Reconsidering Oracle’s antitrust commitments to MySQL

As I mentioned earlier this week, a major research focus for Q1 is the MySQL ecosystem, the positives and negatives of Oracle’s MySQL strategy, and the competitive overlap between MySQL, NoSQL and NewSQL.

It is impossible to think about this without reconsidering the commitments made by Oracle to customers, developers and users of MySQL in late December 2009, which played a significant part in satisfying European Commission concerns about Oracle’s acquisition of Sun.

While the commitments were both welcomed and derided when they were announced, it is worth considering today whether those commitments have been as significant in practice as they appeared to be two years ago.

For example, Oracle’s commitment to and investment in InnoDB – while positive for MySQL users – has arguably diminished the relevance of some of the storage engine-related commitments.

We will be coming to our own conclusions based on our research over the coming weeks, but I am interested in any feedback from MySQL customers, developers and users about how well Oracle has kept to its commitments and their significance in hindsight.

You can find a full list of the commitments here but the edited highlights are below:

1. Continued Availability of Storage Engine APIs.

2. Non-assertion of copyright and no requirement for a commercial license related to implementing the storage engine APIs .

3. Extension of any existing commercial storage engine licenses until December 10, 2014.

4. Commitment to continue licensing MySQL using the GNU GPL.

5. Customers would not be required to purchase support services from Oracle as a condition of obtaining a commercial license to MySQL.

6. Increase spending on MySQL research and development.

7. Commitment to create and fund a customer advisory board.

8. Commitment to create and fund a MySQL Storage Engine Vendor Advisory Board.

9. Commitment to retain the free MySQL Reference Manual.

10. Retention of annual or multi-year subscription renewals for end-users and embedded customers.