Sun Releases First GPLed Java Source

An anonymous reader writes "You can now get GPLed JVM sources from Sun. Everyone seemed to be expecting the desktop version (J2SE) but J2ME has been released first. It looks to be buildable for Linux x86, MIPS, and ARM platforms. Sun now calls it 'phoneME.' Enjoy."

And if you want to play with it now... MIDPath

[Anonymous Coward]

Looks like this was released back in November with the full Java GPL announcement according to the official announcement [sun.com].

And people already started hacking it and combining it with all kinds of interesting existing free java projects to product MIDPath [thenesis.org]

Seems the GNU Classpath, Kaffe, GCJ, etc projects really want to Collaborate [wildebeest.org] and work together [wildebeest.org] with Sun according to their latest release notes [gnu.org]. 2007 might be a pretty interesting year for Java and GNU/Linux (and mobile devices!)

Re:

[LarsWestergren]

Another project which might be fun is this one [pspstart.de] porting J2ME to the PSP. I'd like to see one for Nintendo DS even more, but maybe that is coming. :-)

Re:

[Chief Camel Breeder]

"Seems the GNU Classpath, Kaffe, GCJ, etc projects really want to Collaborate [wildebeest.org] and work together [wildebeest.org] with Sun according to their latest release notes [gnu.org]. 2007 might be a pretty interesting year for Java and GNU/Linux (and mobile devices!)"

If SUN Java is GPL'd, why would anybody carry on with an alternative version? Do they really thing that they can do better than SUN? Usually they do worse.

Kaffe in particular has been a problem for my project because it lacks some of

Re:

[DragonWriter]

If SUN Java is GPL'd, why would anybody carry on with an alternative version?

Isn't that one of the whole points of the GPL, to enable that?

Do they really thing that they can do better than SUN?

Sure. And they probably can, at least for certain uses.

Re:

[Schraegstrichpunkt]

Isn't that one of the whole points of the GPL, to enable that?

To enable it, sure, but we've seen that for the most part, GPL-covered software tends not to be significantly forked most of the time, since the economics of the situation tend to encourage people to pool their resources together. I imagine this observation was a major consideration for Sun in choosing the GPL in the first place.

Re:

[DragonWriter]

I would be delighted if they could build a better Java implementation than SUN have done to date. In what specific ways do you think they can better SUN?

I don't know. I do know that there are a lot of competing trade-offs in implementing most complex systems, and that some users may have interests that are different than those Sun is designing around. So that someone else with access to the Sun source could make better implementations for specialized uses than Sun (even if it means, say, an Linux distributo

Re:

[rpdillon]

Well, I think Jikes beats the hell out of Hotspot, but that is just my opinion. And Jikes was open source before Sun even talked about open sourcing their version. So, yeah, I think there are probably other areas that can be improved upon in Sun's code by the community.

Re:

[rpdillon]

My mistake, I was referring to javac. I always use javac with Hotspot, so they are perhaps more tightly bound in my thinking than they should be.

What?

[Evil Sheep]

phoneME? phone MicroEdition? Some kind of really small phone? Perhaps, a...micro phone?

Re:

[mangu]

phone MicroEdition?

No, I believe "ME" stands for "Millennium Edition".

Re:

[Evil Sheep]

You're thinking of windows. phoneME is the latest name for J2ME [sun.com] or Java 2 Micro Edition, the version of Java that is put on phones and PDAs.

Its 7:00 AM and its slashdotted

[DeadSea]

Its not responding. Are there really enough slashdotters awake at 7:00 AM (EST) to bring down something from Sun?

Or is it just so large that the two people that are downloading it are now sucking up all the bandwidth?

In any case, anybody have a torrent?

Re:Its 7:00 AM and its slashdotted

[Richard_at_work]

Hint: the rest of the world doesnt go on EST. Its not 7am where I am, its halfway through the working day for me - try to think outside your own country, Java usage isnt limited to the US.

Re:

[lahvak]

Well, this is "news for nerds". Everybody knows that USA is the only country that has nerds. We European geeks are all really cool and not at all nerdy, and the same goes for geeks from Asia, Africa, Australia, South and Cental America and Canada, and Antarctica. That's why we never read Slashdot... oops!?

Re:Its 7:00 AM and its slashdotted

[DjReagan]

And here was me thinking that the domain name would be more relevant to where the server was hosted/run rather than where its users came from.

Re:

[Richard_at_work]

Based on the fact that theres very rarely a drop off in number of comments being posted while the US is asleep, I dont see how your argument is valid.

Re:

[joshetc]

I don't see how that is a valid arguement (not saying Slashdot is primarily US based). It is a geek site though, I'd imagine most of us from the US either don't work and are up all night or have IT jobs and are at our place of business / getting ready for work at 7AM EST.

Re:

[Richard_at_work]

Or using Occams Razor - there are as many people outside the US taking part in the Java download as there is within.

Re:

[jZnat]

There's a huge drop in the number of stories posted, and that not only annoys the other side of the world but also people like myself who might be awake at 4 in the morning (GMT-6) with nothing to do.

Re:

[Curtman]

most people who go to slashdot are in the US

Says who? Show me something that says more than 50% of Slashdot visitors are in the U.S. please.

see how it says slashdot.org and not slashdot.eu

And how do you figure .org is exclusive to the U.S. ?

Re:

[minus_273]

"Says who? Show me something that says more than 50% of Slashdot visitors are in the U.S. please."

Maybe the fact that its in American english, on a us server, with a us domain and american editors. but most of all, and this is the real proof, the ads target Americans. If there were really more visitors from other countries dont you think it would be a waste advertising american products and services.

and .org is american just like .com and .net are. Other counties use a national suffix or things .co.uk inste

Re:

[Wudbaer]

Good you tell me that I have to relocate immediately to the US due to having two .net domains in spite of being in Germany. Dude, get a clue.

Re:

[Matt Perry]

yeah but most people who go to slashdot are in the US (see how it says slashdot.org and not slashdot.eu )

Yeah, it says slashdot.org not slashdot.us. So what's your point again?

Re:

[Mikelikus]

You do know there's a whole world that wakes up before 7am EST right?

Re:

[CortoMaltese]

There's a whole world of slashdotters very much awake at timezones other than EST, you insensitive clod!

Re:

[Hektor_Troy]

Now, I realise that I am a rather strange individual, but I tend to be awake when I'm awake, no matter what timezone you make reference to. For instance, if you're in EST, I'm still awake as I write this little post. Same if you're in CET. Or PST. Or GMT+11. Hell, even GMT-11.

But I'm sure you meant something completely different.

Congratulations to Sun and Thank You.

[dwalsh]

Congratulations to the Sun people who have made this a reality.

They are freeing up the crown jewels, and the significance of that fact should not be underestimated. Free as in 'gratis' and free as in 'libre' [wikipedia.org].

I am not a Sun employee, but I am a Java dev., and I would like to remind people of Sun's contributions to open source over the years. While the press communications of executives have muddied the waters, Sun have done more in the past for open source than a certain "Think Free" company. That company pressed for open sourcing Java and then bitched about the choice of the GPL.

I would love to see the source to Websphere (not the Geronimo 'Websphere' product, but the real deal).

... for laughs if nothing else.

Re:Congratulations to Sun and Thank You.

[tygerstripes]

Just wanted to say "hear hear". It took 20 posts before somebody actually had the decency to say Thank-you-this-is-a-good-thing, most of those 20 straying into completely niche related topics. I'm not saying they weren't all relevant or interesting points, but thanks for actually saying thanks.

As far as I'm concerned: the short-term impact of this will be decent as people start getting their teeth into the source (as they have done since November), but the long-term impact will be fucking huge. I don't have a lot of personal experience, but this announcement combined with the fact that so many CS degrees start with OOP by teaching in Java means that people will routinely be encouraged to appreciate the power of FOSS from the start, before they get used to the limitations that its absence imposes.

To reiterate: This-Is-A-Good-Thing.

Re:

[dwalsh]

-- Subtle recursive jokes in sigs are not funny.

I don't get it. Is there some recursion in your sig? If there is, it is too subtle. It is not funny.

Re:

[tygerstripes]

I'd mod you up, but there's no "Fulfilling" modifier.

Re:

[Hatta]

It took 20 posts before somebody actually had the decency to say Thank-you-this-is-a-good-thing, most of those 20 straying into completely niche related topics. I'm not saying they weren't all relevant or interesting points, but thanks for actually saying thanks.

I'm waiting until I can actually apt-get install java and have it work before I'm too thankful.

Re:

[Matt Perry]

It took 20 posts before somebody actually had the decency to say Thank-you-this-is-a-good-thing, most of those 20 straying into completely niche related topics.

That's because most everyone said thanks in the previous stories about open source Java. How many times must it be said?

Re:

[Scarblac]

With Solaris, Java and Open Office, I believe Sun is the biggest contributor to free software now, by far. Thanks Sun!

Re:

[HiThere]

Yes! Congratulations to Sun.

I haven't been taking Java seriously as a programming language because there hasn't been a libre version. Only a gratis one. I'm also rather glad that of the various plausible licenses they chose the GPL, but that's an "optional extra".

Re:

[dwalsh]

If you want the source to Websphere, then you either have to apply for a job with IBM, or come up with several billion dollars to buy its freedom.

Either one will work.

Negotiable at all? I have a bag of magic beans here. Would they take them?

When you apply for a job with IBM, do they show you the Websphere source? Why?

What is this supposed to do for their recruitment efforts?

Something real good I guess!

[freakxx]

I dont have much idea of licensing issues associated with JAVA and GPL but I think it is going to change the things drastically. I guess all the difficulties in making JAVA work properly on a system is only because the open-source vendors can't implement JAVA so freely in contrast with other real open-source things like mono... I hope the JAVA will come properly installed on systems from now onwards and one doesn't need to dig around sun's website to download binaries and then follow some tutorial on internet to set the variables appropriately !

requirements:

[DrSkwid]

To properly build executables for the Linux/ARM target platform, a Linux/i386 build platform must meet the following requirements:

        * Red Hat Linux distribution version 7.2 - 9.0
        * Java 2 Platform, Standard Edition (J2SE(TM)) Development Kit (JDK(TM)) version 1.4.2
        * GNU Make version 3.79.1 or later
        * GNU Cross Compiler (GCC) 3.4.6 or later
        * Doxygen version 1.4.1
        * Development Kit for the Java Card(TM) Platform 2.2.1

To set up the Linux/i386 build environment, you must do the following things:

        * Acquire Monta Vista Developer Tools
        * Set Linux platform environment variables

Acquiring Monta Vista Developer Tools

To build phoneME Feature software for the Linux/ARM (P2 board) target platform, you must acquire the MontaVista CEE 3.1 ADK developer tools. [mvista.com]

Comment removed

[account_deleted]

Comment removed based on user account deletion

Java for OpenMoko a step closer then

[mjrauhal]

This seems good for the upcoming OpenMoko [openmoko.com]-based ARM smartphone; although the project emphasizes native app development, fact is, there's a lot of mobile Java apps floating around. So once this is ported to OpenEmbedded/Moko, it should boost the platform's usefulness for many users.

So thanks, Sun, for this Christmas present. (Now just waiting for the phone to actually come out... :)

Thank you Sun

[wikinerd]

I am very happy that Sun Microsystems [sun.com] open sourced its Java and OpenSolaris products. If I buy my own server hardware, I will certainly prefer Sun. Contrast this with Microsoft, which is known for its Embrace-Extend-and-Extinguish practices, its preference to its own shared source licences for the very few lines of code that they ever made public, their aggressive hiring of some open-source people (why? to silence them with dollars?), and shadowy agreements with GNU/Linux vendors. Sun initially tried to use CDDL, but now took a bold step by adopting GPL and releasing actual, useful, working code under it. This means that Sun has open-minded people in its management.

Re:

[RAMMS+EIN]

``their aggressive hiring of some open-source people (why? to silence them with dollars?)''

Maybe they just want good developers?

JavaOS?

[Doc Ruby]

Does anyone link the Java VM right into a kernel, so it doesn't have to load like an app every time a Java app/let starts? Maybe just as a kernel module? Would that violate the sandbox security? How about a pool of VM daemons? Maybe listening to a network socket into which Java bytecode can be sent for execution, like a webserver that runs against code, not request data. But more tightly integrated to the OS than, say, Tomcat or another JAS.

If not, will the open source of Java help any existing projects do

Smart

[w3woody]

This is smart but for a different reason than "gee, we get to play with the sources." What it means is that phone developers who are looking at putting Java on the phone now have a leg up. While having Java on the phone is pretty much a basic requirement for most cell phone manufacturers, it does mean that if you are a new entrant into the phone market you can get a basic Java environment up and running on your hardware with minimal costs over the development tools necessary to do hardware and software deve

Open Source Java at SACLE

[socallinuxexpo.org]

Matt Ingenthron of Sun will be speaking at SCALE about Sun's new open-source java implementations. SCALE 5x [socallinuxexpo.org] will be Feb 10-11, 2007 at the LAX Westin, in Los Angeles.

Java is not easy to tune

[Serveert]

I hear a number of complaints about Java's performance. Java tuning is an artform. If your app is interactive(web server / GUI), I would use the low pause garbage collector. This will eat up more CPU but be more responsive. Otherwise use the throughput GC which will have noticeable pauses if memory gets low but will be more CPU-efficient. The linux kernel manages this tradeoff for you by measuring how 'interactive' each process is and adjusting things, with java currently you must manage this yourself. Mayb

How does this impact PalmOS?

[antdude]

I am having difficulities finding Java for PalmOS on my father's Palm Treo 680 PDA. He needs Java applets in his default Web browser. From what I read, there is none. Even Palm's technical support says the older versions are incompatible.

Will this source bring the updated versions to work?

Re:Linux is great and all

[EGSonikku]

Pfft, you've got the source, get to work!

And I also want this running on the Super Nintendo this time tomorrow, *snap *snap

Re:Linux is great and all

[AKAImBatman]

And I also want this running on the Super Nintendo this time tomorrow, *snap *snap

To be serious for a moment, I honestly hope that this encourages ports to the Wii, XBox 360, and PS3. Java is an extremely capable game programming language at this point, and could potentially save programmers a great deal of development and debugging time. In fact, the only thing that's been holding developers back from using Java is that it doesn't port to the major consoles. If that were to change...

Re:

[MikeBabcock]

Porting the full Sun Java to the PS3 would be great, but for real development you'd need Sony (or NVidia)'s help getting the OpenGL acceleration working.

I play Wurm Online [wurmonline.com], a fairly involved persistent online fantasy simulator which runs in Java and JOGL and games like it could easily be made to work on the PS3 with PS3Linux, if the OpenGL acceleration were available.

Re:

[AKAImBatman]

Porting the full Sun Java to the PS3 would be great, but for real development you'd need Sony (or NVidia)'s help getting the OpenGL acceleration working.

More precisely, you'd just need the development kit. (Which, granted, is a pretty exclusive club.) Sony already support the micro version of OpenGL, so it shouldn't prove too difficult to port JOGL or LWJGL. Of course, my understanding is that a lot of the graphics programmers develop their own drivers for the consoles. So that part would probably remain un

Re:

[MORB]

and could potentially save programmers a great deal of development and debugging time.

No, They could do that by not writing shitty code like most video game programmers in my experience, do.
Java won't magically make them better programmers. And it has quite some memory consumption issues.
Consoles have a limited amount of memory, you can't carelessly let the heap bloat to hell and back like you can on the desktop.

Re:

[doctor_no]

PS3 already has full-JAVA support via Blu-ray (also you can install Linux)

Blu-ray spec requires that all Blu-ray players have Java(J2ME, JavaTV API, etc.) due to future interactive menus, bonus material, etc will be done entirely in JAVA.

This is the core reason of Microsoft's opposition to Blu-ray and support of HD-DVD(which uses MS's iHD instead of JAVA) being that having a machine that runs JAVA by default and in every home can be very scary to MS.

So you shoiuld be able to make BD-J games on Blu-ray and h

Re:

[homer_ca]

Yeah, how about that RAM usage? The PS3 only have 256MB and the Wii a lot less than that.

Re:Linux is great and all

[IversenX]

You are really, really, really comparing apples to oranges here.

Mono is comparable, yes.

However, Qt, GTK and wxWidgets are all just GUI toolkits! You still need a programming
language (Pascal, C++, Perl, even Java(!)) to use these. Installation will be easier,
though. I'm personally looking forward to "apt-get install sun-java" or somesuch.

Also, it will soon (when J2SE comes out) be possible to write better integration with existing
apps, such as better (faster, more modern) browser applet plugins. That, I'm looking
forward to.

(Oh, and now that the sources aer GPLed, it should be really easy to make this thing run on *BSD if it doesn't already)

Mono is not compareanble either

[brunes69]

While I applaud the Mono team for all their hard work, it is not comparable to Java. Hell, Microsoft's .Net is not comparable to Java yet. With Java, you have a 10+ year old tried-and-true platform. You have 10+ years worth of class libraries written, most Open Source, that eliminate 50%-75% of your workload when writing any application..

Sure, .Net does some things better than Java, like Windowing. But Mono's Windows.Forms is brand new and hardly what I could call enterprise-ready.

Re:Mono is not compareanble either

[alexhard]

While I applaud the Mono team for all their hard work, it is not comparable to Java. Hell, Microsoft's .Net is not comparable to Java yet. With Java, you have a 10+ year old tried-and-true platform. You have 10+ years worth of class libraries written, most Open Source, that eliminate 50%-75% of your workload when writing any application..

Sure, .Net does some things better than Java, like Windowing. But Mono's Windows.Forms is brand new and hardly what I could call enterprise-ready.

And You have 10+ years of waiting for java apps to launch!

Re:Mono is not compareanble either

[repruhsent]

The 90s called; they want their joke back.

Re:

[eno2001]

Sorry... while I might be in favor of opening Java, it IS very late in the game. And Java apps generally suck ass no matter what platform you run the JRE on. I've had to deal with many java apps that are total shit in terms of quality and performance:

1. HP Commandview SDM (for managing an HP VA7410 array)
2. Cisco ASDM (for managing Cisco firewalls. As of 5.0 they finally got it somewhat right, but it took nearly five years for what I consider a critical app. See my rant below.)
3. Rio Music Manager Lite

Re:

[richlv]

the thing, there is no java fault that most developers are shitty. i also have experienced crappy java apps from lucent/avaya, ibm and other "big brands" - on the other hand, i also have seen very nicely coded software that was like "wow. and this is done in java ?".

on example (that i have used only a couple of times) - azuerus. it's fast, it's responsive (especially for java app ;) ).
then there's tribal trouble - 3d strategy game. yep, that's right, written in java.

i am sure there are many more good examp

Re:

[vadim_t]

Azureus? Are you kidding?

It's one of those things I HATE to use. Yes, on the surface it works rather well, but the fact that it is sluggish on a dual Athlon MP 2000+ (which was the heck of a machine when I got it) is extremely unimpressive. The same box runs absolutely everything else flawlessly. Heck, Visual Studio 2005 runs a lot smoother than Azureus.

Re:Mono is not compareanble either

[Roman Coder]

Java is an extremely poor choice of language for desktop applications and high-concurrency network daemons. That is Java's weakness. Do not use it for that kind of application -- it will never come close to C++ in that area. Its strengths generally lie in areas where the overhead of the class libraries do not come into play so heavily -- some server-side work and servlets, and the stripped-down mobile version (that was released today).

As someone who makes his (good) income writing desktop applications in Java SE (Swing) for Fortune 100 companies, I would have to strongly disagree with this (and not just because I want to keep paying my bills either! :p ).

I (as a contractor) come to a customer site, and see crappy Swing-specific code written. Its usually the developer not knowing how to deal with multi-thread programming (event dispatch thread, etc.). I rewrite the app, it goes into production, and the user base loves it. They click on a single web link to start their app, and automatically get updates when new versions come out. They can run it on multiple OSs too (music industry companies use lots of Macs (for example)). Its performance is comperable to other apps running on their OS/desktop.

Java (and Swing, or if you prefer SWT) is more than fast enough to do the job, is very powerful and is allot easier to write to than 3GL languages. But like with any tool (or weapon), you need to know how to use it to use it effectively. And that can be said of any computing language, both 3GL and well as 4GL.

I don't mean to be insulting, but it seems like you really don't know what you are talking about. I would even argue that (especially for businesses) it is the BEST choice of language to write applications in. No idea about using it for writing a game and such, but if you're looking for a 4GL (PowerBuilder) type replacement, its the best out there (even though its really a 3 1/2GL language).

Re:

[frenchs]

With that many parenthesis, I would be more inclined to think your a lisp programmer.

Re:

[WilliamSChips]

C++ is a bigger problem than Java. We should all switch to languages like Java and Python.

Re:

[WilliamSChips]

I actually meant "like Lisp and Python" but I accidentially said "Java". Ah well, I don't need my mind today, it's break.

Re:

[Reteo Varala]

Just remember to return that one, as well.

Re:Mono is not compareanble either

[Shaper_pmp]

"Comparable" != "identical to"

Comparable means you can compare the two things - one might be wonderful and the other total shit, but they exist on the same spectrum.

Apples and oranges are (canonically) not comparable because they're different fruit, so they have different criteria to be fairly judged on.

You can compare a nice apple with, say, a shitty, maggot-ridden one - they both have the same criteria, so comparison is valid.

So, on the basis they're both managed programming environments, both compile to bytecode, both tackle the same kinds of tasks in a similar way, you can compare Mono, .NET and Java.

You might believe one is better than the others, but that doesn't make them incomparable.

Re:

[bytesex]

10+ years ? Are you one of those people that advertise for java developers - must have 10+ years experience ? Java was still called 'oak' 10+ years ago, and all you could do with it was hack a few native gui widgets together. I know - I was there.

Re:

[blincoln]

10+ years ? Are you one of those people that advertise for java developers - must have 10+ years experience ? Java was still called 'oak' 10+ years ago, and all you could do with it was hack a few native gui widgets together. I know - I was there.

Apparently Java 1.0 was released in 1995, so the GP is potentially not exaggerating. I remember my boss giving me a book on it at the time (I was 17 or 18), and I started doing simple work with it in the 1996-1999 era.

It's okay, we're all getting old, and some seni

Re:

[jZnat]

Plus, .Net doesn't have their own Apache.org equivalent for more than enough classes you'll ever need and then some.

Re:Mono is not compareanble either

[molarmass192]

Java is hardly what I would call enterprise ready either.

Man ... that's a +5 Funny if I've ever read one. You obviously don't work in an "enterprise". Take it from someone who does (telco), Java is used in massive deployments where Mono/.Net doesn't even make the faintest blip on the radar. There are production Java apps running with 5-9 uptimes that have been going for years.

Re:

[jimstapleton]

Actually, I do, and I am forced to work with the steaming pile all the time, as it is deployed by many enterprises I've dealt with. Popularity can be because of a good product or good marketing - and I'd blame the latter on Java from most of the software I've dealt with made in Java, including enterprise software.

Actually, the software I administrate is a Java application, it's probably the ONLY Java application I've dealt with that hasn't been a royal piece of shit. Most in house development stuff here see

Re:Mono is not compareanble either

[samkass]

A language and toolkit can't compensate for the people who build and maintain the software. Java is easily the best language out there for writing stable, maintainable systems if you use the right tools and know your domain. .NET isn't bad, but IMHO its one big advantage over Java-- the ease with which one can integrate "native" code-- is also a big weakness in potential stability.

.NET and Java in the enterprise

[jimfrost]

Frankly speaking, .NET is a Java fork. Right down to the bytecode and up through the class libraries. If you are familiar with JDK 1.1 class libraries you'll find practically everything in them in .NET, usually with only package name and method capitalization changes. .NET added a lot, particularly in terms of XML, SOAP, HTTP, and GUI support, and fixed some seriously stupid stuff in Java like classloaders, but it really is a fork.

It's kind of amusing, when you think about it, that what Sun really got out of their lawsuit against Microsoft for their (really, really minor, especially relative to stuff like what Netscape did) modifications to Java was a pure competitor in .NET.

You mention .NET's ability to easily (I'd say "relatively easily") link to native code as a big detriment, but in many .NET implementations that's not used at all. It's easier to work with disparate code like that through a SOAP or database interface. In practice you see a lot of .NET front-ends to traditional servers via a SOAP integration. You see less of it used as a replacement for traditional MFC code, the kind of thing where such integration would be most useful.

But getting back to the enterprise, .NET's largest problem in terms of enterprise software is not that it's less mature than Java (in many ways I'd say that Microsoft took the good stuff from Java and improved it a lot) but rather that it's locked to Windows. Maybe you haven't noticed, but Windows is not a very good server operating system -- not very reliable, not very fast (except in very specialized situations), certainly not scalable. It's all very well and good that you can drop a couple of hundred boxes in there to scale to huge applications, but when you could run the same application on a single Sun you're really not making a cost-effective choice. (I wish I were making that up, but it is actually pretty typical to be able to replace as many as 100 Windows servers with a midsize Sun or two, and that is true not only of stuff like IIS/ASPX versus Apache/whatever that are differentiated by more than OS but also for directly comparable stuff like databases and ETL). Push Windows hard and it will break, often. It's nuts to put it in critical places (although that is done, a LOT, and people pay the price in ongoing maintenance).

Having said that, .NET is probably the single best GUI implementation framework I've seen yet (although that may be damning it with faint praise), and Windows, at least aside from the malware issue, is a pretty fine desktop. In that domain it shows what Java could have been if Sun had been even remotely competent (rather than giving us stuff like AWT and the Swing abomination). We're going to see a lot of .NET on the desktop because it is pretty much best-of-breed. More power to it.

Java is today, and has been since at least the late '90s, often used in enterprise situations. Whether or not it's appropriate in a lot of those situations is debatable, but it is deeply integrated into the core operations of a lot of companies at this point. Personally I feel that JMS is not very good at its job and J2EE as a whole is a steaming pile of dung designed by people who wouldn't know a good application architecture if it ran over their foot, but Java as a whole and these things in particular are out there and being used by a lot of people -- and at least in some cases doing a good job.

It is certainly possible to build robust, reasonably efficient large-scale Java applications. It is even easier to do that in Java than it is in C++, especially if you avoid some of the more ridiculous parts of J2EE. But that doesn't mean it's easy to build that kind of thing, and as you might expect there are a large number of really awful Java applications out there (just as the majority of large applications built on all the other languages out

Re:

[jimstapleton]

Regarding the platform lock: there is mono.

However a lot of corporations won't trust it because it's OSS, and I've found many corporations won't trust OSS.

Re:

[jimfrost]

I don't think it's so much an "open source" issue; corporations don't seem to have much trouble buying into open source so long as it is well supported. We run into Linux a lot in large corporations these days.

.NET is still in the acceptance curve at large corporations. If or when it finds itself securely in place only then will people start thinking about whether or not there is something to be gained by choosing a Mono solution. Today Mono provides almost no benefits, and quite a few drawbacks, compar

Re:

[shutdown -p now]

Having said that, .NET is probably the single best GUI implementation framework I've seen yet (although that may be damning it with faint praise)

You mean the one (WinForms) that still doesn't have proper layouts, and until version 2.0 didn't have owner-drawn listviews? Or the one (WPF) which is so ugly it beats Swing in Java 1.4 (I'd rather look at non-smoothed fonts then suffer the pain of catching a glimpse of what WPF font smoothing does)?

Personally, I rather find Swing to be one of the best, if not t

Re:

[jimstapleton]

It's better, because they don't use a lot of the Java stuff, they actually used a few small parts of the Java features, and built what they needed on top of those, because the fancier stuff in Java ended up not being sufficiently cross platform. The biggest example they mentioned was SWING.

The first two I can think of:
Nationwide's hiring website for the longest time (maybe even still now), would only work with IE, it was written in JSP/Applets. Dunno why, but a couple applets wouldn't work in firefox in Wi

Re:

[computational super]

would only work with IE, it was written in JSP/Applets.

I'll give you that applets never worked worth a damn, but their choice of JSP was irrelevant to whether or not the pages loaded in your browser - JSP generates HTML/CSS/JavaScript and your browser either understands it or it doesn't... the fact that the backend was done in Java is irrelevant in that case.

You're right about applets - although in Nationwide's (possible) defense, there was a time, not too long ago, when it was not obvious what the right

Re:

[jimstapleton]

"Nationwide's hiring website for the longest time (maybe even still now), would only work with IE, it was written in JSP/Applets. Dunno why, but a couple applets wouldn't work in firefox in Windows or BSD"

Sounds like an implementation problem; not a fault of the language.

"The local mass transit website used to use Java for their website. It would always work on IE or Firefox, not both. Whichever it didn't work on would have massive null-pointer exceptions."

Again, not a fault of the language - that's

Re:

[jimstapleton]

Yes, I am a programmer.

And I brought up FF + IE because they were both using the same JVM on windows.

Or would you rather an older example of Netscape on Windows (Sun JVM) and Netscape on Solaris (same version of the Sun JVM)...

I had a couple of those two.

Re:

[jimstapleton]

then don't read it. And yes, they are Java because no other option makes sense.

I can add a few command line java utilities I used as well to the list, except I don't know there names, I just threw them away after findnig the worked well on Solaris, but not linux or windows even though they were supposed to.

Sorry, but if a cross platform API doesn't work the same on multiple platforms, it's flawed. It may work fine 99% of the time, and I may have had the 1% bad experience, but the fact is, I've dealt with th

Re:

[jimstapleton]

different application.

But both the applications I mentioned before sent applets to the browser, not just server side JSP, so the browser was running Java stuff too.

Re:Mono is not compareanble either

[Tmack]

I think it's the steaming pile of bad programmers that you're having problems with, not Java.

> Actually, I do, and I am forced to work with the steaming pile all the time

I fall into the "work for a Telco and have to deal with steaming piles of java" category as well. Yes, the developers have something to do with it, but its also the Java mentality of those developers, as well as a few wonderful quirks of the language and its environment. For starters, it seems, for our dev team at least, anything can be done in the Java world if you throw enough $$ at a "platform" or "Framework" and then spend the next several years with large teams of developers and outsourced help (India) to find that the platform/framework you bought cant actually do anything you bought it for (buzzwords), so another one must be bought to solve all the problems (rinse/repeat). They also like to over develop stuff, writing full-blown "feature" filled aps where a single line of cron and/or 5 lines of perl would suffice, and spend the next few years debugging it and restarting it every time it crashes (nightly for most). Java has also somehow managed to become the ONLY SOX compliant language in the eyes of management, possibly due to the dev team, requiring SOX related stuff (which becomes whatever someone feels is somewhat related to SOX in any remote fassion) to be put into a Java wrapper if its not already Java based so that their Java platforms can tickle it all they want.

As for the Java platform itself, one of the most common things done in my group (system ops) with systems is restarting Java aps and Java engines. Why? The ap breaks or tickles some Java bug. One nice feature in Java (or Tomcat or JBoss) we know about because of specific breakage it causes is that it keeps its own cache of DNS. The only problem being it ignores TTL and the whole thing has to be reloaded to refresh that cache. Then there are the other Java bugs that cause breakage to the bewilderment of our dev team. Load a page, it works, go back and try again a few minutes later and it crashes. Most likely a poorly written ap causing some memory buffer to overflow, but wasnt Java written to handle that sort of stuff internally so the app dev team doesnt have to worry with it??

People might see it as Enterprise App worthy, but I think it has long gone the way of PHP, where most developers have gotten lazy and sloppy. I have used it in the past, though I currently use Perl for a good number of reasons. Like any language, it has its place.

Tm

Re:

[GrueMoon]

Well, gee, thanks for making my point.

>For starters, it seems, for our dev team at least, anything can be done in the Java world if you throw enough $$ at a "platform" or "Framework"...

>They also like to over develop stuff...

>Most likely a poorly written ap causing some memory buffer to overflow...

>...but I think it has long gone the way of PHP, where most developers have gotten lazy and sloppy.

Re:

[Creepy]

I'm stuck in the "write once, test everywhere" world of Java. It's amazing how little things slip through the cracks on some platforms. One bug I hit just last week involving focus listeners on a popup window (dialog) was reported against Windows, fixed in Windows and HP-UX, but never tested or fixed on Solaris and Linux until a bug was written against those platforms. You'd think something like that would get heavy testing, but it still slips into production code.

On the plus side, at least java mostly r

Re:

[rjshields]

I'm personally looking forward to "apt-get install sun-java" or somesuch.

You can do this already in ubuntu and probably debian. Just enable the multiverse repo and do a "sudo apt-get install sun-java2-jdk". You do have to agree to the license, but it's not really that hard.

Re:

[just_another_sean]

Yes, in debian you need to add non-free to your list of repositories.

Slightly offtopic, Etch seems to have lost support for OpenMotif. I used to install this and
then the Citrix ICA Client for Linux. Well that doesn't work anymore.

But, installing Sun Java, Apache and the Java ICA Client works quite well. Sure it's a lot of
closed source but Citrix just happens to be one of those "I'd always use Linux if it would just
run..." apps for me at the moment so Sun's Java really helped me in this case.

A couple more mo

Re:

[marsu_k]

I'm personally looking forward to "apt-get install sun-java" or somesuch

"pacman -Sy jdk" has worked for ages...

Re:

[bky1701]

"Where's the love for FreeBSD, OpenBSD, and NetBSD?"

Have you actually tried to compile it? I compile for my architecture (64) all the time with things that aren't made to support it. Sometimes I have to make small changes to strings, but it can't be THAT much worse installing on BSD.

Re:

[Anonymous Coward]

Neither J2SE or J2EE is very strong anymore. Both for desktop apps and for server apps there are many more better alternatives than Java. However, J2ME more or less has a monopoly on embedded devices. You HAVE to use Java if you want to make applications for mobile phones. Making J2ME free software ensures that the situation will stay that way. It also means that all other J2ME JVM implementations except for Sun's one becomes irrelevant. In this move, they have both killed off all Java competitors and ensur

Re:

[bsharitt]

I agree that J2ME is probably one of the brightest spots of Java crown, but I wouldn't count J2EE just yet, though J2SE is probably relegated to web plugin status alongside Flash.

Re:

[Anonymous Coward]

Sun, you're a bit late.

Fanboys will always be fanboys. Why don't you just say "I like .Net better and that's why I am trying to scare everyone away from Java, so I have a chance"?

Sun did what nobody expected, opensourced its greatest (both in terms of size and of completeness) and industry leading development platform. Now productivity at the grasp of even the most rabid opensource zealot.

Now what? You are going to tell it's "too late"? I will tell you what is going to happen, Mono has just lost any reason to exist and to be use

Re:

[jZnat]

Mono would lose its necessity iff Microsoft were to release the entirety of .Net under the GPL (or LGPL, or another OSI-approved license perhaps) and donated all its related patents either to the public domain or to some other open patent initiative. Since I don't see that happening, uh, ever, we can rest assured that Mono will continue to be relevant as long as .Net is relevant as well.

Re:

[fm6]

Where's the love for FreeBSD, OpenBSD, and NetBSD? :)

The same place as the Amiga support! ;)

This is nothing more than Sun trying to ensure that Java stays relevant, with the greatest stability of other toolkits

I don't think anybody at Sun would deny this. But so what? Enlightened self-interest [wikipedia.org] is nothing to sneer at.

Re:Quality of the code?

[bobaferret]

The quality of the code varies from source to source. Thus making it sub par. Their documentation is okay, since it's produced from the source code and is shown to the public. Atleast on the .java source. But overall I think their code it crap, but not nearly as bad as Mozilla was. The really irritating thing with the rt.jar source code, which has always been viewable, is that they don't follow their own java formatting conventions. There's going to be a lot of available "Janitorial" positions available once all the code get realeased in March (I think they said march). The only thing that really worries me is the JCP process. Linux works well, because in has a benevolent dictator at the top. Translation, it has a vision/direction. JCP's are commitiees, and that will slow down OS/FOSS development efforts. I imagine/hope that ClassPath will stick around and add features/ be the eqivilent of a development branch. There are things I'd like to see added to that language that would never make it through a commitiee (I just can't spell that word this morning, sorry). But by having a development unstable branch, maybe some of these things can be tried out and proven in the field, then added back into the mainline trunk. The JCP seems to work well, but I'm really curious to see if it can keep up with OS development.

Re:

[mhall119]

The name is a trademark, and I suppose Sun want to keep it for compliant implementations, as has been the case since they started licensing Java to other companies for implementation. The problem is that a restriction that you cannot change the APIs to make them incompatible with other Java implementations would not be compatible with the GPL, so the only way around this for them is to change the name for what is released under the GPL.

PhoneME is Sun's name for their implementation of the Java ME specific

Re:

[anarxia]

It's already in the non-free section for Debian in both testing [debian.org] and unstable [debian.org]. So it should be included when etch is released (early next year probably).

Re:

[Laebshade]

Oh you mean because permissions/ownership properties aren't retained? Ah well, it's still fine.

Re:

[Da_Weasel]

I always thought that "bundled tightly" was the condition in which your undergarments got lodged in your nether regions and was beyond any degree digital dexterity and grace to extract. The condition of "bundled tightly" requires that the offending garment be removed and completely reapplied to the body.

Re:

[jimstapleton]

everything compiled with it has to be GPLed as well (as in, binaries containing both portions).

I've seen Linux distros with the Intel 3945ABG drivers (binary blob stuff) and Linux distros with Java...