The last chance to save MySQL from Oracle's clutches is for antitrust authorities to block the deal unless Oracle makes real (not fake) concessions.

Many of you in the community will not be very familiar with antitrust law and merger control. Conversely, antitrust regulators are not familiar with free and open source software, generally speaking. Oracle has tried to use both circumstances to its advantage. It's time to inform.

A big transaction like Oracle buying Sun needs to be approved by regulatory authorities, not only in the countries where those companies have their legal domiciles and headquarters but also in all other countries where they do a significant amount of business. Of course, there are small countries that don't even have any regulator for this kind of thing, and there are others who do but global players would go ahead with their mergers and acquisitions anyway because if it came to worst, they could still abandon some smaller market somewhere (which would hurt those small countries more than the big global players).

There are some really big ones that Oracle certainly depends on. The European Union is even bigger than the US in terms of Gross Domestic Product and population size. China is expected to become the world's number one economy in the foreseeable future. Russia is a big powerful country with natural resources and a very educated population. Those were just the three most important examples.

If the regulator of any major market like that says No to the deal, Oracle will practically have to give up Sun. We don't want to prevent Sun from having a future under Oracle, but the fact that regulators can say No gives them the leverage they need to tell Larry Ellison (Oracle's CEO): "Give up MySQL or you get no deal."

In a legal sense, they wouldn't say it like that. There's a common misunderstanding or oversimplification here. Antitrust regulators can't order companies in a merger situation to do any particular thing. But they can look at what those companies propose and if it's harmful to the market, they can say No unless the notifying party (in this case, Oracle) makes a satisfactory proposal. If Oracle makes no such proposal in time, a regulator can only say Yes or No to the deal as a whole. So the ball is always primarily in the corner of the company that wants to get approval. The regulator won't say "do this, do that" but it can say "this isn't good enough, you have to come back with something acceptable".

Regulators are rightly and seriously concerned over Oracle's desire to buy MySQL. The European Commission issued a Statement of Objections on November 9, 2009. It had 155 pages according to Bloomberg, explaining in detail why customers would be likely to suffer harm in terms of less innovation and higher prices if Oracle gained control over MySQL. In late October 2009 Oracle felt forced to withdraw its antitrust application in Russia, as such news agencies as Dow Jones reported.

Oracle claims there's no antitrust case because, according to Oracle, MySQL is primarily a web and low-end database that can't really compete with Oracle's top-range products. But that's not true. Of course Oracle can offer some features that MySQL doesn't have. As explained on another page here, that still doesn't mean that they don't compete in significant areas. The European Commission's deputy director-general said at a conference that the EU has evidence that MySQL is used not only for low-end and web purposes but also for "complex transactional databases".

Oracle also says that MySQL doesn't have a lot of market share. That claim is made only on the basis of revenues but the key thing is that MySQL is so widely used, probably the most widely used database in the world with an estimated 20 million installations and exerts huge pressure on Oracle because of its open source approach, being available potentially free of charge for many purposes and even if someone pays for different licenses or premium offerings, it's very competitive. The aforementioned EU official pointed out in a public speech that MySQL is far more important in the market than its revenues suggest. This guest blog article explains why in today's digital markets a product or service can be extremely relevant even without making any money at all (such as Twitter so far) or be so extraordinarily aggressively priced that its importance is hugely greater than its revenues (just think of Skype).

In addition to its pricing structure, MySQL's flexible architecture allowing pluggable database storage engines is another key factor to consider.

Oracle seems to know that its arguments about MySQL not being competitive are weak. That's why Oracle says that even if there is a problem, MySQL being available under the GPL is enough to take care of it. That fallacy is addressed on this page or you can also read this blog article about it.

This campaign here will go on until every important market has taken its decision on the deal. There's no such thing as a done deal until the formal decisions have been made. Regulators can make announcements in public about their intentions but if they have a real decision, they will publish it, and until then, let's stand together as a community and fight for MySQL's future.

In the EU, the situation is particularly interesting because Oracle was given a deadline (even extended by six working days) to propose a formal solution. That deadline expired on midnight of December 14. Oracle made certain vague promises in a press release but those are neither a formal proposal (for which they would have to file a particular document, called "Form RM", with the European Commission) nor a solution (because even if formally proposed, those promises would mean the end for MySQL-related innovation and would just make the suffering slightly easier for some of the affected parties).

Oracle had been adamant all the time about not proposing anything real and serious and has let an important legal deadline go by without doing what it should have done. As a consequence of that behavior, the European Commission and other regulators actually have a legal and moral basis to block the whole acquisition of Sun now. Obviously, it would still be preferable to find a solution that allows Oracle to buy the rest of Sun, and the EC as well as other regulators could always find ways to interrupt the process and work out a true solution with Oracle. But if all else fails, meaning if Oracle only offers things that are useless anyway, then MySQL should be considered sufficiently important to the market that Oracle would not be allowed to buy Sun, as a last resort.

Oracle has mobilized hundreds of major customers to support the deal and to downplay the importance of MySQL. Now we need your immediate help, and the help of your friends and their friends, to show to regulators around the world (and to the press, which will play a key role here) how much MySQL really matters. If we achieve that, then we can still win. Oracle has used its means to make pressure, now we have to use ours. Please help. Now.