I get that everyone wants to read and publish everything for free - and they should be able to: all transaction costs are barriers to participation. But we need actual solutions to achieve this, not "solutions" that just shove the costs onto somebody else.

As I've written elsewhere, the proper way for the system to be supported is for funders and research institutions to directly subsidize publishing infrastructure and staff to publish all scientific work in a completely free and open way (i.e. no costs for publishing or access).

These groups currently spend on the order of $8b per year on publishing -- around $5,000 per published paper - mostly via subscriptions, some via page charges and APCs. That is way more than enough to support a completely open publishing system for the entire world.

What we need are for the funders who know the system is broken - and who are taking steps already to try to fix it - to realize that tinkering isn't enough and that a radical restructuring of the way publishing is paid for will benefit science and scientists.

Things like Plan S are an important step in the right direction - but only if they are a step and the current status of subsidizing APCs for researchers they fund is transitional and not permanent (the people behind Plan S get this).