Follow Slashdot stories on Twitter

 


Forgot your password?
Close
typodupeerror
×
Java Programming Software

Sun Completes Java Core Tech Open-Sourcing 141

Posted by Zonk from the the-beans-are-gratis dept.
MsManhattan writes "A year after announcing its plans, Sun Microsystems has made almost all of the core technology in Java available as open-source software under the GNU general public license version 2 (GPLv2). However, some of the code remains 'encumbered'; that is, Sun doesn't have sufficient rights to release it under GPLv2, and the company is requesting the open-source community's help in resolving these issues. Rich Sands, community marketing manager for OpenJDK community at Sun, would not say what percentage of Java's 6.5 million lines of code are encumbered, but explained that it is largely Java 2D graphics technology, such as font and graphics rasterizing."
This discussion has been archived. No new comments can be posted.

Sun Completes Java Core Tech Open-Sourcing More

Sun Completes Java Core Tech Open-Sourcing

Comments Filter:
  • Woohooooooo! :) Now OpenSolaris
    • There's Nexenta [gnusolaris.org], which is basically Debian on the OpenSolaris kernel. But you're right, it would be nice to have a completely unencumbered Solaris.

    • Re:One word! (Score:5, Insightful)

      by eviltypeguy ( 521224 ) on Tuesday May 08, 2007 @12:34PM (#19039285)
      OpenSolaris is open source and *free software* minus a few encumbered portions, just like Java. So I'm uncertain as to what you are referring to.

      • Re: (Score:3, Informative)

        by howlingmadhowie ( 943150 )
        there's been a rumour for quite a while now that sun will move solaris10 from the CDDL to GPL3 as soon as GPL3 is finished. (one should also mention that duke nukem forever will also be released under the GPL3 according to a similar rumour)

        • Re: (Score:2)

          by zootm ( 850416 )
          I'm fairly sure that Sun said that they decided against putting Solaris onto the GPL because they didn't see a benefit, and their userbase had said the same thing. I could be wrong, though.

    • Re: (Score:1)

      by ruffnsc ( 895839 )
      Thats three words and some other characters but who is counting.

  • Not much of a surprise. (Score:5, Informative)

    by AKAImBatman ( 238306 ) * <<moc.liamg> <ta> <namtabmiaka>> on Tuesday May 08, 2007 @11:09AM (#19038077) Homepage Journal

    Rich Sands, community marketing manager for OpenJDK community at Sun, would not say what percentage of Java's 6.5 million lines of code are encumbered, but explained that it is largely Java 2D graphics technology, such as font and graphics rasterizing.

    In case anyone is wondering, this isn't much of a surprise to the Java community. When Sun was creating the latest and greatest Java libraries, they designed the APIs themselves to be generic. However, Sun generally licensed the underlying libraries for their reference implementation rather than developing them in-house. In the case of the Java2D APIs, they used code from Kodak to do all the fancy 2D rasterizations and transformations. This is why many Java coders thought that Sun's reference implementation would never be Open Sourced. (Happy to be wrong, BTW.)

    That code by itself could probably be replaced with a modern 2D rasterizer (similar to the types found in SVG and Canvas implementations), but it would need to be heavily overhauled to backport the VolatileImage support added in Java 1.4. (Basically, the JVM is able to manage the video card memory to store images for faster rendering and backbuffering.) I'm thinking that something OpenGL-based would be the best bet.

    However, that's not the only major library used. JavaSound also uses Dolby Headspace to render sound. It barely uses a fraction of the library's capabilities, but it would still need to be replaced. I don't know what was used for cryptography, but that would be replacable with a library like Bouncy Castle [bouncycastle.org].

    All in all, the final code shouldn't be too hard to replace as long as Open Source equivalents can be found. However, these areas *do* require significant expertise, so don't expect that joe random can jump in the code and make it happen.

    • Re:Not much of a surprise. (Score:5, Interesting)

      by AchiIIe ( 974900 ) on Tuesday May 08, 2007 @11:38AM (#19038469)
      I would love to help myself
      As a longtime java developer, and advocate of the open source java, I think it's time to put my time where my mouth is. Albeit a few questions:
      * Where is the effort being organised
      * Is there a list of the methods that need to be reimplemented?
      * Have I been tainted -- I've seen java code before, If I reimplement something and it might look like the previous code, how do I guard against this?
      * Can we make java better?

      • I would love to help myself
        As a longtime java developer, and advocate of the open source java, I think it's time to put my time where my mouth is. Albeit a few questions:
        * Where is the effort being organised
        * Is there a list of the methods that need to be reimplemented?
        * Have I been tainted -- I've seen java code before, If I reimplement something and it might look like the previous code, how do I guard against this?
        * Can we make java better?

        It seems to me that java.sun.com is the place to start looking for these answers, and not slashdot ;)

      • Re:Not much of a surprise. (Score:5, Informative)

        by AKAImBatman ( 238306 ) * <<moc.liamg> <ta> <namtabmiaka>> on Tuesday May 08, 2007 @12:03PM (#19038825) Homepage Journal

        * Where is the effort being organised

        https://openjdk.dev.java.net/ [java.net]

        * Is there a list of the methods that need to be reimplemented?

        I imagine that the Graphics2D class as well as the VolatileImage class need to be implemented. The BufferedImage implementation may need to be modified to obtain the "Automatic Image" behavior seen in the Sun JVM. Depending on how much Sun has withheld, you may need to implement the GraphicsEnvironment code as well. I recommend asking on the mailing list for specifics.

        * Have I been tainted -- I've seen java code before, If I reimplement something and it might look like the previous code, how do I guard against this?

        The "tainting" problem was always an OSS community invention based on Sun's license wording. As long as the code you've seen is in the OpenJDK, it's quite impossible for you to be "tainted" as Sun has given it to you under the terms of the GPL. Of course, you should always consult a lawyer for the latest in paranoid-protective behavior, but I sincerely doubt that Sun would try and sue you over something they just gave you.

        * Can we make java better?

        You can do whatever the GPL license allows you to do. However, I imagine that OpenJDK won't accept anything more than implementation enhancements. New APIs need to go through the JCP for standardization before they can be accepted into a mainline JDK distribution. Again, ask on the mailing list for more details.

      • Can we make java better?
        Yes: add typedef to the language.
      • You could always help out with Apache Harmony [apache.org] to create a full, open sourced JVM!

      • * Can we make java better?

        Making it stop throwing away type information (generics) would be very, very nice. This is imho one of the greatest current flaws in Java. I fear it will mean changing the JVM, though.

        In case anyone wonders: From the perspective of the JVM, List<String> is the same as List. If you can't see why this is bad, take a breath and think again. Think casts. Think interfaces.

        Besides this, I could make a very, very long wishlist for Java. Though I suppose I'd just end up with C++, minus the old annoying ba

        • Making it stop throwing away type information (generics) would be very, very nice. This is imho one of the greatest current flaws in Java. I fear it will mean changing the JVM, though.

          Which is exactly why Sun opted to make generics a compiler constraint, so that generified code would still run on older VMs. I agree though, that since Java 6 introduced changes to the bytecode, they could have included generic type information at that point.

          Besides this, I could make a very, very long wishlist for Java. Thou

          • I think the problem can be summed up more as "those who think C should be the only language will do everything in their power to denigrate everything else."

            Not as pithy, I'm aware.

          • Making it stop throwing away type information (generics) would be very, very nice. This is imho one of the greatest current flaws in Java. I fear it will mean changing the JVM, though.

            Which is exactly why Sun opted to make generics a compiler constraint, so that generified code would still run on older VMs. I agree though, that since Java 6 introduced changes to the bytecode, they could have included generic type information at that point.

            Exactly. Throwing away type info is a major flaw. Practically speaking, Java is no longer typesafe, though you get warnings if you perform an unsupported operation. But that is like warning you that the dog bites rather than train it not to bite...

            Besides this, I could make a very, very long wishlist for Java. Though I suppose I'd just end up with C++, minus the old annoying baggage.

            Isn't that an old saying? "Those who don't understand C are doomed to reinvent it"? Or maybe it was Lisp.

            Perhaps Lisp, C is very simple (and a completely different game than C++). But perhaps you mean that it would be far better to work on a language that already have lots of the missing features implemented (like C++) or simply more advanced if yet somewhat incomplete (like Haskell). You might be right, then. It is just that I am forced to drag Java along like a block of concrete around my legs, and thus I hope (someone else) will fix some of the more glaring shortcomings. :)

        • Re: (Score:3, Informative)

          by myawn ( 562028 )
          I'm sitting in a JavaOne BOF right now where reification of generics (the fancy way of saying not throwing away type information) is being mentioned as a possible JDK 7 feature.

          There are some real compatibility issues in doing this (I remember Gilad Bracha covering this at a previous JavaOne), but it sounds like they are at least willing to take a look at it again.

        • Re: (Score:2)

          by MyIS ( 834233 )
          Meh. Even if there is an improper cast from a List<String> to List<Integer>, the JVM still won't break; it will just throw an error at access time instead of original cast time. Muddling up Java's clear and simple type system isn't worth it here.

          • Meh. Even if there is an improper cast from a List<String> to List<Integer>, the JVM still won't break; it will just throw an error at access time instead of original cast time. Muddling up Java's clear and simple type system isn't worth it here.

            Yeah, why have compile-time types at all? Let's throw all that away, who needs it? You'll get a stacktrace if you encounter the bug while running. And overloading? Who ever needs to overload on a List<float> vs. List<int>, say? No need for that, we can just sample the elements to see if they are of one type or the other.

    • I agree, code replacement does seem to be the most reasonable way to resolve this issue. However I cannot help but feel that this would prematurely cause forking and general divergence from the original closed Java.

      Additional features and improvements are no doubt going to occur as the result of opening up the code base, but I feel that compatibility should be at the forefront of any design decision when Java is involved.

      After all, isn't the whole point of Java interoperability, and hardware agnosticism (ge

    • Re: (Score:3, Interesting)

      by ciroknight ( 601098 )
      Wouldn't Cairo (cairographics.org) be the perfect fit for a software 2D rasterizer? Might not be the fastest but it's there, it's usable, even Firefox is jumping on the boat.

    • Re: (Score:2)

      by LWATCDR ( 28044 )
      And combine this with JGC and I will be a happy man.

  • Serious question: Java, Apache 2, and GPLv2 (Score:2, Interesting)

    by Anonymous Coward
    Apache currently has many [apache.org] Java [apache.org]-based [apache.org] Apache 2 licensed [apache.org] projects. The Apache 2 license is incompatible with the GPLv2 [fsf.org].

    Does releasing Java under the GPLv2 effectively prevent the use of Apache 2 code (or any commercial code)? Does this mean that the only Java programs that can be used with the GPLv2 copy of Java are those under the GPLv2?

    • Re: (Score:2, Informative)

      by gedhrel ( 241953 )
      No, it doesn't. There was quite a deal of work that went into making sure that that was ok.

      • Re:Serious question: Java, Apache 2, and GPLv2 (Score:5, Informative)

        by mhall119 ( 1035984 ) on Tuesday May 08, 2007 @11:35AM (#19038427) Homepage Journal
        I'm giving up mod points on this article, but given the amount of misinformation you're getting, its better that I post a reply.

        Sun isn't using GPLv2, they are using GPLv2 + the Classpath exception. This is a modification to the GPLv2 pioneered by the GNU Classpath project. It basically makes Java's code GPL, but doesn't require code running on the JVM, or using the J2SE libraries, to be GPLv2 complaint.

        Links here:
        http://www.sun.com/software/opensource/java/faq.js p#g5 [sun.com]
        http://www.javalobby.org/java/forums/t84256.html [javalobby.org]
        http://en.wikipedia.org/wiki/GPL_linking_exception [wikipedia.org]

        • Philosophy of exception: usage IS NOT derivation (Score:2, Interesting)

          by Anonymous Coward
          Since you understand the Classpath exception well, you might be interested in an alternative explanation we've devised, which is both simple and exact.

          For elements licensed under the Classpath exception,

          USAGE of an element by linking does not imply derivation from that element.

          It's this key consequence of the exception that makes it so excellent.

          Nobody who is moderately sane thinks that when we use a car then the car-driver combination is a derived work of the car, and that therefore the copyrights coveri

          • The term "linking" has actually caused a lot of confusion due to the difference between how Java links to libraries as opposed to C/C++. For C/C++, linking is much closer to a derivative work than in Java since (to my knowledge, I'm not a C programmer) when you "link" to a C/C++ library, you actually need the source code (or at least header files) from that library in order to compile your code, therefore including elements of that library's source in your final binary. In Java, you can "link" your code t

          • Re: (Score:2)

            by ajs318 ( 655362 )
            It's important because when you call a non-GPL library from a GPL program, or a GPL library from a non-GPL program, the combination (program + library) in the memory of the computer could be considered a derived work of both components.

            It's also irrelevant, because making a copy of a computer program in memory, from a source which does not infringe copyright in and of itself (e.g. an original CD that you rightfully own) and for the purpose of running the program, is explicitly permitted by copyright l
    • Apache should be able to continue to use the same license it is using now. The Apache 2 license takes a much stronger anti-patent stand than the GPL2, or even the GPL 3 license. This makes Apache 2 less compliant with other open source licenses.

      See the OP's link to FSF for more info
    • From the GPLv2, last published June 1991:

      If identifiable sections of that work are not derived from the Program, and can be reasonably considered independent and separate works in themselves, then this License, and its terms, do not apply to those sections when you distribute them as separate works.


      In addition:

      However, as a special exception, the source code distributed need not include anything that is normally distributed (in either source or binary form) with the major components (compiler, kernel, and so on) of the operating system on which the executable runs, unless that component itself accompanies the executable.


      The second part is tenuous given the wording, but could be made to apply given Java's attempt to be the execution platform. However, the former part carries far more weight. Programming to the Java specification does not imply that you are basing it on a GPLed version.

    • Re: (Score:3, Informative)

      by slack_prad ( 942084 )
      GCC is GPL, but my gcc compiled malware program is commercial software.
      • "malware program is commercial software." You admit you write malware, and use it commercially? On Slashdot? You are one brave soul. ;)
    • Like everything else, the GPL2 only covers the code in question, not the input/output
      For example, Sony uses GCC in the PS3 devkits, yet there is not a single OSS game for the PS3
      If I write a paper in OpenOffice, the paper is not automagically GPL
      etc. etc.
  • Saw the original suggestion last year and mentally wrote it off. Didn't think Sun would drive it through - the fact that it has happened with some areas still to be worked out (the libraries) impresses me the most.

    It'll be very interesting to see where this goes next... let's see if the open-source leveraging effect works in practice for something this big (and this awkward).
  • Being on the fence about using Linux and never really being sure what version, what build and what revision I should go for (hundreds of choices make it intimidating for newcomers to the Linux world), I've always been curious what would have happened if there was a board to approve the general direction, additions and revisions so that be it a platform or a programming language, the tangents wouldn't stray too far from the main group.

    FTA
    ---------
    Open sourcing Java represents one of the largest donations of

    • Being on the fence about using Linux and never really being sure what version, what build and what revision I should go for (hundreds of choices make it intimidating for newcomers to the Linux world), I've always been curious what would have happened if there was a board to approve the general direction, additions and revisions so that be it a platform or a programming language, the tangents wouldn't stray too far from the main group.

      It's called the Linux Standard Base [linux-foundation.org]

    • Being on the fence about using Linux and never really being sure what version

      The number of GNU/Linux desktops running Ubuntu is now one in three and still growing, couldn't do too badly to jump on that bandwagon if only to give it a try.

      Sands: " Open-source developers need to have rules and governance spelled out for them" But maybe not by Sun. The community might choose another group to lead them, especially since Sun hasn't taken the time to build free alternatives to important Sun java libra
  • So, does this mean we'll see better integration of Firefox with Java now? Currently, if I go to a page and the browser freezes for about 15 seconds, I think to myself, "This page must have an applet."

    • Re: (Score:3, Insightful)

      by AKAImBatman ( 238306 ) *

      Currently, if I go to a page and the browser freezes for about 15 seconds, I think to myself, "This page must have an applet."

      That's caused by the JVM startup. The pause can be nearly eliminated by starting the JVM with the browser. Of course, that slows the browser startup and holds onto memory needlessly. So I doubt anyone is going to bother doing that for a dead technology. (i.e. Applets)
    • Firefox supports Java just fine. If you get a delay when you see a Java applet it's probably because Java isn't loaded into memory.
    • Why Sun never recognized this as a problem and did something about it, like working with browser makers to pre-load the JVM, I don't know.

      It's the no1 reason why Applets died.. and because of that we had to wait until the Javascript/XML fightback to get Web 2.0.

      Is this just another example of corporate culture getting in the way of seeing reality?

  • The Sun Experiment (Score:5, Insightful)

    by krelian ( 525362 ) on Tuesday May 08, 2007 @12:05PM (#19038845)
    (I posted this under the wrong story so I am copy pasting here....)

    I think most people in the F/OSS community don't see the importance of Sun's actions in the last couple of years regarding the economic viability of F/OSS.

    This is a company who opened sourced (or is open sourcing right now) a very large important and complex portfolio of it's software. It is also a company which is considered a major player in its field and a lot for these software products where successfully sold for big bucks in the past.

    F/OSS takes a lot of criticism regarding it economic model which most businessmen see as non existent. If Sun can pull it through and improve its financial results after making such a big commitment to F/OSS software, only than will the F/OSS community will have a winner in their hands to show off in front of skeptics. This is not the same as Red Hat who made a business out of F/OSS but a company which is rejecting the old ways of closed source and is taking a big gamble that F/OSS is not only the right thing to do morally, but that it can also become a better business model than closed source software.
    • F/OSS takes a lot of criticism regarding it economic model which most businessmen see as non existent

      That's based on a misunderstanding. FOSS is enormously valuable to businesses, but not as a "business model" in the sense of earning money. FOSS is about collaborative development by end users for the purpose of reducing risk and costs.

      But Sun isn't an end user. They reduce neither their risks nor their costs by open sourcing Java, which raises the question: why is Sun doing this?

      I think there are several
      • I don't know about Solaris but Java is definitely not losing ground. #1 language by any measure.
        • #1 language by any measure.
          bullshit there are plenty of measurements by which it loses. For example if you count in terms of the total number of processors running stuff written in a language C and C++ would probablly win by far.

          java applets have almost completely lost the in browser apps market to flash and ajax (a posh name for javascript/dom based coding that uses the network)

          java applications never really took of in either the commercial or opensource desktop software markets (there are notable exceptio

          • java applications never really took of in either the commercial or opensource desktop software markets

            Now you're talking bull. Search on the major IT job advertising websites by language, and you'll notice that there is a massive demand for Java programmers. Not only to work on J2EE web related stuff, but Swing based applications, particularly for banks. C# and .Net have started to eat into this market to some extent, but there's a serious amount of Java code that would have to be ported or thrown away

            • sorry when i said commercial software i meant software where copies are sold to users.

              i don't dispute that java is very common in the custom internal apps which users will be forced to use market

  • Gnu classpath ? (Score:3, Interesting)

    by Salsaman ( 141471 ) on Tuesday May 08, 2007 @12:09PM (#19038883) Homepage
    Is there some reason why they (Sun Java) can't merge with GNU classpath ? It seems quite likely the latter would be able to provide all of the missing libraries, since they were working for many years on a free implementation of Java.

    • Re: (Score:3, Informative)

      by rabbit78 ( 822735 )
      Speaking as one of the GNU Classpath developers, I agree with that idea. However, there are some issues to be worked out, like the ownership of the code. GNU Classpath's code is copyrighted by the FSF and Sun's code is copyrighted by Sun. Both organizations require that contributors sign a copyright agreement before accepting code. I don't know about details, but I guess Sun and the FSF would have to arrange something before such a thing could happen. Luckily, Dalibor Topic, also a Classpath developer, is i
  • I am a c++ developer who doesn't really like C++, but has held back from switching for years.

    I've been very annoyed at and cynical about Sun's handling of Java. If Sun had opened Java years ago I would have switched (and I'd be much happier right now).

    Now, it's a bit late, but I am satisfied that they have finally done the right thing, in a real and full way. The remaining issues are understandable. Yes, they only did it because of competitive pressure, but I guess that's how it goes.

    I won't be switching

  • However, some of the code remains 'encumbered'; that is, Sun doesn't have sufficient rights to release it under GPLv2, and the company is requesting the open-source community's help in resolving these issues.

    Mozilla had the same issue when Netscape first released it. It's probably the single biggest reason why Mozilla development was held up for years. That may not happen here, but I find it worrisome.

    • Re: (Score:3, Insightful)

      by KarmaMB84 ( 743001 )
      Aside from scrapping nearly the entire Netscape codebase and starting over?

      • As I understood it, the portion of code Netscape open-sourced didn't really even compile into a browser. Netscape was in the middle of rewriting all their own stuff in Java (of all things) already, so the codebase was incomplete and neglected. Mozilla managed to extract Bugzilla and some other interesting stuff, but they did not throw away a functioning browser implementation; they cannibalized a mostly abandoned codebase and used what they thought they could.

        Food for thought: Imagine how different it woul

  • Thank you Sun and thank you RMS for the GPL (Score:4, Insightful)

    by Anonymous Coward on Tuesday May 08, 2007 @01:33PM (#19040277)
    Sun contributes a lot to Open Source and produces very interesting hardware. Thanks for Java, thanks for OOo (which I find to be a huge pig but then I also find MS Office to be a mega pig), thanks for OpenSolaris... And thanks to RMS for the GPL. It's good to see Sun choosing the GPL (with exceptions) as their open source Java licence.

    Trolls are already in the thread saying that OOo is unusable and that (Open)Solaris is the most insecure Unix ever. The trolliest of them all even dare to say that Java is a failure... On this last point nobody should be mistaken: Java is a langage and a VM whose success is beyond any language author's wildest dreams.

    As stated by James Gosling himself not long ago: it is basically impossible to do a financial transaction today without having Java involved at one point or another... And this is a Good Thing (TM) Why is the banking industry so in love with Java? The bullet-proof JVM and portability of the languages. Not a single language before Java had ever offered such an incredibly good VM running on so many different hardware/software configurations.

    There are countries where tens of millions of people are carrying "Java SmartCards" on them. More than 90% of all cellphones sold worldwide have a J2ME VM (agreed: not Java's most glorious example, but nearly 20% of all games sold worldwide are little Java games for cellphones sold to casual gamers... Food for thought to anyone dismissing J2ME).

    Java is a success that has grown beyond anyone's wildest imagination and it just keeps getting better and better. I was an very early Java believer (back in the days where it was really messy to do Java development on Linux) and back in the days there were many misconceptions regarding Java. Nowadays it's simply unexcusable to try to deny that Java is a huge success. Like COBOL today, Java will still be the technology making the real-world work in decades (insert your lame jokes about COBOL here... I personally laugh everytime some COBOL is involved when you're making a real-world transaction [and, yup, I've done COBOL programming]).

    Note that (contrarily to C#/.Net) "Java the language" and "Java the VM" have been chosen because they were very impressive technologies, not because the company creating them had a monopoly on a certain market. C# is a copy (7 years later or so) of "Java the language", with some things done better (thankfully, 7 years later, that they got some things right ;) and some others done worse (*). This is a testimony to Java's quality as a third-generation OOPL (I'm not saying that third-generation language are the be-all end-all of programming nor that OO is the be-all end-all of programming, but simply that as an 3rd generation OOPL Java is very good).

    Thanks Sun.

    (*) C# has then been basically forced down the throats of programmers locked in the MS monopoly. If C# hadn't Microsoft's illegally maintained monopoly (the "illegally maintained monopoly" is a fact, checks your facts if you disagree) backing it wouldn't be anywhere near where it is today.

    • Re: (Score:2)

      by hughk ( 248126 )

      As stated by James Gosling himself not long ago: it is basically impossible to do a financial transaction today without having Java involved at one point or another...

      No, I know at least one major bank where the retail banking system is written in Assembler. It just works and handles an incredibly large number of transactions. I know of exchanges where the backend is written in a mixture of C++/C & COBOL all the way through to the client APIs. You *can* use a Java based GUI on the system, but you coul

    • Very well written, and I personally agree with a lot of what you said.

      I also want to add a few more things which you have missed.

      Java is also part of the Blue-Ray Specification, with Java being used to create some of the more interactive Elements.

      Also here in Europe, Java is used in the MHEG DVB specifications, as well as new iterations of DAB.

      Recently, there has also been some work in creating emulators for other platforms using Java. However, they are not emulators, in the strict sense, as the native code
  • What about the poor folk like me running Linux on Sparc who can't use anything Java? Or some other architecture? Once this is complete it's going to be good for the non x86 Linux community.

    • Re: (Score:1)

      by chromatic ( 9471 )

      Finally, Java might be cross-platform after all!

    • Re: (Score:1)

      by jaiger ( 166690 )
      I use Kaffe (http://kaffe.org) on my Debian Sparc Linux system(s) with OK results. I use the version packaged for etch.

      Sure it has limitations:
      - it's not up to JDK 1.5+ levels for compatibility
      - performance *seems* slow but I haven't compared with Solaris+Sun-JDK on same hardware yet

      I wouldn't say "can't use anything Java" though.

      -Joe

      • On my Sun Blade 2000, Debian wasn't up to a version that had a compatible SILO apparently. Gentoo was the only one that worked. Haven't tried Debian 4 yet, which I presume works with a newer version of SILO now. Another issue was Flash. I wasn't aware of anything I could do about that. I was aware of an implementation like you had described, but was told it didn't really 'work.' Ah well I've decided to sell off my Sun Blade 2000, put it on craigslist. Too many bills and I need a good intel machine for play

        • Re: (Score:1)

          by jaiger ( 166690 )
          My systems are Ultra10s and a Netra T1 100. I've installed Debian sarge and etch on them. I mostly use them as server systems really so Flash and other desktop features are not so useful to me. Kaffe does have compatibility limitations but you can get it to work for a lot of applications. I was impressed that Tomcat and my webapps mostly work. I keep my needs modest at Java 1.4.x code and it largely works.

          The most annoying feature is I found Kaffe on sparc linux to be much slower than my sun jdk on int
          • That's funny because IME Sol 10 on the hardware you're mentioning crawls. I think it's more the JDK that's slow than the OS.

            You're not interested in a blade 2000 are ya? lol

Slashdot Top Deals

It isn't easy being the parent of a six-year-old. However, it's a pretty small price to pay for having somebody around the house who understands computers.

Close