Scott is Wrong About Open Sourcing Java, But we Knew That Going Into This
While it is no shocker, Sun CEO Scott McNealy has made his decision very public, Java will not be open sourced. If you follow the link (above) you can see the report on his speech on Government Computer News. I haven't found the text of the whole speech online, so I'll only talk about what was reported there and elsewhere.
While ESR's comparison to Python and Perl just shows he doesn't understand Java (.net is the obvious comparison in my mind, especially with the very cool, very open, mono toolset), he has a point that Sun not open sourcing Java is detrimental to the future of the language. This doesn't mean that Scott is wrong when he says: "We're trying to understand what problem does it solve that is not already solved". The two concepts aren't mutually exclusive. Scott has not understood what problems open sourcing solves for Java. But I don't want this to be an accusatory article towards sun. It is their language, and it is theirs to decide what to do with and I respect that.
It is clear Scott doesn't agree with Eric that open sourcing a language is vital to its health. I think that it is and that there are benefits of the open sourcing of a language that people recognize as being served by the JCP, but. I suppose that I could be declared an open source eventualist (or realist), and I do believe that , in time, Java and many currently closed programs will be open sourced, whether through wisdom, fiat or in desperation.
It is somewhat puzzling really: Sun did something very good and very important by open sourcing Star Office. I don't see why the wisdom applied to that decision isn't being applied here.
Anyhow, should Java be abandoned by the open source enthusiast? Obviously (read my past articles), I feel that the answer that is best for Linux for now is no. I do hope that Sun does see the light, and one day decides to lead a free Java through its place as the thought leader, not as the IP owner of a mostly open Java.
Java is still very important to Linux. It is important the Linux as a platform enjoys parity with Solaris and Windows and AIX. Java development and deployment must be as good on Linux as it is on these other operating systems for Linux to succeed in the enterprise computing environment. But I've said all this before. Am I off base? What are your thoughts?