Suber-Harnad strongOA/weakOA borderline

Peter Suber (again mysteriously unable to post comments) and Stevan Harnad clarify the weak/strongOA borderline. This is making it clearer to me. Accept that I completely misread or misunderstood the original statement. I’ll comment below. At present I am interested in seeing where the borderline is set, not what the two partitions are called. When we – and hopefully the community – agree on where the border is I’ll comment on whether I think this is a useful way forward. Note that my concern is now what a “permission-barrier” is.
======================= from Peter Suber =======================
Peter: This is incorrect and confusing. The borderline between strong and weak OA is easy to define. Weak OA removes *no* permission barriers and strong OA removes *at least some* permission barriers. (Both of them remove price barriers.)
The fact that strong OA covers a range of different positions which may or may not be easy to distinguish is not relevant to the distinction between strong and weak OA itself.
Please see again the <a href=“http://www.earlham.edu/~peters/fos/2008/04/strong-and-weak-oa.html”>statement</a> in which Stevan and I introduced terms for describing this distinction. (NB: We all agree on the need for new terms. All I’m doing here is clarifying and reiterating the distinction itself.)
====================== from Stevan Harnad =====================
Stevan Harnad Says:
May 4th, 2008 at 12:50 pm e
The Need to Specify a Minimal Lower Bound for Permission-Barrier-Free OA
I agree that Permission-Barrier-Free OA (by whatever name we give it) needs at least a minimal lower bound to be specified, otherwise it is too vague.
Price-Barrier-Free OA (regardless of what name we gave it) does not need an upper or lower bound, because it is not on a continuum. It just means free access online, However, as I have said before, it does need to be shored up a bit by stating the obvious:
(1) The free access is to the full digital document (not just the metadata).
(2) The free access is one-click and non-gerrymandered: Instant download without having to do a song and dance for every page.
(3) The free access is immediate, not delayed or embargoed. (A document is not OA if it *will be* available free in a year, or in 10 or 10,000.
(4) The free access is permanent and continuous: A document is not OA if it is available free for a limited time, say, for an hour, or on even-numbered calendar months.
(5) There is no “degree of free” access: Low-priced access is not “almost” free access.
(6) The free access is for anyone netwide, not just those at certain sites or in certain domains or regions.
For Green Price-Barrier-Free OA self-archiving and Green Price-Barrier-Free OA mandates all of these specifications are dead-obvious, irrespective or what proper name we choose for it. They are spelled out only for the pedantic and the obtuse.
But in the case of Permission-Barrier-Free OA, regardless of the name (and even in the case of the BBB definition), a minimal lower bound has to be specified, otherwise the condition is so vague as to make no sense. It is not just Peter Murray-Rust who is perplexed. Anyone would be. The BBB definition gives examples, but it does not give a lower bound. That is like saying “hot” means temperatures like 30 degrees, 300 degrees or 3000 degrees. That leaves one in perplexity about what, between 0 degrees and 30 degrees, counts as not hot. In particular, does Price-Barrier-Free OA alone count as Permission-Barrier-Free OA? The answer is No, but the only way to give this content is to specify a minimal lower bound for Permission-Barrier-Free OA.
========================== PMR =====================
This is useful. I was going to address the price-free criterion but Stevan has done a good job. I’ll add one or two additional concerns and see if he or anyone thinks they should be included
I hope we can agree on a spectrum like this:
closedAccess || weakOA ||<– strongOA –>| removal-of-all-permission-barriers (BBB, CC-BY)
Here || means a complete, clear barrier while | means potentially inclusive.  <– –> means a range of options. To help clarify:

• closedAccess means either price barriers and/or access restrictions (specific logins) and/or embargo period and/or metadataOnly and/or abstractOnly and/or limited access to multiple documents.
• weakOA means that anyone anywhere can bring up the full document on their screen (as often as they wish). They have no other rights other than “fair-use” (which is not relevant to OA IMO). All possible permission-barriers are in place. [we need to discuss these later.] I would interpret this a meaning there is only one type of weakOA and manifestations differ only in non-OA matters.
• strongOA covers a spectrum defined by the removal of one or more permission barriers. Let’s assume there are N such. Then a document with 1 PB removed and N-1 PBs remaining is strongOA and so is a document with 1 PB remaining and N-1 removed. If all N PBs are removed this is a special case.
• BBB, CC-BY are a special case of strongOA with no permission barriers that could be potentially removed.

If we can agree on this that will be useful. If not I have again failed to understand.
Now weakOA. Here are some additional comments for Stevan’s list:

• weakOA must have the explicit or implicit permission of the publisher. Self-archiving (by individuals or institutions) forbidden by the publisher could lead to take-downs and is not therefore permanent OA.
• I’d replace “gerrymandering” (which is to do with an electoral system) by “restricted technical access”. This includes DRM, explicit logins and restriction of the number or type of downloads.
• weakOA in hybrid journals overrides any contractual obligations provided by the reader’s institution. Thus, as a domestic individual, I may be allowed to access multiple articles in a hybrid journal but if I do this from a university address the system might cut me off for overreaching the limit of downloads

weakOA has ALL conceivable permission barriers. We need to know what they are, because otherwise we cannot define anything as weakOA. Here are some currently enforced permission-barriers.

• you may not store a local copy of the article on your machine
• you may not mail copies of the article
• you may not reproduce any of the article beyond fair use
• you may not index it not publish an index
• you may not use if for data or text-mining
• you may not make print copies of the article

Stevan said:

Most Gold OA today is just Price-Barrier-Free OA.

which implies that this retains all the permissions. However if, for example, we regard the prohibition of storage on disk as a permission-barrier and if, then, a publisher removes it, then it makes the article strongOA. So it is probably easy to claim that almost all OA is strongOA. If so the use of “any permission-removal” to define strongOA makes it almost meaningless as it is always possible to find something you can do without ending up in court.
Note, in passing, that the NIH/PMC system is not even weakOA, so not OA at all. There is a restriction on the number of articles that can be downloaded so that if I access any particular one I may be cut off.
That’s not any form of Open Access