The basic idea is that the research papers should be open access 6 months after publication. 6 months is enough for publishers to recoup their costs of publicat

The basic idea is that the research papers should be open access 6 months after publication. 6 months is enough for publishers to recoup their costs of publication. The reality is that publishers make all their money from site licenses. Harvard, e.g., will spend $15 million a year on subscriptions to various publishers like Elsevier, in return for access to all the new papers, and the back catalog. Most papers that are downloaded, around 80%, are from the back catalog anyway, meaning that universities are incredibly price insensitive with respect to site license subscriptions. Journals could offer the new papers for free, and still universities would pay for the site license subscription. This is what has happened with the arxiv: most new papers in theoretical physics are on there for free, and yet physics departments still subscribe to the theoretical physics journals. Why is that? It's because the departments need access to papers in the back catalog, which are not on the arxiv. The upshot of this is that site license revenue wouldn't be affected by requiring open access on research 6 months after publication (or even 3 months, or 0 months, in my view). There is enough price insensitivity in the system for existing revenues to be unaffected by a change like this. The reason the journal industry hates legislation like this, despite it not posing an economic danger to them, is that they are worried about the general trend of the government getting involved in changing the scientific publishing industry. The journal industry would like to freeze their business model in the form that it was in during the pre-internet days. They believe that it's in their interest to fight every piece of pro-change legislation that the government suggests, in order to slow down the rate of change.