Yesterday I went over to arctic Lincoln for a SUETr (Start-Up and Enhancement Training) repository related event; which was introduced by Steph Taylor (UKOLN). What follows are my notes on the day.
Julian Beckton talked about embedding a dual purpose repository at Lincoln – hosting research and learning objects within the same environment. Initially it got external funding, but there were issues over getting staff resource. It started as an archive for student projects, but had to deal with some unusual teaching materials (3D) models but difficulties of visual browsing. He demonstrated the MACE Project visual object browser as an ideal best practice example, but one far beyond what they were funded to create. In trying to bring about innovation within the institution and the team considered using the Collis & Moonen 4-Es model (1991) and the Rogers Diffusion of Innovation (1962) approaches. Ease of use was a concept that was highlighted in both these models, and something that Julian suggested was not exemplified by repositories.
He contrasted the discovery of vitamin C and the 200 years for this to be adopted as S.O.P. for Navy, as opposed to rapid up take of new weaponry and ship design. People implement what they understand, and this is true for repositories – most academics don’t really understand what they’re for and why they’re there; with the knock on effect that engagement with them is reduced. Lincoln’s repo automatically transcodes many multimedia formats into flash animations for ease of viewing, but you are still able to download the original format. Noted that many learning objects (e.g. powerpoints of lecture slides) have a short life-span, and aren’t an ideal ingest for a learning object repository. Lincoln has added social bookmarking tools (e.g. RSS, commenting and user-tagging).
Next moves are a full launch, repository advocacy (local champions) and statistics. Comments in discussion that followed about the importance of good data in the repo accessible via common protocols, far more important than building a sophisticated local service layer. This is something I have to agree with, the underlying repository and data must work well and be accessible; after all statistics continue to show that few end-users come to a repository via the home page. Access is at the object level.
Steph Taylor talked briefly about her role at Bath as both a repository project person and author. Then she opened the floor to debate turning to look first at policies and S.O.P, a topic SUETr (and myself) will be returning to at the event at the National Library of Wales next week. An interesting comment from one institution where if a research output isn’t in the repository it’s not counted in research returns. Discussion ranged around mandates, author pays funding streams and sustainability.
Sally Rumsey, from Oxford Uni was next and started off by talking about repository branding and the route of access to the objects in the repository. Oxford University is not just looking at the repo as a silo of output, and is very much considering digital archiving for the long time. Lots of large collections available to them that they can put into their storage. Sally talked about the advantage of making local digital collections available globally, and how this is a very desirable thing to the global scholarly community. Their Fedora based system underpins effectively multiple repositories (e.g. an image collection, special collections) as well as the ORA (Oxford University Research and Archive). Notably as the ORA was there first, they are able to drive the standards for these other repositories.
She looked at ROAR, Intute RS and OAIster as resources for increasing visibility of repository contents; and how they are not well known by the common librarian or academic. Then it was looking at OAI-ORE and how it can take entire digital objects from repositories and reuse rather than just the metadata. Sally suggested that this is something that will be occurring more as the repository field matures. More complex objects are being ingested, and OAI-ORE may help with their curation and sourcing. Google remains the primary route into repositories, and making sure objects are exposed to it is a major route to enhancing user and reuse. UUIDs (Universal Unique Identifiers) are being used as these are very unique and persistent Ids that for the foreseeable future they should identify the object in the repository and no other. Possibility to use these as a Google search to ID the item and any citations to the original object, which will have a knock on effect for bibliometrics, though this is only just beginning to happen as most papers are too new to get many citations. Sally went through the statistics package that Oxford uses (PikiWiki) showing that virtually all discoveries of objects within the repository are via Google. GoogleScholar was noted as being more variable in finding repository items, even Oxford is ignored a lot – this seems to a problem across the sector that GoogleScholar is somewhat biased in search results returned.
Next Sally talked about development, focussing on her experiences with Fedora Commons, nothing that Australia and the US are leading in this area. Oxford are developing a semantic web architecture with Talis. She also highlighted the work of CRIG (Common Repository Interface Group) and their active developments – cutting edge but sometimes scary. Oxford are very involved with data archiving, but still early days and exploring the various issues around it is quite challenging. Oxford are involved in a range of projects such as the PRIUS (Publisher and Institutional Repository Usage Statistics) project – one that will be of considerable interest to any bibliometricians. Other projects mentioned include BID, PRESERV, BRII and DataShare Project. Notable Oxford has one full time dedicated developer working for the repository as well a range of other support staff; and that their involvement in these many projects is only possible due to their number of dedicated repository staff.
Sally noted her involvement in the wider repository scene is one way in which the ORA is made more visible globally, as well as learning what other people are doing. Sally mentioned that like Leicester, conferences at Oxford can have their papers hosted on the repository – the onetime non-Oxford academics are allowed to deposit. Finally she looked at the time commitments for the repository staff, and the need to be choosey about what they commit to be involved in within the wider profession. As a result of its successes Oxford University has been involved in all this global activity as a result of their work, not as a goal for their activities.
After lunch Lucy Keating, Newcastle, how to add value to repositories. Lucy spoke as an enthusiast and not an expert, and made a disclaimer that many of the things she was going to be talking about were not going to be embraced by her institutions; rather they were ideas and inspirations for the whole community. Newcastle’s repository began in 2005, but was more fully developed in 2007 with her appointment with a focus on articles and papers (6,000 items of which 25% are full text). Lucy demonstrated the repository, which displays the number of downloads per item for all to see; as a way of encouraging more access as well as transparency of simple metrics. Lucy noted that she has about 44% response rate to her enquiries and requests to academics. The university is developing an in-house MyImpact research information service (working along the same lines as Symplectic). The repository is going to be fully integrated to the RIS – which should reduce the interaction the academics are required to perform to record their research outputs and archive their papers.
She noted the involvement with the RIS and REF has opened doors that would otherwise be shut. She posed the question – what else can our repositories do beyond OA, preservation and description? She looked through some of the widgets that Les Carr at Southampton has suggested. She touched on mandates, and her personal reluctance to engage with them. They needed more thought before they’re introduced; it’s an all stick and no-carrot route to populating repositories. Other things that could be possible enhancers included enabling interaction and allowing others to form groups and make associations (e.g. tagging and rating) not just formal citations. Displaying content in different ways – visualising content – image wall, previews, tag cloud or broadening the context of the ingest. Finally she talked about ensuring the marketing department can link to original articles in the repository from press releases.
Some comments in discussion that engaging marketing staff on the right level, and on message, can be actually quite difficult. Then some challenging questions about how the repository could generate income or save time for the institution? She advocated the idea that the repository should be integrated into the research workflow, not something that happens at the end of the research process. Will repositories replace, supplement or merge with journals (the concept of overlay journals) is a challenging idea, and one that some repositories and academics are already beginning to exploit. A Group work session looking at case studies followed
Finally Mary Robinson from SHERPA talked about the international dimension of the institutional repository. Mary showed the 1,300 repositories in OpenDOAR, of which about 1,000 are IRs. Mary talked on some similar issues to Sally about making your repository more visible. She noted that there is a need to be proactive in marketing your repositories towards some of the international services, rather than waiting for them to find you. She listed certain guides to ensuring how your repository is visible, drawn from work on the OpenDOAR survey. Mary then moved to talk about the DRIVER project, which worked with European Repositories to provide an infrastructure that other services can plug into to aid repository discovery. DRIVER provides tools for subject communities for academics and tools for bringing together groups of repositories through a single access point (e.g. Spain).
She next turned to the DRIVER Confederation which tries to draw together a global voice for repositories, working with agencies and other stakeholders in the OA world. DRIVER online tool can automatically test the DRIVER guidelines, though currently running on the old version of the validation rules. Mary gave an overview of DRIVER’s activities and resources, including the Mentor service – which is something that I am sure we in Leicester could potentially get involved in.
The day finished with wrap up discussions and final points of interest. Slides for the event are available. Also my twitter feed on the event too: http://search.twitter.com/search?q=%23suetr