Optimize the NIH Mandate Now: Deposit Institutionally, Harvest Centrally

The January issue of Peter Suber's SPARC Open Access Newsletter is superb, and I recommend it highly as a historical record of the milestone reached by the OA movement at this pivotal moment. There is no question but that the NIH Green OA self-archiving mandate is the biggest OA development to date, and heralds much more.

There remains, however, an important point that does need to be brought out, because it's not over till we reach 100% OA, because mistakes have been made before, because those mistakes took longer than necessary to correct, and because a big mistake (concerning the locus of the deposit) still continues to be made.

First, a slight correction on the chronometric facts:

Peter Suber wrote:"If NIH had adopted an OA mandate in 2004 when Congress originally asked it to do so, it would have been the first anywhere. Now it will be the 21st."

Actually, if the NIH OA mandate had been adopted when the House Appropriations Committee originally recommended it in September 2004, it would have been the world's third Green OA self-archiving mandate, not the first. And Congress's recommendation in September 2004 was the second governmental recommendation to mandate Green OA self-archiving: The first had been the UK Parliamentary Select Committee's recommendation in July 2004.

(1) The Southampton ECS departmental mandate was (as far as I know) the very first Green OA self-archiving mandate of all; it was announced in January 2003 (but actually adopted even earlier). QUT's was the second OA mandate, but the first university-wide one, and was announced in February 2004. (See ROARMAP.)

(2) The UK Parliament's Science and Technology Committee Recommendation to mandate Green OA self-archiving was made in session 2003-04 and published in July 2004 (i.e., before September 2004, when the US House Appropriations Committee made its recommendation).

Moreover, the recommendation to mandate self-archiving had not only been part of the BOAI Self-Archiving FAQ from its inception in 2002, but the FAQ's contents had actually preceded the existence of the BOAI by several years, with the recommendation itself -- that departments, universities and funders should mandate self-archiving -- already in circulation since about 1999. (The FAQ was also already quite specific at that time about mandating the self-archiving of the author's final accepted draft, rather than the publisher's PDF. Its one glaring error was to advocate central self-archiving -- but that was corrected as soon as the OAI protocol was formulated, making it possible to create the first OAI-compliant Institutional Repository software in 2000, thereby returning to the original distributed, institutional model of self-archiving of 1994.)

In contrast, to see where the precursor to the NIH mandate stood in 1999, one must re-read the original e-biomed proposal of May 1999. There was still a bumpy and meandering road ahead (via the PLoS petition in 2001 and the Bethesda Statement in 2003), with several false starts and dead ends (among them the first NIH non-mandate itself!), before the realization that what had been needed all along was self-archiving and a Green OA self-archiving mandate.

"A Simple Way to Optimize the NIH Public Access Policy" (Oct 2004)

Now NIH's has indeed instantly become by far the most important of the Green OA self-archiving mandates to date in virtue of its size and scope alone, but it still hasn't got it right!

The upgrade from a mere request to an Immediate-Deposit/Optional-Access (ID/OA) mandate was indeed an enormous improvement, but there still remains the extremely counterproductive and unnecessary insistence on direct deposit in PubMed Central. This is still a big defect in the NIH mandate, effectively preventing it from strengthening, building upon and complementing direct deposit in Institutional Repositories, and thereby losing the golden (or rather green!) opportunity to scale up to cover all of research output, in all fields, from all institutions, worldwide, rather than just NIH-funded biomedical research in PubMed Central: an altogether unnecessary, dysfunctional, self-imposed constraint (in much the same spirit as having requested self-archiving instead of mandating it for the past three lost years).

Even the benefits of Congress's wise decision to mandate deposit immediately upon acceptance for publication -- thereby transferring the allowable 12-month embargo to the date at which access to that deposit is set to Open Access, rather than allowing any delay in the date on which the deposit itself is done -- are lost if that deposit is required to be made directly in PubMed Central, rather than in each author's own Institutional Repository (and thence harvested to PubMed Central): With direct IR deposit, authors can use their own IR's "email eprint request" button to fulfill would-be users' access needs during any embargo. And, most important of all, with direct IR deposit mandated by NIH, each of the world's universities and research institutions can go on to complement the NIH self-archiving mandate for the NIH-funded fraction of its research output with an institutional mandate to deposit the rest of its research output, likewise to be deposited in its own IR. This will systematically scale up to 100% OA.

The hope is that -- recognizing that similar mistakes have been made in the past, and that that has cost dearly in years of lost OA, and recognizing that the remedy is ever so simple, with no loss, only gain ("Deposit Institutionally, Harvest Centrally") -- the NIH will still have the sound sense, in the euphoria over the successful passage of the mandate itself, to optimize its mandate now, so it can do the maximal good in the minimal time, across all fields and institutions, worldwide.

"Optimizing OA Self-Archiving Mandates: What? Where? When? Why? How?"

Stevan Harnad
American Scientist Open Access Forum