Exchange in SIGMETRICS with David Wojick, a consultant to OSTI:

DW: "the non-NIH US agencies are implementing the OSTP mandate for a 12 month delayed access program, just as NIH already does."

That the OSTP mandate requires providing OA within a year (at most) is well-known.

But how each agency will formulate and implement its mandate is definitely not well-known, nor even fully decided as yet: it is still being worked on, agency by agency (and I'm sure Peter Suber, Heather Joseph, Alma Swan and others with expertise in OA and OA mandates are being consulted).

The most important practical implementation issues are:

#1 Who must make the paper OA? the fundee or the publisher? Obviously for a uniform, systematically verifiable mandate, it must be the fundee, the one bound by the mandate, and not the publisher, the one that is in conflict of interest with the mandate, and not bound to comply with it (except if paid extra money).

#2 Where must the paper be made OA? Here again, for a uniform, systematically verifiable mandate, it must be in one verifiable locus, and the only locus shared by all fundees, all funders and all institutions (and for both Green and Gold OA) is the fundee's own institutional repository - from whence it can be exported or harvested to other sites, such as PubMed Central, if and when needed.

#3 When must the paper be made OA? (The mandate already stipulates this: within 12 months of publication at the latest.)

#4 When must the paper be deposited? This is the most important question of all, and carries with it the answer to the other questions: the fundee must deposit the final, refereed, accepted draft, immediately upon acceptance for publication -- not 12 months after publication -- irrespective of whether it is published in a subscription journal or a Gold OA journal, irrespective of whether the deposit is immediately made OA or embargoed, and irrespective of whether the journal endorses immediate OA or imposes an OA embargo.

It is #4 that holds the key to a successful and effective OA mandate, the Liège model "Immediate-Deposit/Optional-Access" model (which Peter Suber calls the "Dual Deposit/Release" model). The model has been tried and tested, and has already proven to be more effective than any other mandate model, and is both compatible with and subsumes all the other mandate models.

The key to the Liège model's success is that it is convergent and systematic rather than divergent and anarchic, mobilizing the universal source of all research, funded and unfunded, Green, Grey and Gold, across all disciplines -- the fundee's own institution -- to monitor and ensure timely compliance as well as to tide over any embargo with the repository's facilitated copy-request Button.

All of this depends on requiring deposit, by the fundee, in the institutional repository, immediately upon acceptance for publication, which is the only universal, objective, verifiable calendar date of reference for timely compliance. (Publication dates diverge wildly from both the acceptance date and the actual date of appearance of the journal. Whereas a 12 month embargo is the number to beat, publication date can lead to an uncertainty of as much as two years or more.)

Gargouri, Y., Lariviere, V., Gingras, Y., Brody, T., Carr, L., & Harnad, S. (2012a). Testing the Finch Hypothesis on Green OA Mandate Ineffectiveness. arXiv preprint

Rentier, B., & Thirion, P. (2011). The Liège ORBi model: Mandatory policy without rights retention but....

DW: "If you know of an agency that is doing something else I would like to hear about it. Note that NIH has half of the Federal basic research budget so this is merely rounding out the existing program."

No U.S. funding agency has yet adopted the immediate-deposit clause, but it has been adopted by the FNRS in Belgium, and has been proposed by HEFCE in the UK. It is also implicit (though not yet implemented or enforced) in the Harvard mandate model.

DW: "The only big issue at this point is whether the non-NIH agencies will collect and post accepted manuscripts, as NIH does, but perhaps via SHARE repositories, or use CHORUS and link to the publisher websites."

You leave out the most important option of all, which is that all papers are deposited in the fundee's own institutional repository (and exported if/when desired, to institution-external repositories).

And of course on no account should the depositor or the locus be the publisher (although of course the institutional repository can and will also link to the version on the publisher's site, whether subscription or Gold, OA or embargoed).

I hope all the US funding agencies are likewise taking advice on implementation from those who represent the interests of the research community rather than the publishing community.

DW: "Stevan, I am well aware of your vision. I have read your NRC submission. It just does not happen to be what the US Government is implementing."

It may not be what is being implemented at OSTI, where you are advising, but have you read what each of the other agencies is doing?

DW: "The Brits wanted the US to follow them, but that too is not happening."

And a good thing too, since the Finch/RCUK Policy U-Turn was a disaster. But HEFCE and BIS now look to be fixing that...

DW: "The situation is as I describe it."

Perhaps at OSTI. The rest remains to be seen.

The OA movement has won some and lost some, across the years, but it's not over till it's over...

(1994) A Subversive Proposal

(2001) The Self-Archiving Initiative

(2002) The Budapest Open Access Initiative

Harnad, S. (2004a) Memorandum to UK To UK Government Science and Technology Select Com... Select Committee on Science and Technology Written Evidence

Harnad, S. (2004b) For Whom the Gate Tolls? Select Committee on Science and Technology Written Evidence

Harnad, S. (2007). No Need for Canadian PubMed Central: CIHR Should Mandate IR Deposit.

Harnad, S. (2011) What Is To Be Done About Public Access to Peer-Reviewed Scholarly P... (Response to US OSTP RFI).

Harnad, S. (2011) Comments on Open Access FAQ of German Alliance of Scientific Organi... (Allianz der deutschen Wissenschaftsorganisationen).

Harnad, S (2012) Digital Research: How and Why the RCUK Open Access Policy Needs to .... Digital Research 2012.

Harnad, S. (2013). Harnad Response to HEFCE REF OA Policy Consultation. HEFCE.

Harnad, S. (2013). Harnad Comments on HEFCE/REF Open Access Mandate Proposal. Open access and submissions to the REF post-2014

Harnad, S. (2013) Harnad Evidence to House of Lords Science and Technology Select Com.... House of Lords Science and Technology Committee on Open Access, Winter Issue, 119-123.

Harnad, S. (2013) Harnad Evidence to BIS Select Committee Inquiry on Open Access. Written Evidence to BIS Select Committee Inquiry on Open Access, Winter Issue

Harnad, S. (2013). Follow-Up Comments for BIS Select Committee on Open Access. UK Parliament Publications and Records.

Harnad, Stevan (2013) Recommandation au ministre québécois de l'enseignement supérieur.

Multiple Comments on CIHR Open Access Policy

Multiple Comments on SSHRC Open Access Policy

Multiple Comments on OA Progress in Canada

Multiple Comments on NIH Public Access Policy

Multiple Comments on Harvard Open Access Policy

Multiple Comments on France/HAL Open Access Policy

Comments on H. Varmus's 1999 E-biomed Proposal [1] [2]