Is JBoss Oracles Next Trophy?

Updated: Opinion: If Oracle really does buy JBoss, analysts think it could be the best glue to stitch together all the technology the company's been gobbling up.

Rumors have been flying about Oracles next acquisition target: namely, open-source middleware leader JBoss.

This Java Industry News article popped up on Jan. 27, crystallizing rumors that JBoss has been making the rounds, popping in at Hewlett-Packard and IBM to find the sugar daddy with the deepest pockets.

If the rampant speculation proves correct and Oracle is that sugar daddy, who benefits from the acquisition?

Oracle, by a long shot. With all the acquisitions its made since and including PeopleSoft, it has a hell of a lot of complex technology to stitch together in Project Fusion.

While Oracle does have good middleware, analysts say that it needs to be better than good enough to pull everything together. "If what your goal is, is not just to have good middleware but middleware that aims at being the middleware platform—the glue in the heterogeneous environment—then you need more than good, solid middleware that works on all your various platforms," said analyst Judith Hurwitz of Hurwitz & Associates.

JBoss is more than good enough middleware—its an open-source leader. Thats what Oracle needs to pull itself out of its also-ran middleware status and really be seen as a neutral industry standards player, Hurwitz said.

Some commentators are positioning the potential acquisition as Oracles initiation to the world of enterprise open source. As it is, JBoss LGPL software can be used in business applications in any way users want. It is a wildly popular and constantly expanding application development platform that pretty much owns the open-source end of the market. Oracle would suddenly be in the position of providing customer support for an open-source community that has been spoiled rotten with what analysts call "incredibly good" customer support.

That is, after all, the main premise of JBoss on its business end: to provide first-rate customer support for the open-source product.

"I have yet to talk to a JBoss customer who wasnt satisfied," said Richard Monson-Haefel, an analyst with The Burton Group.

Imagine Oracles sharp-toothed sales force being forced to leave a product user community alone. It brings up images of Dracula sending his undead brides away from a victim.

But bear in mind that this is in fact Oracles second open-source buy within a few months. In October, Oracle snatched MySQLs transaction engine, InnoDB, right out from under its comparatively teensy competitors nose, remember?

I dont know how the InnoDB purchase and the potential JBoss acquisition relate. Is the company bolstering its open-source presence? Undoubtedly. But open-source good intentions aside, there are plenty of sound business reasons to do what its doing in both cases.

On one hand, Im not convinced that Oracle would see MySQL as such a threat that it would need to go to any trouble to cripple it. On the other hand, crippling MySQL was so extraordinarily easy to do. So cheap! So easy! MySQL was sitting on a crucial component and never got around to buying it—Lord knows why.

Who could blame Oracle from wanting to trip MySQL up? If you were standing on a street corner finishing a banana and you saw your ex walking down the street with her new boyfriend, isnt it your obligation to do something slippery?

Im still convinced that Sleepycat is working with MySQL on an alternative transactional engine, at any rate, so instead of a pratfall, the InnoDB loss could well turn into a slight skid. Keep your ears open for that announcement.

As far as the theoretical JBoss acquisition goes, its obvious what Marc Fleury et al. get out of the deal: cash. Lots and lots of greenbacks.

Heres how Forrester analyst Michael Goulde put it to me when I asked him via e-mail what the acquisition would mean:

  • Mark Fleury becomes a millionaire
  • Oracle has minimal interest in open source above Linux—why the change in heart?
  • JBoss customers will need to either create a JBoss open-source community outside of Oracle or find another open-source J2EE platform.

What Oracle gets out of it is just as clear, according to Monson-Haefel. "It would be the smartest move Oracle would have made in a long history of not-smart moves in application development," he said.

As it is, Oracle, along with the other big for-profit application development players—IBM, BEA and, most recently, SAP—have developed what The Burton Group terms a superplatform: an application development platform that goes well beyond an application environments original delivery of operating system, compiler and frameworks to deliver much more sophisticated J2EE functionality, including sophisticated integration brokers, advanced Web services, powerful portals and tightly coupled integrated development environments.

Even though Oracle developed what Monson-Haefel called "a really good platform" that put it in the same arena as IBM and BEA, it still hasnt been able to compete effectively in the market. Its a distant third.

Why? Monson-Haefel, like others, blames it on Oracles being primarily a database company. "Making that segue into the J2EE space just hasnt been that easy," he said.

Meanwhile, companies that are rationalizing their strategy on J2EE just might prefer a bigger, better company such as Oracle, he said. The majority of JBoss users would likely be fine sticking with it, but the high-end players would serve Oracles market position very well.

As it is, IBM last year bought the open-source software provider Gluecode, giving it access to the top developers on the Apache Geronimo application server.

As eWEEKs Darryl Taft reported, that move put pressure on JBoss.

"I think this is a move on IBMs part to counter the momentum of JBoss to have a monopoly in open-source middleware," Bob Bickel, vice president of strategy and corporate development at JBoss, told Taft at the time.

So now the tables have turned, if the acquisition in fact happens. (Oracle hadnt responded to inquiries by the time this story was published.)

Monson-Haefel views the potential JBoss buy as being Oracles ticket to doing what IBM did: namely, picking up a low-end entry point to the product market, hoping to use it as a stepping stone to the sale of higher-cost products.

Nothing to do but wait and watch.

In the meantime, let me know what you hear. Im also curious to hear your thoughts on whether the JBoss community would need to create a JBoss open-source community outside of Oracle or find another open-source J2EE platform, as Goulde suggests, and what you think Oracle is up to with open source.

Lisa Vaas is Ziff Davis Internets news editor in charge of operations. She is also the editor of eWEEK.coms Database and Business Intelligence topic center. She has been with eWEEK since 1995, most recently covering enterprise applications and database technology. She can be reached at

Editors Note: This story was updated to correct a statement about JBoss by Judith Hurwitz.


Check out eWEEK.coms for the latest database news, reviews and analysis.