Why Oracle Overpaid for Sun
I'm sure that Oracle's purchase of Sun is mostly about MySQL - they now own all the pieces of the database (they had previously bought the most commonly used storage engine, InnoDB).
Control is an interesting idea with a GPL'd product though, and I think Matt Mullenweg's comments about MySQL (in the context of how WordPress uses it) illustrate the problem quite well:
Today our servers are running various versions of MySQL, tomorrow they'll be running the same thing, and if need be ten years from now they can run the exact some software. Because of the GPL every WordPress user in the world is protected -- we're not beholden to any one company, only to what works best for us. Today that's MySQL, tomorrow that's MySQL, a year from now we'll see.
Later on he mentions how large the third party maket for MySQL support and add ons is. Combine that fact with the GPL license, and you find that Oracle now "owns".... well, nothing much. Anyone can grab the bits, fork a new version, and stay on the GPL. So if the user community starts getting antsy about Oracle, they can go their own way.
This is why I thought Sun's purchase of the db was dumb, and it's why I think Oracle's purchase is dumb. They could have achieved the same thing by offering stupid amounts of money to the top, say, 10 MySQL core developers. It would be costing them in the low tens of millions to do that...
Comments
not only MySQL
[HKN] April 21, 2009 11:04:13.736
first of all I don't think that MySQL is the only reason why Oracle bought Sun.
The number one reason was probably not letting its rival IBM get total control over Java, a crucial base technology of Orcale's offerings. IBM more or less owns Java already today via the IBM initiated and funded Eclipse project. If IBM would have bought the IP of Java per-se they would have had the sole power over it.
Second even MySQL alone can be good business for Oracle: Sun did stick to the strict dual licensing policy of the original MySQL company with GPL for FOSS projects and a commercial closed source license for all those who do not wish to open source their aapplication. Since it is GPL and not LGPL there is no way to fork a branch competing for these revenues.
[Gordon Weakliem] April 21, 2009 11:23:59.413
I tend to agree with HKN that letting IBM have Java would create a hegemony in the Java world. I think that Java was the bigger prize for IBM. On top of that, while the "offer stupid amounts of money to MySQL devs" sounds nice in theory, in practice I have to wonder how many of those guys can be bought. I'm sure they're happy to have Oracle paying their salary - IF they can still run MySQL their way and not be just another cog in the machine. I don't know any of those guys personally, but knowing the type of person, I'd imagine that their tolerance for corporate BS is pretty low. My feeling with Sun is they were given a lot of leeway to run things their way, we'll see what Oracle does.
Why?
[Mike Wilson] April 21, 2009 11:44:38.968
uhm... why are you sure of that? The notion that mysql would be a reason to purchase sun is fairly thoroughly debunked in your next couple paragraphs (a point you make well.)
You don't think Java might be a bit more important perhaps?
Perhaps if you didn't post a comment every 5 minutes, you could think before writing
[jimmy james] April 21, 2009 13:26:16.853
"I'm sure that Oracle's purchase of Sun is mostly about MySQL" As others point out, that's all wrong.
Re: Why Oracle Overpaid for Sun
[ James Robertson] April 21, 2009 15:57:59.989
Comment by James Robertson
What positive does Oracle get from owning Java? They already have a services business using it; now they just get to add the maintenance cost. I'd call IBM the winner here. They get to keep using Java, while someone else pays the bills.
Yes, MySQL Is The Main Thing
[W^L+] April 22, 2009 0:19:24.546
First of all, MySQL has been a thorn in the side of Oracle and SQL Server's sales for several years. DB2 also. This is very likely why Oracle bought Sleepycat, the maker of Berkeley DB. BDB used to be one of the transactional engines for MySQL. Next, Oracle bought Innobase, the developer of the InnoDB transactional engine used in MySQL. At the time, it was widely said that Oracle was trying to keep MySQL from underpricing their namesake Oracle database. It was also said that MySQL had offered to by Innobase to keep it out of Oracle's hands.
Shortly after the Inno deal, there was a lot of activity around developing MySQL-owned transactional engines (e.g., Falcon) in order to counter expected pressure from Oracle (offering InnoDB as part of their dual-licensing scheme requires that Inno's licensing allows that).
As for MySQL and the GPL, remember that MySQL's development is mostly in-house. This isn't like Mambo, where outside developers were part of the team and could easily leave to form Joomla!. As HKN notes, a commercial fork would presumably not be able to dual-license the codebase. Meanwhile, the in-house development model prevented MySQL from ever getting a development community, so non-commercial forking will take months or years to build up enough to move forward.
For concerned LAMPers, it is wait and see right now, but a potential rewrite for PostgreSQL or Firebird looms in the future.
Anyway, as I said, Oracle has been after MySQL for some time, primarily (in my opinion) as a way to shore up their offering on the lower-cost end of the market.
Then Sun's purchase was shrewd
[George] April 22, 2009 11:26:12.521
Jim,
Oracle thought Sun with mySQL was worth $5.6 billion. You think a lot of that value (certainly well over $1 billion) was because Oracle wanted mySQL. So you're forced to conclude that Sun's $1 billion purchase of mySQL was, in fact, a good idea.
Sun figured there was a greater fool, you figured there wasn't, and Sun was right. (This is, of course, independent of how much mySQL is "actually worth" — a thing is worth what someone will pay for it, no more and no less.)