|
|
MySQL's story is one of immense success. Initially a blazingly fast but loose database, it grew through user involvement to become the preeminent open-source relational database of this world. But now, I believe, it is nearing its highpoint and will eventually trail out, being overtaken by other open-source relational databases. I'm not saying it will go away---it is too entrenched for this---but it will loose significance as more and more new projects start using other databases.
MySQL was a database whose time had come with the Internet. It was fast and simple, ideal for the humongous amount of Internet and Intranet websites being developed over the last 8-10 years or so. Today, it is a mature database that offers much more than what most of its users will ever need. Think "Innovator's Dilemma": MySQL fulfilled a need, and now has outgrown its common user base. Efforts for MySQL clustering show that MySQL (the company) is going for higher-margin business, where they can charge more than the $250 they ask commercial users each year if they want to use their database.
MySQL is based on a dual-license model. Non-commercial use is free and handled by the GPL license. Commercial use costs $250 per year which includes support. MySQL never accepted outside code contributions unless those contributing where willing to sign away their copyright to their code. Being free for non-commercial use gave MySQL a lot of early feedback that helped them improve the database, but the more restrictive commercial license prevented them from capitalizing on the true power of open-source, the code contributions of individuals trying to extend the basic database for their and other people's needs.
Its competitors have caught up. PostgreSQL reputation has been improving consistently, and even Firebird is out in the open gathering a following. Both databases are provided under a non-GPL license (Mozilla, BSD) and are therefore less restrictive than MySQL. If you were to choose from three different databases, each supported by their own community, each supported by commercial services, all of them satisfying your needs, which database would you choose? Which one would you rule out? Unless you have special requirements, MySQL is out. Why would you want to support a database that makes a buck on your behalf?
Recently, MySQL, the company, took on venture capital. It shows. If you look at their website, you get all kinds of one-year-license options, ranging from the basic $250 to multi-thousand dollar licenses. Why? Because they can. Why? Because in traditional terms, they are worth it. Why? Because venture capital requires a substantial return on investment, and MySQL will have to deliver. VCs may be a little more patient these days than during the heydays of the Internet bubble, but MySQL's clock is ticking. I think the VCs got a bad deal. MySQL's growth rate will be declining and hence they are most likely not to reach the valuation they have been banking on.
Will any of the commercial databases like Oracle, DB2, or MySQL go away because of PostgreSQL or Firebird? Certainly not. They are too entrenched. But over time, they will be pushed into the lower-volume higher-margin business. As Christensen has shown in "The Innovator's Dilemma" and "The Innovator's Solution" this is not a place you want to be. Entrenched or not, you don't want to become a dinosaur.
|
|
|