But it does, there's a limited amount of hours in the day and an infinite amount of papers. You need someone to pre-screen the torrent of crap that has an occasional jewel floating in it, and traditionally journal editors have served that role.
How many Open Source projects are managed by an individual, or a small dedicated group? If they can do it - by self organising! - it's baffling that some of the 'smartest' minds in the world are unable to comprehend what's going on in their field on a meta-level.
Peer review is supposed to filter out the crassest nonsense (it usually does). What a journal does is grading. When you are reading stuff where you don't know the relevant names personally a good heuristic is the journal. You don't bother with Hindawi and MDPI and just read stuff from the better-known publishers.
We are all very aware of it, but it's not easy to get out of it if the people that pay you measure you based on those metrics.
Open access is partially solving the problem though and imposing open access to all publicly funded research seems to me a good compromise.
But it does, there's a limited amount of hours in the day and an infinite amount of papers. You need someone to pre-screen the torrent of crap that has an occasional jewel floating in it, and traditionally journal editors have served that role.