"Some forms of open access are, it seems, more open than others, with an American Scientist Open Access Forum debate ending with a consensus that a number of open access (OA) models exist", reports IWR
http://www.iwr.co.uk/information-world-review/news/2172651/whn-open-access-really-open

The debate was in early December. I agree with the statement by Peter Murray-Rust:
http://listserver.sigmaxi.org/sc/wa.exe?A2=ind06&L=american-scientist-open-access-forum&D=1&O=D&F=l&S=&P=100729

See also his blog entry:
http://wwmm.ch.cam.ac.uk/blogs/murrayrust/?p=203

Harnad and others are permanently ignoring that all relevant OA declarations (BOAI, Bethesda, Berlin) have as essential point the removal of "permission barriers" (Suber).

The Berlin declaration had the form of a license (requesting that an OA document should be labeled as such by addition of the text of the declaration) but the growing popularity of the Creative Commons licenses makes them the first choice.

PLoS and BMC's Creative Commons Licenses CC-BY are exactly that what the OA definitions request. Whether excluding commercial use nor forbidding derivative works is in consent with these definitions. Each scholar should understand what he signs. If a scholar had signed BOAI he has explicitely given support to the aim that scholarly output should be really free. This means:
*mirroring the article in other repositories is allowed
*class room use is allowed
*using it for commercial online-media is allowed
*translating in another languages is allowed
and so on.

Suber seems to have decided not to attack Harnad although Harnad's position is not on the ground of the relevant OA definitions. The most authoritative source for OA is Suber's website and his interpretation of OA is unmistakable.

You can find a summarry of Suber's position and my critical comments (in English) at:
http://archiv.twoday.net/stories/320964/

Most DOAJ journals are definitively not OA in the Berlin/BOAI sense.

Should be satisfied with free breakfeast if our aim was to have also free lunch?

*OA advocates should underline all the time they speak about OA that OA is both: cost free and free of permission barriers.

*OA repositories and OA journals should allow (and request) CC licenses for all content.

*The default CC license should be CC-BY.

*Mandatory deposit regulations for funded research should contain the duty to remove permission barriers (according to the last two points)

*If an author or an institution pays a lot of money for OA in a so-called hybrid journal the content should be free of permission barriers (default CC license).

*We need a registry (using OAI-PMH) for scholarly OA content free of permission barriers (whether DOAJ nor OAIster nor the CC search allows to filter such content).

This is also essential for advocates of open content projects like the Wikipedia and for advocates of OA for cultural heritage items in archives, libraries and museums. (And for advocates of open data too, where the remix-ability is a core feature of "Open Science").