The news that the European Commission thinks Oracle should jettison MySQL as part of its deal to acquire Sun is a typical case of bad analysis yielding potentially good results. I have to agree with Oracle’s contention that MySQL + Oracle DBMS does not constitute an unfairly competitive combination, and the EU’s perception to the contrary proves Oracle’s point Brussels doesn’t understand Open Source, or the database market, for that matter.
I also have to agree with the EU that selling MySQL would be a good resolution to the problem, but I also feel that letting Oracle hang on to MySQL would have pretty much the effect that the EU is looking for, albeit not exactly in the manner the EU would have liked.
Here’s how I see it. If Oracle hangs on to MySQL, which has carved out a niche in the new world of Web-based businesses, two things will happen. Oracle will work day and night to convert those MySQL customers to a cloud-based Oracle DBMS environment, and basically try to pick off as many MySQL customers as possible. Meanwhile, Oracle will be happy to collect service and support fees from those customers that remain in the MySQL camp.
Over the long run, this will produce the same effect as Oracle’s acquisition of the old Rdb database from Digital Equipment Corp., or IBM’s acquisition of Informix: dead, gone, and forgotten, with no impact on the top vendors’ market share.
But, don’t forget, this doesn’t actually mean that MySQL has to die: it’s an open source database, which means there are no end of suckers – oops, I mean dedicated database experts – willing to work for free to better a product that drives a significant, VC-funded or publicly-traded company-based service and maintenance business. There will always be open sourcers willing to help make for-profit companies successful, and MySQL, even if it is totally neglected by Oracle, will likely still progress through a reasonable innovation cycle.
Even if Oracle could manage to kill MySQL, there’s plenty of other open source DBMSes to take its place. Ingres and PostgreSQL, among many others. So, open source will continue to thrive, the for-profit service and support vendors for open source will continue to make suckers – I mean heroes – out of the free labor they derive from the open source movement, and not much will change, except the names of the contenders for top open source database.
Just to fill out my thoughts, I also think that Oracle could and probably should jettison MySQL if that’s the only way to get the EU to play ball. Jettisoning the database doesn’t mean jettisoning a service and support business for MySQL (a la Red Hat), which is anyway where the money is in this open source “business”. And, while providing high levels of support and service to MySQL customers, Oracle could pursue its merry conversion business unimpeded by regulators or the messiness of corralling open sourcers to do high-value, enterprise-class work.
So, successful if they do, successful if they don’t, the current flap with the EU is just that, a flap. Oracle can and will prevail in its efforts to buy Sun, and, whether MySQL comes with it or not, Oracle can and will start the process of moving MySQL customers to a cloud-based DBMS world. It’s not a matter of if, just when.
PS: to all of you open source database developers offended by this column, please bear in mind I have your best interests in mind. Stop working for free to make VCs and shareholders rich, your work deserves honest remuneration, not exploitation.
(Cross-posted @ Enterprise Matters)
MySQL is too important a community asset to let a MNC own it. The EU understands that its role is protect freedom from corporate exploitation, and to keep open source in the public domain. If Oracle owned MySQL, they could immediately remove it from hard drives across the world, and THAT’s NOT OK.
“… ….or IBM’s acquisition of Informix: dead, gone, and forgotten, …” you must be kidding right? or is it ignorance?
RIan,
I guess my overly strong language probably deserved yours. I’m neither kidding nor ignorant: Informix is a shadow of a shadow of its former self in the database market, thanks to IBM’s disinterest in doing much beyond maintaining the brand and supporting existing customers. Okay, it’s not dead. But being among the living dead isn’t much better either.
Agree on the MySQL analysis. I agree with Rian Puna on your Informix position. IBM is continuing to enhance Informix. It’s like saying IMS is dead. Both Informix and IMS are doing quite nicely for IBM.
Colin,
You and I were both there in the heyday, when Informix was an independent company and it contended head to head with Oracle, Sybase, Ingres, etc. Where I sit in the enterprise software space, I never hear Informix mentioned by a prospect on a short list for a new DBMS deployment. IBM may keep poking at Informix, but it’s not in the limelight of the market.
Josh
I’m missing highway humor… like this classic, showing the Oracle HQ as background:
“You just passed Redwood Shores. So did we. Informix”
Priceless. Too bad there was no Flickr or even digital cameras to preserve it.
Josh,
Actually, I get paid pretty well to work on open source technology. So do most OSS developers. There’s some kind of myth that most OSS developers are working in their spare time, for free; in reality, the OSS developers who write the most code generally collect generous salaries or are getting equity.
Of course, there are certainly OSS developers who get taken advantage of by VCs, but there are proprietary software developers who get taken advantage of as well, and business people, marketeers, etc. It’s how VC works, and you just have to understand the Faustian bargain you’re signing when you take it.
Josh,
Every briefing I’ve ever had with a venture-backed open source apps company has included the prospect of an army of developers willing to work for free. It’s true some of you are drawing salaries for your work, but the model has always been postulated at the VC level on some degree of free labor.
Josh
I think you just don’t get open source.
The majority of open source development I’ve done has been salaried – fixing a bug, or adding a feature to an existing tool that my company uses. It’s never a large amount of work, and pretty much always cheaper than buying a commercial product instead of using the open source equivalent and having to get stuck in. I win, my company wins, and the open source project wins.
I do also occasionally do this sort of thing in my free time, and again, I do it because I want the tools I use to be better, and I’m not paying for them – so why expect someone else to fix them for me?
Finally, you can’t use proposals to VCs as evidence of the truth. Proposals to VCs are only as close to the truth as they have to be. If open source development is exploitative, I don’t feel exploited.