Mysql Triggers Turn on the Heat
Peter Gulutzan wrote an interesting article describing his Mysql Triggers Tryout with the latest Alpha build of MySQL 5. I've been a long-time fan of MySQL. I first started using it in 1998, and have watched with great interest as it has matured. It seems that it is working very hard to match the features of the "enterprise" level SQL databases such as Oracle and MS SQL. For quite some time, many enterprise database (MS SQL, Oracle, etc.) advocates have drawn attention to the lack of features such as triggers and stored procedures in the MySQL database. Soon, the lines between the features of these competing databases will narrow.
One of my favorite things about MySQL in the past has been its speed and its simplicity. Could all of these new features be a pandoras box of instability and performance drains? In trying to compete with the larger databases, I hope that the database that I "grew up" with doesn't loose its identity, its speed, and its stellar performance. The way I see it, not every application needs the power to run join queries on a 3 Terabyte sized database. For small to medium sized database applications (most web apps), speed and simplicity are paramount. In these smaller applications, good database design can almost always compensate for the lack of some of the advanced features offered by the more advanced databases. As MySQL bridges into the enterprise DB space, will these key benefits remain intact?
I wonder what kind of an ultimate impact the "improvement of MySQL" will have on the Database market, as well as on MySQL itself. MySQL is an open source database, but it can be used either under the GPL or under a commercial license. The commercial license is not free, but gives the purchaser assurances and frees them from some of the limitations of GPL (most notably, doesn't require the licensee to distribute their application under the GPL). As the competition from MySQL picks up steam (and it already is), perhaps the other vendors will allow a free license for their databases under specific circumstances similar to MySQL licensing. The challenge is that the database market is a very high revenue generator for both Microsoft and Oracle, so there is a great deal at stake for both companies. At the same time, if MySQL takes over a significant portion of the market will their licenses change to better take advantage of their improved position and give them more cash? Will the approach of MySQL cause the enterprise databases to invent new innovations to differentiate themselves from the growing competition from MySQL?
Needless to say, it is going to be a battle ahead for the database players. There is a great deal at stake, so I'm sure the battles will be hard fought. Unlike Linux, the MySQL community centers around a private "for profit" company funded by venture capitalists. This could give a great deal of stability, legitimacy, and force-of-will to the MySQL movement; Or it is possible that the need for profits will cause a large rift with its Open Source (i.e. "We want everything to be free") champions. Only time will tell.
*Disclaimer* This posting represents the opinion of the author, and not any of the parties mentioned herein or the opinion of my employer. This post is also subject to the default disclaimer of my blog.
