Proposed schedule change for JDK 9

[Jeremy Manson]

What constitutes a reasonable JEP in this time frame?  Is it basically
defined as one that doesn't add undue risk to the overall release?

I have no particular concerns about the timeline for Jigsaw features, but I
do have concerns about the velocity of OpenJDK development.  If larger JEPs
are not going to be able to target Java 9, and the releases are being put
on hold for 6 months, will this timeline shift have an effect on when the
Mercurial repos are made available for JDK10?  Will larger JEPs be able to
target Java 10 earlier?

We have a couple of JEPs that we've been meaning to write, but we felt like
we missed the Java 9 deadline, so the answer to this will inform our 2016
plans.

Also, will this affect the flexibility in backporting new features to
JDK8?  We've been seeing substantial improvements from JEP-246 (targeting
Hotspot improvements for AES-GCM), to the extent where, for many customers,
AES-GCM went from being unusable to being usable.  "Wait until Java 9"
takes on a very different meaning after this message.

(Perhaps also of academic interest to the community: the more that JDK9
diverges from JDK8, the harder it will be for us either to maintain our
patchsets against both (or to forward port the changes when JDK9 is
ready).  From that perspective, this sort of delay somewhat works against
us.  I imagine this is true of other organizations that maintain their own
forks.  I'm not crazy enough to expect this to affect the decision: I'm
just saying it because I think it might be of academic interest to the
readership of this list.)

Jeremy

On Tue, Dec 1, 2015 at 9:08 AM, <mark.reinhold at oracle.com> wrote:

> The key feature of Java 9 is Project Jigsaw [1], which will introduce a
> standard module system and use that system to modularize both the Java
> SE Platform and the JDK.  This large project consists of a JSR for the
> module system plus five JEPs, for the implementation of the module
> system and for other changes specific to the JDK.
>
> We've made good progress on Jigsaw over the last eighteen months: We
> reorganized the source code into modules in August 2014 (JEP 201 [2]),
> restructured run-time images to support modules in December 2014 (JEP
> 220 [3]), began discussions in the JSR 376 EG last February [4], and
> published a design overview, draft specification, and EA builds in
> September [5].  More recently we presented an integrated series of
> talks on Jigsaw at JavaOne 2015 and Devoxx BE 2015 [6] which were very
> well-attended and motivated many developers to download the EA builds,
> try them out, and send feedback and suggestions.
>
> In the current JDK 9 schedule [7] the Feature Complete milestone is
> set for 10 December, less than two weeks from today, but Jigsaw needs
> more time.  The JSR 376 EG has not yet published an Early Draft Review
> specification, the volume of interest and the high quality of the
> feedback received over the last two months suggests that there will
> be much more to come, and we want to ensure that the maintainers of
> the essential build tools and IDEs have adequate time to design and
> implement good support for modular development.
>
> For these reasons I hereby propose a six-month extension of the JDK 9
> schedule, moving the Feature Complete (FC) milestone to 25 May 2016, the
> General Availability (GA) milestone to 23 March 2017, and adjusting the
> interim milestones accordingly.
>
> As with previous schedule changes, the intent here is not to open the
> gates to a flood of new features unrelated to Jigsaw, nor to permit the
> scope of existing features to grow without bound.  It would be best to
> use the additional time to stabilize, polish, and fine-tune the features
> that we already have rather than add a bunch of new ones.  The later FC
> milestone does apply to all features, however, so reasonable proposals to
> target additional JEPs to JDK 9 will be considered so long as they do not
> add undue risk to the overall release.
>
> Comments on this proposal from JDK 9 Committers are welcome, as are
> reasoned objections.  If no such objections are raised by 18:00 UTC next
> Tuesday, 8 December, or if they're raised and satisfactorily answered,
> then per the JEP 2.0 process proposal [8] this will be adopted as the
> new schedule for JDK 9.
>
> - Mark
>
>
> [1] http://openjdk.java.net/projects/jigsaw/
> [2] http://openjdk.java.net/jeps/201
> [3] http://openjdk.java.net/jeps/220
> [4] http://openjdk.java.net/projects/jigsaw/spec/
> [5]
> http://mail.openjdk.java.net/pipermail/jigsaw-dev/2015-September/004480.html
> [6] http://openjdk.java.net/projects/jigsaw/talks
> [7] http://openjdk.java.net/projects/jdk9/
> [8] http://cr.openjdk.java.net/~mr/jep/jep-2.0-02.html
>