The weekend was tiring but fulfilling. Muscles (what pass for muscles, at least) sore. K’s move is now complete. Had dinner at the Blue Fin club in Inala with K’s parents to celebrate the culmination of weeks of hard work.
The coming weeks and months ahead will be dedicated to thesis writing. But first I have to help RGW write (or modify) one last paper. I’ve also been roped into doing three hours of coms3200 pracs per week, but that’s nothing. Thesis progress is looking up, as I’ve recently made significant inroads into formulating a mathematical description of the ant-foraging part of my protocol. I had kind of left this bit for last because, well, it involves maths. But it’s looking good now. The mathematical model enables me to calculate expected results against which I can compare the results from my experiments. This worked beautifully for the wide-area part of protocol. The results agreed perfectly with the mathematical model. There would have been something very wrong with either the maths or the implementation if this wasn’t the case, since the wide-area protocol is deterministic, and should therefore be directly calculable. If nothing else, the comparison shows that the wide-area implementation has been implemented properly. The local-area ant-foraging protocol is a completely different kettle of fish. The way in which resource descriptions propagate through the network is non-linear and stochastic. The mathematical model, therefore, is probabilistic and far more complicated than the wide-area model. But I’m pretty sure I’ve reduced this to a mathematical expression that adequately describes the behaviour of the protocol. So, for the last week, I’ve been writing up the analysis chapter of my thesis. It’s going well, I think.