The MySQL Story
By edort on Feb 27, 2008
You've heard the story. Two young guys have an idea about computing, turn it into something tangible, become wildly successful, and change the computing culture in the process. No, I'm not talking about the story of Apple, Inc. I'm talking about the story of MySQL. When MySQL co-founders David Axmark and Michael "Monty" Widenius started writing code, they were teenagers. They wanted to make some money on what they did, but had no inkling of the tremendous impact their work would have.
In his keynote talk kicking off Day 2 of the Sun Tech Days event in Hyderabad India, Axmark recalled that the first database code that he and Widenius worked on was written in 1995. That code would ultimately become part of the world's leading open source database management system, MySQL. Axmark said that when the code was initially released as a usable package in 1996 it immediately drew a lot of interest. "In the space of a few months there were several hundred people on the mailing list and thousands of downloads. It was amazing."
Thousands of downloads is certainly a good start for two young and ambitious developers, but it doesn't necessarily imply a wildly successful outcome. Today's reality is that 66% of the world's enterprises use or are planning to use MySQL in their applications and some of the most popular sites on the Web, including Google, Wikipedia, and Facebook, use MySQL. It's a light year's jump from the early days. So huge a jump that Sun Microsystems recently bought MySQL AB, the company that owns MySQL, for $1 Billion. Axmark asked the rhetorical question "How did we go from that humble start to this pretty hefty price tag?" The answer is simple: innovation. In the talk, Axmark outlined many of the innovations in MySQL, some of them made by people who worked for MySQL AB, but many created by the community.
Those innovations were fostered by some fundamental philosophies. Axmark stressed that from the beginning lots of people working on MySQL really cared about databases, especially databases for web sites. There was also more focus from the very beginning on implementing a few features that worked well and were supported well compared to lots of features that almost worked. Ease of installation also was ingrained early into the work ethic. Axmark said "We had this 15 minute rule. Back in 1996 we tried to make a package that would install and compile in 15 minutes. We always had the idea that if you wanted to get MySQL and try it out, and if you spent a couple of minutes and got it working, you might use it 10 years later. If you didn't get it working in the first 15 minutes, you're likely to go somewhere else."
Features like optimization for the Web made MySQL a natural for today's Internet sites. In fact, Axmark pointed out that MySQL has always had a faster connect time than traditional database systems. This is very important for web applications, considering that a web application needs to connect to a database much more often than a traditional application.
Some of those features took on a life of their own. MySQLProxy is a programmable layer in MySQL between the client and the server. Axmark noted that people came up with so many ways to run this feature that it became a product in its own right.
But it's the huge community of MySQL users that really does the heavy lifting. Axmark showed a chart the listed all of the languages that MySQL supports -- it's a long list -- however, only a few of those langauges are maintained by MySQL AB. The rest are maintained by the community. Axmark also noted that this thriving community enables a lot of significant advances. He cited the case of a student who developed a JDBC driver for MySQL. Not only did the driver work, but it outperformed commercial JDBC drivers in competitive tests. MySQL AB knew it had a good thing going and hired that student. In fact, one of the benefits of an engaged community is that it identifies the people who have the skills that MySQL AB needs. Axmark stated that "We can hire people who have shown that they already know the code."
Axmark and Widenius took a big business risk in the early days. "Back in 1996, we decided to do business in a very different way." That way was to give away everything for people who wanted to use MySQL internally, but charge for people who wanted to include MySQL in their closed source solutions. Clearly that risk payed big dividends to the founders, their team, and to the community. Axmark stressed that a free database system such as MySQL gets better all the time because the community is attuned to coming up with fixes to problems, fixes that go into a public bug database. Some other payoffs are faster developer feedback than closed source, testing by the community of all the code, and a low total cost of ownership.
With a $1 Billion buyout, you would think Axmark would be worry free, but he ended this engaging talk by saying "My biggest challenge now is working in a huge company [that is, Sun]." No doubt Sun is very happy that he's here.
Learn more about Sun Tech Days.