Forgot your password?
typodupeerror
Java Apache

Apache Harmony Moves To Apache Attic 120

Posted by Soulskill
from the giving-up-the-ghost dept.
think_nix writes "After the resignation of Apache from the Java SE/EE Executive Committee, the time has now come for Harmony to be added to the Apache Attic. Harmony was 'the project to produce an open source cleanroom implementation of Java.' An open vote was taken within the Project Management Committee, which resulted in a 20-2 majority to discontinue development."
This discussion has been archived. No new comments can be posted.

Apache Harmony Moves To Apache Attic

Comments Filter:
  • It lives on (Score:1, Flamebait)

    by ttong (2459466)

    It lives on in the form of Android, more specifically the java. parts of the Dalvik core library.

  • It is a tragic day. First I find out that someone was working on an open implementation of Java, then I find out that it's cancelled? Oh well. Here's to hoping that the universities speed up the inevitable transition to Python so we can see the glut of Java programmers disappear.
    • Re:Harmony what now? (Score:5, Informative)

      by sitkill (893183) on Friday November 04, 2011 @01:59PM (#37950872)
      uh, not sure if you even read the original article but..

      There IS an open implementation of java, the openJDK [java.net], which is why this is being shut down (IBM who was the main contributor to Harmony has moved it's resources to openJDK).

      But I guess it's more in line with slashdots javahate if we ignore those facts.
    • Universities wouldn't switch for a non-typed language as a main platform. C++/Java will continue to be entreched, with a little C# sprinkled on. Also they don't follow flavor of the month.

      By the way they embraced Java before it was opensourced, so I don't think they'll make a boycott about Oracle dicking around with its software patents.

      • by meustrus (1588597)

        I only suggested Python because I heard some universities have already started using it for introductory courses. Of course I'm not suggesting it be the primary platform but there has been success in teaching basic programming principles without dealing with Java-specific issues immediately. I would say the progression ought to be:

        1. Python (or any other simple, untyped language)
        2. C++ (so the students can be all "whoa that's way harder")
        3. Lisp (hey wut this language is nothing like the rest)
        4. C# or Java (combined w
        • Lisp (hey wut this language is nothing like the rest)

          LISP?! We already have enough trouble attracting quality CS students - are you trying to drive everyone to becoming Philosophy majors?

        • Re:Harmony what now? (Score:4, Interesting)

          by Lulu of the Lotus-Ea (3441) <mertz@gnosis.cx> on Friday November 04, 2011 @03:00PM (#37951512) Homepage

          Aaaghh!

          Python is, of course, a STRONGLY typed language (not untyped). It is also DYNAMICALLY typed rather than STATICALLY. But these issues are orthogonal, and languages exist in every quadrant of the type system grid implied.

          • Exactly such as C being weakly, but statically typed.

          • by meustrus (1588597)
            OK, so I've never actually even seen Python code. I've only heard of how amazing it must be and I assumed it was weakly typed because it's a scripting language.
        • What about PROLOG instead of LISP? Not quite as abusive of your bracket-keys but equally abusive of your "WTF?" when seeing the results :)
          • In my university we studied Prolog and SML. Now they're changing SML to Erlang. You can also take up Constraint logic programming for 3 credit. (Europe)

        • by vbosch (1552029)

          At my university UPV ( Polytechnic University of Valencia) we studied:

          Pascal - 1st year

          C with some assembler - 2nd year

          Mathematica - 2nd year

          Octave- 2nd year

          C++ - 3rd year

          Java - 4th year

          Prolog - 4th year

          Lisp - 4th year

          Haskell - 5th year (depending on your specialization)

          Shell script was just basic as we did most our coding on the Linux distribution favored that year by the sysadmins.

  • by bmo (77928)

    The "write once run everywhere" model of Java was just a poor imitation of Forth's anyway.

    --
    BMO

  • OpenJDK? (Score:4, Insightful)

    by bigsexyjoe (581721) on Friday November 04, 2011 @02:24PM (#37951088)
    I'm not completely up to speed on these issues. But is there anything about OpenJDK that people are unhappy with? Is it not open enough in some manner?
    • Re:OpenJDK? (Score:5, Insightful)

      by Jonner (189691) on Friday November 04, 2011 @02:53PM (#37951396)

      It seems that IBM started Harmony to produce a complete, permissively licensed implementation of the Java specifications, while OpenJDK is Copyleft. A year ago, IBM shifted their support to OpenJDK, so there's no longer a major sponsor for Harmony. That, in addition to the fact that Apache is no longer participating in the Java Community Process seems to be the reason there's little development going on in Harmony.

      The reason that Apache resigned from the Java Community Process Executive Committee is that the Java specifications are completely controlled by Oracle, which doesn't allow their compliance test tools to be used freely. Harmony will remain in a similar relationship with Java that Mono has to DotNet rather than being a first class implementation. The significant difference between Java and DotNet seems to be that Sun/Oracle have released most (all?) of the official implementation as Free Software, but the specifications that software implement are still proprietary.

      The main beneficiary of Harmony seems to be Google, which is using some code from the project as part of Android, though Android was never intended to be a complete Java implementation. They've made it clear that they want to use only permissively-licensed code so that they can release it as proprietary software whenever they want. That's exactly what they did with Android 3. While I'm glad IBM is now supporting a Copyleft implementation rather than a permissively-licensed implementation, a mature programming language system needs alternative implementations to keep the spec-writers honest. Oracle's handling of the specification process continues to prevent Java's use as a completely open system.

      • Thanks for the info.

        Does copyleft mean that if you use something what you release has to also be open source and free? Does that mean that Oracle is actually suing Google to be more open in their implementation rather than shut it down?

        Do we know why they won't open source the compliance tools? Does that effectively prevent other implementations of Java from existing? Do we know why Oracle wants that?

        • by vadim_t (324782)

          Does copyleft mean that if you use something what you release has to also be open source and free?

          Not if "what you release" is something that runs on the JVM.

          The copyleft part only applies to you if "what you release" is the modified JVM, or something that reuses parts of its code. The license specifically contains a linking exception, so linking to the JVM (eg, using the class library that comes with it) doesn't fall under the copyleft license.

        • There is a field of use restriction on the TCK tools preventing other certified Java implementations from being used in a mobile environment. Ostensibly this would be done because there is money to be made by charging for mobile implementations.

        • Does that mean that Oracle is actually suing Google to be more open in their implementation rather than shut it down?

          No, that's not it at all. It has nothing to do with openness and everything to do with ensuring that there can never be a JVM that competes with the one Oracle bought from Sun.

          Do we know why they won't open source the compliance tools? Does that effectively prevent other implementations of Java from existing? Do we know why Oracle wants that?

          Yes, yes, and yes. Your second question is the answ

          • No, that's not it at all. It has nothing to do with openness and everything to do with ensuring that there can never be a JVM that competes with the one Oracle bought from Sun.

            You may have your tin foil hat on too tight. The suit has more to do with Google trying to circumvent Sun's license by making a "clean room" implementation of JVM for mobile use. Sun was never happy about Google doing an end run on their sale of embedded Java.

            That said, nothing prevents us from wondering "what if" Google had based th

            • You may have your tin foil hat on too tight. The suit has more to do with Google trying to circumvent Sun's license by making a "clean room" implementation of JVM for mobile use. Sun was never happy about Google doing an end run on their sale of embedded Java.

              That's pretty much exactly what I said, though. They're using patents to prevent competition.

              And it's kind of hard to say Sun was "never happy" when they published press releases congratulating Google on releasing Android and promised Android support

              • It would suck. The JVM is a lumbering undead monstrosity. I say this as someone who's livelihood depends on it on a daily basis and (unfortunately) spends about 70% of his time writing Java code, now. It's just bad.

                JVM is not Java. Scala, Jython, and JRuby run within the JVM and JVM7 has better support for these dynamic languages.

                And I'm listing that as a negative, not a positive. So much of the legacy Java 1.1 stuff is just BAD and completely broken, but it stays around and causes problems for everyone. S

                • JVM is not Java. Scala, Jython, and JRuby run within the JVM and JVM7 has better support for these dynamic languages.

                  Are you suggesting the JVM should be used to run Scala, Jython, and JRuby on a mobile platform?

                  That's a feature that separates a mature language from a flavor of the month language. With your examples, I doubt the authenticity of your points.

                  Refusing to fix broken, braindead APIs sounds more like a mark of senility than maturity to me. :)

                  • Are you suggesting the JVM should be used to run Scala, Jython, and JRuby on a mobile platform?

                    Why not? I'm seeing the same thing done with Dalvik. JVM is ahead of the game when it comes to JIT and dynamic language support.

                    I didn't mean to sound too harsh on the "authenticity" comment.

        • by peppepz (1311345)

          Does copyleft mean that if you use something what you release has to also be open source and free?

          In this particular case no, because OpenJDK is released under the GPLv2 license but with the GNU Classpath Exception:

          As such, it can be used to run, create and distribute a large class of applications and applets. When GNU Classpath is used unmodified as the core class library for a virtual machine, compiler for the java languge, or for a program written in the java programming language it does not affect the licensing for distributing those programs directly.

          Do we know why they won't open source the compliance tools?

          I think that certification was one way for Sun's

        • Does copyleft mean that if you use something what you release has to also be open source and free?

          Yes, specifically OpenJDK is under the GPLv2.

          Does that mean that Oracle is actually suing Google to be more open in their implementation rather than shut it down?

          While I wouldn't put it that way, technically yes.

          However, the real reason is Oracle wants Google to pay royalties to them like all phones that use JavaME do.

          Do we know why they won't open source the compliance tools?

          Because they make money from it?

          Does that effectively prevent other implementations of Java from existing?

          No, other implementations exist, such as IBM's version [ibm.com]. JRockIt was another, but BEA was bought by Oracle before Sun was.

          Do we know why Oracle wants that?

          See: Sun Microsystems, Inc. v. Microsoft Corporation

          In other words, Sun (and their successor in interest Oracle) want to control the platform to prevent fr

          • by Jonner (189691)

            Does that mean that Oracle is actually suing Google to be more open in their implementation rather than shut it down?

            While I wouldn't put it that way, technically yes.

            However, the real reason is Oracle wants Google to pay royalties to them like all phones that use JavaME do.

            Oracle is not suing Google to make anything more open. They are suing Google for patent infringement in code Google has already released as Free and Open Source software. Oracle is attempting to use their patents to extort money out of Google. I'm sure they'll do the same to anyone else with enough cash for them to think it's worth it.

        • by Jonner (189691)

          Thanks for the info.

          Does copyleft mean that if you use something what you release has to also be open source and free? Does that mean that Oracle is actually suing Google to be more open in their implementation rather than shut it down?

          Do we know why they won't open source the compliance tools? Does that effectively prevent other implementations of Java from existing? Do we know why Oracle wants that?

          I forget that not everyone here is familiar with Free Software and Open Source licenses. Copyleft [gnu.org] is "a general method for making a program (or other work) free, and requiring all modified and extended versions of the program to be free as well." Since most of OpenJDK is under the GNU General Public License [gnu.org], the most common and best known Copyleft license, if you use any OpenJDK code as part of something which you release, it must also be released under the GPL. If you add some features to OpenJDK and relea

      • To add to the above, I believe the specific disagreements were in regards to licensing clauses of the TCK preventing the use of certified Java implementations in a mobile environment.

        The particular licensing restrictions were not in compliance with the Java Community Process rules, but that unfortunately has not resulted in the license being changed. The entire fight seems to be directed specifically at Google, and Harmony is a casualty of the battle Oracle is waging (possibly at the behest of Steve Jobs w

Nothing happens.

Working...