Even if the "publish fewer papers" plan for improving academia were possible, it doesn't make any sense to me. Even if there were fewer papers, presumably the same amount of work would get done, and maybe more since it wouldnt be punctuated by so many long writing and revision cycles.

Whats really the problem anyway? Probably that most results talk past each other and dont build any sort of cumulative understanding. Maybe that we are driven to publish rushed and sloppy work by the prestige economy. I dont see how decreasing the volume of papers while keeping their form and the rest of the system intact addresses that.

They're different, but for the sake of illustration, is it a problem that there are 150-200k edits to Wikipedia every day? Is it a problem that there are ~8 million pages, the vast majority being very low quality? No. Its because things have a place, and you can find things that are related from any given place. Arguably there should be more edits and more versions of pages bc the stodgy Wikipedia mod class.

Imagine instead Wikipedia was structured such that every edit had to be a self contained monograph, organized by the moderator that received the edit rather than by the page topic. Then it would really be a problem, and we might be tempted to plead to slow down editing because we couldnt keep up with all the edit monographs coming in.

IMO too many papers is a symptom of how knowledge disorganization is obscenely profitable. The journal system makes about as much sense as charging for every wiki edit and incentivizing more edits by resisting consensus on a topic. The problem is the form of journal articles and how they're organized, not that there are too many of them. 

 @ab7cf35b yeah this is precisely my issue with the "slow science" idea.