Thursday, May 31, 2012

Who is preventing the release of Java 1.7?

I recently attended a talk by a Sun engineer, Charlie Hunt, regarding performance. The talk was interesting enough but one question was regarding the release date of 1.7.

He said it's delayed as there are parties who are refusing to sign off JSRs they own and thus preventing the 1.7 release. It apparently has something to do with the cost of determining your Sun compliance.

What is the full story or a pointer in the right direction?

What triggered my question was the amazing long release notes for 6u18 .

Source: Tips4all


  1. Although Sun open-sourced Java SE, they did not open-source the test suites required to claim conformance. This caused a conflict with the Apache foundation. Many members of the JCP executive committees, whose votes are required for JSRs to become final, support Apache in this conflict. Consequently, many JCP executive committees have agreed not to approve any of Sun's JSRs until the license terms are "fixed". Sun has avoided filing JSRs that they know will be shot down. Without an umbrella JSR for Java SE 7, there is no Java SE 7.

    So depending on whose side you take, the answer to "Who is preventing the release of Java 1.7" is either Sun Microsystems or the Apache Foundation.

    Incidentally, Oracle was a strong supporter of the Apache foundation, and Oracle has now purchased Sun Microsystems. That may provide hope for breaking the logjam.

    Stephen Colebourne has written about these issues:

    Shedding new light on No Java SE 7 JSR
    The JCP doesn't exist!
    No Java SE 7 - The Oracle perspective

  2. The basis of the problem is the release process itself. The Java Community Process was meant to bring a more democratic process to JDK development, but it's turned into an awful, bureaucratic mess, with too many people having too much of a say on what happens. It doesn't take too many of those people to dig their heels in, and the whole thing grinds to a halt.

    The current expectation for Java 7 release is Q4 2010, but I'd fully expect that to slip further.

    I follow Alex Miller's blog (RSS here), where he publishes links to all blog and news items referring to the progress of Java 7. It's painfully slow, but things are moving along.

  3. I believe the real show stopper was the decision to open source Java 6. That has apparently taken a LOT of effort.

  4. Can't confirm it, but this might make sense regarding the Java release schedule.