OOPSLA 2006, what is next

This week I am visiting the OOPSLA, promoting OSGi, and it is kind of quiet. Today’s OOPSLA’s are not what they used to be in the heydays of 3500 attendants. Over the years the OOPSLA has lost the magic to attract big crowds; we are currently around 1200 attendants. Obviously the competition is fierce; the number of software conferences is staggering. Then again, the number of people working in software has soared.

So why has the attraction of the OOPSLA diminished? A key reason is of course the fact that objects are no longer a big deal. My first visit in 1987 was terrible exciting because I heard also those wonderful ideas of those bright and brilliant people. I still can feel the excitement when I understood the MVC paradigm for the first time. That excitement has returned several times over the years. Patterns, aspect oriented programming, and extreme programming are just a few of the highlights that were first presented on the OOPSLA. However, the benefit of the OOPSLA runs much deeper for me. I currently work with the OSGi Alliance specifying a component based framework based on Java. It never ceases to amaze me how often I can leverage OOPSLA contacts or introduce ideas that I was first confronted with many years ago on an OOPSLA.

That is, OOPSLA has never really been about objects. OOPSLA has been about discovering new ideas about programming. Talking to people here you always sense that they are convinced that there should be a better way to program than we do today. Sometimes there is just a sense of a direction we should explore (the onward! Track) sometimes it is as concrete as optimizing the control flow of Aspect Oriented Programming in Java VMs.

I regularly visit other conferences but most of the time the attitude is different. Most people go to those conferences to learn a technology. I come to the OOPSLA because they want to discover a technology.

Like all exploratory processes you cannot expect grand ideas every year. And maybe the lull in attendance is caused by the lack of great ideas in programming? The dominance of Java has stifled a lot of innovation. The Smalltalkers were always a great source of inspiration but they seem on their way to become an extinguished species with the overwhelming presence of Java. This is also visible in the research papers. The majority do research related to minute details of some Java VM aspect.

Do we need to save the OOPSLA? Parkinson’s law states that any organization will over time try to expand itself. Did OOPSLA have had its time and should it be left to die? Difficult question because the OOPSLA mission has been more than achieved; objects are everywhere. To fold now would be honorable.

However, which conference should I then go to? I skipped the OOPSLA for three years because I felt that I was spending too much time among friends and too little time in the technical tracks. Returning this year was like a fresh breath of air. You meet all those people that are eager to progress the state of the art of programming into new directions. Where would those people have to go to meet so they can cross-fertilize each other’s ideas? Where do you meet this mix of academics and business?

So maybe we have to realize that OOPSLA is about furthering programming. It is not about Java, Ruby, extreme programming, patterns, aspect oriented programming. These subjects were proposals to further the state of the art of programming. I think the key for the coming OOPSLAs is to define a limited number of challenges our industry faces and let the industry and academia come up with the next thing.

The OOPSLA is at a difficult stage in its life. It might be time to fold honorably because we have achieved our goals. However, I would hate to loose the only conference where academia and industry meet to work on moving programming forwards.

	Peter Kriens