> So assuming for a second that the publishing hegemon is destroyed, what will happen next? Will all the up and coming star scientists happily publish on undifferentiated internet platforms where all that matters is science? Some maybe, but my more cynical guess is that a thriving internet status economy would soon emerge that would inhabit the exact same niche that publishers have now. Because the exclusivity of publishers is not the tool they wield against the public or scientists, it's the very commodity they are selling.

That's a very good analysis, I think, but it overlooks that an internet attention economy is a better state.

In the field of machine learning, it has already come to pass. All the top publishing venues (the Journal of Machine Learning Research, the conferences NeurIPS, ICLR, ICML, ...) are already free of charge and open access for everyone. There are many problems with reviews in those venues (mostly growing pains from the rapidly increasing number of submissions, and problems stemming from the fact that there is 1 single round of review), and indeed the conferences, JMLR and Twitter are now the "attention economy" of the field.

But it has massive positive externalities, namely, you don't need to pay (or have your university pay) for access to the research anymore. The system works as badly (or as well) as it would with the publishers, but without giving them a cut.

The thing about (closed-access journal) publishers that ticks people off is just how little of the value they create for their cost. They get the papers they publish for free, they get the peer review for free, they even get most of the editing for free (academic editors are generally volunteer, although copy editors, who check for spelling and formatting, are generally employees).

As for why people keep paying, the answer is industry lobbyists. Whenever there is a movement to require open access of research, industry lobbyists shut it down. Although in many fields like physics and mathematics, people are bypassing journals (closed or open) in favor of preprints.

> what a publisher is, is an institution that sells reputation and attention

If that is all publishers are now, then they are no longer what they once promised to be. Initially many respected journals were published by non-profit learned societies. (In some fields, like certain branches of linguistics, they still are.) For-profit publishers originally told those learned societies that if they handed their journals over to the corporation, the corporation could perform more high-quality editing, proofreading, and typesetting and do it more economically.

Fast forward a few decades, and the for-profit corporations are no longer providing those things. Proofreading and copyediting is now all on the unpaid editorial team (or even on the individual authors). Typesetting is often on the unpaid editorial team, and the publisher wants the unpaid editorial team to simply provide a camera-ready PDF.

So, yes, in the end the for-profit publisher is just providing printing and distribution (which even the non-profit learned societies managed to do just fine) and a vague “reputation and attention”. Sounds like a raw deal.

Many researchers are abandoning Elsevier and for-profit journals, but unfortunately two of the major non-profit technical societies in our field - ACM and IEEE - still seem to view digital libraries as a cash cow that can milk as much as possible to pay for unrelated activities, and they also still charge fees to authors for open access. These societies are hard to escape since they actually organize some of the best and biggest conferences.

USENIX is open access I believe, which is great, and I even think some of ACM's SIGs - SIGCOMM for example - make their publications, such as CCR and conference proceedings, available immediately via open access. (I think they are no longer published in print format, so that may save money.)

Hopefully many other SIGs (as well as IEEE societies) will follow suit, and hopefully government open access requirements will improve the situation as well. It doesn't make sense for taxpayer-funded research publications to be locked behind a third-party paywall.

I'm not sure what's supposed to be wrong with "thriving internet status economies" or "reputation and attention." The problem is the "exclusivity" where people have to pay e.g. $45 to read an article often partially or completely funded by taxpayers and with absolutely no value-add other than "reputation and attention," in order to discover it is irrelevant to what they're researching.

In my view, it would be ideal if in an open-access world some editors and/or organizations endorsed and vouched for particular papers, and academics competed intensely for those endorsements, if those endorsements were career-making or career-killing, and those editors/organizations made a living from charging scientists for their consideration and review.

That's not the bad part. If the output is available to everyone to read, I don't see the tragedy.

They don't literally hold anyone at gunpoint, but they do figuratively. If you stop paying a publisher then you don't get access to the papers your researchers need to read in order to perform their work.

Because by copyright law they hold the keys to the kingdom (of scientific literature)

> But this is strange, because if it was true, everyone would just stop paying them, they don't literally hold anyone at gunpoint.

In a world where scientists careers didn't depend on publications, you'd be right.

Change is hard, common knowledge attacks are easy, publishers are like dictators, defectors are punished (publishing in a worse journal, basically only a minority of researchers can even flirt with the idea), and even if the global optimum is not a dictatorship it's hard to get there.

That is indeed a fair point: Why do people go to some Elsevier journal to look for an article instead of just going to the corresponding arxiv.org section? Because there is trust in the curation of those articles.

The problem with scientific publications is that there has not been a Spotify, Steam or Netflix disrupting company that provides the same service in a better way.

EDIT: Thinking more about it, I think such a service would be fair to say have a raw collection of journal articles (it could even be based on Arxiv) and charge for the "curation" layer on top of it. Now the only question is how to kickstart that curation layer in a matter that is trusted by scientists. Maybe it could be something more distributed where also curators (reviewers) and writers get some profit.

A reasonable request is not to shut all prestigious publishing monopolies down, but to ask/beg/fight them to be less greedy. As you mentioned, publishers run market places and sell distribution channels. They do not need that high margins to run the business. Where the profit goes to? Not the science community, but heir owners and executives high up on the rank who do not contribute much but get the most cash rewards. I believe this is what worth fighting for.

I would go even further and say that these prestigious institutions don't need any execs. They are just a pressure tool for corporate interests. There's no shutting down needed, just dismantling their bureaucracy. If there is one community where self-organizing is genre consensually known to work it's the academic community. There are already tons of fields where this is the case: some top-notch-international and most local conferences and journals alike in CS and math are already being run collegially by universities and unions. As always it's always where there are big corps that thing go awry (looking at you biology and medicine).

This! What the journals are selling is reputation.

The authors can choose to publish wherever they want - but there is value in publishing in Science, Nature. And their academic overlords acknowledge it too. Who gives a shit if a professor publishes a dozen papers in some obscure journal no one reads?

If you want to solve this issue stop violating the journals copyrights and start attacking the academic leaders who demand professors publish in them.