News for hackers. Stuff that matters.

A comment on Challenge to scientists: does your ten-year-old code still run?. Parent comment

Re:Challenge to scientists: does your ten-year-old code stil... by dekhn on Monday August 24, 2020 @01:45PM (#24263220)
As an ex-scientist who used to run lots of simulations, I really fail to see a truly compelling reason why most numerical results (for publication purposes) truly need to publish (and support) deterministic seeding.

We've certainly done a lot, scientifically speaking (in terms of post-validated studies), without that level of reproducibility.

[ Parent | on HN ]
Re:Challenge to scientists: does your ten-year-old code stil... by jnxx on Monday August 24, 2020 @02:08PM (#24263508)
If nothing else, it helps debugging code which tries to reproduce your findings.
[ Parent | on HN ]
Re:Challenge to scientists: does your ten-year-old code stil... by dekhn on Monday August 24, 2020 @02:22PM (#24263684)
The code I work with is not debuggable in that way under most circumstances. It's a complex distributed system. You don't attempt to debug it by being deterministic- you debug it by sampling its properties.
[ Parent | on HN ]