if reviewers work free, where does the bulk of the money go?
there is so much publicly-funded research bottled up in journals like science. how do we break this model so that taxpayer-funded research is more like wikipedia and available to all?
In 2019 the Journals division reported $7.3M revenue against $5.6M expenses. Direct-paid salaries appear to be a minority of expenses; the AAS journals are hosted by IOP publishing (https://iopscience.iop.org/), so presumably IOP charges for its journal hosting platform are a significant proportion of the remaining expenses.
Unless their fee involves paying for a granite-imprint of the paper to preserve it for centuries, the marginal cost should be near-zero.
I also think that journals are too expensive, but that's why I only publish in society journals.
What guarantees about preservation do journals give, though? (Hence my remark about papers being carved into rock)
It isn't officially available anywhere today. There's no invisible hand to save those old publications from being lost to time. Happily, with the help of strangers online, I was eventually (after many months) able to get my hands on an electronic copy. The formal system of publishers was of absolutely no use, and the fear of copyright trouble is likely the reason the paper isn't easily available from some online archive.
Doubtless there's no profit in preserving such papers formally.