Re:FBI subpoenaed Apple for data of Sci-Hub founder
by MacsHeadroom on Monday May 10, 2021 @06:02AM
(#27104366)
It's not that older studies tended to be more accurate. It's that peers worked together in seclusion via lots of back-and-forth letters and collaboration over years, only "publishing" when they and their peer group were sure they had something worth sharing with the wider community.
This is how science was done up until some time in the 1970s. This is how real groundbreaking science is still done. The publishing prestige economy is of little actual value.
The "reproducibility crisis" is that we recently realized that ~half of published results in many fields fail to reproduce, presumably because many of them are false. My impression was that this goes back as far as you care to look, and that old results are just as likely to fail to reproduce. So the only new thing is that we noticed this, which is a step forwards, not back. Is there reason to believe that older studies tended to be more accurate?