Re: Suggestion: port Dalvik JIT to Maemo
 

It would maybe even make more sense to port Dalvik than porting JavaME to Maemo5. There is those improvements in Dalvik which makes it more suitable for mobile devices.

Re: Suggestion: port Dalvik JIT to Maemo
 

I dont say people here a not smart something like that. I am just a realist.

Lets give another Example: the open flash that I have forgoten the name of can it compete with Adobe closed one? No it cant and probadly never will. Same goes for an Dalvik clone too N900. And again why should developers here bother try it when those smart developers can do better apps directly in QT/Quick and make it work far better in QT than in Dalvik clone? Dalvik on N900 would be slow and then again people here would start complain and say "hey why is N900 so slow HTC is much faster blablabla..". Seriously it is an impossible mission make it work smooth..

If ýou or another developers can show I am wrong fine for me. But atm. I dont see any good reason to port it.

My tip to all android fanboys who want that X/Y/Z app is to hint the Maemo developers about it (maybe setup a new forum thread like:"Android apps I want to see in Maemo/Meego") maybe some of them have some time to port it to QT/Quick and even make it better than the original Android one :-D Imho its far better use time for this than try to make a dalvik clone. Because I am sure it will never will work.

Btw. about twitter here is an app that maybe is what N900 one example what N900 is lacking atm but will progress faster now when QTQuick is getting more stable :-D

http://thehandheldblog.com/2011/01/1...-qml-download/

I am sure we will see more 2011..

Qt/Quick is here to stay ;)

Re: Suggestion: port Dalvik JIT to Maemo
 

You are comparing a rewrite of a closed commercial app, to the port of the open source app with all sources open?
Don't you see the dissonance here?


How many high quality apps would one write from scratch in a time needed to port open source Dalvik and getting access to all Android apps in existence?
And then having to do nothing for more Android apps to appear?

Your point of view. OK. Just don't proclaim it a POV of the whole Maemo community, please.

Calling people names does not make you a partner for discussion.

EOT

Re: Suggestion: port Dalvik JIT to Maemo
 

Not with the momentum those two are gaining. There's at least a good 5 years of prosperous developing in both.

Out of a job, with C# or java? That will be the day.

Re: Suggestion: port Dalvik JIT to Maemo
 

Buy an android phone? Why? I'm waiting 2 month. Then the audio problem with NitDroid will be solved. Some Android phones are still on Android 1.5. n900 has already Gingerbread!!!

I like very much maemo and qt.
Android is only a toy for teenies, no quality, no performance, no respectable GUI, no stable main applications.


But the developers of NitDroid did a very impressive job. Congratulations and all my respect and my best wishes!

Maemo risks that the users of their only important device go Android.
Is this a good publicity for maemo? Is this a future for maemo programmers?

I like maemo. And because of this I find that Dalvik on Maemo is absolutely necessary.
With my vision NitDroid on n900 offers supplementary options to maemo and will not replace maemo.

If there will be no dalvik Maemo will lose users. And users is the only thing that maemo is missing.

Google offers much applications that are interesting:
Where is streetview on maemo?
Where is flash 10.1 on maemo?
...

The few programmers of Maemo will not be able to follow the speed of google for a long time.

Yes! Maemo applications are much more stable and performant than Android ones.
But for the moment the speed of application development is much higher on google side.

We NEED maemo users to have more applications and feedback.
Users want to have access to all modern applications.

Dalvik is a MUST.

Re: Suggestion: port Dalvik JIT to Maemo
 

The problem is not to know what platform is better for mass public/business/geek... But how many times a company or a single developer will have to fully re-write their applications instead of just "porting".

GNU platform is a great success because it runs on almost all hardware/operating system from the same POSIX C source code. Small directives and compilation chain switches do the job !

Mobile market is so volatile that it is still war-time to define the "standards".

At the moment, to target as many users as possible, an application must be written in JavaME for old/new Symbian versions, in ObjectiveC/Cocoa for iOS iPhone, in Java/Dalvik for Android (or POSIX C for Linux), and Qt/C++ for new MeeGo/Symbian. [what about Samsung Bada ?]

Well, far enough, it is too much work (first dev, tests, fixes and maintenance to keep up-to-date !) and requires so many different competences, some targets are simply excluded. Hope comes from cross-compiling or porting required virtual machine.

For me, cross-compiling requires too much work, potentially on each application (at least if sources are available). Porting the VM is done once for all, applications run as soon as the used APIs are available.

I have started the job but I lack knowledge about Android build system and its specific sub-systems.
http://ymartin59.free.fr/wordpress/i...et-n900-maemo/

For those who want to help, you're welcome: first get android or android-x86 sources from Git and follow my track to get it compiled for your Linux or ChromeOS.

Re: Suggestion: port Dalvik JIT to Maemo
 

open or not but what I meant it is a huge job making it work on other platforms than android.

My point is you will neverl ever get full access to all andoid apps or get them working on n900 anyway. So why give endusers false hopes??? Android is using Linuxkernel yes but the rest of the os is far different than maemo it doesnt even use libc and the graphical system is not x11. how do you rewrite this in a good way? I am far from expert in this area but I dont think it just do it. also dalvik probadly is more memory and cpu hungry and as we all know N900 is outdated in today android hw specs.

So instead of dalvik port hint the developers what apps they want from android ported to maemo instead of dreaming of a working dalvik on n900.

I can as I said before understand if endusers want some cool android apps (or even iphone) but as I said better let developer know WHAT to port t instead of bashing about android has xxxx more apps. I mean is that really intresting to now? do people really install all those xxxxx apps?

I dont say the rest of the community/developers agree some will maybe try port dalvik and.

and what did you mean about "Calling people names does not make you a partner for discussion."

what name? cause i said Android fanboys? I heard people call me and other nokia fanboys or payed by Nokia could I care less. I just give my opionion. and fact is maemo forum seems to have alotlpeople here now basing about how bad maemo is and how good android is. so to mee there is alot of android fanboys/girls to ofcourse ;)

Re: Suggestion: port Dalvik JIT to Maemo
 

You are thinking exacly the same way as symbian engineers inside nokia atm ;) "why change we now symbian is best"

my point is changes happens, world changes... maybe it is Kina who comes with a totally new os then we use that instead.
Maybe not in two/three years ofcourse but still.

Re: Suggestion: port Dalvik JIT to Maemo
 

Do I care less if people leaves maemo in favor of Android? better this happens than reading a forum where 75 percent is bashing about how bad maemo is how bad Nokia is and so on. But to the point, I really dont think dalvik would help maemo in any way.

And about the developers it is far more important that Nokia fokusing on QT now both on symbian/maemo/meego. thinking dalvik would help the developers is not what I think, maybe if they prefer going Android but not if they prefer working with QT in that case it would only hurt...


...

About streetview so you think Google would alow people to install theyr app on a branded dalvikvm? I dont think soo.

Its not only a problem getting dalvik to n900 it also is some legal rights etc...

But when was flash related to dalvik? do yout really think porting dalvik would ge flash 10. to n900 :O

I am sure you can give better app examples than this ;)
so you like maemo but want dalvik well arent you afraid dalvik would kill maemo at the end? I dont think it is possible to port and make it fully work... But lets say it is possibe why would any developer then wanna write core maemo apps then? Why not go Android all the way then?

Dalvik would just KILL maemo/meego and thats bad imho.

But yes Maemo/meego needs more professional apps! But thats a totally different story imho. atm. Nokia need to get some stuff done to make it happen. They have to get the damn QT4.7(version already in n900) upgrade done on theyr new phones like n8, c7, e7 that way we probadly also will se some of those apps on n900 too.

because I am sure many developers is pending to release some apps until damn ovistore supports QT 4.7 and the phones are synced with latest release!


HELL I am to much in this forum I should start write some cool QT apps for Maemo/Meego instead :D

Re: Suggestion: port Dalvik JIT to Maemo
 

I'm more confident in maemo than you.

If there is a good maemo application nobody will use the Dalvik one.

Dalvik is only for the applications that are missing.
This is a supplementary option and will help maemo and not kill it.
Maemo is performing much better. Dalvik is no real concurrent. It's only the door for applications that have not be ported to maemo yet.