We've definitely inherited something from a pre-digital age, where it took non-trivial capital and staff to actually publish and distribute stuff.

But on the other hand, there are still aspects that make it unclear exactly how to change the 'market'.

Someone else in this thread said what the publishers provide is 'prestige'. I don't think that's exactly it -- the editors and peer-reviewers, which are who provides the prestige, are largely unpaid (or very trivially paid), even at commercial publishers. That's not where the money is going.

It still takes some staff to organize and provide administrative services for the actual prestige-providing volunteer editorial staff. Receive papers, send the out, copy edit them, get them in the right format for publication. I'm not actually sure how much staff, not being on that side of the industry.

But it's also not actually free to publish things online, at least not if you do it right (which some publishers do and some don't). Someone's got to write and maintain the software and deal with the operations, pay for the hardware and network, etc.

The costs are presumably less than for print distribution, but we on HN know they are not non-existent. The barriers for a startup may be lower than they used to be, but you still can't scale to serving millions of customers without any income at all. It costs money to put up something on the web, in this case the something being hundreds of thousands (millions?) of articles, that you keep up forever, with good metadata, etc.

So who will pay for it, and how? Formerly, largely libraries paid for it as customers of publishers, and they paid for it because if they didn't pay their subscription they didn't have access to it. They were customers.

Who's going to fund the endeavor when you will get access to the content whether you pay or not? University libraries, who pay for it now, don't generally have budgets (and the people who set the budgets) which allow them to pay for things just because it's good for society. They are getting their budgets squeezed too. Even though it's the universities who pay the exhorbitant rates to get access to the content produced by people who's salaries they already pay. (the authors, editors, and peer-reviewers), and they would benefit from a 'market' that worked entirely differently -- nobody's yet figured out how to create that market.

Maybe governments will pay. Perhaps via university systems in other wealthy countries where, unlike the U.S., universities are almost all public and part of a centralized organization. It sounds like that's what the Dutch are trying to do. People are trying to figure it out. In the U.S., government grants are starting to maybe include line items to pay for publication, but we're still figuring out what the actual right cost for that is, or how to create the organizations to organize it without charging extortionate rates. And meanwhile, existing faculty, without who's buy-in no change can happen, have a fear of major changes to the existing precarious system which may be a mess, but existing faculty know how to deal with it (or they wouldn't be existing faculty). Like people scared of change everywhere.