Keeping data safe in an educational repository: cost-benefits analysis

KeepIt course module 2, Southampton, 5 February 2010
Tools this module: KRDS, LIFE
Tags Find out more about: this module KeepIt course 2, the full KeepIt course
Presentation referred to in this blog entry Costs, Policy, and Benefits in Long-term Digital Preservation (Slideshare)
Presentations and tutorial exercises course 2 (source files)

18 stalwart repository friends gathered in the School of Electronics & Computer Science, University of Southampton. Most people who came had attended Module 1 a couple of weeks before.

Keeping Research Data Safe (KRDS): an activity model

Module 2, entitled “Institutional and lifecycle preservation costs”, was launched by Neil Beagrie, on KRDS2 – Keeping Research Data Safe – an activity model for identifying the benefits of a repository and the corresponding allocation of costs to those benefits.

We followed the same pattern set out in Module 1 of the course – our guest gave a presentation which set out in a very clear, structured and helpful way, the rationale for the tool. We, as training course pilot participants, were interested in exploring the applicability of the tool and the benefits of following the processes set out by KRDS. Subsequently, we undertook two practical exercises which enabled us to focus on specific aspects of the tool and to understand in more detail the processes involved.

Since EdShare is an educational repository, we were interested in understanding what the likely applicability of the KRDS processes would be to the domain of learning and teaching? As a matter of fact, I agree with Neil Beagrie’s reply to this same question: not everything would be applicable, but there may be some relevant aspects for the domain.

In the practical exercise: in four separate groups, we were asked to categorise, according to the KRDS2 Benefits Taxonomy, which benefits could be costed. We were then tasked with identifying three of these benefits and identifying what information would be required in order to assign specific costs to them. Time was allocated for each of the four groups to report back to the whole, a summary of their discussion and any specific questions which were raised.

The KRDS 2 Taxonomy, structured costs in this way:

• Dimension 1. Benefits – direct and indirect
• Dimension 2. Time – near term, longer term
• Dimension 3. Private/public – benefits

Early in the group activity, one of our members identified the complementary relationship between benefits identified within each of the dimensions. So, within Dimension 1, for each of the Direct Benefits identified, it appeared that there was a corresponding Indirect Benefit. To some extent this could create some redundancy within the task we were set, or otherwise this could be treated as a verification step within the process to provide additional support for the work undertaken.

For the specific categories identified within the Taxonomy, some of us struggled to understand what exactly was intended in their use – so, “Skills Base” was intended to refer to the specific skills of the researcher(s) within any Project or linked to the production of a Data Set. We, on the other hand, felt that it could refer to the skills of those involved in the repository or preservation activity, or indeed skills of a wide range of stakeholders identified within the research process at many levels. From this discussion, we suggested that clearer articulation, preferably with specific examples listed would be helpful throughout the Taxonomy.

Our practical session provided a series of feedback comments to Neil and his team, which were broadly positive and enthusiastic about the possibility of assigning complex costing models to summarised benefits. In addition, however, issues were raised for cases in which confidentiality was required/mandated by the funder for instance, where national security or commercial sensitivity were involved. There were many examples of specific benefits offered by preservation and repository work where economic costs could not necessarily be assigned, but where we would wish to indicate reputational, peer recognition and general community benefits for work undertaken.

Applying KRDS to EdShare

In the context of the educational interests of EdShare, within the Taxonomy, it would be appropriate to edit references to “research” (although maybe not always – there could be specific pedagogical research work linked to the development of educational resource sharing, especially as it develops over several years).

The Taxonomy might look more like this.

We will need to work through this in more detail in the EdShare case and develop better understandings for these costing activities. As educational repositories develop nationally, within institutions as well as across subject disciplines, we will have more opportunities to explore what form the appropriate Taxonomy takes on. There may be different aspects which will need to be addressed dependent on whether resources are shared openly or restricted to specific communities.

In all, this session on the KRDS was both interesting and stimulating. There are undoubted benefits in this approach for better understanding of the opportunities and benefits provided by all the repository and preservation activities our community is involved in.