Open Access Archivangelism

---

SUMMARY: The Australian Research Council (ARC) has proposed to mandate OA self-archiving by its fundees. SPARC has advised ARC (1) to require retaining non-exclusive rights as well, (2) to require OA self-archiving within 6 months of publication, and (3) to earmark ARC funds for OA journal publication costs, so OA can be immediate. I suggest instead (1) that to mandate self-archiving it is neither necessary nor desirable to mandate retaining non-exclusive rights at this time, (2) that deposit should be mandated immediately upon publication, with any allowable 6-month delay applying not to the timing of the deposit itself but only to the timing of the setting of access to the deposit (as Open Access rather than Closed Access), and (3) that it is neither desirable nor necessary at this time to earmark ARC funds to pay to publish in OA journals for immediate OA: Institutional Repositories' EMAIL EPRINT REQUEST button will be sufficient to tide over user needs during any 6-month embargo interval between deposit and OA. (Australia's OA specialist Arthur Sale concurs.)

---

Across the years, SPARC has often been a great help to the Open Access movement. But SPARC could help so much more if it could take advice, in addition to giving it (sometimes with insufficient information and reflection):

- A Role for SPARC in Freeing the Refereed Literature (Jun 2000)
- SPARC reply
- Comments on the SPARC Position Paper on Institutional Repositories (Aug 2002)
- New SPARC/ARL/ACRL Brochure on Open Access (Jun 2004)
- Eprints, Dspace, or Espace? (Oct 2004)
- "Life After NIH" (Apr 2005)
- A Keystroke Koan For Our Open Access Times (May 2005)
- "Disaggregated Journals" (Jul 2005)

SPARC has given the Australian Research Council the following advice:

(SPARC's advice in boldface, followed in each case by my comment, indented, followed by Australian OA specialist Arthur Sale [AS] commenting on my comment, in italics, double-indented)

SPARC: "Research funders should include in all grants and contracts a provision reserving for the government relevant non-exclusive rights (as described below) to research papers and data."

Fine, but this is not a prerequisite for self-archiving, nor for mandating self-archiving. It is enough if ARC clearly mandates deposit; the rest will take care of itself.

AS: "A sensible fundee will take this action; how sensible they are will remain to be seen. The unsensible ones will have some explaining to do. ARC could have given advice like this, but didn't."

SPARC: "All peer-reviewed research papers and associated data stemming from public funding should be required to be maintained in stable digital repositories that permit free, timely public access, interoperability with other resources on the Internet, and long-term preservation. Exemptions should be strictly limited and justified."

That, presumably, is what the ARC self-archiving mandate amounts to.

AS: "Exactly. And every university in Australia will have access to such a repository by end 2007. 50% already do."

SPARC: "Users should be permitted to read, print, search, link to, or crawl these research outputs. In addition, policies that make possible the download and manipulation of text and data by software tools should be considered."

All unnecessary; all comes with the territory, if self-archiving is mandated. (The policy does not need extra complications: a clear self-archiving mandate simply needs adoption and implementation.)

AS: "Totally agree..."

SPARC: "Deposit of their works in qualified digital archives should be required of all funded investigators, extramural and intramural alike."

Yes, the self-archiving mandate should apply to all funded research.

AS: "It does."

SPARC: "While this responsibility might be delegated to a journal or other agent, to assure accountability the responsibility should ultimately be that of the funds recipient."

Not clear what this refers to, but, yes, it is the fundee who should be mandated to self-archive.

AS: "Yes the onus is on the fundee(s), and especially the principal investigator who has to submit the Final Report."

SPARC: "Public access to research outputs should be provided as early as possible after peer review and acceptance for publication. For research papers, this should be not later than six months after publication in a peer-reviewed journal. This embargo period represents a reasonable, adequate, and fair compromise between the public interest and the needs of journals."

The self-archiving mandate that ARC should adopt is the ID/OA mandate whereby deposit is mandatory immediately upon acceptance for publication, and the embargo (if any, 6 months max.) is applicable only to the date at which access to the deposit is set as Open Access (rather than Closed Access), not to the date of deposit itself. During any Closed Access embargo interval, each repository's semi-automatic EMAIL EPRINT REQUEST button will cover all research usage needs.

AS: "ARC is silent on timing, but I expect a quick transition to the ID/OA policy by fundees. Anything else is a pain - it is easier to do this than run around like a headless chook later. The Research Quality Framework (RQF) will encourage instant mandate because of its citation metrics. NOTE ESPECIALLY THAT THE ARC GUIDELINES DO NOT SIT IN A VACUUM BY THEMSELVES. The National Health and Medical Research Council and the RQF are equally important. "

SPARC: "We also recommend that, as a means of further accelerating innovation, a portion of each grant be earmarked to cover the cost of publishing papers in peer-reviewed open-access journals, if authors so choose. This would provide potential readers with immediate access to results, rather than after an embargo period."

The ID/OA mandate -- together with the EMAIL EPRINT button -- already cover all immediate-access needs without needlessly diverting any research money at this time. The time to pay for publication will be if and when self-archiving causes subscriptions to collapse, and if that time ever comes, it will be the saved institutional subscription funds themselves that will pay for the publication costs, with no need to divert already-scarce funds from research. Instead to divert money from research now would be needlessly to double-pay for OA; OA can already be provided by author self-archiving without any further cost.

AS: "This recommendation will certainly be disregarded, correctly in my opinion. ARC has never funded publication costs and does not intend to start now. Australian universities are already funded for publication and subscription costs through the normal block grants and research infrastructure funding. All they have to do is redirect some of their funding as they see fit. The recommendation might accelerate innovation, but it is not the ARC's job to fund innovation in the publishing industry."

Stevan Harnad
American Scientist Open Access Forum

Here are actual and projected growth rate statistics for France's national OA Repository, HAL, kindly supplied by HAL's architect and helmsman, Franck Laloe. France's annual research output is about 12,000 articles per month, so HAL's present spontaneous deposit rate of 1600 articles per month is about the same as the baseline of 15% for spontaneous (unmandated) self-archiving worldwide today.
Monthly Deposits in HAL: Two Extrapolations
[Linearlog/log fit and projections done via Origin]
Spontaneous self-archiving in HAL seems to have begun in about 2002, so it is not clear whether the monthly deposit rate will continue to accelerate or was simply catching up with the baseline at which all other unmandated self-archiving rates have been idling for years now. If HAL's monthly deposit growth rate is indeed exponential, then HAL will reach 100% self-archiving in 5 years without a mandate; if it is a power curve ("puissance") it will take 15 years; if (like Arxiv) it is linear, it will take even longer. (Arxiv's power exponent has been unchangingly quadratic for 15 years, HAL's so far seems ternary)

Franck Laloe: [translated from French] "There are also some small research institutes in France which are already self-archiving 100% of their research output, for example IN2P3, a component (high-energy physics) of CNRS. A team of 3 documentalists deposits 100% of IN2P3 article output in Hal-IN2P3, because on a small scale this is possible. Another example is IFREMER, a small institute for research on seas and oceans. They have a small, well-done archive containing 100% of their output. As to my own field of research, it has been self-archiving at 99% in ArXiv for a long time…"

There seem to be two morals to this story: (1) Even a centralised national archiving system in a centralised country like France, cannot succeed without a national deposit mandate; (2) until France adopts a national deposit mandate, it too, like all other countries, will have to rely on individual institutional (and research-funder) mandates.

Pertinent Prior AmSci Topic Threads:

Are things otherwise in France? (began May, 1999)
INIST/CNRS : nouveautés du site Libre Accès (Dec 2004)
France's INRIA Registers Commitment to Implement Berlin Declaration Self-Archiving Policy Recommendation (Mar 2005)
France's CNRS Registers Commitment to Implement Berlin Declaration Self-Archiving Policy Recommendation (May 2005)
Hélène Bosc et le progrès en accès libre en france (Mar 2006)
Guide juridique CNRS (Nov 2005)
CNRS position on OA : new details (Jun 2006)
Forthcoming OA Developments in France (Jun 2006)
France's HAL, OAI interoperability, and Central vs Institutional Repositories (Oct 2006)

Stevan Harnad
American Scientist Open Access Forum

Executed in gouache on black Canson Mi-Teintes pastel paper using William Mitchell metal nibs, brushes and airbrush.
See: Full Size Original 600 x 420 mm.

Artwork: Alma Swan
Text: Stevan Harnad