We should place attention on developing a theoretical framework of Psychohistory and this is where most of my effort is going. I think there is plenty of room for the models you suggest (“black box”), for within a framework it has a place. I have heard some really far out ideas recently; super computers simulating billions of individual people for example. Aside from the practical problems, we need to keep in mind good model design. A model is as good as a model does. I have spent some time pondering underlying assumptions and expectations for my models and this has resulted in some changes. One thing funny about models is how easily they disconnect from reality. We need to place models in the painful process of testing and refinement. I have seen several web sites that claim to be “Psychohistory” and found them so conceptually ludicrous as to be painful to read. Psychohistory must be a practical discipline with a definite purpose; not a fantasy of how we might “predict the future” with some quasi-deterministic set of assumptions. I have suggested in another site that we are not even able to verbalize the purpose of Psychohistory and this statement met with a good deal or resistance. This really is our starting point, refining our “mission statement” if you will.