I think I’m doing the current part of my Ph.D backwards. Here I am trying to figure out how a complex systems analysis of service discovery protocols can be useful before I’ve finished creating my own model for service discovery. What I should be doing is concentrating on developing a service discovery protocol and then performing a complex systems analysis of it. It would take all the guesswork out of what I’m doing at the moment. The reason, I think, I started doing it this way was because I actually wanted complex systems theory to have a bearing on what my initial protocol looks like. But this probably isn’t smart. If there’s anything I’ve learned from talking to Kai and others, you need to have some sort of working model or a simulation so that you can discover patterns and attractors and the like.
This situation presents a problem because Jaga and I were wanting to write a complex systems based paper for MDM 04. I wonder what I’m going to do?
