=0A=0A

It is clear that the issue of the Science Code Manifesto=0Astirred= a lot of feelings. I must say that I found this topic almost as=0Aenjoyabl= e as the match between Fortran and C++. I learnt a couple of small=0Athings= with this discussion, but all in all I still subscribe to the same ideas= =0AI wrote some days ago:

=0A=0A

= =0A=0A

=E2=80=9CI like the idea of open software, a= nd in principle I=0Asupport the proposal of having access to the software c= ode=E2=80=A6 However, I feel=0Athat this is more wishful thinking than real= scientific life.=E2=80=9D

=0A=0A

=E2=80=9COpe= n or closed software (usually) doesn=E2=80=99t make me any=0Adifference, ex= cept from a philosophical perspective.=E2=80=9D

=0A=0A

I think of myself as a = pragmatic guy. That means that as=0Amuch as I like GPL and open software, I= am prepared to pay for a close program that=0Aworks better. And I dare to = say most of the computational chemists try not to=0Adive into the code anyw= ay, as we only want to obtain an energy value, and not=0Aunderstand how the= Roothaan=E2=80=99s equations were implemented.

=0A=0A

Let me rescue one thought that appeared (sorry I don=E2=80=99t=0Ar= emember the name of the author): =E2=80=9CReproducibility lies in the possi= bility of=0Aobtaining similar values, not necessarily with the same program= =E2=80=9D. So, if for=0Aexample I wrote down in the paper the functional an= d the basis set, that should=0Abe enough for reproducibility of the geometr= y (although a good=0Aresearcher should help his community by giving much mo= re information). If=0Aprograms X and Y give very different bond distances, = then clearly one of them=0Ais doing wrong its work. Who cares about the cod= e if it=E2=80=99s a crappy code anyway?=0AIt is not reproducible, and that = is death according to the scientific method.

=0A=0A

I still think that we must co= mpare ourselves with the=0Aexperimental chemistry world, and don=E2=80=99t = make a special epistemological rules for=0Acomputational science. If I use = a specific NMR, nobody will ask me to publish=0Athe designs of the equipmen= t including the Fourier transform algorithm and the=0Acrystal structure of = the superconductor (which I don=E2=80=99t have anyway); I must=0Aonly write= down the model and frequency of the machine. Should we ask more from=0Athe= theoreticians?

=0A=0A

Let me be repetitive:

=0A=0A

=E2=80=9CI like the idea of open software, and in principle I=0Asupport = the proposal of having access to the software code=E2=80=A6 However, I feel= =0Athat this is more wishful thinking than real scientific life.=E2=80=9D=0A=0A

=0A=0A

Now back to my calculations. From time to time we must stop=0Aarguing an= d make a bit of real work.

=0A=0A
Best,
Sebastian

--270156429-1988330702-1319045960=:31035--