Sometimes I really wish people who try to write very confidently about a subject would at least make a token effort at trying to research the topic first. Or, failing that, perhaps because they don't have the time, at least try not to sound so damn Dunning-Kruger about it.
The listicle website Cracked recently tried to explain why science is broken, in six bullet points (hat tip: RationalWiki):
6. Publication bias towards positive results.
5. Scientists not having to make the data underlying their studies available.
4. Scientists having to pay publication fees, which creates perverse incentives and restricts access to publication venues for scientists from poorer countries.
3. Big publishing companies earning obscene profits by charging journal subscription fees in addition to publication fees, thereby restricting access to the results of research.
2. Scientists being forbidden from giving out reprints of their papers because they signed over the copyright to aforementioned publishers.
1. Predatory journals publishing nonsense science.
Except for the first point (#6), all of this is either misleading or self-contradictory.
At least in my area there are subscription based journals and open access journals. In the former case, the problem is that people without institutional access or in poor institutions cannot access the publications, which is bad (#3). In the second case, the problem is that scientists from poor institutions cannot publish, and further that there are perverse incentives if journals earn more money for accepting sub-par articles; and that is also bad (#1, #4). But this post makes out as if both is happening at the same time. It isn't.
It is a classic trade-off situation. Because somebody has to pay for publication somehow, you have to pick one of those problems, whichever you think is the lesser. Unless somebody can convince the voting public to make all journals tax-financed open access non-profit public utilities. I would totally be on board with that, but good luck.
What is more: you can't hand out a copy (#2)? Nonsense. The copyright agreements always include a clause allowing the author to ... hand out copies. We constantly do that. You send me an email requesting one of my papers, you get it. That's how we worked fifteen years ago, that is how we work today. How can somebody write a post like that cracked one without understanding this simple fact?
Finally, #5 is just wrong in my area. You couldn't get a paper published without showing a data table, having your data matrices and trees submitted to TreeBase, or having your sequence data submitted to GenBank - whichever is the most appropriate to your specific case. Maybe there are areas where it doesn't work like that, but systematic biology definitely does not have that problem, so one can hardly say that all of science has it.