thank you for pointing this out – do you have any idea how Google are determining that a particular version is OA?

I’m just wondering how applicable this might be to other types of open content (in repositories and elsewhere)

cheers,
John

I guess that if they can view an associated file (PDF etc) and it isn’t on a publisher site where they have an agreement to harvest publications but not cache them, then they assume it is OA. Just a guess though.

Thanks,

Stuart