News

Oracle's Sun Acquisition Spells Uncertainty for Java, Database Developers

• By John K. Waters
• April 21, 2009

Though Oracle has a storied history of absorbing large software companies (most recently BEA Systems), uncertainty over the fate of Sun's key software assets will reign for some time.

"The question now on the table for developers is which pieces and parts are going to be supported going forward," said Jonathan Eunice, principal IT advisor at Illuminata. "JavaFX is a good example. Will Oracle want to be in the rich-media, Java plug-in space? What about Sun's IDE? Should I be continuing to develop with NetBeans? Should I continue to develop for OpenOffice? What are Oracle's intentions with regard to the future of these technologies? Which APIs will be supported? We all want to spend our time, money and attention on platforms that are going to be an important part of the future." 

Early signs suggest that some of those technologies won't make the cut, said Forrester Research senior analyst Jeffrey Hammond. "They're talking about the importance of Java and Solaris, but not saying much about MySQL or NetBeans," Hammond said.

Hammond said he expects NetBeans to be among the casualties of the acquisition. Sun has bundled the NetBeans IDE and other products with its Java platform offering, and Oracle, which currently supports both BEA's Eclipse-based tooling and its own JDeveloper tools, might find that three IDEs is a crowd.

"There's no reason for Oracle to support three IDEs," Hammond said. "I think they'll stick with JDeveloper, because it's thoroughly integrated with its Oracle Fusion strategy, but I can't see long-term support for NetBeans."

Another Sun technology with an uncertain future under the Oracle aegis is the Java EE-based GlassFish application server. GlassFish, created by Sun and then open sourced, has its fans in the developer community, but RedMonk analyst Michael Coté pointed out that it becomes redundant in the Oracle product lineup. "Oracle has all of BEA and its own app servers," Coté said. "Does it need to support another one?"

Yet despite the changing of the guard for Java, Coté said the acquisition will likely have a minimal impact on the developer community at large. "It's a hard ecosystem to control," he said. "I think the Java world has evolved to that place where buying a trademark won't have a huge impact. Besides, Oracle is already a big part of the Java world."

There is even more debate as to what Oracle's acquisition of Sun will mean for the MySQL open source database platform, which Sun acquired last year for $1 billion. "It's probably bad news in the short term," said George Reese, co-founder of Minneapolis-based cloud infrastructure management tools maker enStratus. "MySQL has made significant progress in the past five years toward becoming an enterprise-capable database, and I doubt that Oracle will want to compete with it at that level."

But in the long term, the acquisition provides some much-needed clarity to the MySQL picture, added Reese, who has been around both Java and MySQL for a while. He helped review the initial JDBC specifications in 1996 and wrote the first JDBC driver. He's also the author of Java Database Best Practices and Managing & Using MySQL. 

"MySQL has languished under Sun's ownership," Reese said. "They never really figured out how to manage it and be a database company, and as a result it has been forking. With Oracle, at least there will be a clear direction -- though I doubt it will be the direction a lot of people were hoping for."

Potential Benefits
Despite questions about numerous likely changes, Eunice said the acquisition could work to the advantage of developers. "Sun has been poorly operated," he said. "[CEO Jonathan] Schwartz has had some inventive strategies. Open sourcing Java and Solaris, buying MySQL -- those were good calls. I just wish that they'd had a better crew of hard-[nosed] operations managers to back up the strategically interesting choices they made."

MySQL in particular is likely to fare better in Oracle's hands, Coté said, though he agreed with Reese that the company is unlikely to fulfill the hopes of its most ardent proponents. "The beating heart of Oracle has always been its database," Coté said. "And Oracle has already figured out how to cope with MySQL by emphasizing the high-end, enterprise use of the Oracle DB. MySQL had positioned itself as a lightweight database, rather than a massive numbers-cruncher. Now, instead of losing that lower end of the market, Oracle has it, at least potentially, with MySQL."

The change of management for the technologies that survive the acquisition is almost certainly a positive for developers, said Yankee Group analyst Zeus Kerravala. "Oracle has been dying to get its hands on Java which will make it even more competitive with Microsoft," he said. "The good news for developers is that it's now in the hands of a real software company."

Reese disagreed with Hammond's prediction that the long-term future of NetBeans doesn't look promising. "Oracle might do NetBeans some good," Reese said. "It's always been a niche tool for Java. I could see Oracle putting more muscle behind it and leveraging its distribution channels to make it more widely relevant to the market than it is today."

Java developer Dag Blakstad, senior consultant at Oslo-based Webstep AS, is taking the acquisition with a grain of salt. "I'm hoping for the best," he said. "The MySQL community is very big, and I would be very upset if Oracle did anything bad with it. I'm a little worried about Oracle doing proprietary things to Java. Oracle is sometimes a company that likes to do things their own way, not following standards. But it's very early to start worrying. At this point, I'm hoping that they will keep Java alive and well."

In the end, Oracle's stewardship of Java is probably less of an issue for developers than the general constriction of technology options an acquisition like this represents, Hammond said.

"That's the biggest fundamental impact on developers," Hammond said. "It's the Sun deal on top of the BEA deal, and so on. There's now less choice among app servers, among IDEs, and to a certain extent, among databases. There are advantages to consolidation, but it's not necessarily a good thing for developers to narrow their choices."