Literate, Reusable, Geodynamic Modeling

In the practice of open scientific research, there is a principle that nothing goes to waste. Work is freely available for anyone to read, review, and reproduce. Negative results are valuable because they illuminate directions that others can avoid, and they may turn out to be very useful in the light of later results and theory. Computational models would seem to be very easy to fit into this paradigm, but, in practice, the fate of most computation is to be reduced to a graph and discarded. Open or not, computational results are often so bound up with the implementation details and idiosyncrasies of installation or execution, that models are simply set aside. I am advocating for the reuse of computational geodynamic models as much as possible and I suggest that a closer connection to the mathematical description imbues the models with meaning that will allow others to find models and build up on them.