What Sun+MySQL says about open-source business models

On the heels of yesterday's Steve Jobs keynote at Macworld, Apple may be the tech company that's top of mind for many readers. However, from an enterprise computing perspective, Sun Microsystem's announcement that it is acquiring MySQL is far more pertinent. News.com's Martin LaMonica summarizes the announcement thusly:

Sun Microsystems will pay $1 billion to buy MySQL, the provider of a popular open-source database.

Sun said Wednesday that it will pay about $800 million in cash for MySQL's stock and take on about $200 million worth of options. MySQL CEO Marten Mickos will join Sun's senior executive team after the transaction closes.

The acquisition is a bold move for Sun, which has embraced open-source software and development practices in an effort to garner more revenue from its software business. Until now, it has sold support services for a competing open-source database, PostgreSQL.

MySQL is one of the most successful open-source companies founded in the past five years. It's part of the popular combination of open-source development products referred to as LAMP, for Linux Apache Web server, MySQL, and the PHP development language, which is broadly used on the Internet and within companies.

Here, I wanted to focus in on one specific implication.

MySQL is the clear category leader in open-source databases; it's the "M" in the LAMP stack that also includes Linux, the Apache Web server, and the Python, Perl, and PHP scripting languages. And LAMP underpins a huge portion of the open-source software world. As a result, MySQL--like JBoss before it was acquired by Red Hat--made a nice little business of selling support subscriptions for its software. Indeed, it was one of the more successful of the more-or-less pure standalone open-source companies.

If that sounds like damning with faint praise, it is a bit. Because so few end users tend to buy support contracts relative to the number of people that use the product, pure open source has been a challenging business model for its practitioners. That's not to say that there aren't companies successfully taking such an approach, but there are no pure open-source Oracles, Microsofts, or VMwares raking in the dough.

Small software companies get bought by larger companies all the time of course. Open source or not, enterprise customers often appreciate the sort of global support that large vendors are better prepared to offer. And the ability to put together sets of products that address broad business problems is more appreciated. However, in the case of open source specifically, the fact that a large vendor can leverage open-source products to sell other software and even hardware creates far more revenue opportunities than when the only thing a company can sell is a support contract on a single piece of software.

In the Sun and MySQL case, for example, one can imagine Sun eyeing the vast population of MySQL users not so much for the opportunity to sell MySQL support contracts but as an entree for selling other Sun middleware, Solaris, and Sun hardware. One can imagine a conversation like this repeated many times: "Oh, you need better performance out of MySQL running on Linux? Of course, we're happy to help. But you might think about Solaris because we have this DTrace tool. We also have this ZFS file system. And, oh, have you heard about Thumper?"

It's not so much that there aren't workable business models around open source. But life is so much easier when those models can include pieces that people have to pay for as well.