Tuesday, November 17, 2009

Can Oracle Kill MySQL?

Yes

Because MySQL is open source, Oracle cannot cease development without effectively handing the project's reigns to someone else, but Oracle can easily turn MySQL's development arc so it never intersects that of its flagship product. If the EU recognizes the Oracle-MySQL merger, the database industry leader could not only quash the potential of MySQL's open source version, but it could create a no-cost commercial version with a feature gap sufficient to kill development of earlier forks.




Everyone I know who is watching the EU vs. Oracle battle over the future of MySQL wants to know the answer to one question: Can Oracle Kill MySQL?

Right now, this is the single most relevant question the human race can ask about its future. We will either have a hierarchical global economy or a neural global economy depending on whether we, the people, can create something neural before the elites create something hierarchical. Those of you who are fighting for no global economy are unwittingly helping the elites.

An economic system requires communication, distribution and accounting. An entity that can keep you from achieving one of those three requirements has control over your economy.

Communication: Three weeks ago, the FCC declared itself to be an enemy of the United States by voting to grant itself the right to control communication on the most popular brand of global network. The Internet brand consists of an insignificant "backbone" (less than 1% of global connectivity) controlled by Washington DC and a wide array of private networks that agree to use the protocols of Washington DC's "backbone." Some of those networks also use different protocols. Washington DC hates it when a private network uses a different protocol because it makes it hard to spy on the users of that network.
The FCC is illegally claiming authority to revoke communication licenses unless private networks exclusively use protocols approved by Washington DC. AT&T, Verizon, Comcast and Charter telecoms are willing to comply with the FCC's illegal coercion. They have taken over the "last mile" of wire running from the homes of most citizens of the States to the web of private networks. In order to restore uninhibited communication, we are going to have to connect directly with each other. We can no longer rely on the big telecoms to connect the "last mile" of wire to our homes because they can be coerced through Washington DC's licensing scam.

Distribution:
Under the guise of controlling unregistered immigrants, the Fed has placed hundreds of customs checkpoints in an area one hundred miles wide around the perimeter of the United States. You can google hundreds of stories and videos demonstrating that the Fed's "border patrol" is much more interested in trying to control the "black market" (trade without Fed Notes) than unregistered immigrants or drugs. In order to restore uninhibited distribution, we're simply going to have to look beyond the fancy uniform of the Fed's "border patrol" to see them for the criminals they are. They have no legal authority whatsoever. If they attack you, they are committing a criminal act and you may legally defend yourself by whatever means you deem necessary


Accounting: Beyond uninhibited distribution and communication, a global neural economy also requires an uninhibited scalable method of detailed and automated accounting. By 2005, our good friends in Sweden had provided us with a robust open source relational database foundation for such accounting: MySQL. Unlike most open source projects, however, the Swedes did not fully give MySQL to the world. They retained ownership of the software under a license that let them do whatever they wanted with it while granting a second license to the open source community. In other words, all derivatives of the free copy of the software must remain open, but the Swedes could make independent derivatives of the copy they kept that they could close if they so desired. When you take brand name recognition into consideration and the fact that MySQL relied on third party storage software (that Oracle started buying up in 2005), it became feasible that the owner of MySQL could effectively kill the product or make an advanced, stagnant commercial fork more attractive. 

The potential for disaster enabled by MySQL's joint licensing was disconcerting to some developers who gave their time freely to promote the project's evolution, but these coders felt confident the Swedes would never sell out to what they called "the evil Oracle," MySQL's only viable competitor after it had surpassed Microsoft and IBM in both its growth curve and number of installations. Instead, the Swedes sold MySQL to Sun Microsystems, a behemoth legacy computing company in its death throes that had nothing to gain by purchasing an open source relational database company except to make itself more attractive for purchasing.

Sure enough, Oracle immediately found Sun Microsystems to be very attractive and offered to buy the realistically worthless behemoth for billions of dollars. Yet, Oracle now promises that it is not shelling out all this dough for the sole purpose of killing MySQL. Its founder, Larry Ellison may have been the richest man in the world before MySQL cut into his massive relational database monopoly from the bottom, but Ellison assures us that his intent is not to kill the world's only viable hope for future Oracle competition. Let's see if Oracle's stock holders believe him.

Oracle and Sun officially announced their pending marriage on April 20, 2009, but they had been discussing the idea, making agreements and finalizing details for some time before that. Looking at the stock price of both companies, it is pretty easy to see when the inside traders got the word.


Sun and Oracle were in a general downward spiral until March 10, 2009, but on that exact date, both companies simultaneously experienced a profound reversal of fortune. Investors suddenly found both Sun and Oracle to be very alluring. Yet nothing Sun had to offer Oracle would increase the attractiveness of its database. When investors suddenly moved to revalue the companies 57.4% and 125.78% higher, did they smell monopoly?


It's nice to think a corporation would be so humanitarian that it would try to replace its hundred billion dollar flagship database with a free one, but realistically, corporations have to do what is best for the bottom line, or their investors would leave or sue them. Ellison himself owns less than a quarter of the company he founded. While he has more power than any one person at his corporation, he does not have the ability to override the monetary good of the shareholders. As long as Oracle owns MySQL, if investors can make more money by letting it whither on the vine, that's what Ellison has to do. Even if he so desired, Ellison could not do the right thing through his company. He would have to purchase MySQL with his own money and change the dual licensing into an exclusive general public license.

Proponents of the merger note that MySQL is not in a position to unseat Oracle's high end database monopoly because no other software has the capabilities of the Oracle database. Yet MySQL's current technological position is insignificant compared to where it is going and how long it would take to get there. On its current growth curve, an uninhibited MySQL would unseat Oracle in less than ten years. I am sure that Ellison is smart enough to realize this.

While I believe that humans should have a method of holding corporations accountable to individual measures, the mechanism for doing so has been subverted in the United States and in most of Europe. Being forced to complain to the EU is a horrible method of accounting. Yet, without the three legs of a scalable neural economic system, it is the only method currently available.

If Oracle's acquisition of MySQL is recognized by the EU, Ellison could effectively kill open source database technology in three different ways:


1: Because of the dual licensing of MySQL, Ellison could create a commercial fork far enough ahead of the open source fork to prevent it or any other attempt at a high end open source database from evolving. For example: if Oracle gave a freely distributed, but commercial (not open source) version of MySQL one of its high end toys, everyone who uses MySQL would prefer the commercial version. Without users, the open source version would no longer be developed. Without competition, the commercial version would also cease to evolve.

2: MySQL uses a variety of storage engines. Oracle has been buying them up since 2005. With the acquisition of Sun, Oracle would own even the Falcon engine that is being developed to free MySQL of its reliance on third party storage software. Ellison could effectively kill MySQL by commercializing the storage engines MySQL needs. Once again, the dual licensing of these engines enables the owner of the software to endow a commercial fork with enough improvement that the open source version cannot compete. If Oracle did not own the software, it could not license a commercial fork to kill the open source version. Granted, even without buying any engines, Oracle could create a MySQL compatible storage plug-in from scratch for the express purpose of killing MySQL, but that route would be considerably more difficult and the antitrust issues would be more obvious.

3: Sometimes there is only one right way to improve software. If another vendor has already improved a commercial fork along the right path, he can claim ownership of the only right path of improvement. Civil courts tend to frown on claims to own the only logical growth path, but proving there is only one path can be a legal nightmare. Whenever legal nightmares can be produced, the court can be used as a weapon. Oracle's team of legal experts could effectively kill forks of MySQL or its engines by tying their development up in civil court with claims of copyright infringement. Because Oracle and MySQL have taken fundamentally different development paths, claims of infringement between the two would be frivolous, but if Oracle owned a commercial fork of MySQL, the similarities between it and the open source fork would be close enough to make a case.

It would be best for Ellison's company and for his personal fortunes to control and suppress future development of MySQL. Yet, sometimes when a man holds the fate of the world in his hands, he can transcend his personal bias and even the bias of his small group of friends to do what is right for all mankind.



If your name is Larry Ellison, I want to congratulate you on owning the most awesome private boat I've seen. Using boats as a standard, you have certainly proven your worth, but I hereby challenge you to meet an even higher standard, one that Bill Gates has spectacularly failed to achieve despite his best efforts: that of giving something to the world that it desperately needs without having to control it. I would recognize you as the greatest human being in my lifetime if you would take a small percentage of your billions, purchase MySQL from your company, and give it to the human race.

Throughout history, only a few men can say they made a contribution that kept evolving and enriching humanity forever. You can be such a man.