Surely there has to be a cost to the infrastructure of publishing and curation though. And possibly all the work of setting up and organizing the peer review process. So they probably charge the institutions or authors submitting the paper instead of their readers. But perhaps we should treat scientific journals as a public good, like libraries, or at least have a publicly funded option. Or have universities and institutions fund it for the public good.
But it's mostly a scam. The costs don't remotely compare to the revenue. Reviewers time is not paid, and there's a price to both publish and access. It's all about the prestige to publish. If you contact the author directly they'll typically gladly send you the article for free.
Not to mention that system started about four centuries ago, long before the Internet was invented. I'd assume that back then, the costs and effort of operating a journal really did justify the prices they charged.
Oh absolutely. I agree. I don't think anyone's disputing that something about it needs to change. Even given that things cost money to run, for profit journals that can basically act as gatekeepers means there's also going to be excessive price gouging and profiteering and that needs to change.
The issue is partly that, over time, private entities are going to price gouge or take similar measures (see: "enshittification") in order to keep growing as profit falls over time. That's just how the profit motive works, it eventually optimizes everything for profit, not just what you are comfortable with having turned into a vehicle aimed solely at making money.
So yes, this and many other important things should be treated as public goods.