Two aspects of Stevan Harnad's vision for Open Archiving of postprints
concern me.

The first is a practical problem from a researcher's perspective of
using self-archived postprints. Harnad recommends that the author's
final, peer-reviewed draft is what is self-archived. But how, then, do
authors cite a version that does not share the pagination (or volume or
issue) of the published paper? Won't they still have to obtain the
published paper, anyway? How then does this version of OA solve the
access problem?

My second concern is how links will be made from bibliographic databases
to institutional repositories (IRs). Will link resolvers have to have in
their databases every IR in the world? In theory that can work, but in
reality how expensive will it be to maintain the database of links, and
how reliable can such a distributed resource be? Perhaps IRs do not have
to be linked from bibliographic databases with controlled vocabulary and
careful indexing, but then users would have to check OAIster or Google
Scholar instead. Is that cure worse than the disease?

Steve Black

Reference, Serials, and Instruction Librarian

Neil Hellman Library

The College of Saint Rose

392 Western Ave.

Albany, NY 12203

(518) 458-5494

blacks@strose.edu