Institutional repository?

During the project, we have come across the problem time and time again, of trying to define just what it is we are talking about when referring to an institutional repository. It turns out that we’ve failed to be consistent about how we define this, even in the context of our own project. The problem is magnified when asserting our definition against the 20 participating HEI in Scotland, all of whom have their own definition (and understanding) of what an institutional repository actually is.

I have been asked by the project team therefore, to try and come up with a working definition that we can apply to the project.

I’ve been looking at the definitions from a number of sources, and have even asked the question of the wider community on the JISCPM answers site but will start off here by looking at the definition contained on Wikipedia (as good a place as any to start on :-))

It states;

“An Institutional Repository is an online locus for collecting, preserving, and disseminating — in digital form — the intellectual output of an institution , particularly a research institution .

For a university , this would include materials such as research journal articles, before (preprints ) and after (postprints ) undergoing peer review , and digital versions of theses and dissertations , but it might also include other digital assets generated by normal academic life, such as administrative documents, course notes, or learning objects .

The four main objectives for having an institutional repository are:

• to create global visibility for an institution’s scholarly research;
• to collect content in a single location;
• to provide open access to institutional research output
• to store and preserve other institutional digital assets, including unpublished or otherwise easily lost (“grey”) literature (e.g., theses or technical reports).”

And do I agree with this?

In the above definition, the Institutional Repository is defined as the ‘locus’ for outputs. I just don’t believe this to be the case. The Institutional repository should not, as a general rule be treated in the singular.

Its the same problem that I came across in my previous work at the National Library when trying to describe what a ‘Trusted Digital Repository’ was – i.e. that the ‘traditional’ Library is in effect a trusted print repository that had been developed over years and years and comprised many services, facilities and specialisms. Without fail, staff made the assumption that TDR was a thing that you bought or installed, rather than an objective to be met in the future following period of intense technological and cultural internal change.

The repository services that an institution provides bring together all the components that act as enablers for the institutions overall policies of collection, preservation and dissemination. This is the Institutional Repository.  In this repository landscape, the term institutional repository as is currently used  more truly describes a single element of these services, which is the provision of an enabling system that  could more readily be described as an ‘open access repository’ designed and operated on the basis of maximising its acquisition of full-text holdings of ‘formal’ research outputs and making them easily discoverable and openly accessible.

We are also unable to make the assumption that the institutional repository is the store of objects taking the definition above; ‘materials such as research journal articles, before (preprints) and after (post-prints) undergoing peer review , and digital versions of theses and dissertations, but it might also include other digital assets generated by normal academic life, such as administrative documents, course notes, or learning objects’

Many institutions maintain separate systems to manage different object ‘types’ such as post-print materials or e-thesis, but for their own, independently valid reasons, often associated with workflows, or security and accessibility due to copyright/IPR issues. It is equally the case that some institutions choose to combine this material together in a single system, based on their own needs and requirements.

Definitions as used in the ERIS Project

The ERIS Project (Enhancing repository infrastructure in Scotland) has unfortunately generate a mixed definition of repositories.

Take the example of the first of the ERIS project objectives;

‘the project has the core objective of enhancing the level of researchers engagement with repositories, with a view to achieving a more sophisticated understanding of what repository functionality is needed.’

This is a good example of a problem that we have set ourselves.

For this objective, the project has embarked on the assumption that we are referring to the narrower definition of institutional repositories as singular systems designed to maximise the holdings of more ‘formal’ full text research outputs.

Note: what we find from our interaction with users is that one of the primary methods to enhance their levels of interaction will be by ensuring that there is a wide breadth of integrated services encompassing all repository ‘types’, of which the open access, full text repository is just one. The ability to link records and objects together from across this landscape of solutions and services is, and will become increasingly important.

The second of our objectives – ‘to enhance curation and preservation processes within institutions with a view to strengthening the credibility about the longevity of repositories amongst researchers’ takes the much broader definition of repositories as being part of an overall landscape supporting research‘.

This, as it happens, has always been the case on the project, but we have caused some confusion amongst the community by referring to the more assumptive definition used for the first objective. Hence the need to create a preservation policy framework, to allow the core policy to be adapted as required by the user and relative to their circumstances.

Recommendations to the project

So, to conclude. My recommendations to the project are that we shouldn’t use the term institutional repository as far as is possible when referring to the singluar repository set up.  As an ambiguous term,  Institutional Repository should refer to the the ‘repository landscape’ which makes up the research repository services provided by an institution overall, an approach which allows for individual scoping as required.

I will also recommend that for the currently assumed definition of institutional repository, we refer to the ‘open access repository’ designed and operated on the basis of maximising its acquisition of full-text holdings of ‘formal’ research outputs and making them easily discoverable and openly accessible. This will not be a one size fits all definition, but is explicit enough for the project to use when needed.

I will present this argument to the project delivery team at our next meeting (the 9th March) and let the discussion ensue. I’ll expect to update this post over the next couple of days as a result.