[FranchuFranchu]

I think that this has good intentions, at least. Science should be easily reproducible. But using the median for this is very very silly, and I don't think the results from this are very meaningful.

[cleandreams]

The claim that science should be easily reproducible is wrong. Take collider experiments, or huge data set analysis problems.

[PaulDavisThe1st]

The actual claim is that it is better if science is easily reproducible, but sometimes you just have to make do with what's feasible.

[uniqueuid]

Well, the goal is fair: It would be nice to have science reproducible. There are simply a lot of pragmatic (and ethical, and legal ...) reasons why it often is not.