Even established, tenured scientists will often need to obtain grants and will still be judged by where they publish. Though there's also, of course, quite some institutional inertia and people just not caring, since it doesn't affect them personally. And of course, there are also a lot of (established) academics who do care.
If someone decides that their group won't publish in top journals, they're hurting their own students, which probably discourages people from doing so.
I didn't say they get paid, I'm talking specifically amount the money flow to these journals. Does the funding from the research/scientists come from the profit the journals make, or from outside funding?
Funding flows from those organisations to the researchers, and from the researchers to the publishers. And from the publishers to their shareholders.
One data point: we were 3 authors who worked for several months on a chapter of an authoritative book in our field. We got €100 to split. Of course, we get fuck all for our papers.
No. And most of the time, if you want a paper that's behind a paywall, email the corresponding author and they'd be delighted to know that someone on the planet actually cares, and then email you a pdf, copyright be damned.
Honestly, sci-hub is the best website of the last decade; I use it every day. It works better than the publishers' websites (that I have legitimate access to). It's just excellent. It's founder should be made a saint.
What? Do the journals get copyright over the authors too???
Many (most?) books through publishers don't earn out their advance which is probably low 4 figures in most cases. In practice, writing books is either a hobby or it's a reputational side-gig for you day job whether self-employed or employed by some organization.
Also, as you mention, newcomers need the reputation given to them by journal, but I don't understand how this is a stable system. In theory, well established scientists, who give the journals its reputation, should be able to easily migrate to an open journal, and therefore make the whole system collapse, no? Do these large publication have any sort of deal to give big researchers on their journals?