4 ms·
Maybe there should be a "reproducability risk" metric, calculated with the transitive closure of all published results that reference a paper that has yet to be
by infogulch 4y ago
Maybe there should be a "reproducability risk" metric, calculated with the transitive closure of all published results that reference a paper that has yet to be reproduced. This could help researchers calibrate how much they should trust a result and indicate when a foundational paper really needs to be reproduced.