In the later BOF session, Mark elaborated a bit. He also elaborates in
that blog entry. Specifically:

1. a single CPU core is not going to become any faster. That's right;
a single thread is never going to run any faster in the future. We've
reached the end of the line as far as clock speed and on-chip
parallelization is concerned. From here on out, it's cores. In a few
years, you'll likely buy a computer with ~100 cores in it. Various
tactics have been tried to make working with that many cores easier in
the JVM, and they all failed. STM has been tried, and the experience
has been lukewarm at best. synchronized works, but sort of sucks in
the general area where fork/join and ParallelArray excel. Therefore,
PA and fork/join are of crucial importance. They both could do with
closures, particularly PA.

2. JDK7 has been delayed from feb 2010 to at least sep 2010, due to
reasons he didn't want to elaborate on, but it's not because of
language features. So, what to do in the extra time given? Well,
closures is an option.

3. For the same reason (delayed), as well as oracle/sun merger likely
slowing down JDK8 even more, closures suddenly turned into a now-or-
virtually-never proposal, with 'never' defined as: JDK8, which is a
loong way away (personal estimate: no earlier than the tail end of
2012). This wasn't the case before.

4. It's been fermenting at sun for a while, with various key players
all liking closures but not sure if it is a good fit for java, which
already has tons of libraries that do things in a different fashion,
and with language constructs already in place that don't necessarily
play well with closures. However, recently, these key players have all
become convinced java is better with closures than without them. What
caused this process is unclear, but I get the sense from Mark Reinhold
that he got convinced due to the reasons mentioned above, and pushed
for it.


Just trying to recollect what was said at devoxx.

On Nov 25, 9:11 am, etzel <[email protected]> wrote:
> From this blog post by Mark Reinhold (http://blogs.sun.com/mr/entry/
> closures) it seems that the primary reason to introduce closures in
> Java 7 is to remove boilerplate code from the fork-join framework
> (also in Java 7). Isn't this quite selfish? I mean, I know two or
> three frameworks that would benefit from closures as well.

--

You received this message because you are subscribed to the Google Groups "The 
Java Posse" group.
To post to this group, send email to [email protected].
To unsubscribe from this group, send email to 
[email protected].
For more options, visit this group at 
http://groups.google.com/group/javaposse?hl=en.


Reply via email to