Oracle, What About Java and mySQL?

In all the hubbub and hoorah about Oracle's leaving Open Office to its own devices , and The Document Foundation's decision to become that device, Brian Proffitt of IT World was first to ask the right follow-up question.

(To the right, one of a number of proposed logos for Libre Office, from the organization's Wiki. Personally, I like it. If you do, let them know.)

What about Java? To which I might add, what about mySQL?

Java is the more pressing issue, because it's the closest analog to OpenOffice.org. Java had a community process. Oracle blew it up. Apache walked away. That was, in its way, as significant a move as the decision to create The Document Foundation.

But there are two key differences:

1. There is a real Java market. There is no real OpenOffice.org market. That just existed to cut Microsoft's monopoly rents on Office.

2. Java has a business community built around it. The Apache guys got jobs.

With money around to play with, Oracle needs to ask where the bigger profit lies. Should it be making nice with Apache, unifying the code base again, or should it be trying to extract maximum short-term profit from its present position in user desktops and business servers?

What's in Oracle's best long-term interest?

This is a question Larry Ellison seldom asks himself. He's more a Louis XV “apres moi, le deluge” kind of guy. Anyone getting that much work done (and let's face it, kids, he's the male Joan Rivers) is keenly aware of time's passage, and deeply resentful of it. The idea of seeing past the horizon, to a time beyond the Age of Larry, isn't in his nature.

But that's what open source really asks of us. That, and not the profit implications, is what's toughest for many businessmen to get their heads around. You really have to ask, what's in the best interest of the code base, and treat the community as a city unto itself, separate from your corporate interests.

So I think some rapprochement with Apache is called for, but because these guys have jobs, and employers, it should be more achievable than it presently is with TDF. Business issues and personal issues are different. A good businessperson doesn't get too emotionally wrought over business issues. It's just business.

In the case of Apache, it means proper attention to process can go a long way. Bringing Apache back into the Java Community Process (JCP) should be possible. And that's important in terms of making Java as universal as possible.

Universality is the key to Java's long-term profit, and given the enormous size of the installed base, given the amount of business attention that has been paid, I don't think the split even now is irrevocable.

The situation with mySQL is different. That software is losing its position as clouds become the new paradigm. A cloud can download a single community version of mySQL, run its own support off it, and limit the market. But in doing that you also make your business dependent on mySQL. So it could be a short-term approach there will bring more money than a long-term approach – there may not be a long term there.

The lesson to me is this. The more money there is in an open source project, the more businesses are treating it in a businesslike manner, the more room there is to resolve disputes and move forward. Once emotions get involved, and money is no longer part of the picture, the more trouble you're in, the more likely it is you'll just have to walk away.