In last December's column -- An Opening for Open Source? -- I made a blanket statement about open source technology that I would like to revisit. I opined that open source in the enterprise would remain the bailiwick of the techie side of the business for some time: "Just don't expect your executives and business managers to be using it anytime soon," was my conclusion.

After sitting down with a couple of open source database companies -- MySQL and Greenplum -- I'm ready to concede that there's a burgeoning open source portfolio that isn't just for geeks and nerds, and, if well-applied, may actually be a means for the business side to reduce its reliance on the techies.

There's a caveat, however. Don't expect an open source database to replace your existing Oracle, DB2, or SQL Server implementation. Despite the obvious cost advantages of open source (license costs are often one-tenth that of a closed-source database, development enhancements are "funded" through an open source community, etc.), the complexity and cost of a "rip and replace" strategy will keep the incumbents in place for some time.

But for new, greenfield applications -- and facilities -- starting with an open source database might be one of the smarter things you can do. This is especially true if you're implementing enterprise applications (except those that are tied to a specific relational database, as in Oracle and Microsoft Dynamics). For the most part, other than Oracle and Dynamics, few if any enterprise applications make full use of the rich features in top closed-source relational databases. The reasoning: If your application is designed to run on multiple databases, then you have to use a least-common-denominator approach to database functionality.

This fits in perfectly with the open source database model. Not to imply that MySQL is not functionally rich. But if you're not using a top-tier vendor's bells and whistles, it might make sense to go with an open source database that provides needed functionality at a significantly lower price.

An open source database also may make sense for the non-traditional, non-packaged application functionality you may be considering developing. A look at MySQL's customer base and usages shows a wide range of complex, custom applications running on the MySQL database. It's really part of the same low-cost equation: custom development is expensive, so why not save some money on the commodity database that lies underneath?

There's a similar message in Greenplum's pact with Sun to deliver a multi-terabyte, turnkey data warehouse system starting at $250,000 -- easily 10% of the cost of a similarly sized data warehouse using a "name brand" database. Again, unless there's a compelling reason to choose the name brand -- and there aren't a lot in the data warehouse world -- then a Greenplum/Sun solution should suit most of your data warehouse requirements.

The bottom line is that there is definitely room for an open source alternative in your business, depending on the application or function you're trying to implement. I still think you'll be buying a lot of commercial applications. But the MySQLs and Greenplums of the world are showing that there is also an "open source" way that might make sense for your executives and business managers to consider. If not now, then soon. Very soon.