Integrating University and Funder Open Access Mandates

[Stevan Harnad <harnad@ecs.soton.ac.uk>]

                      ** Apologies for Cross-Posting **

There is a simple, natural, universal way to integrate (a) funder 
open- access mandates and (b) university open-access mandates, 
reconciling the NIH and Harvard OA mandates, as well as making 
the two kinds of mandate synergistic and mutually reinforcing:

(i) Separate the deposit requirement from the open access 
requirement.

(ii) Separate the issue of the locus and timing of the deposit 
from the issue of the locus and timing and copyright conditions 
for providing open access to the deposit. Both universities and 
funders should mandate immediate deposit of the peer-reviewed 
final draft (postprint), in the author's own university's 
Institutional Repository (IR), immediately upon acceptance for 
publication, without exceptions or opt-outs, for institutional 
record-keeping purposes.

Access to that immediate postprint deposit in the author's 
university IR may be set immediately as Open Access if copyright 
conditions permit it; otherwise access can be set as Closed 
Access, pending copyright negotiations or embargoes. (Only the 
metadata are visible and accessible webwide, not the postprint 
full-texts.)

All the rest of the conditions described by universities and 
funders should accordingly apply only to (ii) the locus, timing 
and copyright conditions for providing open access to the 
deposit, not to (i) the depositing itself, its locus or its 
timing.

That way:

(1) there will be a systematic (and natural) common locus of 
direct deposit for all research output worldwide;

(2) university mandates will reinforce and monitor compliance 
with funder mandates;

(3) funder mandates will reinforce university mandates;

(4) legal details concerning open-access provision, copyright and 
embargoes can be handled separately on a case by case basis, 
according to the conditions of the mandate (instead of needlessly 
making (1)-(3) contingent on each case);

(5) opt-outs will apply only to copyright negotiations, not to 
deposit itself, nor its timing; and

(6) central OA repositories (like PubMed Central) can then 
harvest the postprints from the authors' IRs under the agreed 
conditions at the agreed time.

Right now, the NIH mandate requires that the postprint must be 
"submitted" immediately upon acceptance for publication (which is 
excellent!), but it does not specify how or where to submit it!

The obvious solution is that the postprint should be directly 
deposited, immediately upon acceptance for publication, into the 
researcher's own university's (or institution's) IR -- possibly 
as Closed Access rather than Open Access, depending on copyright 
and embargo conditions and negotiations. (NIH can then be sent 
the URL, and given access privileges.)

The recommendations of the SPARC/Science Commons/ARL joint white 
paper "Complying with the NIH Public Access Policy - Copyright 
Considerations and Options" by Michael Carroll are all excellent: 
Their only flaw is in not separating those valid and helpful 
considerations and options from the question of the locus and 
timing of the deposit itself. That locus should always be the 
author's IR, and the timing should always be immediately upon 
acceptance for publication. None of the copyright considerations 
are pertinent to the deposit itself: They apply only to the 
provision of open access to the deposits.

In exactly the same way, the Harvard mandate is excellent in 
every respect except that it too conflates the deposit itself 
with the copyright and embargo considerations and options -- 
which should only apply to to the time when open access to the 
deposit is provided, not to the making or timing of the deposit 
itself. The Harvard mandate offers the option of opting out 
option from the requirement to negotiate copyright retention. 
That makes the Harvard mandate into a non-mandate unless the 
copyright requirement, with opt-out, is separated from a deposit 
requirement, without opt-out.

The solution proposed here is simple, natural, solves both the 
NIH and Harvard problems at once, makes the funder and university 
mandates complementary and convergent, and provides an 
integrated, synergistic OA mandate model for both funders and 
universities that will systematically scale to all worldwide 
research input.

I hope that funders and universities will give this integrative 
proposal serious thought, rather than just pressing ahead with 
the current NIH and Harvard models, both of them welcome and 
timely, but both in need of this small yet crucial revision to 
ensure their coherence and success.

It is noteworthy that three recommendations were made to NIH 
three years ago: (1) mandate immediate deposit, with no opt-out, 
(2) specify direct deposit in the fundee's university IR, and (3) 
harvest into PubMed Central.

Those recommendations were not followed, and after three years 
the NIH policy was acknowledged to have failed. Because of that 
failure, the policy has very recently been upgraded to an 
immediate-deposit mandate (1). But there are already signs (from 
the very similar Wellcome Trust mandate) that 
systematicmonitoring mechanisms are needed to ensure compliance 
with funder mandates.

University mandates are the obvious means of reinforcing and 
monitoring compliance with funder mandates (as part of the 
fulfillment conditions for receiving the grant overheads and 
indirect costs allotments). Moreover, university IRs are also the 
natural, convergent locus for direct deposit of all research 
output, the universities being the providers of the research, 
both funded and unfunded, with a direct institutional interest in 
archiving, recording, assessing, showcasing their own research 
output as well as in and maximizing its uptake, usage and impact.

Funder mandates like NIH's will naturally reinforce university 
mandates, like Harvard's. The two mandating parties simply have 
to agree on separating the universal issue of immediate deposit 
(and locus of direct deposit) from the independent issue of 
copyright, embargoes, the timing of the provision of Open Access 
to the deposit, and whatever further central repositories may 
wish to harvest the OA deposit or its metadata.

Stevan Harnad
American Scientist Open Access Forum