BEA Systems, Inc. announced it has acquired the object persistence company, SolarMetric, Inc., today in a move that will give it an early lead in the race towards Enterprise JavaBeans(TM) (EJB) 3.0 and Java(TM) EE 5 compatibility.

Solarmetric is the vendor of the popular Kodo JDO implementation.

See the full release on Yahoo! news: BEA Systems Acquires SolarMetric.

Kodo is a top quality product, so kudos to Patrick Linskey and the entire Solarmetric team.

Dion

http://ajaxian.com

Now that all the BEA people have congratulated themselves. I thought JDO was basically stillborn along with EJB 3.0, but what do I know. I will still use Hibernate and Spring. If I need remote access I will use a front controller with a session bean or web services.

I really don't understand BEA's strategy. They still think people buy software, not solutions. IBM practically gives away software and sells professional services. JBoss's approach isn't very different, to me BEA is headed in the wrong direction and is relegating itself to eventual irrelevancy. Art Technology II?

Now that all the BEA people have congratulated themselves. I thought JDO was basically stillborn along with EJB 3.0, but what do I know.

Indeed, you are mistaken. To quote BEA on the vote on the final draft of the JDO 2.0 specification: "After further review, BEA has concluded that there is a vibrant JDO community". I have no doubt that many JDO users will migrate to EJB 3.0, however, there will (I believe) still be a niche for JDO. This is because JDO is a more general persistence mechanism than EJB 3.0, as it includes the useful ability to handle non-relational stores, and does not require the latest Java implementations to run. My view is that both APIs are likely to remain useful for a long time to come, and products like Solarmetric's Kodo that supports both are important.

Now that all the BEA people have congratulated themselves. I thought JDO was basically stillborn along with EJB 3.0, but what do I know.

Indeed, you are mistaken. To quote BEA on the vote on the final draft of the JDO 2.0 specification: "After further review, BEA has concluded that there is a vibrant JDO community". I have no doubt that many JDO users will migrate to EJB 3.0, however, there will (I believe) still be a niche for JDO. This is because JDO is a more general persistence mechanism than EJB 3.0, as it includes the useful ability to handle non-relational stores, and does not require the latest Java implementations to run. My view is that both APIs are likely to remain useful for a long time to come, and products like Solarmetric's Kodo that supports both are important.

Steve hits the nail on the head here - Kodo supports *both* JDO and EJB 3 so it provides excellent flexibility for its users. I can't see many people complaining about being able to choose between 2 standards :)

Rob

BEA Systems, Inc. announced it has acquired the object persistence company, SolarMetric, Inc.

A very good strategic move .. Solarmetric has a full EJB3 implementation and some great engineers.

Peace,

Cameron Purdy
Tangosol Coherence: The Java Data Grid

BEA Systems, Inc. announced it has acquired the object persistence company, SolarMetric, Inc., today in a move that will give it an early lead in the race towards Enterprise JavaBeans(TM) (EJB) 3.0 and Java(TM) EE 5 compatibility.Solarmetric is the vendor of the popular Kodo JDO implementation.See the full release on Yahoo! news: BEA Systems Acquires SolarMetric.

Correction in announcement.

"BEA announced it has acquired SolarMetric...that will give it a *self-proclaimed* early lead in the race towards EJB 3.0"

Bill

It's about time ;-)

Congratulations Patrick !!

Congratulations to Patrick, Neelan and the rest of the SolarMetric guys!

Even though "early lead" is obviously a marketing insertion, this is good news for EJB 3 and EJB 3 users. Now we can look forward to an EJB 3.0 implementation from BEA sooner rather than later.

BEA has never had their own decent persistence offering since they tried and failed to develop their own alternative for all of the folks that are using TopLink with WLS ;-)

Congratulations Patrick and all at Solarmetric!

It's great news for Solarmetric, for BEA, and more broadly for EJB3 and Java EE 5 :-)

Patrick and all - congratulations! It's a great year for deals under $100M!

geir

Allright Patrick! Woo hoo! You're no geek, no, you're the man! Patrick! Patrick! Patrick! Patrick! Patrick! Patrick! Patrick! Patrick! Patrick is the man!

Congratulations to Patrick and his team. You have worked relentlesly these last years and deserve all recognition you can get.

Now we have Oracle, JBoss and BEA with EJB3 Persistence solutions - can't wait for the spec to go final so we can start to put them to some serious use.

can't wait for the spec to go final so we can start to put them to some serious use.

... because you clearly can't use them seriously now

Well done indeed - Patrick and the guys are certainly deserving. A great team of engineers, a great product and great party hosts.

Persistence 'seems easy' until you sit at a table with Gavin and Patrick when their conversation disappears down the rabbit hole of complexity.

However, I am very worried about what this means for the future of the Tangsolarmetric JavaOne parties. I notice this got left out of the official statement. I'll assume this was an oversight and will be rapidly corrected? :)

m
--
ATLASSIAN - http://www.atlassian.com
"Still not part of the under $100m club"

Mike :

I am very worried about what this means for the future of the Tangsolarmetric JavaOne parties.

One thing I discovered about going to a large company is that the resources available are on an order of magnitude greater than what you have at a small company, so I'm really bullish about the party next year. I do wonder if they'll have to name a drink "AquaLogic" or something to get marketing to pay for it....

geir

This is excellent news and it makes me happy to see another example of good people being rewarded for their efforts.

Here we had a bunch of talented engineers who dedicated themselves to supporting object persistence in the Java platform (via JDO), despite all the politics that stould against JDO in favour of persistent components (EJB).

Your courage and forsight both had an impact on the community and has now come back to reward you. Congrats!!!

Each participant in this community can have impact in ways we can't imagine. I'm sure that SolarMetric's success provided Sun & the JCP EC with real evidence that there are real paying customers out there who want an object persistence standard, and it's not just a vocal minority who was making such a fuss about JDO. SolarMetric may have been an important force in helping "the powers that be" finally decide to promote an ORM standard in Java via EJB 3.

I hope the yearly Java One parties will still be part of the budget now that SolarMetric is at BEA. :)

Floyd

This is excellent news and it makes me happy to see another example of good people being rewarded for their efforts.Here we had a bunch of talented engineers who dedicated themselves to supporting object persistence in the Java platform (via JDO), despite all the politics that stould against JDO in favour of persistent components (EJB).Your courage and forsight both had an impact on the community and has now come back to reward you. Congrats!!!

Indeed. We need future support for both JDO 2.0 and EJB 3.0. Solarmetric have worked hard to provide this. A combination of BEA and the Solarmetric team should ensure this continues. Congratulations are definitely in order.

Nicely done Patrick and Neelan. I've personally seen this product grow through the stages since JDO was just a beta spec. I am excited about this move.

Congrats.
Raj

BEA is back in news doing good things. First there was an announcement about explicit support for Spring. That might not mean much for many TSS developers,but, it is a big thing for corporate IT managers who need that support comfort to deploy a popular open source framework in enterprise. As the replies indicate Soalrmetric was the best JDO implementation, but they did not get the claim bcoz of internal JDO politics within Sun. Now we can say at least BEA App server has solid O/R implementation. O/R was there only real weakness. I wonder when Sun will make a smart acquisition like this.

Congrats to Patrick, Neelan and the rest of the Solarmetric team - you guys really deserve this.

Rob

Congratulations to SolarMetric folks! But I am not so sure if this is a good news for Java communiy, just think about Oracle and TopLink. Seems EJB is closer to seal JDO, am I correct?

Are there any plans to donate the code to the Apache project?

Fantastic Job Patrick!

Congratulations to Patrick, Neelan and all the other folk at SolarMetric. Kodo has always been an outstanding O/R mapping engine, and SolarMetric's technical and commercial success has been well-deserved.

I think this is a very smart move for BEA. They have filled a major gap in their stack with a very high quality solution. Kodo has always had particularly good performance, so this should play well with BEA's high end customers, in particular. Two limits to Kodo adoption--resistance to the JDO API, and the fact that SolarMetric was a relatively small company--have now disappeared, so the competition in the persistence space is going to be very interesting. (Hopefully it will now be largely on technical merit...) SolarMetric already had quite a large customer base, and this should take it to another level.

Rgds
Rod

So when will Spring have a JDO template?

http://www.springframework.org/docs/api/org/springframework/orm/jdo/JdoTemplate.html

Since:
    03.06.2003

Ouch, that fixes that - I hadn't seen that. I'll look harder next time! :)

yes, EVERYTHING must be wrappered in Spring.

yes, EVERYTHING must be wrappered in Spring.

Not EVERYTHING, but whatever makes sense and makes programming life easier.

Cheers,
Dmitriy.

yes, EVERYTHING must be wrappered in Spring.

Not at all. Have a look at the ORM chapter in our current reference manual:

http://www.springframework.org/docs/reference/orm.html

As elaborated there, Spring's support for a specific data access strategy always comes as separate, fine-granular components: for resource setup (e.g. LocalPersistenceManagerFactoryBean), local transaction management (e.g. JdoTransactionManager), and DAO implementation support (e.g. JdoTemplate).

The DAO implementation support is completely optional: You can always code DAOs against the plain JDO / Hibernate3 / TopLink API instead, while still participating in any kind of Spring-managed transaction (with appropriate setup around). Your DAO will throw native resource exceptions then, though, not Spring's DataAccessException hierarchy.

The same applies to our forthcoming support for JPA (Java Persistence API, also known as EJB3 persistence). This is essentially gonna be analogous to our existing JDO / Hibernate3 support, with the same two options for DAO implementations: coding against Spring's JpaTemplate or coding against the native JPA EntityManager API.

BTW, congratulations to Patrick and co :-) You deserve it!

Juergen

I hope that BEA acquisition of SolarMetric means that JDO is still important for the enterprises, and that there is still going to be a strong competitor for Hibernate.

KODO JDO is great, thanks for reversemappingtool.

"Additionally, the Kodo engine will be integrated into an updated version of the BEA WebLogic Server® release, currently planned for the second half of 2006."
That's a long time to wait for EJB3 on WLS! A lot of people will be using HB with WLS.

"Additionally, the Kodo engine will be integrated into an updated version of the BEA WebLogic Server® release, currently planned for the second half of 2006." That's a long time to wait for EJB3 on WLS! A lot of people will be using HB with WLS.

Bear in mind that you can use Kodo + WLS today -- we have lots of customers doing so! What the press release is talking about is the tight integration of Kodo as the EJB3 Java Persistence API provider in a full production-supported Java EE 5 release (including session beans, MDBs, web services APIs, etc.).

It's also possible to deploy and use Kodo via the EJB3 APIs within WLS9 right now, via the EntityManagerFactory bootstrapping APIs.

-Patrick

--
Patrick Linskey
http://bea.com

Thanks for all the well wishes, everyone! (Yes, including Race Condition. I'm going to imagine that there was no sarcasm there.)

-Patrick

--
Patrick Linskey
http://bea.com

Patrick what is the next step?

Historically, after BEA people going to Google.com
Do you foresee this in the long run?

Ruslan :

Patrick what is the next step?Historically, after BEA people going to Google.comDo you foresee this in the long run?

My personal theory is that it's not a question of if someone will work for Google, but when...

geir

Google is definitively not the absolut Heaven.. it is just right now the only company able to hire 800 people on a last quarter... google hired people from BEA or others.. are just there for money!!... Considering.. the old "brain drain" story at BEA it just confirms that BEA hires the best..!!!

Oracle now has one less reason to merge with BEA.

On a tangent subject: does anybody of you ejb3 knowledgable people here know wether there's any possibility that ejb3 will support full-blown dependency injection before becoming final, instead of just simple types?

As a customer of Solarmetric and a long time user, i am welcoming this step but with mixed feelings. I hope they keep up the great responsiveness and support and i hope that they keep being visible in the OSS community. I allways liked to communicate with solarmetric because their staff is motivated and there is some sort of durability in terms of employee turnovers.

All the best...

Marc Logemann
http://www.logemann.org

On a tangent subject: does anybody of you ejb3 knowledgable people here know wether there's any possibility that ejb3 will support full-blown dependency injection before becoming final, instead of just simple types?

While the EJB 3 spec started the ball rolling within Java EE by adopting dependency injection it is now defined (where it should be) at the platform level in the Java EE spec.

In any case, the answer to your question is that injection of a known set of resource, ejb and persistence types (which are not necessarily simple) is supported, but generalized injection of random JavaBeans/Java objects will not be in the Java EE 5 release. It is being looked at for the next release, however.

-Mike

generalized injection of random JavaBeans/Java objects will not be in the Java EE 5 release. It is being looked at for the next release, however.

So that means there'll be an appserver implementing it available in 2009? Hmph....
But anyway, thanks for your reply Mike. Looking forward to using JEE 5 and EJB3.

Good job! Congratulations, well done guys.

Best,
Michael Vorburger

Congratulations Patrick, Neelan, and all the other SolarMetric guys. I know it's taken a lot of hard work to get here, over many years. Back in 2001, when I was working at a VC/incubator, one of our startups was was using Kodo, then at something like a 0.9 version, and I remember seeing even at that time the long hours and dedication everybody was putting into producing a top-quality product.

Regards,
Colin

--
Colin Sampaleanu
Interface21 Principal Consultant
Spring Training, Consulting and Support - "From the Source"
http://www.springframework.com

This news makes my day. Nice guys don't always finish last...Neelan, Patrick, you're two of the great ones.

BEA Systems, Inc. announced it has acquired the object persistence company, SolarMetric.

Great new!