Institutional Repositories vs Subject/Central Repositories

Beth Tillinghast wrote on the DSpace list:

"I have just run into my first case where I am finding our IR in competition with a Subject Repository... I am wondering if others have run into this dilemma and can provide me with many good reasons why submission should take place in an institutional repository rather than a subject repository?"

The dilemma has a simple, optimal and universal solution, with many, many good reasons supporting it:

Direct deposit should be in each researcher's own institution's IR. SRs and CRs can harvest from IRs.

That's what the OAI protocol is for. Institutions are the (distributed) research providers. They are the ones with the direct stake in the record-keeping and showcasing of their own research output, in maximizing its accessibility, visibility, usage and impact, and in assessing and rewarding its research performance. Institutions are also in the position to mandate that their own research output be deposited in their own IR; funder mandates can reinforce that, and can benefit from institutional monitoring and oversight (as long as funders too mandate institutional deposit and central harvesting, rather than direct central deposit).

Distributed, convergent institutional self-archiving makes sound sense and scales systematically to cover all of research output space, whereas divergent self-archiving, willy-nilly in SRs and CRs is arbitrary and simply produces confusion, conflict, frustration and resistance in researchers, if they need to deposit multiply.

(Before you reply to sing the praises of SRs and CRs, recall that their virtues are identical if they are harvested rather than the loci of direct deposit. The overwhelming benefit of IR deposit is that that is the way to ensure that all research output is universally self-archived.)

(And before you reply that seasoned Arxiv depositors will resist institutional deposit, forget about them: they are not the problem. They are self-archiving already, and have been for a decade and a half. Arxiv self-archivers' habits will be integrated with those of the rest of the self-archiving community once self-archiving mandates prevail and institutional self-archiving becomes universal. For now, focus your attention on the 85% of researchers who do not yet self-archive at all, anywhere. They are the problem. And convergent institutional [and funder] self-archiving mandates are the solution.

THE FEEDER AND THE DRIVER: Deposit Institutionally, Harvest Centrally

How To Integrate University and Funder Open Access Mandates

Optimize the NIH Mandate Now: Deposit Institutionally, Harvest Centrally

Swan, A. et al (2005). Developing a model for e-prints and open access journal content for UK higher and further education. Learned Publishing 18 (1). pp. 25-40.

Stevan Harnad
American Scientist Open Access Forum