So I read your links. How is Angela Kerr's job any different than a spam filter function? We've got production grade techniques for handling that now, and without going into technical details everything you mentioned as being in her set of tasks could be implemented using ML and NLP techniques, most of them not even that cutting edge. In many ways, an algorithmic approach could do her job better because some "crackpot" articles will not be crackpot but genius and if you give researchers the ability to tweak the filter settings by adjusting sensitivity / specificity to their own tastes you're mathematically guaranteed to increase the likelihood of getting the next breakthrough out there.

The actual problem, which I think you implicitly identify when talking about Nature and Cell, is the "prestige" factor. As long as researchers are motivated by a prestige level of a journal b/c they may fear being ostracized by their peers or not getting recognition for their research - which has monetary and career costs involved - I think it will be very difficult to convince anyone to switch, regardless of how effective the platform could be.

I haven't thought about that problem before so I'm not sure how to address it as of now.

Some things may be better shown by example.

Here's the first paper off the top of today's submissions to gr-qc: https://arxiv.org/abs/1803.11224

Is it right?

How do you know?

To whom should you send it in order to get an expert review?

Which of those people has an ax to grind with one of the authors?

Of the remaining people, who do you think you could convince to invest a day or more to carefully evaluate the paper?

Okay, so you got replies back from two of your carefully selected referees (after two months of badgering one of them):

Referee A thinks the paper is great, insightful, and advances the field, but wants extensive changes.

Referee B thinks the paper is derivative drek and should be rejected because his friend C has already done something similar.

Your journal publishes twenty similar articles a week but receives three hundred a week. The careers of the authors are partially on the line, as is the prestige of the journal, the attention of the readership, and the future submission of articles by prospective authors.

Good luck training a neural net to do this well. I suspect a neural net can be trained to reject the worst crackpots, but little more, without rejecting insightful but unique/important papers.

The problem is that most of the (substantial) work you highlight above is done by the academic community (mostly paid by the public purse), while most of the (substantial and above-market) profits go to the private publishers.

I described the job of the editor -- for the good for-profit journals, they are, to the best of my knowledge, paid:

e.g.: https://www.nature.com/nphys/about/editors https://www.nature.com/nature/about/editors

I read from the linked comments:

> In addition to basic copy-editing, she prevents crackpot articles such as "Darwin's Theory of Evolution is proven wrong by Trump" from reaching the journal editors and wasting their time. (She may inadvertently forward some bad articles but she has enough education to reject many of them outright to minimize the effort by the journal's scientists.)

Do I read it right that an Elsevier manager with no degree in the subject rejects research papers without consulting the editors? To me as scientist this is a big red flag and concern about journal's quality. While the example given here looks obvious (and probably contrived), most real examples are less so. And while recognizing those takes little time and effort to a trained eye, this task certainly should not be left to a subject non-specialist. It will not save much time and create more concerns than help.