Thoughts on the Oracle/Sun merger
The Oracle/Sun merger is old news, even the approval of it is old news. All the same, I am a bit surprised by how little people fought against it. Look, if it was Microsoft buying Sun, there would be picketing outside the SEC by a zillion people begging for the deal to be halted, on the basis of monopoly. But somehow the protests seemed rather muted. In any event, it is now a done deal, so let’s look at the results.
With the acquisition of Sun, Oracle has just made an end run around IBM to become the big Java shop out there. The big difference is, Oracle hasn’t capitalized well on Java so far. All of their Java related offerings fall under “also ran”, “never was”, and “wannabe”. In fact, the same could be said for just about everything else that Oracle makes, other than their core database product. They are a lot like Sun, in that regard. Other than Java, has Sun been a market leader in anything for the last ten or so years? Nope. That’s why they were going under. Oracle stays in business by extracting gigantic license fees from their mostly unwilling victims… I mean… customers. Given the choice, which do you prefer? Oracle, or Microsoft SQL Server, MySQL, or PostgreSQL? Probably not Oracle. Would you rather deal with Solaris or Linux, a BSD, or Windows? Probably not Solaris. Which makes more sense, using “real” Apache or IIS, or Oracle’s hacked up version which is always ages behind? And so on.
Another thing the two companies have in common in the utter ugliness of their products, in terms of living with them. The only installer worse than Solaris’ is the Oracle installer. You know, the installer that for years puts the wrong permissions on the files it installs under Windows, and makes you wrestle with a million Java errors just to get the install going. I have never met someone who actually said, “I like Oracle”. Every single person I spoke to despises it. The only positive thing people have to say about it, is that it is so hard to use that they can justify charging $250/hour as an Oracle consultant. The situation is about the same for Solaris, except few people will be a consultant for a whole OS; instead, Solaris admins get paid such a premium that no one wants Solaris in their server room.
So what we have here is the merger of two really ugly parents who have some good looking kids with behavioral problems (I’m thinking of a mix of The Brady Bunch and House of 1000 Corpses).
And then there is the MySQL end of things. Oracle just went from being one of three major DB players to two of three. That’s pretty scary, and almost makes me want to buy a SQL Server license just to keep things fair. All joking aside, it says a lot about MySQL that folks are considering it just as big of a prize, if not bigger than Java. The fact is, Sun messed up with Java pretty bad. When they kept the development close to themselves, they did the usual “Sun’s magic touch” where they created a marvel of technical wizardry that met no one’s needs in reality. Somehow, Java survived long enough for folks like Spring and Apache to bring enough inovation to the table to make Java an attractive platform, despite all of its problems. Meanwhile, Microsoft, the 80,000 lb. gorilla, has been able to do enough interesting things with .NET that there is relatively little third-party open source for .NET because there just is not enough necessity to birth any invention. Luckily for Oracle, Sun didn’t have MySQL long enough to wreck it.
All the same, one has to wonder what will happen to MySQL with the awful combination of Sun and Oracle’s collective ineptitude when it comes to customer satisfaction. I would not be surprised to see a fork of MySQL at some point, but at the same time, you have to realize that very few people work on MySQL for fun or in the spirit of open source, relative to the number of developers working on it. It’s just one of those projects where people are paid to work on it, because everyone hopes to exploit it somehow. Hopefully, enough people will consider MySQL to be a critical piece of infrastructure (and it is, despite all of its flaws) that they’ll manage to keep Oracle from trashing it or reserving all of the important features for paying customers.
J.Ja
