10.3. Feedback and Evolution

Ted Codd's original paper on the relational data model was published in the open literature and influenced people outside IBM such as Larry Ellison, and Mike Stonebraker's group at UC Berkeley. Was this process similar to the "open source" model? How did external visibility affect the development of SQL?

Don: In the 1970s, the relational data model was a new idea. It was a subject for advanced research and prototyping and it was not generally available commercially. SQL was developed as part of an experimental research project called System R that was independent of IBM's normal product development process. The research division at IBM traditionally publishes the results of its exploratory research in the open literature and that's what we did with the SQL language and other parts of System R.