I remember once asking Marten Mickos to participate on a panel - "The Battle of the Databases" - for Linuxworld a year or two back. He declined. At the time, I was mildly annoyed at his sense of camaraderie - he didn't think it was productive to try to set MySQL against Ingres against PostgreSQL. A bit later, I can appreciate his stance. Yesterday I exchanged emails with Boris Kraft, founder of the Magnolia CMS I remember once asking Marten Mickos to participate on a panel – “The Battle of the Databases” – for Linuxworld a year or two back. He declined. At the time, I was mildly annoyed at his sense of camaraderie – he didn’t think it was productive to try to set MySQL against Ingres against PostgreSQL.A bit later, I can appreciate his stance. Yesterday I exchanged emails with Boris Kraft, founder of the Magnolia CMS project. I regularly exchange emails/IMs with Dries Buytaert, founder of Drupal. Both are competitors in one sense, but in reality we’re focused on different opportunities. The market is too vast to get overly concerned about fighting over a few select customers. I have profound respect for the work that these entrepreneurs/developers do, even if it’s sometimes competitive (or perhaps precisely because it’s competitive).Which brings me back to databases. I was just reading a blog post by Lumen Software that declares that “PostgreSQL More Functional for Commercial Open Source SaaS.” Highly intriguing, especially given all the traction that MySQL has been getting in this very same market (like its recent win with Bookings.com. Granted, Lumen’s post suggests PostreSQL is a better choice for small- to medium-sized businesses utilizing open source SaaS software (and MySQL is perhaps more focused on the Googles of the world?). Just as in my ECM world, I have profound respect for both sides – MySQL and PostgreSQL (as well as the companies that surround them, like Lumen, Greenplum, and EnterpriseDB in the Postgres world). I’m not sure there’s a right answer to the “Which?” question. In some cases, you’d answer that with a “both” and in others you might answer it with “neither.” Usually, though, the answer is going to come down to what, exactly, you’re doing (fit the proper tool to the problem at hand – hammers don’t do well on screws) and, frankly, to which database you’ve first encountered. There is no once-and-for-all “Vote for MySQL/Postgres!” decision to be made in databases, just as there aren’t in many aspects of our lives. Passion on either side of this technology and licensing debate is for the better, but at the end of the day the business decision comes down to which software/service meets an enterprise’s needs best. Lumen has given this some thought, as the blog suggests:We have re-evaluated MySQL several times over the years. We attended the MySQL conference last year and heard about many changes MySQL was going make during the coming year, so we just recently assembled the team for another review (early 2007). In summary, we’re sticking with PostgreSQL.I’m less concerned with Lumen’s final vote than with the process that led up to it: open source enables a truly competitive market where IT buyers can first be IT trialers. Lumen didn’t need to make a decision based on marketing gimmicks, a sophisticated demo, etc. It chose its database based on actual, hands-on experience. Can you do this with proprietary software? Yes, but proprietary companies aren’t in the business of giving away free licenses…. Open source provides a fundamentally different way of finding, testing, evaluating, and then deciding on and using software. Lumen can take advantage of the LAPP stack, with a “P” for PostreSQL, if that’s what serves its customers best. The LAMP stack, with an “M” for MySQL, is ubiquitous these days, but only because people try it out and decide on the merits. Open source is about choice. Informed choice. Whatever the ultimate decision. Open Source