Simply Statistics A statistics blog by Rafa Irizarry, Roger Peng, and Jeff Leek

It's like Tinder, but for peer review.

I have an idea for an app. You input the title and authors of a preprint (maybe even the abstract). The app shows the title/authors/abstract to people who work in a similar area to you. You could estimate this based on papers they have published that have similar key words to start.

Then you swipe left if you think the paper is interesting and right if you think it isn’t. We could then aggregate the data on how many “likes” a paper gets as a measure of how “interesting” it is. I wonder if this would be a better measure of later citations/interestingness than the opinion of a small number of editors and referees.

This is obviously taking my proposal of a fast statistics journal to the extreme and would provide no measure of how scientifically sound the paper was. But in an age when scientific soundness is only one part of the equation for top journals, a measure of interestingness that was available before review could be of huge value to journals.

If done properly, it would encourage people to publish preprints. If you posted a preprint and it was immediately “interesting” to many scientists, you could use that to convince editors to get past that stage and consider your science. More things like this could happen:

So anyone want to build it?