How to create a top journal by accepting (almost) everything

More than the open access aspect of the issue, it&#x27;s the reviewing practices that appear to be the real success here.<p>Papers in EECS area are typically published in so called &quot;top tier&quot; conferences. These conferences like to have acceptance rates of 20% or lower. Supposedly this ensures that only the best of the best papers are published in these venues. In practice, the 20% or lower criterion ends accepting a motley crowd of papers with all sorts of biases. IMO papers submitted by PC members are favored, certain &quot;hot&quot; subfields tend to be favored, papers written by well-established research groups are favored over papers from &quot;unknown&quot; groups. I&#x27;m sure there are many other biases.<p>From the authors&#x27; perspective, we end up playing all kinds of &quot;positioning&quot; games to try and increase the chance of acceptance. Maybe I really have a technique that I designed to increase performance, but the hot new thing program committees are looking for is reliability. So I&#x27;ll try and sell my paper as a reliability enhancer with a side-benefit of better performance. Or maybe I have some technique that works really well in practice but is just a combination of two previously known ideas. If my papers says so in plain English, there&#x27;s almost zero chance of acceptance a top tier venue. So instead I&#x27;ll go to great lengths to obfuscate the connection between the prior art and my work and spin it as brand new revolutionary insight that just so happens to be vaguely related to these previously known techniques.<p>The original point of peer review was: (a) catch unsound experimental practices and methodologies and (b) provide authors feedback to help improve the paper. The competitive nature of modern peer review seems to have lost sight of these original goals. Instead it&#x27;s being used as some sort of ranking system for estimated future impact&#x2F;novelty based on necessarily limited current information. The review practices in the OP seem to be going back to the original goal.

&gt; Our little publishing experiment was not supposed to go like this. Largely by accident, we created a peer-reviewed outlet with impact that matches the top journals in linguistics, with time-to-publication that is almost implausibly fast relative to peers, that saves institutions money, and that saves authors lots of stress.<p>in the movie version, a sr executive at Elsevier, talking on his iPhone says &quot;liquidate these two with extreme prejudice&quot;

Recently I had the idea of a FIFO review process. That is, when you submit a paper, you must then submit a review of the paper that has been sitting in the same queue the longest. Thus every researcher winds up reading as many papers as he&#x2F;she submits; papers written by groups would presumably jointly submit reviews. This, I think, is the closest one can get to having one&#x27;s peers in research review your papers without having your specific peers (i.e. colleagues at a university) reviewing your papers (and the likely corruption that it entails).<p>I haven&#x27;t thought about whether this process would be better or worse than the currently common peer review process, but it is at least interesting in its simplicity.<p>This FIFO process could be chained with the traditional panel-based review process. I am not sure whether that would undercut or improve upon whatever benefits that a FIFO review process would provide.

Fascinating glimpse into the state of the art in open access publishing, which only a few years ago was thought of as the academic equivalent of a vanity press. Seems very clear that with acceptance increasing so rapidly in academia, traditional publishers are looking like draft horse breeders c. 1909.

If only this spread to the social sciences. The time to publication there is ridiculous. I&#x27;m also a partisan of publishing everything that&#x27;s valid and letting readers figure out if it&#x27;s important.