Institutional vs. Central Repositories: 1 (of 2)

Chris Armbruster wrote (in the American Scientist Open Access Forum):

CA: "Much hope and a lot of money has been invested in institutional repositories - but, for example, in the UK the significant mandates are now research funder mandates and all the life science RCUKs have joined UK PMC. It would thus seem important and urgent that IRs reconsider their strategy and take a closer look at the idea of being a research repository or joining forces for building a national (or regional) system."

Armbruster, C and Romary, L (2009) Comparing Repository Types: Challenges and Barriers for Subject-Based Repositories, Research Repositories, National Repository Systems and Institutional Repositories in Serving Scholarly Communication

(1) It is not at all clear that the "significant mandates" are the funder mandates, especially in view of the past year's burst in institutional mandates (UCL, Harvard, MIT, Stanford...). See Alma Swan's latest graph of mandate growth in ROARMAP:

---

Click here for Latest Growth Graph Update

---

(2) The ones who need to reconsider their strategy are the (few) research funders who have needlessly and counterproductively stipulated that locus of deposit should be central rather than institutional.

(3) Institutions are the universal providers of all research output -- funded and unfunded, across all subjects, all institutions, and all nations.

(4) Institutions have a vested interest in hosting, monitoring, showcasing and archiving their own research output.

(5) OAI-compliant Repositories are all interoperable.

(6) Either funders or institutions can in principle stipulate any locus of deposit for a mandate, either institutional or central.

(7) But mandates are still growing too slowly, and one big reason is that no one wants to do -- or mandate -- multiple deposit.

(8) There are potentially multiple, diverse and divergent central loci for any piece of research output: subject collections, national collections, funder collections, multidisciplinary collections, etc.

(9) The metadata and/or full-text deposits of any OAI-compliant repository can be harvested, exported or imported to any OAI-compliant repository.

(10) The natural, economical, rational and systematic solution (one-to-many, unitary-local -- multiple-distal) is for all researchers to deposit locally, in their own institional repository -- and for distal central collections to harvest, import or export -- not the reverse (many-to-one, distal to local, willy-nilly, with institutions having to back-harvest their very own output from here, there and everywhere!), or both, or neither.

(11) The only thing that stands in the way of that optimal solution -- whereby institutional and funder mandates can collaborate, converge, and mutually reinforce one another instead of diverging and competing -- is the arbitrary and ill-thought-through requirement by some funders (but by no means all) to deposit centrally instead of institutionally.

(12) This obstacle is neither a functional one (it has nothing whatsoever to do with the relative functionality of institutional and central repositories -- they are interoperable and equipotent in every respect) nor a "cultural" one (since self-archiving culture is still very new and all too rare): the problem is simply the needless adoption of arbitrary and ill-thought-out locus-of-deposit requirements by some of the initial funders.

(13) The solution is to fix the funder locus-of-deposit specs, not to switch to central locus of deposit.

(14) Prediction: The notion of a "central repository" -- new as it is -- is already obsolescent: Is Google a "central repository" or merely a harvester of local content?

Stevan Harnad
American Scientist Open Access Forum