Marten Mickos is less known for his current gig as CEO of Santa Barbara, Calif.-based cloud management platform, Eucalyptus Systems, than he is as the guy who ran MySQL, the open-source database company that put the M in the LAMP stack.

MySQL, which mirrored the web renaissance that started in 2003, was snapped up by Sun Microsystems for $1 billion in January 2008. Since then, Oracle has gulped Sun and now owns MySQL, the database that threatened the Oracle’s core offering. Last week, Mickos stopped by our offices for a conversation that covered a variety of topics:

• The rise of cloud computing and its impact on open source movement.
• Will there be a LAMP stack for the cloud?
• Will Oracle destroy MySQL and how does NoSQL movement impact MySQL?
• Philosophical aspects of open source.
• The open-source business models.
• Most common mistakes he has seen startups make, especially open-source startups.

Mickos also took a trip down memory lane and talked about some lessons he learned from his days at MySQL:

• We learned many things from the many mistakes we made at MySQL. One thing we learned: You must become the category leader because that drives so much of your momentum. That’s how we could withstand the pressure from Oracle and Microsoft.
• Another thing, which dawned on us early, was that we thought we were in the database business. But it turned out that most people used us and still use us as a web platform. Technically, we were in the business of relational databases, but we were viewed as web technology, not database technology. This was a big realization!
• Mixing of business desire and open source. We told everyone that this company (MySQL) produced cash and code and there was no compromise on that. That commitment to revenue growth kept us growing with very little capital.

He also discussed some of the big changes in open source since he started worked for MySQL.

• It has changed a lot since I joined MySQL in 2001. At that time it was very grass roots driven, driven by projects. Today it is very much driven by corporations. Success is that open source is everywhere, which makes it boring because it is everywhere. No one calls it a cancer. It is much more institutionalized.
• Some open source developers have no business interest and no desire to build a business. They need to just build the code and get it out there. If you are going to build a business, you must make sure there is revenue so you can hire more people and produce more code.
• That is always going to be a complicated issue in the history of open source. I have always believed that the best way to serve open source is to make money, because when you make money, you can hire more people to produce more code, and you can challenge closed competitors and win over them.

You should set aside some time to watch this video.  It’s about 25 minutes long and is full of insights, something you should obviously expect from the man Chris Albrecht, who produced and edited the video, describes at the tallest man to walk into our humble offices. Roll tape!