Oracle buys BEA, Sun buys MySQL – it must be the post Christmas shopping season…

It was serious business yesterday. Within 10 minutes I received word of the Oracle BEA acquisition and the Sun MySQL take over. That was a lot to take in – for both Oracle and Sun, but for me also. What is happening here – and what will the consequences be. For customers, the industry as a whole, other vendors and last but not least my company and myself.

Of course I am not a market analyst or Gartner guru. And I know that those that are, often write an ‘objective analysis’ that is extremely hard to write without one’s one agenda and prejudice. I read for example an interesting piece by Rod Johnson, the Spring Framework founder, who stated “As an open source company, we are thrilled to see MySQL rewarded for their effort.” (read his blog entry here) I find it hard to not read into this: “wow, I would like my open source company to be bought for one billion dollars too!” And who can blame him? But from an open source point of view – is that the best thing that can happen? He writes (cites) that “money cannot buy communities”. So what does money buy when you buy an Open Source company?

An interesting observation is that Sun – (especially) under Jonathan Schwartz – is taking open source so seriously. And is building perhaps the most compelling JEE open source stack around: NetBeans, GlassFish and no MySQL. What will this mean for JBoss?

AMIS focuses on both Open Source JEE development as well as Oracle Java technology, and tries to combine the two whenever possible. Our focus Open Source stack contained both JBoss, MySQL and Eclipse. However, a lot of our energy in this area has gone into Glassfish and NetBeans as well (including several contributions to the Glassfish community). We will have to reconsider which stack to consider the premium one.

I am wondering whether Oracle’s aquisition of InnoDB will come back to haunt anyone – like Sun for example. Sun and Oracle used to be best of pals. Will this acquisition by Sun sour that relationship in any way?

Oracle buys BEA

They did it again: see http://www.oracle.com/corporate/press/2008_jan/bea.html or http://www.oracle.com/bea/index.html . After buying themselves into second place in the ERP space (and wiping out most of that space as it was), Oracle now probably goes into the market leading position with the combined marketshare of Oracle (OC4J) Application Server and BEA WebLogic, overtaking IBM’s WebSphere platform.

And now the speculations start. Depending on who you ask you may here several opinions:

• Oracle could not build up market share for OAS of their own, so now they buy it
• This spells the end of OC4J as WebLogic is the superior product
• Both products will continue to coexist, gradually taking over each other’s strengths before finally blending into one (look at what Oracle does with the ERP products it bought)
• This means a quick end to WebLogic and AquaLogic: the best pieces are absorbed by OAS and Oracle SOA Suite and the rest will vanish – customers forced to migrate to Oracle’s own products

and many more.

Most of these statements seem to be driven by sentiment or hidden agendas and opportunistic reasoning. And of course I am not free of bias myself. My experience with Oracle’s acquisitions of the years is that they buy good products that usually (starting with RDB – acquired in the mid 1990s and still around despite all claims that it would rapidly be replaced by Oracle’s own RDBMS, PeopleSoft and the slough of other ERP products), TimesTen, Hyperion (so far), Berkeley DB, Tangosol Coherence, and many more) continued. Some, especially the smaller ones, have been absorbed into the Oracle stack – such as Oblix and Collaxa-, typically improving a lot along the way.

My hopes are that initially the two products will coexist, with Oracle benefiting of the open, plug-and-play architecture of its SOA Suite, that will allow use of the best of breed from the Oracle and BEA stacks for various components in the SOA infrastructure. I would expect that even more than before Oracle products that run on J(2)EE Server will easily deploy and run on WebLogic – like again the SOA Suite, but also for example ADF Applications, the latest releases of Oracle Applications (?) and maybe (though highly unlikely in my opinion – Oracle Portal and Oracle Form Server. Support in JDeveloper for WebLogic will be further extended. At the same time I expect BEAs WebLogic Workshop to do the same for Oracle technology. Interoperability of the two stack will improve. And they will continue to coexist for quite some time to come.

Oracle will think liquid – and Rock It! And will I – having not had a lot of exposure to BEA products in the past, I have just started to browse www.bea.com. I feel a bit liquid today…

Putting things together, I expect yesterday’s events to ensure more clarity, better software in my hands and some pressure on IBM to show their hand – as I find their JEE strategy and product portfolio not always as clear (and their participation at many developer’s conferences and events poor to non-existent). And I think other vendors and parties with a vested interest will try to distribute FUD, scaring Oracle and BEA customers and developers, trying to make the best of what they may perceive as a threat to their business.