Providing codes is imperative; they are part of the methods and should be available for examination to ensure the integrity of the science. Those conducting research funded with public monies should be required by the funding agencies to release the products of that research, not just the results, but the data and codes, too (absent truly compelling reasons, such as national security). Eventually, I expect funding agencies will indeed require this for astronomy, just as they do for some other sciences; journals could help the field along, could improve the transparency and reproducibility of research, by requiring code release upon publication.
Absent funding agencies and journals insisting on code release and the moral argument of reproducibility, what incentives would help convince code authors to release their software?
Providing codes is imperative; they are part of the methods and should be available for examination to ensure the integrity of the science. Those conducting research funded with public monies should be required by the funding agencies to release the products of that research, not just the results, but the data and codes, too (absent truly compelling reasons, such as national security). Eventually, I expect funding agencies will indeed require this for astronomy, just as they do for some other sciences; journals could help the field along, could improve the transparency and reproducibility of research, by requiring code release upon publication.
Absent funding agencies and journals insisting on code release and the moral argument of reproducibility, what incentives would help convince code authors to release their software?