Open Source Java SE: Who gives a fig ?

Who cares ? Five days after our announcement on Monday, I felt it was time to digest the reactions around the Java world to the news.

Missed opportunity for word play

I was all geared up for a slew of press headlines marrying 'Sun' with some combination of 'rise', 'set', 'shine', 'burn', 'spot' or 'screen'. And, you know, a graphic of a sun cresting over a choppy sea full of joyous penguins or something.

But instead we've had a real treat of a large number of articles that mostly accurately represent the plan we announced on Monday. For example: CNET, Internetnews, eWeek, TecnNewsWorld, DevX, all with sober headlines.

(OK, one of them couldn't resist.)

You did ask...

Much of the complexity in Sun making this move is for us to do so in a responsible manner with regard to to the other vendors who have embraced Java technology. I imagine BEA will be pleased at this latest news given their past beseechings, and current 'blended strategy' on Open Source (chirp...). A hasty reaction from some folks from IBM appears to have been mostly made facing upwind, except for some free product feedback they got.

More helpfully, and critical to our getting this right, have been some very constructive assessments and advice from people like Geir, Dalibor and Tom who really know this stuff because they have or are working on open source implementations of Java SE already.

OK, but just don't break it for us

For me, the most interesting reactions have been those from Java developers. Perhaps in the long anticipation of this completion of opening up the development of the Java platform codebases, much of the heat and fire has been dissapated. So I have to say that the volume of debate in the fora that I have been following has not eclipsed other burning issues of the day. But the reactions appear to have been largely focussed on the mechanics of rolling out the program, rather than on its merits. Though of course there are some colorful exceptions. Debates on the central issue of the choice of licence appear as a popular discussion point, though there does not appear to be a consensus. My own suspicion as to why being that the consequences of the choice are complex to divine. Law school anyone ? There have been a couple of sinister theories that I probably won't lose any sleep about, and some misinformation about Java EE [Update: That got corrected since I posted this]. And there I was, thinking everyone knew that our implementation is already open source. But I was happy to see that one of the potential benefits that Java SE will go where no Java SE has gone before, is being discussed by others too.  Of course, what I suspect many developers feel on this topic is indifference to mild concern: that they do not care how Java is made, but just that it's there and that it works. No surprise then that there are some worries that it might not.

We do well to keep them at the front of our minds.


Too little, too late. The only real beneficiary will be IBM (remember J9 compliance? ) While Java development is still "active", inroads by lightweight active scripting languages, not to mention .Net, Java will be the COBOL of the 21st century. SOA diminishes the value prop of Java or any heavyweight language, there are easier, more efficient languages for the development of services. Not to forget EE, JavaEE will continue to collapse under its own weight (why not Spring instead of EJB 3 or 5 or whatever?) If you want a real language platform, maybe you (Sun) can dust off Self (a real innovation in language and platform especially when compared to Java, much of which found its way into Squeak) and make it freely available to the open source masses (like me!). Open source Self is more interesting to me, you can keep Java for all I care! What is the Self license? Java open source, the interest has diminished. Much ado about nothing. Irrelevance.

Posted by Richard on August 23, 2006 at 06:35 AM PDT #

"Java will be the COBOL of the 21st century" Cool I am glad to hear this ! Let's invest more in Java then ...

I mean, if Java EOD developement is 2099, this gives me still more thant 90 years of sucessfull business to come !

There is no doubt that soon or later a better solution than Java will come, no technology in computer science is ethernal but what I am sure is that the technology that will supercede Java has not already invented.

The problem about .net, is that there is no major advantages (appart from the church reason) because all the initial revolution that were part of the project COOL (like business oriented language) were dropped because the upcomming language (read C#) was aready way a too big pill to swallow for a VB guy ... good market decision, bad strategic decision IMHO (decision was to stop Java devs and fork the platform so that coontrol will be in seatle suburbs only), but history will judge my thoughts.

About Java RI opensource license, I think a dual license model is the killer solution, if Sun choose to brin SE RI under GPL+ComercialLicense (or GPL+BinaryLicense depending on their orientation), then I think you can say "bye bye" to the some commercial contenders. And only a realy innovative next generation of platform will be able to push Java to retirement ... Is technology Armagaedon comming from Sun's ?

Posted by Testman on August 31, 2006 at 02:50 AM PDT #

Post a Comment:
Comments are closed for this entry.



« March 2015