The solution is on the funding side and the reader side, not the publisher side. Scientists have to publish in order to get other scientists to read their work

The solution is on the funding side and the reader side, not the publisher side. Scientists have to publish in order to get other scientists to read their work and (hopefully) refer to it in their own work. This business of referring to other scientific work is what drives academic careers and the allocation of research budgets. Most of the better universities look at metrics that take this into account. The problem is all the second rate universities haggling for funding trying to boost their metrics. Promoting scientific work is essentially an SEO problem. How do you get other scientists to take your work seriously? Well, they need to find your work when they are doing research for their own work. Getting the PDF distributed is the easy part of the problem. That's a solved problem since the nineties. It doesn't really cost anything worth talking about. It's not 0$ but close enough that the difference isn't really worth talking about. Publishers don't matter for distribution any more; it's all digital now. Any website that Google can index will do for distribution. The goto scientific SEO process hasn't really changed in centuries. You submit your article to some exclusive groups of researchers, anonymous peer review is arranged, and if deemed acceptable the article is included in some publication. Which these days just means they put it on their website. Actual printed journals are a side show. Some publishers still do it. But mostly it's a waste of paper. The process beyond that is simple good old SEO. You drive traffic by promoting your work at conferences, workshops, writing books, by talking to others, etc. This is also how you become part of these exclusive groups. You'll be invited to take part in peer reviewing, editorial work, conference organizing, etc. And in return, you are expected to refer back to work of others in the same group. The odd thing is that scientists don't really do any traditional SEO. They aren't buying keywords from Google. They aren't messing with their meta data. And they aren't optimizing their pdfs for readability, AB testing various click baity titles. Nor are they using analytics tools to measure how other scientists engage with their work. They aren't running social media campaigns either. This is super odd. You do all this work and then you throw it over the fence and hope for the best. Why?! This is stupid. The value add of publishing to some paid publisher scam journal that nobody reads and where the peer review is basically bribing the "editor" is of course very low. The key problem is that universities don't scrutinize their metrics enough. That's because they aren't independent. Like researchers, their funding depends on public sources that look at metrics to judge how well the money is being spent. That process is super corrupt and bureaucratic. A lack of standards and integrity in this process is the root cause of the problem. Scientists referring to work from dubious publications is a giant red flag that they are manipulating their metrics. That giant red flag needs to be taken into account during grant applications for public funding. This isn't rocket science. Like plagiarism is frowned upon, referring to work from disreputable sources should be frowned upon as well. It signals scientists trying to mess with the numbers and not taking their own work and integrity seriously. This should be flagged by editors of reputable publications because it threatens their own reputability. Grant application reviewers should be flagging this as well. Metrics should be created to automate this. Simply flag some publications and calculate the least reputable authors based on their citation habits referring to work in these publications. That will sort things out in no time. If you can calculate an impact factor, you can also calculate a reputability factor.