It is important for those of us who have been advocating for open access for many years to remember that there is still much work to do educating people about what open access is; from time to time, let's remember to get back to the basics. As open access mandate policies are developed, we need to watch for, and correct, misunderstandings.
Here is the definition of open access, from the Budapest Open Access Initiative:
By "open access" to this literature, we mean its free availability on the public internet, permitting any users to read, download, copy, distribute, print, search, or link to the full texts of these articles, crawl them for indexing, pass them as data to software, or use them for any other lawful purpose, without financial, legal, or technical barriers other than those inseparable from gaining access to the internet itself. The only constraint on reproduction and distribution, and the only role for copyright in this domain, should be to give authors control over the integrity of their work and the right to be properly acknowledged and cited.
The only element missing from this definition, is that true open access means that an item is available open access immediately on publication, not after a delay period.
The literature that should be given away for free, is the literature that scholars have always given away for free. Scholars traditionally give away their peer-reviewed journal articles. Peer reviewers are not paid for their work, either.
There are two basic types of open access:
Open Access Archiving (or the green approach): the author (or someone representing the author) makes a copy of the author's work openly available, separate from the publication process. That is, the article may be published in a traditional subscription-based journal. The version of the article that is self-archived is the author's own copy of the work, reflecting changes from the peer review process (all the work that is provided for free), not the publisher's version.
Open Access Publishing (or the gold approach): the publisher makes the work open access, as part of the process of publication.
Research funders' policies requiring open access always apply to the recipient of the funding, the author. Policies either allow either approach to open access, or they specify the open access archiving, or green approach. The green approach is broader; an article published in an open access journal is available for deposit in an open access archive. Therefore, a green open access policy provides for either approach to providing open access.
Thanks to Peter Suber and Stevan Harnad for pointing out the need to focus on restating the basics of OA.