At the Web 2.0 Expo, Geir Magnusson delivered his prediction of the upcoming demise of relational systems: "The Sequel to SQL: Why You Won't Find Your RDBMS in the Clouds." This reminds me of the old saying that the rumors of my death are greatly exaggerated.

Around 15 years ago, a variety of object database vendors were the upstarts with names like Objectivity, Poet, Versant, etc. Some are still around, and their promise was to eliminate the mapping from object oriented software to relational database. When I look at public tech companies, I see Oracle, Microsoft and IBM at the top in terms of size and sales, all with relational databases as major products. Looking at open source, we see MySQL and PostgreSQL. All of these have thrived while the object databases and other specialized databases are either in the ash heap or very limited by most any measure.

With my background in technology, I am intrigued and excited about all of this new stuff including those database related ones like SimpleDB, Bigtable and Magnusson's 10gen products, but they are unlikely to make a big impact against relational databases any time soon. Even in the cloud, I expect relational databases to remain the leaders for many years to come. Is this the recognizable logo and name that businesses use for their critical data? Not the ones that I am aware of.

* Update 9/23 - Amazon Web Services announced that Oracle has certified their database to run on EC2 and Sun is supporting MySQL Enterprise running on EC2. Sounds like maybe the old relational workhorses will live on for a while longer.