Needed: A common Distributed-Object Platform

Richard Husha, Interactive Home Systems Last year, a colleague and I decided we needed to pull our heads out of a very large home-automation project we’d been buried in and go see what the rest of the intelligent environments world had been up to. We attended the AAA1 Intelligent Environments Symposium at Stanford in the spring of 1998. By the end of the conference, I understood that a great deal of energy is going into building supporting infrastructure just to get to a point where the interesting work can actually begin. I saw all these really smart people spending lots of energy on the same kinds of plumbing-related work just so they could begin focusing on their areas of interest. What a big waste of gray matter! For the past five and a half years, I’ve been implementing the control system for a very large, very complex private home. The system is based fundamentally on a distributed-object model. Back when we were trying to figure out just how to structure the project, I spent a lot of time nailing down the core constraints and issues: general, well-thought-out software platform that would stand the test of time as more and more functionality was piled on it. These facts drove us to spend a high percentage of our development resources on the construction of such a platform. Given our experience, I very much believe that if such a platform were available to researchers and developers working on intelligent-environment projects, it would greatly amplify their efforts. This essay argues for such a software platform and further discusses some of the issues common to complex distributed applications such as those aimed at intelligent-environment behavior.