Java 2 for Linux Released & Blackdown Gets Creds 211
burner writes, "After quite a wait, JDK1.2.2 is released for Linux. You can grab the final release from Sun's site. Sun has also put up bios for the Blackdown guys. Sun's been acting pretty flippy lately, but this is great news. I've been using Blackdown's latest release candidates lately, and they're excellent, but now there's a final release. Nice work guys! "
"Polish" and credit (Score:2)
Please, Do Not Code Games In Java (Score:4)
Great work! (Score:2)
Now onto java 1.3, 3 or whatever it will be called
Re:"Polish" and credit (Score:3)
Re:Please, Do Not Code Games In Java (Score:1)
Ha ha! Nothing funnier than hundreds of dead and wounded innocent people in a remote part of the world! Look at those poor people! What a great joke!!!
Re:Great work! (Score:2)
Isn't Visual Basic at about 6? Based on industry trends, Java 1.3 should be released as Java 7.
- - - -
Cross-platform... (Score:3)
Re:linux was mentioned... (Score:2)
Java development is growing daily. The speed with which Java applications and servlets can be developed is unprecedented. The speed issues with Java are decreasing with each release, and 1.2.2 is pretty quick. The 1.3 early-release 1.3 JRE is quite a bit faster, and Blackdown is already working on the port.
This is very good news for Linux, which, as you may have noticed, a few
Time for trolls to show up (Score:1)
Let's see how many we get on this story:
Offtopic: If you don't have an X-server, JavaLobby posted a story about a very nice GPL'ed Xserver/Esound/Truetype server written in Java at called WeirdX [jcraft.com] Even runs as an applet. Very nice if all you have is a Windows/Mac box, or are at a public terminal/cybercafe and need to remote-X from your Linux box. :) Mostly impressive because a single guy wrote an X server from scratch in a short period of time.
How long were you waiting to post this up? (Score:1)
:)
Re:Great work! (Score:1)
JIT and Java 1.3 (Score:5)
There was a JavaLive chat [sun.com] yesterday about the Java on Linux stuff. They haven't put up the transcripts yet though.
For Java 1.3 from Sun, the Windows version will come out first, then Solaris then Linux. However, they do want to syncronise all releases together and should do this at or before Java 1.4 - might happen first for a maintenance release.
Blackdown and Sun/Inprise JDK different (Score:1)
Ok, so... (Score:2)
That's not to insinuate that there's anything bad about Sun's version of Java. Other than it has a history of being the slowest. (Jikes leaves Sun's v1.1 compilers trailing in the dust.)
Also, now that Blackdown have it ported to Linux, will it be ported to different Linux processors? Or just ix86?
I can't see why it should be anything beyond a simple recompile, to get binaries for all the Linux platforms, and compilation speed isn't an issue, as you're not looking to debug the logic. Emulators, such as the ARM emulator, and cross-compilers, should be fine in producing Java 2 for Linux for every platform it runs on.
Re:Great work! (Score:1)
Java 1.2 vs Java 1.8.8 (Score:2)
I have been using IBM's Java 1.1.8 for basically powering my backend web applications on linux and frankly it has been working so well I have not seen a need to move to Java 2.
I don't use any of the Java EE beans or really anything major complicated, but my team has built some fairly complicated web sites that use multi-tier architecture with a great deal of success.
From the benchmarks I have seen I wouldn't argue that 1.2 is not really any faster than IBM's JDK? Actually, the benchmarks I have seen argue strongly the opposite.
From what I understand IBM will have Java 1.3 ported to Linux Q2 this year. Is Java 2 really worth it on the backend on linux?
Does anyone know what HotSpot is and what advantage that is going to bring us on using Java where it belongs, on the Server.
Great, BUT... (Score:4)
I've done a lot of Java development under Linux this year, and I've noticed several things that prevent me from doing serious work with Java (under Linux). For example, rmiregistry crashes without fail for any type of heavily loaded RMI project. Another example is that Thread objects break just as easily, or refuse to start altogether when you spawn multiple Threads (even if there is plenty of memory available). Luckily I have access to a cluster of Ultra5's to test my applications on, which execute almost flawlessly (I've noticed a few quirks with Threads under Solaris as well, but not nearly as bad as under Linux).
I see the state of the JVM under Linux as being close to a toy. I know all the Blackdown people will probably find this insulting, but unfortunately, I can't do hard-core Java the way I can under Solaris. I do not blame Blackdown in any way for this however. I feel that this is solely Sun's fault.
Thats my $0.02 on the issue. I love Blackdown, I love Java. Bugs suck.
Long overdue... (Score:1)
If they would've known the amount of time I had to struggle with getting GCC loaded and config'd for the HP, just so I could then compile Apache, just so after that I could compile JServ, just so I could *then* use an "Officially Sanctioned" JDK....
As I said, long overdue...
Re:Cross-platform... (Score:3)
About the "faster" part, it will probably never be suitable for OS development, but you CAN get platform-specific compilers for it.
Regarding the "pervasive" & "outside the browser" parts, I think you're just not looking.
Admittedly the start was slow, due to the fact that Java's original target platforms were toasters and TV's. But there are no shortage of Java applications you can't see until you look for them.
- - - -
Sun JDK Experiences (Score:1)
Jini and Java Spaces? (Score:1)
Re:Please, Do Not Code Games In Java (Score:1)
Re:Great work! (Score:1)
> Java 1.2 are synonymous...)
JDk 1.3 is also a Java 2 platform release.
Re:Great work! (Score:1)
Re:Great, BUT... (Score:1)
Sun's JDK is not the Blackdown JDK (Score:1)
My experience has been that Blackdown's port was faster and more stable than Sun's. Now that Sun's port has been officially blessed, I imagine it will have the stability edge for a little while, but I will continue to use the Blackdown port for its superior performance.
Re:Great, BUT... (Score:3)
I think the implication of your post is much more interesting than just 'java doesn't run well on linux.'
What this (teir-1 support for solaris, goat-blowing support for linux) mostly accomplishes is continued degradation of a core-value of java: write once, run anywhere. Not only does it not work, it hasn't ever worked, and even if it does work, it doesn't work well enough to be useable in the Real World.
So, as usual, instead of disparaging linux and the blackdown crew, I think it's important to keep the blame right where it belongs: on Sun. If they REALLY wanted Java to be a ubiquitous standard, they'd release it all and let the people who do the work have at it.
All in all, NOTHING works perfectly across all platforms, not
--
blue
Re:Great, BUT... (Score:2)
Re:Long overdue... (Score:1)
I hate to have to break this to you, seeing the amount of work you've already had to do, but JServ isn't it any more. The new reference implementation for servlets is Tomcat from the Jakarta Project [apache.org].
My condolences...
- - - -
Yes it's great news BUT... (Score:2)
I don't want to sound ungrateful, but... when are you going to drop the other shoe, Scott???
Great JVM (Score:3)
JIT and Threads (Score:1)
Re:Great, BUT... (Score:1)
I must have an old version of NOTHING. Mine only works on ETHERnet.
- - - -
Re:Great work! (Score:1)
Carifications: (Score:3)
An important distinction between the two is that the "official Sun JDK" does NOT support native threads and in fact recommends NOT running it on SMP machines, while the Blackdown release does native threads and SMP just fine.
-=-=-=-=-
Re:Great, BUT... (Score:2)
Re:Cross-platform... (Score:2)
there are lots of embedded Java things coming out now that run embedded Java. remember that the original demo for the language (if I remember correctly) back when it was called Oak, was a bulky TV remote control.
Lea
Warning: the Polybot link will take you to a page that my Mozilla M13 will render and then turn blank grey. I have no idea why, but you might want to try Netscape instead. (or IE, if you're that sort of person)
JVM performance bad on MOST platforms (Score:1)
The problem: It doesn't run the same on any 2 platforms that I've tried. It doesn't run the same twice on any single platform. You never know which thread will run first. Sometimes the sound comes before the pictures appear (which is not the order it's programmed). Often, it doesn't load the images when I tell it to, and sometimes it doesn't load one of the images at all. The only way I could keep it from playing 2 sound clips at the same time was to specify a sleep the duration of the 1st sound clip. This happens with 4 different JVMs and operating systems (no Sun platforms) that I've tried. And, of course, it often crashes.
I have given up on Java. The point was to make a cross-platform program that non-Linux users could use, but if it only runs well on Sun products, forget it.
Re:Best way to put JDK 1.2 into a Debian box? (Score:1)
I have three different jdk's installed under
Re:Ok, so... (Score:1)
A history of being slowest, yes, but Sun's HotSpot dynamically optimizing VM is very nice, competitive or superior to anything else out there, at least for Java 1.2 and greater.
But Seriously... (Score:1)
"The axiom 'An honest man has nothing to fear from the police'
How Sun Lost Us As a Java Customer (Not a troll) (Score:3)
Performance may now be acceptable, but at the time we dumped the product even a small, simple data entry application was too demanding of the JVM at the time (even on Sparc 10's running Solaris, much less Linux). The choice Sun gave us was stark: run the Java VM under Windows or Solaris on a high end sparc, or suffer. We chose Linux, adopted a more open development environment, and now having dumped the product we will not, in the future, ever consider going back (c.f "sun sucks" and "slow"). Using GNU configure and its associated utilities, we are able to get all the cross-platform support we require, even if it involves a quick rebuild of the sources (typing "./configure" and "make install" isn't terribly difficult) with the performance our users demand and languages we can hire developers for (c.f. "use Perl" and "Java sucks").
I enjoyed using Java (despite the, even now, still horrificly screwed up date and time classes) as a language, but the drawbacks were too severe and too critical for too long of a time, and Sun's current and future motives with respect to the openness of the standard and support for Linux, FreeBSD, and whatever other platform we may, in the future, chose to deploy, has eroded our confidence in the product too much for us to seriously consider any future use of Java. Put simply, the stumbling blocks Sun until recently put in the way of development on anything other than their "blessed" platforms far outweighed any advantage the language itself offered (and those were not inconsiderable for those of us coming from C++, with Java's simpler memory management and garbage collection and other features).
Alas, the promise of "write once, run everywhere" quickly became (and IMHO remains) "write once, run where Sun would like you to." At present Sun has chosen to become mildly friendly towards Linux. This is great! However, I would not expect this to remain a long term strategy on their part, unless there are some serious changes in the mentaility of Sun's upper management. (c.f. "blah blah blah").
Re:Time for trolls to show up (Score:2)
Java never seems to work right for me.
At work we are using a java program (running through X in this case), and it crashes about 50% of the time when you try to start it. This is running on HP/UX.
In other words, I'm not impressed with Java so far. But that weirdx LOOKS really awesome. What am I doing wrong?
Re:Great work! (Score:1)
And now, submitted for your disapproval, the dark path I have trodden since 1993: The History of Visual Basic (short form).
- May 1991: VB 1.0. Drag-n-drop window layout, VBX visual component architecture, p-code-based runtime. Simple by today's standards, but remember: It takes MS at least three versions to get it right.
:-) - Nov 1992: VB 2.0. ODBC, Multiple Document Interface, and object variables.
- June 1993: VB 3.0. Access DB engine, OLE Automation, Crystal Reports. Reaches Critical Mass, even though it's not really ready.
- Oct 1995: VB 4.0. Parallel 16- and 32-bit development, rudimentary classes, ActiveX Controls (then called OCXs) replace VBXs, extensible IDE. Almost ready for prime time.
- April 1997: VB 5.0. Goodbye, 16-bit. Native x86 and Alpha compilers, events in classes, API callback support, COM Component and ActiveX Control targets, vastly improved IDE (IMHO).
...and there was much rejoicing. - Oct 1998: VB 6.0. Dynamic form building and lots of web candy. Little more than a service pack to VB5, if you ask me.
- Q1 2001: VB 7.0. Major redesign. Real objects with inheritance, polymorphism, constructors, &c. Structured exception handling (Try...Catch...Finally instead of On Error Goto Hell). Free threading. Essentially, everything we've been begging and screaming for since VB4!
Note: VB7 info is based on Steve Ballmer's keynote at VBITS San Francisco yesterday. So if any real improvements are dropped before RTM, we know where the lynch mob will form.Can I moderate myself -1: Offtopic?
Keith Russell
OS != Religion
Thanks, but... (Score:2)
JMC
Re:Great, BUT... (Score:3)
It doesn't surprise me that Java on Solaris would be more stable than Java on Linux. I run my (large!) RMI server app on Solaris under Java 1.2 and under Linux under 1.1.7/8, and it works fine in both places.
Perhaps you are doing very fine-grained RMI object exporting? One thing I learned fairly early on is that it is important not to have all of your RMI objects inherit from UnicastRemoteObject.. if you do that, then your objects are automatically registered for export on creation. Much better to use the static UnicastRemoteObject.exportObject() method to export your Remote-implementing object only when needed.
Actually, now that I think of it, I only actually register a single object in the RMI registry.. my top-level server object. All further RMI activity is done using returned references to RMI-exported objects.
What sort of architecture are you using in your program? Which versions of the JDK have you been seeing problems on Linux with? What version of glibc do you have installed?
java for mozilla ? !!! (Score:3)
Multithreaded Programming (Score:1)
You may want to brush up on multi-threaded programming- what you describe is exactly the way it is supposed to work (i.e. the order of thread execution is not guaranteed).
It sounds like you've seriously overdeveloped your app- if you're doing things sequentially, why put the actions in seperate threads? If you actually need threads for something, can the UI functions you describe be in one thread? Perhaps you should sychronize things using events.
YMMV, but I daily do development between 98, NT, Linux, and Solaris (on an UltraSparc) with Java and my code works the same on all platforms. Of course, I'm only doing Server Side applications and am not using AWT or Swing.
I'm not saying Java is all the hype made it out to be. In fact, all of the hype probably did more damage to Java over the long term than helped in the short term, but overall as a language it's not too shabby. As always, you should use the right tool for the job- maybe it's Java, maybe it's C, maybe it's Perl, maybe it's VB or Delphi. Depends on what needs to be done.
For the record, I've never been able to get a client side application using AWT or Swing to work the same between implementations.
It could be worse ... (Score:1)
Finally.... (Score:1)
Free Java Software (Score:2)
Now I can move it forward to Java2!
Re:Great, BUT... (Score:1)
Optimization cannot be irrelevant. Yes, it would be nice if the Linux OS and JVM scaled in all the same ways as a certain JVM on Solaris or NT does, but that's not realistic. The Java Language Specification makes no guarantees about threading semantics, let alone scaling and performance issues.
Write Once, Run Everywhere works surprisingly well, all things considered, and it's getting better. It's not a panacea, though. You do have to test your software across several JVM's and in various memory, threading, and loading conditions.
All of which you know already, of course. Just don't curse Sun (too much)for not attaining perfection when they've got a better portability story than anyone else does today.
Re:Cross-platform... (Score:1)
I wasn't very explict I guess. What I'm really lamenting is a pervasive, cross-platform, in-sync environment to RUN java apps in. I do understand where it has been really successful as a language/platform. As another example, it will be exciting to see if Mac OS X really succeeds in fully integrating it as a first-class development language in a popular OS.
However, many spiffy add-on APIs (not only from Sun) are only really supported on those platforms that Sun supports. Admittedly some are inherently difficult to support well cross-platform (Java3D for example), but without even core Java in sync across platforms even simpler packages quickly fall away from non-Sun platforms.
As a mostly high-level language programmer, I really enjoy programming in Java. It also had (has?) the potential to close the cross-platform app gap. Aside from some gee-wiz items and small utilities, that vision (delusion?) seems to be mostly dead for now. It certainly has been a success, but just not where I was looking
mh
Re:Cross-platform... (Score:2)
Non native java code, compiling to native code does not offer much performance advantage (contrary to what people seem to believe) since it does not eliminate the real bottlenecks (garbage collection, memory allocation, synchornization, etc.)
But.. (Score:1)
Tomcat isn't ready for primetime yet. It's lacking many features that JServ already has (and has had for a while now). Just an example, but it still doesn't support servlet reloading. JServ does.
I'm running both side by side, and while I would like to use some of what Tomcat has (ex. support for Servlet 2.2), right now I can only deploy to JServ. Maybe when 3.1 is out I can switch.
HotSpot (Score:2)
Re:Free Java Software (Score:1)
Re:Now for a godawfully stupid question: Netscape? (Score:3)
Re:Free Java Software (Score:1)
Nethack!! (Score:1)
Re: Threading Semantics (Score:1)
Let me correct myself. The JLS obviously does say a lot about threading semantics, just not enough to specify whether threads are even pre-emptive under Java, let alone map to native threads or across processors on an SMP box.
If you are extraordinarily careful and use synchronization primitives with care, you can write code that will function (however slowly/poorly) even in a very limited threading environment.
Re:Time for trolls to show up (Score:2)
Re: Threading Semantics (Score:1)
just waiting for JDK1.3 (Score:1)
Re:Great work! (Score:1)
Actually, I want them to get on the ball with all the supplimental stuff -- Media Framework, Advanced Imaging, 3D, and the plug-in (and make sure the plug in and mozilla milestones work together.
Basically, there are still a number of "Java" pieces that have native code that needs to be ported. If Blackdown doesn't have the resources to do the work, then someone else needs to step forward to get it done.
Re:Great work! (Score:1)
-jarno
Re:Please, Do Not Code Games In Java (Score:1)
Its no fun til someone loses an eye!!
Seriously...? (Score:1)
Wish list (Score:1)
MODERATORS, please rate the suggestions!
Please, only one request per posting or you will mess up the rating.
Lets work together and make better software!
Re:Great work! (Score:1)
"The axiom 'An honest man has nothing to fear from the police'
Re: Threading Semantics (Score:1)
Preemption is allowed, yes, and there are guaranteed atomic operations (reference and integer assignment, at the very least). That's why I backpedalled and said that the JLS did indeed have guarantees on thread semantics.
It just doesn't have any guarantee as to when a particular thread will be run, nor whether a thread will ever be switched unless the active thread comes to the end of its execution or to a wait() call. In practice, every JVM that I care about will do those things, and I don't bother seeking out a Palm Pilot or Win 3.1 JVM to test my code on, but in theory it's something to be aware of.
Re:Nethack!! (Score:1)
work seems to be dreadfully slow, but maybe someone can lend him a hand or something...
BTW, more roguelike java projects can be found on Roguelike News' Links Page [demon.co.uk]. Enjoy!
-Ravagin
"Ladies and gentlemen, this is NPR! And that means....it's time for a drum solo!"
Why a bottleneck? (Score:1)
Re:Please, Do Not Code Games In Java (Score:2)
-Ravagin
"Ladies and gentlemen, this is NPR! And that means....it's time for a drum solo!"
Re:Multithreaded Programming (Score:3)
It is possible to force your main thread to block until images load, but he obviously doesn't know (For the previous poster, look into the MediaTracker interface).
Doug
Re:Long overdue... (Score:1)
Hell, at this point I'd settle for "already installed"... (grin)
Re:JVM performance bad on MOST platforms (Score:1)
I've been programming in Java ( using CORBA and RMI ) on Linux for ~2 years now and it's been great. It's had its problems to be sure but it's not as bad as some here are making it out to be. We've had significant loads and high numbers of objects going back and forth from our servers and Java has worked like a champ.
cheeser
GUI stuff doesn't work for me (Score:1)
public class Test {
private static Frame frame;
public static void main(String[] args) {
frame = new Frame("Test");
frame.show();
}
}
compiles fine, but segfaults every time I run it. Something isn't right here...
Re:JVM performance bad on MOST platforms (Score:1)
Re:Please, Do Not Code Games In Java (Score:1)
Re: Threading Semantics (Score:1)
Re:Now for a godawfully stupid question: Netscape? (Score:1)
ftp://iodynamics.com/pub/mirror/linux-jdk/JDK-1.2
I haven't been able to find a final one... anyone else?
The sun is going down, I say we follow it out of town- We've been here for far too long.
Re:JVM performance bad on MOST platforms (Score:1)
Re:java for mozilla ? !!! (Score:1)
How can this be classed as interesting for Godsake.. funny maybe, but interesting?
what information does it give?
bugger all thats what.
Now I can go Java (Score:1)
Thanks Blackdown and Sun!
Re:Please, Do Not Code Games In Java (Score:1)
Re:Great, BUT... (Score:1)
Check out the Macintosh support. Not Open Source at all.
If Java actually ran on multiple platforms with any performance, it would make the question of Platform old fashioned. However instead they work hard to push it on the majority platform, Windows.
D
Looking into the gift horse's mouth... (Score:2)
Really, the README says that only green threads are supported and running under SMP kernels is discouraged. Given the resources Sun has to do things right, this can mean only one thing: it's their way of saying "Linux is OK for single-user toy usage but for high-end SMP stuff just get Solaris, OK?". This is more a PR release "We do Linux, we want our stock to go up" than the REAL thing.
I hope IBM will bring out a SMP-supporting JDK2 SOON! Their 1.1.8 is wonderful and fast. Are you listening, IBM?
Tuomas
Re:Please, Do Not Code Games In Java (Score:2)
Maybe in a few years..
Re:Please, Do Not Code Games In Java (Score:2)
Just ask Chris Schoeneman.
Should I be Excited? (Score:2)
It is about time Sun got decent Java support to Linux. I know that they'd rather spend effort on working on their own OS, but if they truly want a language that is portable to all systems, they need to make sure it actually works on all systems.
Now for the negative side. My experience with Java screams that the language is in need of much work. It runs slowly. It requires you to look up the APIs constantly anytime you want to do something useful, and it just doesn't get the job done as well as C/C++. It has a place, but I'm not sure that it is really the greatest tool for Linux programmers when compared to C/C++/Perl/Whatever. Still, the idea of "crossplatform" is nice.
GNU Classpath (Score:2)
It's in a early stage right now (version 0.0), but should more people hack on it, we can say bye-bye to Sun.
Re:Time for trolls to show up (Score:2)
Re:Nethack!! (Score:2)
Re:Cross-platform... (Score:2)
Re:Cross-platform... (Score:2)
A closer examination of this benchmarks reveals that it tests socket IO and threads. The performance of both is very much dependend on the OS.
Furthermore, I think a benchmark that stresses the OO usage of Java a little more (i.e. creating destroying objects) would show a different picture.
This benchmark does not proof much to me. A benchmark that doesn't take typical OO behavior of a program into account is not realistic, even for server side programs.
Re:Yes it's great news BUT... (Score:2)
the stewardship of Java standards/apis is not open So? The specs haven't been kept up to date (check for yourself). And have always been incomplete. The specs somewhat lame in a lot of places, too. For example, hardware api's sucked from the beginning, they still suck, and from the look of things, they are going to continue to suck forever unless something drastic changes in the API development process. About the only hardware Sun has created an api for is the mouse. Show me how you support a joystick in Java, for example. It's a big world out there and Java as it stands just can't deal with a lot of it. Don't talk to me about native code interface either: hardware support is pretty much useless in Java unless it exists on every platform, and it's now so late in the game that even if the api's were fixed it would be immensely difficult to deploy the new support widely enough so that application writers could write to it with confidence.
Sun's unwillingness to let go of their baby and let it grow up into an adult. We haven't got that much time to wait. At the current rate, Java is going to be a dead issue by the time Sun wakes up. It's going to be killed by something more useful, and more open, without one, self-interested company trying to dominate it. All I can say is, thanks Sun for giving Bill a few sleepless nights, but really, another proprietary solution is not what the world wants and it's not going to accept it. Many programmers are now using Perl and Python for applications where they would have looked seriously at Java. Python and Perl both work better than Java on Linux, they're faster to develop with, they start faster, easier to debug, more useful libraries, more code developed by other people that you can leverage, etc. etc.
Do you really feel ISO's process has been to the advantage of the language? A single, steering body is always better, whether that body is called Sun or Guido von Rossum.
Then lets have a single steering body, but let that body be chosen democratically, and by merit. As far as I'm concerned Sun has dropped the ball.
We can fix all that, but not under the current conditions. Kaffe is nice. Kaffe is fine. I admire everybody that works on Kaffe, and Blackdown too for that matter. But Sun is passively resisting Kaffe and other Java-clones (by withholding important specifications and compliance tests) to the point where it is very difficult to be sure whether Kaffe is Java-compatible or not. That's evil.
This annoys me, because I'd use Java for a number of things if it weren't such a sluggish, unreliable memory hog. (Which is exactly what it is when you try to run, for example, any Swing program.) Personally, I'm not contemplating using Java for any serious development at this point whereas two years ago I was full of enthusiasm for the idea. Perhaps there are others like me?
It seems to me that you just want to keep things the way they are. Well, it isn't working, that's plain for all to see. Look at the way Linux is leaving Java in the dust as a cross-platform vehicle for delivery of net-enabled applications. That's what Java was supposed to be. Linux has already stolen a lot of Java's thunder, and it will steal the rest unless Sun smartens up fast.
Re:Cross-platform... (Score:2)
Re:Nethack!! (Score:2)
Re:Nethack!! (Score:2)