They forgot that part of #3 is peer review, which is managed by the journal, but actually performed by academics for free. Oh, and the researcher usually has to pay a fee to get the paper published once the journal accepts their paper. So, most of the hard work in #3 is free, and the academic still has to pay the journal (usually from a government grant).
Obviously for your work to be validated it must go through the process of peer review. However, I am suggesting that even after they become published in a scientific journal that the researchers should go about publishing their articles online themselves or through a third party site. I just don't see how the data can be easily locked away behind a pay wall if researchers can do whatever they please with their work. If they want to make their work available to the public there are many ways to do so.
When publishing at one of the paid journals the researchers must give away their copyright to the journal too. This means they cannot republish the paper on a different place (sometimes there is an exception for the personal website though).
You might not believe this at first but yes, its really that stupid.
