My previous post has attracted some comments about the academic publishing model, why it is that academics submit to commercial journals that make (in many cases very substantial) profits from publishing their pieces and so on. This broad set of issues has been debated here and on other sites over the last few years. I’d like to throw out a more focused question, aimed primarily at the academics among our readers (although other commenters should feel free to chime in, as always). Starting from the assumption that most of you submit most or all of your work to traditional journals: what would it take for you to switch to publishing through other means (specifically, free-access online paper repositories)???

My own switching requirements (which I imagine are shared by some but not all of you) would be twofold. First – that any alternative means of dissemination provide some sort of credentialling that is acceptable for purposes of internal review. While most of us do our research because we are interested in our topics and think that they are independently worthwhile, we also do it because we would like to keep our jobs (some might also or instead want to find better jobs elsewhere). Second – that the alternative mechanism provide some analogue to the kinds of focused criticism that we get (when we are lucky) from anonymous reviewers. This not only allows for gatekeeping and quality control on the aggregate level, but also typically leads to pretty substantial improvements in individual papers when the reviewers are on target. Obviously, some bad goes along with this system (the implicit incentives of journal publication make academics less likely to take risks and write on out-in-left-field topics than they might in an ideal world), but it’s hard to see how getting rid of it altogether would be a good thing.

If there were a system that provided these two desiderata for social scientists, I’d jump ship in a heartbeat – on every other reasonable criterion I can think of (perhaps there are some that I am missing) open systems are likely to beat closed ones. Obviously there are some very important economic issues too – arXiv, which is the closest analogue to such a system that I can think of, costs a fair bit of money to keep going. But it seems to me that the basic question of what we should want (or, more precisely, what we would absolutely need; wants are potentially infinite) in such a system should be asked before we ask how it should be funded. So what are the benefits and problems of such a system from your perspective, and what would it take to get you to jump over?