Wednesday, November 11, 2009

The EU's real problem: MySQL and Oracle do not compete!

I think that most people are missing the point, Oracle included. The main objection of the EU is not that Oracle is swallowing up a major competitor.

To understand this you have to read between the lines of the EU decision:

"The regulators see a major conflict of interest in the world's largest commercial database company owning its largest open-source competitor"

This should actually read: "the world's largest commercial database company owning the largest open-source database"

The database market is divided into 2 parts: the back-office and the online world.

And now you know what I am going to say ... Oracle has an near monopoly in back-office and MySQL has a monopoly in online applications.

So let's do a little maths:

If we assume that back-office and online applications divide the database market into 2 equal parts, and that Oracle owns 60% of the back-office, and MySQL 90% of the online world.

This means that Oracle controls 30% (60% of 50%) of the entire database market today, but after the acquisition this number will be 75% (30% + 90% of 50%).

Something to think about.

8 comments:

Tony Bain said...

Hi Paul, to me your argument seems to support Oracle rather than the EC.

It is not illegal for a company to be dominant in two different market segments. However, bodies like the EC exist to ensure competition isn’t removed from a market through acquisition. By stating that Oracle & MySQL are in different markets, then the competitive landscape remains the same post acquisition. To me this would suggest there is no reason to block such a deal taking place.

The idea the Oracle could use this position to force people to purchase Oracle products by removing MySQL from the “online” market seems unsubstantiated. There still remains plenty of choice including a number of alternative open source offerings as well as the MySQL forks of course. It isn’t clear how this could be achieved.

Mark Callaghan said...

Thanks for describing the issue in an objective manner.

What fraction of deployments in the online market are also customers of MySQL? What do they have at risk? Many of them are able or willing, but not always able and willing, to use MySQL without becoming a customer. Do non-paying users deserve the same level of protection as paying users?

Unknown said...

The math is also weird. Even if MySQL did have 90% of the "online" market (by which I assume you mean web sites), websites do not represent 50% of overall database usage. I doubt if websites represent more than a single digit of database usage. So 50% of 5% does not improve Oracle's overall market share much at all.

Paul McCullagh said...

Hi Marc,

Some interesting questions, and as you may suspect, I cannot really answer most of them.

Basically the point of this blog was to say the following:

I believe the EU is more concerned about Oracle building a monopoly in the database market than the lack of competition as a result of the acquisition.

If the job of the EU commission is to prevent monopolies, then I think they are correct with the objection. If the commission is just their to ensure adequate competition in the database market, then they are out of line.

Many say: "But it is open source, nobody can control it". I don't agree.

The evidence makes it clear that this is absolutely not the case:

1. What did Sun buy for $1 billion?

2. What is it that Oracle wants so badly, that they risk the Sun deal?

3. Why did Oracle buy InnoDB so many years ago?

Firstly, it is about ownership. And most important: ownership means control.

And MySQL can be owned, unlike other open source projects like Linux and PostgreSQL where the copyrights and trademarks are not all in one hand.

And control of the market is what Oracle is after.

They will fight the EU decision with everything they have because this control is worth one heck of a lot!

Anonymous said...

Hi,

I support your opinion. Even though Oracle and MySQL are used in different segments of the database market, the border of the segments is highly likely to blur in near future. Because MySQL is used as a web back-end DB, while Enterprise application is getting built as a web app.

In addition, while they are used in different segments, but they are both DB products. Some functionalities are overlapped.

Oracle acquiring MySQL is similar to that the top desktop OS vendor acquires the top mobile OS vendor. Imagine the world where MS dominates both desktop and mobile OSs. No one can escape from or break such a monopoly.

Paul McCullagh said...

Some very good points from Anonymous.

Could this be a Sun employee who is not actually permitted to comment ... ?

Mark Callaghan said...

Paul,

I hope the EC does something other than fight. If they do fight with Oracle there will be nothing left of Sun/MySQL at the end. I am not sure if that is a good outcome for customers.

If this about a market in which MySQL has a monopoly, then we are debating a market in which most of the users don't pay. Are they still consumers? Are businesses obligated to fund products that are not profitable? I suspect this is a new issue for them and all of the economists who are analyzing the issue.

Anyone claiming that MySQL must be freely available in the future must also claim that the user-base must pay enough to support the investment.


Oracle will own the copyright, trademark, docs and bug database when this is done. But I hope the EC's case is stronger than debating whether or not Oracle would control MySQL. Maybe that is good for the press, but the issue is much more technical. They need to define a market and show the size of the market.

The unanswered questions you list as evidence make your case stronger only when you speculate with bad intent on the part of Oracle. I won't engage on those and I hope the EC has something better than that.

The comparison to Postgres is interesting. The GPL version of MySQL is not much different than Postgres with one exception. Development would be done by a small community and somehow time and money would have to be found for the effort. The exception is that some businesses in the Postgres community use it because of BSD. But from my experience with Postgres those companies don't contribute much back in comparison to the community effort. I think the same is true for the non-GPL storage engine vendors with two exceptions -- Infobright and Calpont. So if you are all for community Postgres, I don't understand why you aren't all for community (GPL) MySQL.

In terms of delaying this deal I don't care about those who need a non-GPL license to MySQL. If we are to talk about markets, that is close to non-existent compared to other usage. And I suspect that doesn't count for much of the revenue at MySQL and the total revenue is also not much. So if we claim that MySQL has a huge market, and we must to delay the deal, then I don't get how this motivates delaying it.

I also don't care much for the concerns of those who require non-GPL MySQL because in doing so they want MySQL to be made available to them, but they don't want their source available to others. I think that is unreasonable.

Paul McCullagh said...

Mark,

As you mention, technically/economically the argument is hard to make.

But, I do think I am putting my finger on the core of the matter.

I am just trying to get into Larry's mind...

I agree with you on the insignificance of the non-GPL version. Although it burnt Oracle and other database developers to loose this business, the total market must be small in comparison the entire database market.

So why does Larry want it so badly if it is just copyright's and bug database, which are useless because everyone gets it for free anyway?

I am definitely not speculating with the bad intent of Oracle. I actually believe Oracle will do a good job, and that the community will spur them on and help. As we have seen this happen with InnoDB (you yourself have played a major role in this).

So what does Oracle want? They want you and everyone else in the world to do business with them, even if they don't earn a cent from the "business".

In other words, if it has to be free, then Oracle would rather be the provider than allow the competition a foot in the door.

As you know, Google, Skype and YouTube have shown the value of users that don't pay. So yes, I think that we need not distinguish between paying and non-paying users as far as market economics is concerned.

All users are points of contact that can lead to an introduction to other produces and services. So MySQL extends the reach of Oracle into the database market, and we can practically ignore the fact that the product is free.

And, this is a fact, whether I am for or against community (GPL) MySQL. Think of it this way, if ownership was irrelevant to influence, then Oracle could have forked and made its own version of MySQL a long time ago. The purchase of InnoDB would have been a good time for this.