4 ms·

> This happens because it's almost impossible to write a paper that includes every single detail needed to reproduce your work and fit within any journal's page

• TeMPOraL 7y ago
  It's not just that. The problem is similar to one known in tech companies. There's lots of operations knowledge that exists only in internal e-mail archives (or God forbid, Slack archives), random scripts on someone's work machine, or that's passed around verbally from employee to employee. People don't have enough time or sense to document this and then maintain the documentation (and like with backups, you don't know if a process is truly documented until you try to recreate it just from that documentation). Then somebody leaves and suddenly work grinds down to a halt for a week, because people try to recreate a small but crucial piece of lost tribal knowledge. Now imagine that, but across more serious domains like metallurgy or jet engine design, and imagine trying to recover that knowledge after the company no longer exists and most employees are either retired or dead. In this there's a sense that "unused muscles atrophy" applies to technology. As a human race, we currently do not have the capability to put a man on the moon. We had it once, but it's gone now.