How should we handle Preprints?

[danielBingham]

Right now, preprints are handled as a flag on each paper. If the flag is set to "true", the paper is public and anyone may review it. This somewhat mirrors existing preprint and crowdsourced review - somewhat.

In actual existing preprint servers there is an editorial team and a desk rejection stage. The desk rejects keep out amateur and "crackpot" papers, written by people with little actual expertise. Those running existing preprint servers could easily use either the public journal model or the open-public model to run those servers.

But then we're presented with the questions of multiple submission, and how to differentiate journals from preprint servers. The primary thing that differentiates preprint servers from journals right now is peer review. But there are also journals experimenting with different kinds of peer review. And journals that do only minimal peer review. They can all use the same tooling, the issue communicating which is which and handling multiple submission.

If we keep preprints as a separate flag, preprint editorial teams aren't a thing. Anyone can flag a paper as preprint and it becomes public and reviewable. But then there's no amateur filter.

A third alternative would be to place a "preprint" flag on the journal that allows a journal to make itself as non-peer reviewed publishing preprints. We could then differentiate submission to preprints and journals - if they don't allow multiple submission, you can do one of each.

And there's the question of respositories and the funder platforms. We'll need to figure out how to fit them all together.

How should we approach this?