IR Workshop and DSpace User Group Meeting 2006: Bergen, Norway
The University of Bergen hosted a lively and entertaining conference for DSpace users from 19th-21st April (though I only narrowly avoided registering for a Norwegian Fishermen’s conference being held in the same venue). [1]
Many speakers laid particular emphasis on the theme of Open Access to academic research. Professor Nilsen of Bergen University set the idea of open access electronic repositories in a Scandinavian context: Norwegians are proud of their freedoms – access to sea, countryside, longboats, etc. – though the threat of commercialisation is ever-present. Academic research may be seen in the same light: public-funded research is undertaken for the public good and needs protection. The Berlin Declaration and the Messina Declaration on open access to knowledge in the sciences and humanities were frequently cited. [2]
A need to counter conservatism in the research community was also recognised. Attitudes to publishing, licensing, self-archiving need updating to take account of the electronic information revolution; a self-archiving culture needs to be fostered, even obligated, among researchers. Universities exist to enable high quality research, and to ensure it is read and used for the public good; publication in high quality journals is also important, but for research impact, not royalties.
Institutional Repositories (IRs) are therefore key to achieving better academic impact, as well as strengthening competiveness between and among research communities. To small, scattered populations, such as in Norway, the advantages of distributed, electronic repositories are particularly compelling. “Knowledge on shelves” is inaccessible, journals costly. As the knowledge management hub of research universities, libraries should have overall responsibility for ensuring excellence, availability, academic impact, academic relevance, and to do this by promoting and managing electronic IRs for the research departments.
Theo Andrew, University of Edinburgh Repository Manager, described the need to clearly define workflows and services required when establishing an e-Repository; the University library fills the role of an aggregator/publisher. Discrete collections can be merged, or indexes aggregated for electronic searches (see, for example, the White Rose Eprints consortium: http://eprints.whiterose.ac.uk/). Robust solutions to issues around access and authentication for individuals and institutions (such as multiple passwords and logins) have also gained ground (e.g. Shibboleth).
Jill Walker, a researcher, specialising in post-post-modern hypertextual studies, seemed a less than typical representative of the research community, but made many valid points about the importance of usability in encouraging self-archiving Institutional Repositories. In particular, the language and procedures relating to licensing - “all that legal stuff” - need to be made as simple as possible. Complications relating to publishing and self-archiving need to be eliminated - if possible, by enabling and encouraging researchers to assert their rights to self-archive materials, when submitting to publishers. [3]
Nevertheless, researchers should be enthusiastic about seeing their work available through IRs such as Bergen’s BORA (http://bora.uib.no), and further aggregated in national and international level services (Oslo’s FRIDA (http://wo.uio.no/as/WebObjects/frida.woa); OAIster (http://oaister.umdl.umich.edu)). Federico Meschini from CASPUR in Italy considered Google Scholar (http://scholar.google.com/) a particularly useful “carrot” to encourage researchers to deposit; other institutions adopted the “stick” of mandatory deposit in the Institutional Repository.
Flickr (http://www.flickr.com/), Del.icio.us (http://del.icio.us/), and CiteULike (http://www.citeulike.org/) were cited as examples of online systems that have not only gained widespread use through simple but effective UI design, but also spread a culture of personalised metadata management through tagging. When combined with traditional approaches to classification and taxonomies, this “folksonomy” is a powerful tool for arrangement and discovery, as well as underpinning online communities. Its growing acceptance is another useful tool in helping researchers “discover” Institutional Repositories. [4]
Many repositories nervertheless had to deal with confidentiality issues, not always commercially-driven. Several speakers had had to deal with the conflicts between archiving and access. Jim Downing from Cambridge described a system that included an “escrow” repository - a second instance of DSpace without GUI and public access, where private work could be safely deposited, but automatically migrated to a public repository once embargoes were resolved.
Many implementations dealt with the need to achieve early acceptance and credibility through populating the repository with “e-people”, using staff-lists or by integrating with administrative or authentication systems (e.g. SAP, LDAP), and by pre-loading materials, be they full articles or bibliographic records from other online sources. In cases of uncertain attribution or incomplete metadata, DSpace’s workflow system makes it possible to leave to individual researchers the task of claiming, amending, or rejecting objects that the preload process has assigned to them.
Ease of export of objects and metadata was also essential and widely exploited. Researchers and institutions can benefit from auto-generated bibliographies, both through pre-print templates, and through interfaces to bibliographic software such as RefMan, EndNote, ProCite. The University of Leuven has, further, developed reports based on access log analysis, enabling departments and researchers to see statistics on submissions and downloads. The IRRA project of Southampton and Edinburgh universites (http://irra.eprints.org/) is developing systems and procedures based on electronic IRs, to support submissions to the UK Research Assessment Exercise in 2008.
Interoperability is also key: IRRA is working with both Eprints and Dspace repositories. The use of heterogeneous IR applications among institutions, yet supporting common, open standards and approaches at all application levels, was seen to be a valuable strength in the global academic community, and the best guarantee of ongoing accessibility and future-proof preservation of materials.
Self-hosting DSpace is not a trivial undertaking: it requires not only serious consideration of organisational structure and institutional workflows, but also a input and support from experienced web, database and systems administrators. For institutions without significant IT skills, Marianne Josserand demonstrated Open Repository from BioMed central (http://www.openrepository.com/), which offers customisable installation and hosting of DSpace-compatible repositories.
A number of technical developments in DSpace were also discussed, many implemented with the 1.4 release due in the Summer, or through add-ons. These include RSS news feeds, delegated administrators (i.e. per community) [5], and the Lightweight Network Interfaces for programmatic interaction with repositories [6].
Among the many impressive DSpace implementations we saw were BORA (the site is in English and Norwegian; a simple search for “hypertext” will discover several of Jill Walker’s articles) and Earth Prints from the Istituto Nazionale di Geofisica e Vulcanologia (http://www.earth-prints.org/).
Notes and links
[1] Programme, profiles, papers, etc., are available online at http://dsug2006.uib.no/
[2] Italy has also championed open standards at the highest level: the Consolidation Act on Administrative Records (dpr 445/2000) made the use of XML metadata mandatory in official records management systems.
[3] Sherpa has a useful online resource relating to publishers’ policies on self-archiving (http://www.sherpa.ac.uk/romeo.php?all=yes).
[4] As an example, participants in the workshop are sharing their photos from the event on Flickr (http://www.flickr.com/photos/tags/dspacebergen2006/), and weblinks on Del.icio.us (http://del.icio.us/tag/dspacebergen2006).
[5] This add-on available from Federico at CASPUR.
[6] More info on the DSpace Wiki, or from Will Reilly at MIT.
