Marten Mickos, who was chief executive officer of MySQL prior to its billion-dollar sale to Sun Microsystems (s JAVA), has decided to leave the company, the second high-profile MySQL executive to do so in as many days. On Thursday, Michael “Monty” Widenius, MySQL co-founder and original developer, left Sun to set up a consulting firm, Monty Program Ab. According to published reports, Marten was frustrated by the bureaucracy at Sun. MySQL is an open-source database software that is commonly used by web companies, including high-fliers such as Facebook.
“It is the sort of typical situation with (a) chief executive of an acquired company,” he said when I called him, diplomatically deflecting my question as to why he was leaving. “I love being the CEO of a small company and proving unproven business models.” He said he’ll work through the end of March, then take some time off. “I decided to leave without anything specific in mind, but I want to work on stuff that has a big impact,” he said.He said the Sun acquisition has helped MySQL in many ways, including providing it with new resources that have helped extend its reach. The acquisition has proven to be a good one for Sun as well, and it has helped them reach markets in which people paid little or no attention to Sun. On their most recent conference call, Sun executives pointed out that: “MySQL and Infrastructure software billings grew 55 percent year-over-year.” Sun CEO Jonathan Schwartz added that, “We are again seeing great growth in opportunity around MySQL and our open-source operating system and middleware platforms.”
Marten and I first met in 2004. At the time, MySQL was about to hit its tipping point, as I noted in my post, MySQL – The Real Broadband Brain. Since then, the company has grown manifold, even flirting with an IPO.
After years of closely tracking MySQL, one thing I have learned is how closely entwined Marten and Monty are with MySQL and its community; I think their exits are sure to rock the boat there. But Marten disagrees. Companies survive long after their founders and key executives have left, he said. “I have built this for eight years and we have built a model. I have people who report to me who are much smarter than I am and they are trained to do this,” he said.
“It’s great to be a figurehead, but many times figureheads get too much attention,” he said. “Most of the times, the culture and the pulse and the execution of an organization is spread among a number of key individuals.”
In other words, MySQL is now much bigger than the sum of its individuals. I kinda agree with him, and from personal experience I can tell you that with good management, companies do march on. And as we’ve already pointed out, we are entering a new phase, one in which the importance of databases and data warehouses will only grow. From that perspective, Marten says, “The power is not in our product, but it is in our ecosystem.” He pointed to companies like Infobrite, which are using MySQL but building a new kind of data warehouse. “That is how MySQL will continue and evolve.” Of course, it also depends on Sun and how it treats its database business.
Photo of Marten Mickos courtesy of Wikipedia.
The exits of the MySQL key men, its acquisition by Sun, and the past year’s Web20 scaling failures all over the map have also highlighted a growing understanding that we have cultured a Web architecture and workforce that builds inherently unreliable Web applications.
MySQL ushered in an era where application servers (Apache) could easily be connected to relational, sophisticated, free databases; it was a revolution. Innovative ORM systems broke the mold on old SQL’s stale expressions.
But we gained a devil’s bargain – easy to create Web Apps that could not grow without arcane replicationa and clustering. Now we have a plethora of new sharding techniques, and all sorts of grid solutions for getting away from the column / row paradigm. Or, on the other hand, we preserve the illusion of columns and rows, and create other, more robust storage methods that scale better under the hood.
MySQL will grow with the times, but it has reached a point where it will compete with DB4O, and other, newer, and in my opinion, more naturally robust data storage architectures.
And as I said, it is just a timing thing that the honchos are taking break as these limitations are giving way to new alternatives.
I love MySQL. But I am also tired of Marten’s failure to fix some of the obvious problems Alan references, and which people like Don Macaskill at SmugMug outline here:
http://tinyurl.com/8a7quf
Marten is a typical slick sales exec, who lost touch with the community that handed him the business model he and his fat cat partners turned into a windfall.
“Marten was frustrated by the bureaucracy at Sun…”
Doesn’t surprise me. Many small upstart/maverick people who sell to big companies end up feeling the same way. The corporate culture shock of the sale is too much for some.