latest model: - model normalisation as a function of fitness
- model seasonsÂ
- model reproduction as a function of space
- make simulation into a 'game' where player can take the side of the producer, the consumer, or the message broker and gain points based on the parameter they are trying to optimise i.e. fitness, utility, or maybe message impact?
- add message impact
reflections: - there seems to be a point where it is most natural/beneficial to move from NL to M4A
- when i want to increase agent heterogeneity along the lines of distinct prototypes
- when i want to control scheduling
- when i want to talk with the collaborators about alternative models e.g. adding different behaviours for the same kind of action (e.g. movement)
- is this model a good tool for linking concept in sociology with those of other subjects
- there will be an optima in terms of the number of people alive and the preference scale between 1 - 9 ...and how fast they oscillate i.e. if messages with 1 and 9 preference are in the simulation the agent population will suffer
- it is interesting to think about situations where the production base collapses too i.e. supply-preference = 0
|
|