In any fucked up system, one should look first to the incentives, right? Which takes us to the journal system itself. To me it seems hopelessly outdated, at le

In any fucked up system, one should look first to the incentives, right? Which takes us to the journal system itself. To me it seems hopelessly outdated, at least in some areas. So here's a thought experiment: Science (especially now) seems to be largely about gathering data, doing a simulation, or inventing a method (to either gather or analyze data). The data-gathering papers would be more useful as commits to a giant database in the sky, the simulation papers to a giant VCS, and the methods to a giant wiki. Indeed, within the labs I've seen, something like this is in place internally. It would free up the journals to publish articles about the interpretation of analysis results, and reviews of larger bodies evidence to compile them into deeper theories - which rightly belong in journals. Peer-review would be done by upvoting (!) So what would PhDs be optimizing for? Commits x Votes. Yep, I just suggested that academia should be redone in the image of HN (I'm probably not the first).