I’ve always thought the IRs were temporary. They are a messy beast: part archive for Electronic Theses and Dissertations (ETDs), part publishing platform for grey scholarly material, part journal back-end. We will always need archives, but will Libraries need to put resource into storing duplicate, inferior, copies of scholarly journal articles?
What I’m writing about are the victory conditions for librarians who thought that, like I do, IRs were a tool to challenge the business model of profiteering academic publishers. I don’t have a problem with a commercial service working to provide services for academia, but I do have an issue when the world’s knowledge is being artificially restricted to only those who can pay, in a system of ever increasingly higher prices.
So, when should you pull the plug on IRs?
1) When there are no article database subscriptions anymore. The move to Open Access is gathering pace, and when the end user doesn’t have to pay to read any article, then there is no need to hold a separate copy in the repository.
2) When journals pick out articles already available, edit and curate them. There will always be a use for an editor. Overlay journals that combine already existing OA articles into volumes and issues based on specific themes and disciplines will carry the seal of quality that top journals do now. They may even be the same journals. Peer review will be both crowd sourced, like reddit, and confidential, as the ed board might have a few suggestions to make before they publish your work.
3) When being an active article reviewer is the main way to improve your early academic career. Sorting through the volume of material on big article and scholarly output aggregators like figshare, arXive, PeerJ, PubMed and F1000, where the economies of scale will muscle little IRs out, will be the role of the junior academic and senior student. Evaluating method and conclusions, reanalyzing datasets, and generally having a critical mind will replace mindless regurgitation of accepted truths from high-impact journals.
I’m describing a a new take on academic scholarship – large disciplinary buckets of scholarly outputs that are then picked over for gems. The ability to publish data, and have others analyse it, or reanalyse it. Having professional academic editors reuse outputs and channel them into streams so that busy researchers get the best bits. There is no real place for the IR here. Grey material goes to figshare. The only thing institutions need to do is archive ETDs, and most IR software is far too complicated to manage just to do that.