Open Source Climate Change

There was a post in slashdot a few days ago, about some of the politics and debates that get started by the competing models for how climate is changing. The post, implies one group of researchers is criticizing another, for not publishing the source code they’re using for their model, but only the general algorithms.

The source code is the plain text which can be turned into a computer program you can run, the algorithm, simplistically would be the idea of what the source code implements. The critics in this case seem to be alluding to the idea that it’s difficult or impossible to replicate and thus verify the results of the simulation if one cannot reproduce the software used to generate the results. It’s an interesting discussion, should researchers be forced to publish the source code when presenting data? Many of the concepts ‘science’ is based on, rely on being able to independently reproduce results, which may be nearly impossible if one has to recode the algorithm, as for any somewhat complicated algorithm, there are likely to be many subtly different ways to implement it.

Comments are closed.