> only a few percentages of research literature is actually worth saving True, but we never know which papers and ideas are in that percentage without some h

> only a few percentages of research literature is actually worth saving True, but we never know which papers and ideas are in that percentage without some hindsight. > loss is not as big as it seems. It's an opportunity loss. Even poor papers are useful in meta studies and archival research, because they show the errors, and state of the art in a particular milieu. > it is out there in substantial quantity. It's articles will not be lost any time soon. Problem is, there is no coordination. And while old scientists curate wonderful personal collections, old scientists die, and so do their hard drives. We are at constant war with the forces of entropy. One of the biggest chaotic forces is the publishers who fight legal battles against science by laterally spending millions in "spite money" to take down collections.