Does MySQL's proprietary Workbench signal that the future of open source is...closed?

MySQL, one of the world's most successful open-source companies, has released the Standard Edition of its new MySQL Workbench product under a proprietary license. The company gives several reasons for doing so, but I suspect the core reason is that MySQL is experimenting with ways to ensure more of its production customers pay it for the value they derive from its products.

Is this the future of open source? To get ubiquity through open source and then cash through proprietary source?

I emailed Marten Mickos, CEO of MySQL, about the change and he stressed what MySQL's FAQ already notes: This is not crippleware:

The open source edition is the code base the commercial edition is built on. It has the full feature set that is needed to efficiently design database schemata and is not crippled in any way. The commercial editions only adds modules to the open source edition to help the user save time. [Emphasis mine]

I don't think MySQL stresses this point enough. It's not about adding features that affect the core utility of the product. It's about adding features that cut the time required to develop using Workbench. It really would have been better offered as part of its Enterprise service. More on that below.

The company notes:

Everything that is possible in the Standard Edition (commercial version) can also be done with the OSS Edition. You only trade saving time and ease of use for money....

If you are a MySQL expert who has the knowledge and time to manually perform some steps, MySQL Workbench OSS will be the ideal choice for you. If you want to be able to do more in less time, you would like to have the additional safety-net or you simply want to give something back to the MySQL team - the Standard Edition is your logical choice.

As noted, MySQL would have done better to brand Workbench Standard Edition as a service, perhaps as part of its Enterprise Network offering, much as Red Hat's Network offering is proprietary. The real value of open source is in the software that you run in production, and the code you modify to get there. Workbench is a tool - a service - that is designed to get more people into profitable production. Brand it and license it as such.

MySQL could have heightened the value of this service by letting an open-source version of Workbench run wild in the community, with all sorts of third-party modifications, plug-ins, etc. The company would then cut back on features, functionality, and complexity to provide a safe, simple, and ongoing subscription to the service for its paying customers. The community and MySQL's customers would then get more value, not less, by getting more and less, respectively.

In fairness to the company, it arguably has created this problem for itself by giving away a certified, tested version of its database for years, such that the need for support in production is muted by the quality of the code and it has had to cast about for modified business models. Had MySQL immediately gone with the time-delayed bug fix model that it currently uses, it likely wouldn't need to resort to this kind of tactic. Or, rather, this strategy would have been a tighter fit with its existing model.

Regardless, MySQL is not signaling that its future is proprietary with Workbench Standard Edition. It's just demonstrating that while its core database will be open source, the services around that database need not be. It's a permutation on the Red Hat model: the bits are free, the service is not, both in the sense of free beer and free speech.