Oracle's recent acquisition of a Finnish open source database company has given new life to the ongoing question of what role open source will have in the enterprise software market. The answer: Don't bet the farm on open source, yet.

To be fair, there are lots of ways in which open source technology may have already insinuated itself into your company: Apache servers and Linux operating systems are examples. There may even be a MySQL project -- similar to the open source code recently acquired by Oracle -- hidden somewhere in IT.

In other words: your geeks may already be using open source technology. Just don't expect your executives and business managers to be using it anytime soon.

The reasons your mission-critical applications won't be built on an open source model anytime soon involve the limitations of the open source development model and the increasing requirements for highly verticalized, industry-specific functionality. Alas, the two concepts don't really mesh.

Open source development has worked well in the systems and tools world precisely because of the extraordinary wealth of knowledge and expertise now present in these formerly arcane domains. Credit the Unix and open software movements, the predecessors of open source, for seeding a global market of technologists who are both knowledgeable about and interested in contributing to an ostensibly free open source development effort.

Meanwhile, enterprise software has moved more and more toward industry-specific knowledge about how complex processes work in different companies. This process-centric view requires deep domain knowledge about best practices -- knowledge that is both in short supply and highly valuable. With the pool of expertise relatively small -- and legitimately tempted by the opportunity to capitalize on that talent -- for-profit software companies have taken the lead in enterprise software.

In between these two extremes -- generalist commodity systems software and highly specialized vertical applications software -- are a lot of grey areas that might someday lend themselves to an open source solution. Those parts of the enterprise software portfolio that are both well-distributed and highly commoditized -- such as general ledger, HR, and contact management -- could legitimately end up as part of an open source offering.

But it is no coincidence that these are the very domains that are leading much of the on-demand offerings in the industry today. And in my opinion, if cost-cutting and simplification are what a company is looking for, then a hosted model makes more sense than open source. Hosting what isn't mission-critical, while retaining in-house what is, removes the onus of maintaining an unnecessary IT staff . Replacing mission-critical applications with open source applications still requires a core competence in IT, and that cost could easily obviate the gains from open source.

The bottom line is that as long as the intention of enterprise applications is to deliver genuine innovation and impart a competitive edge, it is unlikely that they will be made available as part of an open source project. The stuff you care about, the stuff that makes your job better and wreaks havoc on your competitors, will have to come to you the old-fashioned way. When it comes to true innovation, there really is no such thing as a free lunch. Or free software.