What will change for Berkeley DB users under Oracle? The biggest change is the fact that Sleepycat had 25 employees and Oracle has 56,000. Customer support used to be provided by the engineering team backed up a very small support team. About half the cases would get handled by support, with the other half sent back to [the] engineers. We expect that model to continue. But now we have a team around the globe that can take cases in local languages. Round-the-clock coverage is easier and less expensive now.

Developers will continue to interact with customers. That's one thing Larry [Ellison, Oracle's CEO] specifically said he liked about open-source -- the easy communication between the consumers and developers of software. We hope and expect that there will be no negative changes [to that process].

Users often choose Berkeley DB specifically because they don't want to have to deal with a full-blown SQL database. Does your new role feel weird? Not at all. If you need the full power and flexibility of SQL, I used to have to turn you away. Oracle used to have a much harder time answering nonrelational requirements. Now, if you don't need SQL or ad hoc query [capabilities] but just want a lightweight, embeddable engine, Berkeley DB gives users that. It's not the case that there's no overlap among these products, but their target is different customers.

What is the development road map for Berkeley DB and the other products you oversee? In general, we're concentrating on continuing to serve our existing niches and customers. One notable change is that because we're now part of a much larger organization, we want all the products, including TimesTen and Berkeley DB, to interoperate with others in the Oracle lineup [so they will] behave predictably if you're an Oracle DBA.

We're about to release Version 4.5 of the Berkeley DB core engine, which was already specced out [before the acquisition]. But what our much larger customer base wants now is more interoperation among the products. So that's driving our thinking.