26-Apr-2024 07:16 GMT.		 UNDER CONSTRUCTION
Anonymous, there are 35 items in your selection
[News] JAmiga Amiga Java Project Status UpdateANN.lu
Posted on 08-May-2004 21:59 GMT by Raffaele35 comments
View flat
View list
There is a new report on the current developing situation at the site of Jamiga Amiga Java. The project was all ported to GCC. A forum about jamiga opened (just click here), and the first donation of 100 Euro arrived.
JAmiga Amiga Java Project Status Update : Comment 1 of 35ANN.lu
Posted by Hagge on 09-May-2004 08:48 GMT
Seriously I don't understand how anyone can actually belive this will result in java for the amiga in some way.
JAmiga Amiga Java Project Status Update : Comment 2 of 35ANN.lu
Posted by Andre Dörffler on 09-May-2004 09:00 GMT
In reply to Comment 1 (Hagge):
Because it will!
JAmiga Amiga Java Project Status Update : Comment 3 of 35ANN.lu
Posted by Kjetil on 09-May-2004 09:52 GMT
In reply to Comment 1 (Hagge):
Every thing takes time,
JAmiga Amiga Java Project Status Update : Comment 4 of 35ANN.lu
Posted by Marcus Sundman on 09-May-2004 10:59 GMT
In reply to Comment 1 (Hagge):
> Seriously I don't understand how anyone can actually belive this will result
> in java for the amiga in some way.

If they would use libjit and GNU Classpath then there is a chance. Otherwise it will never be more than a hobby research project for the few people involved, and certainly not usable for normal users.
JAmiga Amiga Java Project Status Update : Comment 5 of 35ANN.lu
Posted by Kjetil on 09-May-2004 11:09 GMT
In reply to Comment 4 (Marcus Sundman):
Did you check the web page, Java jit compiler works for 040 and 060 cpu's
JAmiga Amiga Java Project Status Update : Comment 6 of 35ANN.lu
Posted by Marcus Sundman on 09-May-2004 11:23 GMT
In reply to Comment 5 (Kjetil):
> Did you check the web page

Yes, I read the forums and the FAQs, but I didn't read the announcements until now.

> Java jit compiler works for 040 and 060 cpu's

To have it working is one thing, but to have it working well is another thing altogether. A good JIT compiler (or any other compiler, for that matter) requires several orders of magnitude more research and development then a simple one. If you use a library that many other projects also use then many more people will be updating/fixing/enhancing the system than if you make your own.
JAmiga Amiga Java Project Status Update : Comment 7 of 35ANN.lu
Posted by Marcus Sundman on 09-May-2004 11:27 GMT
In reply to Comment 2 (Andre Dörffler):
> Because it will!

This has to be one of the most stupid things anyone has ever said. How on earth could anyone be so stupid as to believe something can prove itself? "A is true because A is true?" You have to be kidding me! (And you probably are and I just don't get the joke.)
JAmiga Amiga Java Project Status Update : Comment 8 of 35ANN.lu
Posted by Raffaele on 09-May-2004 12:31 GMT
In reply to Comment 6 (Marcus Sundman):
Mr. Marcus Sundman wrote:

> Java jit compiler works for 040 and 060 cpu's

>To have it working is one thing, but to have
>it working well is another thing altogether.
>A good JIT compiler (or any other compiler,
>for that matter)requires several orders of
>magnitude more research and development
>then a simple one.

Well... I have java as a "fix nail in my mind" from the time that a former amigan who passed to PC's cheered at us amigans for we don't have Java on Amiga...

I started writing my first post on ANN to ask about some development of java for Amiga, if them exists, if them are currently developed...

And during my wanderings I found a lot of thingies, strange news, JIT programs realized and then aborted...

Some of these projects are made from decent software-houses, and even if I think that open-source groups of programmers works fine, I think also as you Sundman, that research and development by a software-house which works in the market are a sure "boost".

Obviously not better than open-source project (they could be of the same quality). But sure these projects are boosted by money.
JAmiga Amiga Java Project Status Update : Comment 9 of 35ANN.lu
Posted by Richi on 09-May-2004 12:33 GMT
I'm optimistic!
Well done guys!
JAmiga Amiga Java Project Status Update : Comment 10 of 35ANN.lu
Posted by Jupp3 on 09-May-2004 12:48 GMT
In reply to Comment 7 (Marcus Sundman):
>How on earth could anyone be so stupid as to believe something can prove itself? "A is true because A is true?"

That reminds me of error message my Macintosh Powerbook once displayed.

It went like this (translated from Finnish): "An unknown error has occured becouse an unknown error has occured."

Nice to know why it crashed though
JAmiga Amiga Java Project Status Update : Comment 11 of 35ANN.lu
Posted by Hagge on 09-May-2004 12:51 GMT
Well, I just know about those old like hello world examples, to get text on the screen is one thing, to fix swing, network and database support and so on another. The real problems however is probably support for threads and stuff like that. I don't see how you can fix such a huge project yourself. For a company like sun or ibm? sure, for 5 hard working amigans on an old outdated os? no.
JAmiga Amiga Java Project Status Update : Comment 12 of 35ANN.lu
Posted by Don Cox on 09-May-2004 14:14 GMT
In reply to Comment 11 (Hagge):
"For a company like sun or ibm? sure, for 5 hard working amigans on an old outdated os? no."

A team of five Amiga programmers can accomplish a great deal. And this is a port - they are not coding the whole language from scratch. Many people have written whole languages for Amiga.

I am quite optimistic about this project. It is certainly something the Amiga needs, so it is worth the effort.
JAmiga Amiga Java Project Status Update : Comment 13 of 35ANN.lu
Posted by Raffaele on 09-May-2004 14:33 GMT
In reply to Comment 12 (Don Cox):
Mr. Don Cox wrote:

>A team of five Amiga programmers
>can accomplish a great deal.
>And this is a port
>- they are not coding the whole language from scratch.
>Many people have written whole languages for Amiga.

And they could also benefit from the efforts, the code (and from some remaining discussions of the developing in certain forum-sites) of the people who preceded them, such as GeekGadgets Java-Koffee code...
JAmiga Amiga Java Project Status Update : Comment 14 of 35ANN.lu
Posted by Raffaele on 09-May-2004 14:37 GMT
Oh... I forget to mention...

Are new Amiga PPC based machines capable to run at an acceptable speed existing Koffee for Amiga (68040 and '060 based), even thru ixemul compatibility...

I know that at those times on a simple '040 the speed of some geekgadgets were unusable...

But up now it could be just right another thing (and a good one)...
JAmiga Amiga Java Project Status Update : Comment 15 of 35ANN.lu
Posted by Amon_Re on 09-May-2004 15:49 GMT
In reply to Comment 1 (Hagge):
Why the negativity? Really, why be so negative about a project like this that actually has some momentum? Critisising is one thing, but downright saying they'll never acheve anything is another.

Cheers
JAmiga Amiga Java Project Status Update : Comment 16 of 35ANN.lu
Posted by Marcus Sundman on 09-May-2004 15:59 GMT
In reply to Comment 8 (Raffaele):
> I have java as a "fix nail in my mind" from the time that a former amigan
> who passed to PC's cheered at us amigans for we don't have Java on Amiga.

I would very much see a modern JVM on the Amiga, partly because java is my language of choice for most projects. I believe it would be best to join forces with other projects, and that means going with libjit and GNU Classpath. Unfortunately libjit is still at a very early stage, but since many other projects seem to be switching to it it seems to be the best choice. I could be wrong, but I don't think so, unless libjit is somehow suddenly abandoned. A profiling and optimizing JIT compiler is very hard to implement.
JAmiga Amiga Java Project Status Update : Comment 17 of 35ANN.lu
Posted by Marcus Sundman on 09-May-2004 16:10 GMT
In reply to Comment 16 (Marcus Sundman):
> I would very much see a modern JVM on the Amiga

Err.. "I would very much WATN TO see a modern JVM on the Amiga"
JAmiga Amiga Java Project Status Update : Comment 18 of 35ANN.lu
Posted by Marcus Sundman on 09-May-2004 16:11 GMT
In reply to Comment 17 (Marcus Sundman):
> > I would very much see a modern JVM on the Amiga
>
> Err.. "I would very much WATN TO see a modern JVM on the Amiga"

Argh! "I would very much WANT to see a modern JVM on the Amiga"
JAmiga Amiga Java Project Status Update : Comment 19 of 35ANN.lu
Posted by Anonymous on 09-May-2004 18:14 GMT
In reply to Comment 18 (Marcus Sundman):
We get the message :-)

So would I!
JAmiga Amiga Java Project Status Update : Comment 20 of 35ANN.lu
Posted by damn on 09-May-2004 18:20 GMT
what a waste of time. make a jvm for morphos and not the obsolete 68k amiga.
JAmiga Amiga Java Project Status Update : Comment 21 of 35ANN.lu
Posted by jools on 09-May-2004 22:35 GMT
In reply to Comment 20 (damn):
How can you be so rude ! (And ignorant)

The source is there available. YOU make a version for morphos.

Well done to the developers. keep up the great work. I certainly find this project very interesting. The fact is there is no decent java implementation for the amiga. So this could be a useful tool.

As for previous comments, it seems people are jumping the gun and not actually reading the progress. threads are implemented as is networking code. they ARE using the GNU classpath. Did you download the code and see already the amount of work put into this project!? the JIT is functional so why should they switch to another JIT implementation?

Especially something like libjit which doesnt even support m68k (Nor PPC!) in its current state.
JAmiga Amiga Java Project Status Update : Comment 22 of 35ANN.lu
Posted by jools on 09-May-2004 22:55 GMT
In reply to Comment 21 (jools):
cd to jamiga/src

and

$ find . -name "*.cpp" -exec cat {} \; | wc -l
33296 (lines)

hmm i`d say thats a fair amount of code (And not the latest version either)

the header files come to about 3500 lines too!
JAmiga Amiga Java Project Status Update : Comment 23 of 35ANN.lu
Posted by Andre Dörffler on 10-May-2004 03:04 GMT
In reply to Comment 7 (Marcus Sundman):
Well, here is the long version: Because it will result in Java for the AmigaOS in some way.
Actually it already IS Java for the Amiga in "some way".
Well I am on the development of this project and I was worried about the pointless (and very negative) first post. So I just wrote an pointless answer.
Besides: Since we Amiga programmers cannot develop full time on our amiga projects, we do have jobs. And incidently all of the JAmiga development team members are professional software developers.
This is a hobby project though, and we do it for ourselves, just to proove, that we can do it (And we have n doubt that we can make it, since we are professionals ;-) ).
So as we were approaching a working version, we thought, it might be a good thing (tm) to share our efforts with this community.
I never thought, that such a project would create this kind and amount of negativity and discourageing feedback. It is a shame, but I guess, these ar just a few people, that do not represent the community. At least all people of the community, that I know personally are very positive ybout the JAmiga project.
On another side note I would like to mention, that the main author of JAmiga has never been an Amiga programmer, so I think it is very impressive piece of work.I some of the people that do talk about projects never beeing finished could do the same.....
JAmiga Amiga Java Project Status Update : Comment 24 of 35ANN.lu
Posted by Anonymous on 10-May-2004 04:25 GMT
In reply to Comment 23 (Andre Dörffler):
I happen to like JAmiga project :)
When will we see a bit of AWT, though?
JAmiga Amiga Java Project Status Update : Comment 25 of 35ANN.lu
Posted by Don Cox on 10-May-2004 05:34 GMT
In reply to Comment 23 (Andre Dörffler):
"Besides: Since we Amiga programmers cannot develop full time on our amiga projects, we do have jobs. And incidently all of the JAmiga development team members are professional software developers."

There is a myth around that people who program for Amiga are amateur or beginner coders learning AMOS. This is not true - most of them seem to be professional programmers doing something creative in their own time.


"This is a hobby project though, and we do it for ourselves, just to proove, that we can do it (And we have n doubt that we can make it, since we are professionals ;-) ).
So as we were approaching a working version, we thought, it might be a good thing (tm) to share our efforts with this community."

Sensible people are grateful.


"I never thought, that such a project would create this kind and amount of negativity and discourageing feedback. It is a shame, but I guess, these are just a few people, that do not represent the community."

I think these are people who have left the Amiga for other platforms and want to prove they were right. There is a continuous stream of remarks about "old OS", "outdated programs", "can't compete with real computers", etc etc. They also think that because Microsoft use hundreds of programmers it is not possible to accomplish anything with a small team, and that the only possible way to design an operating system is to imitate Unix.

(That's not entirely their fault - the college textbooks on operating systems give the same impression.)
JAmiga Amiga Java Project Status Update : Comment 26 of 35ANN.lu
Posted by Jon on 10-May-2004 07:55 GMT
Another good news! Thanks to the team.
JAmiga Amiga Java Project Status Update : Comment 27 of 35ANN.lu
Posted by Fabio Alemagna on 10-May-2004 08:29 GMT
In reply to Comment 16 (Marcus Sundman):
> A profiling and optimizing JIT compiler is very hard to implement.

Bernd Meyer seems to have got it right all by himself, just like many others before him, without need for a "libjit" effort. The fact such an effort exists, doesn't mean that no one can do it by himself, if he has the abilities. And as for classpath: of course JAmiga is using it, no one in their right mind would think of reimplementing by himself the whole java foundation class.
JAmiga Amiga Java Project Status Update : Comment 28 of 35ANN.lu
Posted by Marcus Sundman on 10-May-2004 10:02 GMT
In reply to Comment 23 (Andre Dörffler):
> I never thought, that such a project would create this kind and amount of
> negativity and discourageing feedback.

I hope any feedback I give would not be considered as discouraging. I apologize for my harsh words earlier.
Some of the negativity you experience might be because people don't want to get their hopes up and then be disappointed again.
JAmiga Amiga Java Project Status Update : Comment 29 of 35ANN.lu
Posted by Merko on 10-May-2004 10:18 GMT
Heh. I don't believe this will actually result in anything very useful
either, but.. for one thing I don't believe those people who says you
need a huge team and enormous resources to do something.

Just look at Rebol for example. It's not directly comparable to Java,
but does a lot of things far better IMO, and many of them things which
are supposedly so much work to implement in Java (I'm talking about UI
stuff for example). And it was mostly done by one guy.

Anyway.. in the worst case, a few people will get some valuable
experience.
JAmiga Amiga Java Project Status Update : Comment 30 of 35ANN.lu
Posted by Marcus Sundman on 10-May-2004 10:36 GMT
In reply to Comment 27 (Fabio Alemagna):
> > A profiling and optimizing JIT compiler is very hard to implement.
>
> Bernd Meyer seems to have got it right all by himself, just like many others
> before him, without need for a "libjit" effort. The fact such an effort
> exists, doesn't mean that no one can do it by himself, if he has the
> abilities.

I'm not very familiar with Mr Meyer's work - I've neither tested his JIT compiler(s) nor seen any code he's written. Nor have I made a JIT compiler myself. However, from what I've read about Sun's and IBM's efforts to optimize their JVMs it seems like it's very hard to know what can be inlined where and what have you. There have been many inefficient java bytecode JIT compilers over the years, but only a very few efficient ones, and afaik all those have been the result of lots and lots of research. Of course you will be able to piggyback on research already done in the field, but you'll still need all the help you can get if you want an implementation that won't get outdated immediately.

I know from experience how nice it is to use OSS libraries that are also used by other people. You will automatically get bugs fixed and features and enhancements made, and all you have to do is to share your own fixes and enhancements. Developers of hobby projects usually get bored when the hard stuff is done (I know I do), and then all that boooring bugfixing needs to be done. And then when the product is considered "good enough" it is usually more or less abandoned in favor of some new interesting project, except for the occasional bugfixes. It is at this point that it pays off most to be using libraries that many other also use. Those other people will be at some other stage in their developments and thus there will be enhancements made to the library and you can just sit and watch your own program get better without having to do anything.

Think of programs using datatypes for loading images. What if the developers would just say that they don't want to use datatypes because there is only need to load JPEGs and they have already created a JPEG decoder that is faster than all the existing JPEG datatypes? Sure, having their own decoder might be better now, but when they have lost their interest in the project and a new version of JPEG (or a better JPEG datatype) is released, then they might be very sorry they didn't build support for datatypes.

> as for classpath: of course JAmiga is using it

Good.
JAmiga Amiga Java Project Status Update : Comment 31 of 35ANN.lu
Posted by Anonymous on 10-May-2004 11:56 GMT
In reply to Comment 1 (Hagge):
Tell me atleast one reason why it could not happen ?

Lot's of things were said never to hapen and still we now have lot's of that stuff, both software and hardware. It's about people who actually do something, instead of just talking bullshit in the forums.
JAmiga Amiga Java Project Status Update : Comment 32 of 35ANN.lu
Posted by Anonymous on 10-May-2004 12:04 GMT
In reply to Comment 23 (Andre Dörffler):
Don't care about those negative comments.

There is lot's of us who really do appreciate your work. I still hope you continue your JAmiga project, because no matter what those guys say your project looks quite promising.

Amiga really needs guys like you, who really do something instead of talking here.
JAmiga Amiga Java Project Status Update : Comment 33 of 35ANN.lu
Posted by Anonymous on 11-May-2004 08:21 GMT
In reply to Comment 32 (Anonymous):
I am a software developer that works a lot in Java, and I for one am very happy to see this project. Not that I'm an active amiga user nowdays, but I still like the amiga a lot. A working jvm adds very much to a platform. Even if it never gets as completed as suns jvm, it will most likely open a huge door to using the tons of great java applications out there on the amiga.

I really don't understand the negativity. These guys are working hard to give you something great, and all for free. What's the fuzz about?

Anyway, I wish the best of luck to you guys, and I hope to port some of my own java applications to the amiga in the future.
JAmiga Amiga Java Project Status Update : Comment 34 of 35ANN.lu
Posted by Anonymous on 11-May-2004 12:18 GMT
Thank you for your effort guys! and forget the few negative posts... some people complains all the day about evrything... damn moaners! :-D
JAmiga Amiga Java Project Status Update : Comment 35 of 35ANN.lu
Posted by someone on 11-May-2004 23:30 GMT
Whats your problem?
they are PORTING not developing ...and even if it will be slower than suns or ibms ...better a slow port than no port ;-)
Anonymous, there are 35 items in your selection
Back to Top