jump to navigation

EE6 has been approved, yet… December 3, 2009

Posted by Sacha in English, IT, JBoss.

It is with GREAT pleasure that I’ve learned that the JCP EC had approved the final ballot for Weld and EE6!

This really finishes the work started with EE5 and clearly positions EE6 as the most powerful and yet the easiest to use Java runtime environment, clearly ahead of Spring and its XML logorrhea. Furthermore, the definition of a powerful Web profile without the hassle of some of the most legacy EE specs (IIOP anybody?) will give fresh air to the too many IT teams building (and attempting to maintain) their own Tomcat on steroids.

Yet, a spec with no implementation is hardly usable (for my own sanity, please don’t mention the EE5/JBoss5 drama), and so I’m very impressed that JBoss already released JBoss 6.0M1 which contains some of the most important EE6 bits. Jason’s blog entry here, Dimitris’ here. Congratulations guys, the move away from 5.2 was a smart decision.

Yet, when closely looking at the Weld and EE6 votes, only one company didn’t vote for both JSRs: SpringSource/VMware. They didn’t politely cast an “Abstain” and explain the reason why in the comments, no, they simply didn’t cast a vote at all. How to interpret that non-move? Well, maybe they “forgot” – this happened to me at JBoss for a couple of “less critical” votes – and so I went to look at SpringSource’s official news and PR page, SpringSource’s official blog and even Spring “.org” page: nada, nothing, no official position (and Rod Johnson hasn’t blogged in more than 3 months – so no help from there either).

I don’t think this silence is clever as this will make people speculate that either i) SpringSource is still debating internally what their official stance should be or, worse, ii) that they will retain from giving an official stance for as long as possible to keep all options open and possibly change their opinion down the road as they see fit. Or maybe SpringSource didn’t cast a vote because VMWare refused to sign JCP’s JSPA agreement for some reasons related to their IP. Who knows?!? (disclaimer: I don’t)

In order to keep the worse rumors at bay, I think that SpringSource should clearly state ASAP i) why they didn’t vote and ii) what their official stance on EE6/Weld/etc. is. and iii) whether VMWare/SpringSource is still part of the JCP.





1. Mark Little - December 4, 2009
2. Tony - December 6, 2009
Sacha - December 6, 2009

Tony, this article doesn’t explain why SS didn’t vote. If they din’t endorse this spec, a NO vote would have been more appropriate, don’t you think?

3. Cay Horstmann - December 6, 2009

JBoss did a great job with Weld, but if you want to see EE6 in action, give GlassFish v3 a try. It contains Weld from JBoss and all the other EE6 goodness from other sources (Sun, Apache, EclipseLink). It’s very fast and easy to use.

Sacha - December 7, 2009


Possibly, but my DNA prevents me from going there 😉 And you’ll admit those are not the best days for the GF team.



4. Blogging Techstacks - December 7, 2009

The Fight Continues!! – JBoss vs. SpringSource…

While trolling through my feed subscriptions, I ran across this post entitled: “EE6 Has Been Approved, Yet…” It’s been a while since I’ve read anything in “print” regarding my imaginary Sun Microsystems vs. JBoss vs. SpringSource war so regardles…

Sacha - December 7, 2009

And yet… you don’t explain why they didnt’ simply vote NO 🙂 and this is not just for JSR-299, but all recent votes.

5. Chris - December 7, 2009

Heh. I thought it was pretty funny post “as-is”.

Obviously, I’m blowing all of this out of proportion for fun but at the end of the day, I’m just a customer of both redhat and springsource. I have no insider “insight” but one reason seems fairly obvious in hindsight: The ASF votes “no”: No press or blog coverage. SpringSource doesn’t vote: Even the JBoss blogs are writing about them. Perhaps this is an attempt at “Succès de scandale”?

Sacha - December 7, 2009

Hello Chris,

I don’t think so. The Apache Foundation has been voting NOT to all Sun-led JSR for more than 2 years no! Consequently, it comes at no surprise they have voted NO to EE6 since it is a spec led by SUN. However, they voted YES to JSR-299. In the case of SS, they haven’t voted to any of those specs. And I just canot see this as a mere oversight.



6. ChrisW - December 7, 2009

I don’t think there is any need to make a fuss out of SpringSource’s nonparticipation in this vote. This is a Java EE standard. SpringSource is the JEE *alternative*: its whole purpose in life is to be the not-EE product. For SpringSource to vote on a Java EE standard is like Microsoft voting on any Java JSR: it’s in poor taste.

Sacha - December 7, 2009

Then they should not have participated in the EE6 expert group in the first place. As for the EE6 vote, this is really a responsibility of EC members and SS is an EC member.

ChrisW - December 7, 2009

I think SS got what they wanted, which is to harmonize the @Inject annotations with JSR 330. JSR 330 standardizes non-EE DI implementations like Spring and Guice, so it makes sense from SpringSource’s perspective to harmonize the DI annotations between 299 and 330. That makes their participation meaningful: to ensure EE and non-EE source compatibility for DI. Once that is accomplished SpringSource’s interest in JSR 299 ends.

7. Mark Little - December 7, 2009

The harmonization between 330 and 299 was driven by Google on the 330 side, with Red Hat, Oracle and Sun on the 299 side. Nothing from SpringSource at all.

8. Mark Little - December 8, 2009
9. rainwebs - December 8, 2009

When do we ignore something? If it’s not important to us. I can’t remember that newer JEE specifications were important to SpringSource in the past. We all know why Spring was invented ;-).

Nobody is able to ignore the Spring ecosystem these days. But, there are a lot who already ignore the JCP. The JCP was a brilliant idea at the beginning. But, JCP lost the lead. Rod was one of the first who recognized this and delivered an enterprise-ready alternative. I’m quite happy that I have this alternative these days.

The market share of Spring on Tomcat deployments already foreshadows the future of heavy-weight application servers. The continuous implementation of new JCP standards will not stop this ;-).

Sacha - December 9, 2009

rainwebs, if EE and its related specs didn’t exist, Spring wouldn’t exist. Remember that Spring is a wrapper around EE API and that portability is provided by those APIs. Spring has never been a full fledge runtime in itself.

As for the JCP, SS are member of its Executive Committee. If they thought the JCP was so bad, they would have already left it I’d imagine.

10. Tony - December 13, 2009

Agreed Sacha. I thought the article might give more insight into why they didn’t vote.

11. GAE+VMW: co-petition in the cloud-era « Sacha’s Weblog - May 26, 2010

[…] VMSource gets to amplify their desire to see Spring as the defacto Java programming model of the cloud (remember, they seem to fully ignore EE6!), […]

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: