Beyond the six model attributes detailed before, there are several other distinctions that are important to consider.


Necessarily, a model is a representation of very limited aspects of the thing or events being modeled. In that sense, it can be said that all models are wrong with respect to the full reality of the slice of nature being modeled. Consider a global map of the world. The globe is not the same as the real world. The globe is a very different size. The globe has different colors to identify countries. The real world is not colored the same as the globe. The globe shows distances between cities, but the distances are not the same as those of the world. What is the same are the relative distances between cities and the proportional spatial relations of rivers and country boundaries. That is all. But as such, this model is very useful and aids understanding with regard to those particular attributes.

Increasingly, science and technology and government and industry are being driven by models. In physics, for example, our understanding of the universe is largely based on model extrapolations well beyond what we can observe directly, and huge experimental efforts are made to verify the models (e.g., the hunt by particle physicists for the Higgs boson). Social science is definitely progressing, and in the future may well be aided by progress in neuroscience, but has not come close to that ...

Get Modeling Human–System Interaction now with O’Reilly online learning.

O’Reilly members experience live online training, plus books, videos, and digital content from 200+ publishers.