Side Effect Of Blunt Paper

How good are the models? Mindmaps

Please note, part of this post is a free translation from the last paragraph of " Verification, Validation, and Confirmation of Numerical Models in the Earth Sciences " Posted By Naomi Oreskes , Kristin Shrader-Frechette and Kenneth Belitz in Science February 1994. Recently

the journal Diversity and Distributions "(sorry I do not put the link, it is my sad retaliation: P) has rejected a job dealing with potential climate change effects on plant diversity in Andalusia (see Chapter 8 of my thesis ). Rejection is a normal thing in this world (even gives worth getting used to), and would have no importance, but for one of the associate editor's comments:

" .. You Should Not Be over-confident with your results and models. A model is a model, NOT the truth. "

I never had confidence concealment in the models that work (though perhaps not been adequately reflected in the article), but with my "work" I have learned what their weak points, and I've been critical of them (especially the last chapters of my thesis.) Although I have not personally, these words do have made me re-think my understanding of "what" are the models and what they are. And I have remained as before, but I remembered this article by Oreskes, that ends with a section that I love. Here it goes:

As the philosopher Nancy Cartwright models are "works of fiction" because some properties attached to objects in the model can be genuine properties of the modeled object, but others are simply desirable properties.

Oreskes et al. continue to say that as a novel, a model may be convincing, you can "real sound" if it is consistent with our experience about the world. But just as we ask the extent to which fictional characters are drawn from real life and how is fictitious, the model should ask: how many of the components of the model are based on observation and measurement, and how much is based exclusively on suitable values?. (Oreskes et al. 1994)

Indeed, returning to the case, when we model the distribution of a species and project the model into the future, there is no possibility of knowing how close the model a reality yet has been given, and therefore, the result can be considered pure fiction (or at least "unverifiable.") But these models are not made with the intention of determining the exact locations of the species Fulanita fulanoides be gone the year 2058 at 4 pm because that is not possible. These models are made with the intention of determining the upper and lower limits of the potential situation, and to document the range of possible outcomes.

that's all folks!