3 ms·
If the point of the exercise is to allow public scrutiny of models then IMHO the code should not be changed "for public release". The original code as used to
by mytailorisrich 7y ago
If the point of the exercise is to allow public scrutiny of models then IMHO the code should not be changed "for public release".
The original code as used to produce simulation results that were in turn used for reports/papers/policy should be made public.
- lbeltrame 7y ago> If the point of the exercise is to allow public scrutiny of models then IMHO the code should not be changed "for public release". Some code I've used from scientific papers in the past made a git tag at the time of publication, and said "Use version XXXX and the provided sample data to get the results in the paper", or something of the sort. This could allow review and improvements post-publication.