtreading sample anywhere?
Dec 29 17:13:49 <jpmatrix> hi
Dec 29 17:14:06 <jpmatrix> anyone with Android on his Zaurus here ?
Dec 29 17:33:18 <zgold> Is there anyway for a service to call back to a specific activity (say, its creator?)
Dec 29 17:38:03 <cybereagle> as far as i know the accepted way is for it to send an Intent to it
Dec 29 17:38:26 <cybereagle> there was some discussion on here before about wishing the IPC went both ways, but it doesn't lol
Dec 29 17:38:55 <tulcod> zgold: if it's a different process? no
Dec 29 17:39:03 <tulcod> zgold: but you could try using a file...
Dec 29 17:39:49 <cybereagle> i suppose you could use a shares object of some kind, but it would be a bit messy, not sure if it would work with the process life cycle either
Dec 29 17:39:53 <cybereagle> *shared
Dec 29 17:41:11 <zgold> What about the getApplication() function?
Dec 29 17:42:38 <zgold> Hm this could indeed get messy if I have to use intents only
Dec 29 17:42:47 <zgold> are they efficient at least?
Dec 29 17:42:59 <zgold> It'll be an explict call with a custum URI i imagine
Dec 29 17:43:51 <cybereagle> yeah a while back we were talking about the efficiency of intents
Dec 29 17:44:03 <cybereagle> i havent done any tests but its the basic way for components to talk
Dec 29 17:44:16 <cybereagle> it seems to work ok for my stuff so far
Dec 29 17:44:56 <cybereagle> they make it nice for the whole late binding thing, but sometimes it'd be nice to have a more "internel" lightweight way to communicate upwards
Dec 29 17:44:58 <zgold> I'm writing a messenger client and the networking thread obviously needs to talk to the activity (msg recieved etc.).  So if its not gunna bring things to a halt to throw an intent everytime that happens, thats how i'll do it :).  Thanks cybereagle
Dec 29 17:45:13 <cybereagle> lol thats exactly what i'm doing in fact :P
Dec 29 17:45:21 <zgold> hehe
Dec 29 17:45:33 <zgold> I'm not really doing it for the android challenge, doing it to learn how to do it :)
Dec 29 17:45:47 <zgold> so no competetion amoungst peers, eh? :)
Dec 29 17:45:48 <cybereagle> a bit of both for me
Dec 29 17:46:03 <cybereagle> i think to win anything i'd prob need am ore imaginative app anyway
Dec 29 17:46:25 <cybereagle> my first thought was "i need an irc client". then i figured i may as well add IM networks to it lol
Dec 29 17:46:39 <tulcod> cybereagle: ...
Dec 29 17:46:46 <tulcod> cybereagle: since when do we have a spiritual connection?
Dec 29 17:47:01 <cybereagle> lol!
Dec 29 17:47:06 <cybereagle> are you doing the same thing? :P
Dec 29 17:47:12 <tulcod> pretty much
Dec 29 17:47:15 <cybereagle> haha
Dec 29 17:47:17 <zgold> Heh yeah, im trying to make it work for aim/yahoo/msn etc.
Dec 29 17:47:21 <zgold> guess it was a common idea =P
Dec 29 17:47:25 <cybereagle> i think its an irc junkies first reaction
Dec 29 17:47:30 <tulcod> :p
Dec 29 17:47:40 <tulcod> well, then I'll just rapidly finish this :p
Dec 29 17:48:05 <tulcod> got a couple more ideas :D
Dec 29 17:48:06 <zgold> I'm having a minor issue with my Aim library blowing up the heap but asides from that, looks good =P
Dec 29 17:48:20 <cybereagle> if we all used a standardised set of intents, they could all be interchangable, swapping ui's and services and stuff
Dec 29 17:48:50 <cybereagle> christmas held me up, i havent got that far yet. i'm also waiting for them to fix nio
Dec 29 17:48:59 <tulcod> nio?
Dec 29 17:49:07 <cybereagle> java's non-blocking io
Dec 29 17:49:07 <zgold> That'd be a good idea actually
Dec 29 17:49:17 <zgold> but is there any formal way to do that?
Dec 29 17:49:25 <tulcod> cybereagle: why the heck would you want non-blocking io
Dec 29 17:49:26 <cybereagle> but the harmony lib included right now hasnt implemented the network nio channels
Dec 29 17:49:31 <tulcod> multithreading for president
Dec 29 17:49:38 <cybereagle> but they say it'll be in an upcoming release
Dec 29 17:49:46 <cybereagle> lol thread overhead for the lose
Dec 29 17:49:56 <tulcod> it's multithreaded anyway
Dec 29 17:50:35 <cybereagle> yeah, but you'd be making a lot more context switches with threads for each socket
Dec 29 17:50:53 <tulcod> but you'd use 100% cpu with nio
Dec 29 17:51:15 <cybereagle> though my original reason for switching was an annoying thing trying to interrupt the blocking calls. i forget what call it was but it didnt throw an exception like sun's vm does, and even though it worked, it wasnt as clean
Dec 29 17:51:17 <tulcod> and if 2 out of 3 threads are blocked, I don't think you'll make a lot of context switches
Dec 29 17:51:45 <cybereagle> it shouldnt use 100% cpu if you do it right
Dec 29 17:51:59 <cybereagle> but i';ve yet to test it obviously lol
Dec 29 17:52:00 <tulcod> yeah, if you timeout 0.1 seconds constantly :-/
Dec 29 17:52:31 <cybereagle> the load averages i've seen in other situations arent that bad
Dec 29 17:55:25 <cybereagle> anyway, thanks to the open source way, we get to have both done by different people and see which one works best
Dec 29 17:55:25 <cybereagle> woot
Dec 29 17:55:26 <cybereagle> lol
Dec 29 17:56:07 <cybereagle> most of my apps are probably gonna be primarily for the purpose of saying "duuude, look what my phone can do" hehe
Dec 29 18:14:54 <zgold> android's networking is picky
Dec 29 18:26:47 <zgold> Developing an AIM app is officially annoying: you get rate limited as you debug :)
Dec 29 18:33:01 <Lede> getting layouts right is annoying
Dec 29 19:59:32 <Lede> there is something wrong with my EdiText box, if it gets filled it wraps and everythings stays normal, as soon as i load it again with the given text everything resizes horizontally
Dec 29 19:59:50 * Lede pulls hair
Dec 29 20:17:24 <Lede> i got it working
Dec 29 20:20:42 <Lede> i don't know why though .)
Dec 29 20:37:43 <Garett> anyone around?
Dec 29 20:39:19 <Garett> I'm looking to change my application's icon to indicate the status... has anyone tried this, or have any idea about where I should look?
Dec 29 20:57:22 <zgold> Yes!!! Finally signed in via the emulator
Dec 29 20:59:27 <Garett> irc app?
Dec 29 20:59:48 <jasta> stupid power outages :(
Dec 29 21:00:12 <billmeltsner> I was writing an IRC app for android at one point
Dec 29 21:00:15 <billmeltsner> RFC compliance is a pain
Dec 29 21:00:29 <Garett> anyone know how someone could change their package icon?
Dec 29 21:00:41 <Garett> programatically?
Dec 29 21:00:43 <zgold> aim app
Dec 29 21:00:43 <jasta> Garett: change res/drawable/icon.png
Dec 29 21:01:01 <zgold> hard part is done, phew.  Now just to attach a UI
Dec 29 21:01:22 <Garett> jasta... is that a physical path on the device? I thought it was compiled into the app
Dec 29 21:01:39 <jasta> it is compiled into the apk, yes.
Dec 29 21:01:52 <Garett> can you change it after it's deployed?
Dec 29 21:01:53 <jasta> i don't know of any way to runtime adjust your icon.
Dec 29 21:01:59 <Garett> ok
Dec 29 21:02:00 <Garett> damn
Dec 29 21:02:12 <billmeltsner> Well
Dec 29 21:02:15 <billmeltsner> Hm.
Dec 29 21:02:37 <Lede> Garett: even if you figure out how to dynamically update/change icon.png, the underlying "window manager" wouldnt update it
Dec 29 21:02:44 <Lede> :/
Dec 29 21:02:47 <cybereagle> might be a way by hacking at R.java, changing where it accesses it, but i'm only guessing
Dec 29 21:03:10 <Garett> it would be great to show your app's status with the icon...
Dec 29 21:03:21 <Garett> for example... and email app that would show that you have new messages
Dec 29 21:03:35 <cybereagle> mention it to a google person when they get back in here after new year, they might just add that feature
Dec 29 21:03:35 <jasta> Garett: that would be done with the notification area, i'd think.
Dec 29 21:03:38 <cybereagle> it would indeed be cool
Dec 29 21:03:43 <cybereagle> especially given the app life cycle they have
Dec 29 21:04:00 <jasta> but i see where you're getting at.  having the e-mail app have a little tag with the number of messages or something would be neat.
Dec 29 21:04:03 <cybereagle> yeah, that kind of thing would be a Notification
Dec 29 21:04:23 <cybereagle> but it would still be good to have it change for stuff like whether the app is running and connected or not
Dec 29 21:04:35 <Garett> ya... i know you can do it with the notification area
Dec 29 21:04:38 <cybereagle> or yeah a message count, they;d be sweet
Dec 29 21:04:46 <Garett> ya... blackberry does it for their email app
Dec 29 21:05:03 <Garett> and shows you the "new messages" icon in the notification area
Dec 29 21:05:06 <Garett> it's pretty useful
Dec 29 21:05:46 <Lede> would be nice..
Dec 29 21:05:53 <zgold> Does anybody know what google plans to do with the home screen?
Dec 29 21:05:56 <Garett> ok... so google people hang out in here?
Dec 29 21:06:02 <zgold> They can't possibly plan to leave 90% of it blank (just wallpaper)
Dec 29 21:06:06 <cybereagle> yeah, they just left for the holidays
Dec 29 21:06:14 <Garett> what are some of their nicks?
Dec 29 21:06:20 <cybereagle> you mention a bug here and they hear they usually get it fixed pretty quick
Dec 29 21:06:21 <zgold> i bet they'll be ops
Dec 29 21:06:26 <Garett> ah, ok
Dec 29 21:06:45 <cybereagle> not actually seen any ops, but one of the most active people is "morrildl"
Dec 29 21:07:00 <Garett> ah, ya, he's the guy in the video
Dec 29 21:07:03 <cybereagle> no one ever seems to op up
Dec 29 21:07:28 <zgold> I'm off for now, my emulator is on aim which makes me happy.  good luck folks
Dec 29 21:08:00 <Garett> so, how are apps going to be deployed on real devices?  OTA?
Dec 29 21:08:14 <Garett> point a browser to an apk?
Dec 29 21:09:38 <cybereagle> i dont know to be honest, i would assume you could download an apk though
Dec 29 21:10:15 <Garett> ok.. i'll do some diggin around
Dec 29 21:10:23 <billmeltsner> I gather there's some sort of installer app that handles a .apk file, makes sure all the special things get authorization, and then pushes it to the applications directory
Dec 29 21:10:46 <billmeltsner> s/\'s/\ will\ be/
Dec 29 21:11:27 <Garett> ok... it would be awesome if users could simply click on a link from a webpage to install an app
Dec 29 21:11:36 <Garett> hopefully that would work.. i'll try it out
Dec 29 21:15:17 <Garett> nah, doesn't work yet
Dec 29 21:15:24 <Garett> i guess that can be another feature request
Dec 29 21:16:19 <cybereagle> i'd imagine they'll be planning that anyway, it would blow a hole in the openness if it was hard to install apps lol
Dec 29 21:17:09 <Garett> ya :)... maybe they are hoping someone makes an app that does that?
Dec 29 21:57:10 <jasta> yawn
Dec 29 21:57:31 <jasta> oops, forgot to create /dev/kbde in my vnc server :)
Dec 29 21:57:36 <jasta> *fixed* :)
Dec 29 22:20:20 <zgold> well
Dec 29 22:20:29 <zgold> installing apps is just putting the file in a folder
Dec 29 22:20:39 <zgold> seeing as its linux... you could install an app with one ssh command
Dec 29 22:28:52 <Garett> but how would you invoke that from the browser?
Dec 29 22:30:11 <Garett> you would have to create a browser capable of handling apk files
Dec 29 22:32:15 <billmeltsner> well, the browser would probably send a message when you tried to download an apk
Dec 29 22:32:30 <billmeltsner> and the installer would pick up that message and handle things
Dec 29 22:33:08 <Garett> the current browser you think broadcasts an intent when a user tries to download an apk file?
Dec 29 22:33:28 <billmeltsner> I dunno if the current browser does
Dec 29 22:33:32 <Garett> or are you saying that would be the ideal solution
Dec 29 22:33:37 <Garett> ahh, yes.. that would make sense to me
Dec 29 22:33:38 <billmeltsner> that seems like the ideal solution to me
Dec 29 22:33:57 <billmeltsner> since having an independent installer means you could get apks from other sources too
Dec 29 22:34:01 <billmeltsner> like if someone emailed you one
Dec 29 22:34:07 <Garett> i wonder how hard it would be to build a browser with the webkit stuff?
Dec 29 22:34:08 <billmeltsner> the installer could handle it from the email program
Dec 29 22:34:32 <Garett> and wrangle it to do what we need
Dec 29 22:34:56 <jasta> i suspect that google will already consider the need for installation of packages via the browser.
Dec 29 22:35:27 <Garett> i sure hope so... i really don't want to spend too much time worrying about it
Dec 29 22:37:47 <jasta> man i love winter coding
Dec 29 22:37:59 <jasta> i've got a cup of hot chocolate, my robe on, and fuzzy blankets everywhere
Dec 29 22:42:01 <Lede> you pimp
Dec 29 22:46:21 <jasta> i'm gonna start on a new project now
Dec 29 22:46:27 <jasta> how about an android solitaire game.  sure, why not
Dec 29 22:47:56 <cybereagle> if any of the handsets have an integrated camera, how about strip poker ;)
Dec 29 22:47:57 <jasonlee> i'd play it =)
Dec 29 22:50:28 <jasonlee> i meant solitaire, not strip poker tho!
Dec 29 22:50:35 <cybereagle> hehe bad timing
Dec 29 22:52:08 <jasta> the tough part of a solitaire game (any game, really) is trying to design the interface to work optimally on all the screen sizes and orientations.
Dec 29 22:52:42 <billmeltsner> well, we have to design our apps to assume we have a touch screen
Dec 29 22:52:46 <billmeltsner> so no keyboard for you!
Dec 29 22:53:01 <jasta> we do?
Dec 29 22:53:41 <billmeltsner> aye, with the latest SDK update the default emulator skin is the keyboardless red one
Dec 29 22:53:52 <billmeltsner> which is what we should assume our target handset is for coding things
Dec 29 22:54:28 <jasta> hmm.  that sucks.  i hate touchscreens.
Dec 29 22:55:49 <jasonlee> it'd be nice if i could get a job doing android stuff right now, i'm getting so sick of J2EE
Dec 29 23:06:58 <cybereagle> i've seen at least one guy in here whose getting paid to explore android
Dec 29 23:07:00 <cybereagle> forget who
Dec 29 23:23:38 <jasta> sigh...
Dec 29 23:23:48 <jasta> http://code.google.com/android/devel/resources-i18n.html#AlternateResources -- anyone gotten this feature to work?
Dec 29 23:24:00 <jasta> it seems that specifying dimensions in your layout resource directories is not possible.
Dec 29 23:35:44 <billmeltsner> you can't do drawable-port-320x240?
Dec 29 23:35:50 <jasta> you cannot.
Dec 29 23:36:13 <jasta> it claims that 320x240 portion is invalid.
Dec 29 23:36:20 <billmeltsner> :\
**** ENDING LOGGING AT Sun Dec 30 02:59:57 2007