Android Orbiter

[jimbodude]

I just picked up a Motorola Droid - it's sweet.  The only thing it can't do is be a LinuxMCE Orbiter...  Has anyone started work on this?  I can volunteer to do testing or work along side someone over the next few weeks.  I probably won't have time to do hardcore development work until after winter.

[kyfalcon]

I plan on it , as soon as I get my system back up the way I want it. One MD do go. I think I am going to look into using wireless Lan instead of bluetooth. I will let you know in the next few weeks.

[jimbodude]

Ideally, we should be able to choose.  Personally, I don't care which is used.

[kyfalcon]

I agree and if I can get one working I can try the other. But I want to get one working first.

[jimbodude]

Yes. Have you looked into Android development at all? I'm planning on setting up a dev environment next week if I have time.

[kyfalcon]

Started messing with it. It is java which is my area of expertise.

[jimbodude]

Do you have an idea of what needs to be done to get JavaMO running on it? Should be pretty straight forward to get it to run, and then put some Android wrappers for install and launch I would guess.

[kyfalcon]

Haven't really looked at it on android yet. I messed with it a little on the nokia 5800. Bluetooth is the big issue, that's why I may try something with the wifi if possible.

[jimbodude]

Ok. We'll see where it goes then. Keep eachother up to date on progress.

[jimbodude]

I got an Android development environment, and I think I've got Android development figured out for the most part - it's pretty nice actually.  I looked at the JavaMO code - it seems that it's not going to be terribly useful, except as a reference for a complete rewrite.  JavaMO is heavily dependent on JavaME libraries which are not available in Android.

Supporting WiFi is a very different track than supporting Bluetooth.  The Bluetooth Orbiters are mobile orbiters, which are very different than networked orbiters.  That means none of the JavaMO will be helpful for a network orbiter, which would require a Java-based DCE library.

I'm continuing my investigation into other methods.  If I turn up nothing, I plan on porting JavaMO to Android.

[darrenmason]

Jimbodude,

I had a quick look the other day at all this as well and I actually came up with the opposite opinion on it.

I didn't think that there was a big dependancy on the JavaME libraries really... what did you think the problem bits would be?

Is your aim to write a complete orbiter in Android or to get JavaMO working - as these are completely different tasks. I am interested in both but I think getting the JavaMO working first is probably a good first step. I went to have a go at this but my dev machine's HD seems to have failed.

Supporting Wi-Fi (or just socket connections rather than Bluetooth) should not be very hard and probably just needs another device created that makes use of the BD stuff. The BD stuff actually has code already in it that allows it work (for testing) over sockets. This would not support proximity detection etc that Bluetooth does however.

I will try and help you out in whatever way I can. I think that the Android dev libraries provide a lot of potential for implementing a complete orbiter but will either need java based DCE comms or some form of proxy device arrangement.

Will update if I make any progression....

Regards
Darren

[jimbodude]

Almost every class uses the javax.mobileedition packages, which are not available in Android.  These are used for Bluetooth comms, for rendering, and for hardware control (such as backlight power).  To replace them would seem to require a similar, but different, implementation using Android libraries.

I think a lot of this dependency could be removed, however, to make a more generic implementation of the core features that could support both JavaME and Android, with some specific implementation bits created for each platform.  I'm not sure if it's worth the effort, though.

I have some other things going on for the next 2 weeks, so I'll have to put this aside for a bit.  Don't let that stop anyone from having a go at it - the code is not terribly complex, and I would bet that someone with some Android dev experience could pull this off in a few days or less.

[kyfalcon]

I'm on vacation right now but when I get home I have 2 priorities.

1. Get my system working with a single remote control. (Wife comes first).

2. Start working on the android mobile orbiter. To be honest I'd like to get remote functionality working first on the droid and then add the fluff of album, currently playing, etc. If this takes a complete rework I think it maybe best to break it down into different functional steps/components.

[jimbodude]

The code is not nearly as complex as you are thinking... Take a look - the phone isn't doing much computation work outside of drawing the image on the screen and figuring out what buttons have been pressed.  Excuse me for my bluntness earlier - by "complete rewrite", I didn't mean it would be a difficult one, nor should it require much design work - unless you want it to, I guess.  From what I can see, the work pretty much consists of changing out the JavaME libraries for Android ones...  All I meant was that JavaME and Android probably can't share a version given the way this is written now.

I may have over-exaggerated by saying "almost all classes depend on JavaME libraries" - in fact, less than half of the classes depend on JavaME (about 25 out of 60), and they are mostly not too complex.

I also mis-typed the package name - its "javax.microedition" not "javax.mobileedition".

[darrenmason]

Went through the javamo code and it should be fine as a basis for writing the equivalent using the android framework.
I agree, probably more effort than its worth to share same codebase - but have not discounted it yet.
Will have a go at it on the weekend.