This time, it's MIT. Whether or not you wish to buy MIT's avowal of a principled stand for open access, we can at least acknowledge that there is a pattern developing. I am under the impression (based on meetings I've been part of in the last couple months) that my university system is getting ready to do likewise - although more for budgetary reasons. See, a subscription to Elsevier's stable of journals is very expensive. And yes, Elsevier does have some open access options, but when you're looking at spending $3000 or more per published article, that's untenable. That would practically bust my departmental and College budget (for publication of one article!), given the cuts we are facing this coming fiscal year. The STEM College at my university will be most directly affected, but so too will those of us in the behavioral and social sciences in my College. There are workarounds. Interlibrary loan still works. No patience for that? Use a browser with a VPN (like Opera) and use Sci-Hub. You'll get your article either way. No muss. No fuss.
Here's a radical idea, although hardly novel or original: Open access should be the norm, rather than the exception. In the meantime, if you can, make preprints public. There are many options available. What we produce as scientific workers is intended to be a public good and should be treated as such. Here's another radical idea (also not novel or original): reclaim scientific publishing as a public good that is operated within the public sector rather than privatized. Maybe we can give non-profit organizations that respect open access a pass, but otherwise, the fruits of our labor should be driven by the needs of our citizens - from the peer review process all the way to the finished published article, and archived data and code. Finally, yet another radical idea: do away with all these metrics that give journals "prestige". What matters more is the work itself. Where it appears is of less importance. Finally, we need to provide public support to post-peer reviewers. That in itself can be a full-time job for those who have the interest and talent to pursue it. Published work can and should be scrutinized. Heck, mine has. I'm grateful for that. The initial peer review process is adequate for the purposes of initial filtering, but insufficient for catching all the many potential flaws with the papers that are reviewed. There are folks like Elizabeth Bik, Nick Brown, James Heathers, etc., who do invaluable work data sleuthing on a very regular basis. If anything, we need more who are willing to jump into the fray. To do that, they need a way to make a living and do so with adequate job security.
Ultimately, the goal is supposed to be in the spirit of George Miller's notion of giving away the science of Psychology in the public interest. Our system that exists - the one where the public pays taxes so that scientists can submit research to journals run by for-profit companies that farm out peer review to people who do so for no compensation, and upon publication the same public is charged yet again for access to that information - is flat out insane and unsustainable. If we can come out the other side of this already turbulent decade with a system of scientific publishing that is publicly owned and is freely available to the public, at least one thing will have gone right.