> The paid journals take research which is mostly paid for with public or non-profit money, and hide it behind paywalls. You can't avoid this fact.

Commercial publishers have paywalls, but for new research they no longer have to be the only source for publications.

Nowadays, I think it’s often the indifference/laziness/whatever of the authors that prevent accepted manuscripts (same text, but different layout) from also being freely accessible.

For example https://www.elsevier.com/about/policies/sharing allows putting accepted manuscripts on arXiv.

The way I read it, the main limitations are:

- you can’t put your paper on a commercial site.

- you have to mention the DOI, which, I guess, resolves to Elsevier’s site.

Reading https://www.elsevier.com/about/policies/hosting (“Sites or repositories that provide a service to other organizations or agencies, even if those other organizations or agencies are themselves non-commercial entities, are considered to be providing a commercial service, and this service activity will also require a commercial arrangement with Elsevier”), they make special exceptions for arXiv and RePEc.

So, that’s not optimal, but (for new papers) also not as bleak as it typically is described.

Who is paying and who has access is only one part of the puzzle. Most people only care about access, but they don't understand why that access continues to be limited.

Why do journals exist? It's not to provide information. Ever since the Internet was invented, we can distribute information virtually for free. Everybody knows this. Yet the journals persist for decades. So their purpose is not to provide information.

Yes, of course, their reputation is invented and legacy. It's been shown time and again that papers published in "reputable journals" can be quite problematic. But everybody knows this too. It's not like academic institutes are completely brainless. They know they could have somebody "less reputable" publish their information and it would have the same scientific merit, or that they could even publish it themselves on a blog. But they don't; they publish on the "reputable journals", even though the reputation is clearly not impacting the research results.

So why do these publishers exist? The true purpose of paid, "reputable" journals is to provide an excuse for research institutions to dole out money to people who meet a quasi-arbitrary barrier to the money. They know they don't have any good system of how to assign money, or who to promote, because in general it's hard to quantify. So they hide behind "the reputable journal" and thus the "reputation" of their researchers. This way they can receive more money (because "our scientists are published by reputable journals") and they can dole it out just as easily.

Opening access and reducing cost is a great idea. But shunting money away from journals will result in the entire research industry scrambling to put together a replacement that will allow them to continue being funded, determine how to dole out that funding, organize journals "for free", and retain some sort of rigor/due-diligence/quality from the publishing process. Can this be done? Sure! But we should make that our goal and not ignore the big pink elephant in the room, which is that journals are still a necessary evil for research funding. If we want to get rid of paid journals, let's actually think through the resulting impact and build a resilient system to replace them. Imploding them and "hoping for the best" is just going to hurt research.