Oh no, Onno

by Daniel H. Steinberg

Related link: http://java.sun.com/developer/technicalArticles/Interviews/kluyt_qa.html



Jan Heiss interviewed Onno Kluyt, senior director and chair of the JCP on Sun's site in an article titled Ensuring Speed and Openness at the Java Community Process. I was at EuroOSCON at the time enjoying Onno's home country and have only now gotten a chance to read the interview carefully.

Onno begins by addressing what he considers to be the misconception that JCP standardization is slow. He says that from February 2000 until today the JCP has delivered three versions of J2EE with a fourth expected by the first half of next year. Onno counters "In that same period Microsoft still has to get around to delivering Longhorn."

I'm baffled.

Why is that a counter argument to the speed or lack thereof of the JCP. He could equally have said "In that same period Apple has delivered five versions of Mac OS X and four versions of Mac OS X server." But that would be no more relevant. J2EE is not an operating system. He could just as well have said "In that same period GM has delivered five lines of cars" or "the price of gas has doubled". None of these address whether or not the JCP is slow.

Further, the fact that there are regular releases of J2SE and J2EE is not really the complaint with the speed of the JCP. The complaint is how long it takes features to make it into a release. Onno answers this with the statistic that "In the first three years of the JCP's existence, JSRs took 100 to 150 days longer to complete than JSRs that were started in the last three years." That's a wide range 100 to 150 days - and what is the average time to completion now?

I'm not saying Onno's job is easy. There are a lot of JSRs that are never completed and there are JSRs which are mired in politics. I just don't think that this addresses the issue of standardization being slow. The traditional complaint is that we build a JSR before we have working code. We don't often take something that's working and build a spec around it. Imagine what we'd have if Spring or Hibernate had to go through the JCP before being released.

Onno then addresses sources of FUD around the JCP. His answer concludes, "When a company outside the Java ecology tries to confuse its audience about Java technology and the JCP, that is expected. But for companies that benefit from this ecology, it's disappointing when they only manage to find negativity." I'd love to know who he's calling out here.

At JavaOne 2004 there was a panel on the JCP where very serious issues were raised. I give Sun a lot of credit for hosting this discussion at their developer conference, but it was clear that individual members, corporate members, and non-members all had concerns with the JCP. I thought it was healthy for the community that these were raised in a public forum and have been heartened that many of them were addressed. There may be some companies that fall into the "negativity" category, but I think most members are more interested in improving the entity to which they belong.

Slow or no?