Here is Shieber's post.
Here is Harnad's full response.
And pasted below is Harnad's summary of his rejoinder, taken from the American Scientist Open Access Forum (post # 2009-117):
Date: Fri, 12 Jun 2009 17:13:33 -0400
From: Stevan Harnad
Subject: The Argument Against (Premature) Gold OA Support
** Cross-Posted **
I have written a response to
"The argument for gold OA support" by Stuart Shieber.
The full response is at:
http://openaccess.eprints.org/index.php?/archives/590-guid.html "The Argument Against (Premature) Gold OA Support"
Here is just the summary:
What is needed in order to provide universal OA as quickly and surely as possible is for universities (and funders) to mandate that their own researchers provide (Green) OA by depositing their articles in their institution's OA repository immediately upon acceptance for publication. It is both a strategic and a conceptual mistake to think that money has to be spent at this time on paying for publishing in Gold OA journals. Gold OA journals' time will come if and when universal Green OA makes subscriptions unsustainable. Then publishers will cut costs and downsize to just providing the service of managing peer review, paid for by institutions out of their windfall subscription cancellation savings. Universities and funders should not be either distracted or deterred from mandating Green OA now by thinking that they first need to provide funds to pay for Gold OA.
(Once they have adopted a Green OA mandate, this is no longer a distraction or deterrent and they can of course do whatever they like with their spare cash.)
(1) Any needless cost at all associated with adopting and implementing a Green OA mandate is a deterrent to arriving at consensus on adoption, not an incentive.
(2) Minimal costs for Harvard U are not necessarily minimal for HaveNot U.
(3) The way to explain the possible eventual transition to universal Gold OA is via its causal antecedent: universal Green OA.
(4) The way to allay worries about Learned Society Publishers=92 future after universal Green OA is to explain the simple, straightforward relation between institutional subscription collapse and institutional subscription cancellation savings, and how it releases the funds to continue paying for publication via Gold OA. (And remind faculty that if their institutions really want to keep subsidizing Learned Society publishers' "good works" (conferences, scholarships, lobbying) as they are now through subscription-fees, they can certainly continue to do so through publication fees too, as a surcharge, on the Gold OA model, if they wish.)
(5) Reserve any plans for promoting pre-emptive payment of Gold OA fees for those institutions that have already mandated Green OA (and preferably only after we are further along the road from 85 mandates to 10,000!).
(6) Pre-emptive payment for Gold OA before universal Green OA just retards and distracts from providing and mandating Green OA. Moreover, it is incoherent and does not scale ("universalize"): It is like an Escher drawing, leading nowhere, even though it seems to.