Oh boy.
Let me give you a quick rundown on how academic publications work.
So, you are a good academic and want to get some good boy points to advance your rank. You need to prove that you've been a busy bee last couple years and need something to show. So, you write a paper on your scientific investigations you've done, and want to publish it somewhere. But just anywhere won't do; after all, any quack can go and spit out some graphs and figures and such, nobody reads that anyway, so who will figure out that it's bullshit. Luckily, there's academic journals, with their editors and peer reviewers (other academics who are willing to sift through mountains of bullshit to call it out). So you send out a manuscript and wait for response. Usually, the peer reviewers will discretely hint that you might want to cite this and that paper which oh just so happens to be by a certain group they're probably somehow affiliated with. But generally you'll get the paper accepted (provided it's not easily demonstrably garbage) and are now waiting for a free slot in a future issue. But how are those journals funded? Well, if you want to be a serious scientist, you want to read what other scientists are publishing. And sometimes, just going to every conference possible isn't enough. So your academic institution is subscribed to a couple of those journals. Historically, they would be sent by mail, now they're increasingly just online subscriptions. This is where the other option comes in: not just pay for a single article, buy a subscription! It's extra cheap! Just 200€ for this journal per year! (you get literally hundreds of articles in the package) But most of that is technical bullshit that nobody cares about. Except for other academics in that field, that is. However, every now and then, there's some company or an entrepreneur who might need to read a paper to figure out how to solve their technical problem. And what does the journal do? That's right, charge $40 per view. (Are you sure you don't want to buy the whole subscription?) The main point is that some journals are more reputable than others, and having published there gives you more academic good boy points. And the more often your paper gets cited – even more good boy points. And most people prefer to read the reputable journals, because all the important stuff is in there. So you want to publish there, because it's more likely someone will read it, and more likely someone will cite you. But they can therefore also charge more. Well, there's also the other option: open access journals. But in this case it's even more jewish: you pay to have your paper published. It's essentially an advertisement you buy. From then on, it's a human centipede down to reddit. Nature also runs spotlight articles, editorial notes, etc. which are more aimed to a general reader (but still try to be exact); then you have the I-fucking-love-science websites which bullshit you about the latest paper (zomg! scientists CONFIRMED [bullshit that's not even in the paper they're claiming to cover]) and only rarely bothers to actually link the original publication (it's paywalled anyway lol), and then it trickles down to reddit where dumbasses soyface about how science is based or something.