It sounds like a classic case described in Clayton Christensen's "Innovator's Dilemma." Oracle is continuing to build out the high end capabilities of its enterprise software product line while innovation is occurring at the low end, in startups, web companies, and open source projects. As others have also pointed out, open source, cloud computing, and SaaS are notably absent.
Similarly, the connection to the industry did not receive a lot of attention. Yes, support for standards such as SCA and OSGi are there, but it takes some digging to discover what they're being used for. In both cases the descriptions in the announcement materials and executive quotes make it sound as if SCA and OSGi are used primarily for their benefits to the product developers rather than things customers can directly use.
In a kind of classic proprietary software model, the primary driver for the work done for the release appears to have been essentially internally-driven. I don't see much, if anything, that sounds like a new and different reason to use the Oracle middleware product line, or a point of connection with other vendors' products. It has also been reported that Oracle seems on a path toward a "one-stop shopping" business model, which seems to be validated by this announcement. Other vendors' server environments were mentioned, but it seems clear that the benefits of the suite are only completely available in an all-Oracle environment.
Sanjiva Weerawarana blogged about the size of the environment, highlighting the complexity of a the comprehensive set of inter-dependent tools and runtimes, and pointing out that the size and complexity creates a barrier of entry for a new customer. But as I said previously, I don't think that's the point. And I also I think Oracle's products encourage the use of industry standards as long as that use is consistent within the overall motivation for using the Oracle suite in the first place. In other words, the suite seems designed to try to keep developers within its "four walls."
This is of course a completely understandable and tried and true approach to enterprise software. But it runs contrary to the open source trend, for example, where product plans and feature development are available for anyone to see and comment upon. Look at this week's Google OS announcement for an interesting contrast. This is an open source project for which external comment and input is explicitly requested.
It's unfair to characterize the entire announcement as lacking in innovation or without features that would attract new customers – for example the integration of Coherence with WebLogic, and the use of the OSGi deployment model for each creates a pretty interesting and compelling story for the kind of high performance, memory-based transaction processing applications emerging from new web site architectures. But you have to look hard to find it in the big picture being presented.
This was first published in July 2009