JISC

Panel Session

The panel session featured speakers from the previous sessions in the strand and was chaired by Richard Ovenden. He opened the session by summarising the broad themes which emerged during the conference:

• Organisational issues – value, justification, identifying priorities
• Technical issues: a) Service model, collabortaive/consolidated, b) Structure of preservation activity
• Professional issues – identifying and assessing risk; integrity/trust.

Richard then asked panellists to reflect on the conference so far and comment on two opinions extracted from the keynote presentations:

1.      Andy Powell – preservation doesn't matter

2.      Professor Drummond Bone – preservation is heavy burden of responsibility for IRs

Andrew kicked off the response from the panel, saying that we need to convey the value of digital assets with repositories, especially to senior management. Third party services are a pragmatic solution and an identification of repositories' limitation. Further discussion on risk and issues of resources/skills would be useful. Perla added that repositories can learn from archival practices. Value is a familiar concept as we assess value in our daily life. Some method in financial management and daily life could be useful to repositories.

Richard asked Gareth and Steve how they see their technical developments relating to the philosophical consideration of value. Gareth replied that once the value of the data is recognised, the next step is to preserve the data. This is a great responsibility. Sherpa DP and Preserv have developed different infrastructures and ways to start taking on that responsibility. Steve added that there is a need to link preservation services and repositories: services cannot be created in a vacuum and must respond to the needs of repositories. The two extracts from the keynotes demonstrate awareness of preservation as an issue but they appear to want someone else to address that problem for them. This requires input from stakeholders to help shape and provide the kind of services that they want.

Susan added that different repositories will require and use services in different ways. It's important to communicate your requirements so that service providers can understand and meet them.  However, the lack of long term funding long term sustainable funding for preservation remains a problem.

The discussion then focused on the custody of digital resources. The library community is still struggling to come to terms with the fact that they do not have physical ownership of resources anymore: you lease content instead of owning it. Therefore, a distributed model for digital preservation such as the LOCKSS model seems to make sense. In keeping with the notion of a distributed model, James Currall found preservation services very useful.

The chair asked the panellists whether JISC is paying enough attention to digital preservation and whether preservation was prominent enough in the repositories roadmap, inviting the audience to also join the discussion.

Matthew Woollard from the UK Data Archive (UKDA) commented that the assessment of value is made more problematic by the sheer quantity of digital materials, which is growing incrementally. We do seem to pay enough attention to what happens pre-ingest. On what basis should a pre-ingest selection decision be made?  There is often conflicts between stakeholders about what is important and he asked the panel to discuss the importance of the selection process.

Steve Hitchcock responded that this came back to repository policy – we have to put general repository policy in front of preservation policy. There should be a selection policy on what an institution wants to include. It seems that at this conference, people have different ideas about what a repository is for. It is therefore difficult to make a generic statement on selection. Furthermore, if we take the decision to allow something in, should we therefore be committed to preserving it?

Susan Thomas noted that in our anxiety to populate repositories we may have overlooked the issue of selection, particularly in terms of value assessment. Richard Ovenden responded that once content gets into a repository, its value does not appear to be re-assessed. Retention schedules are therefore very important to deal with this problem, especially in terms of the digital deluge.

Andrew McHugh commented that an important element of the toolkit was to define what repositories are for, particularly with regards to their mandate. James Currall added that subject-based repositories seem to have clearer drivers than institutional repositories, which are commonly regarded as 'buckets' into which digital stuff can be thrown. Perla asked whether re-use is an appropriate measure of value, to which Matthew Woollard responded that re-use should be considered within the wider context of the user community and specific stakeholder groups.

Richard concluded the session by recommending JISC sponsor more work in the area of selection. In addition, we need to consider how to present – or 'sell' - preservation in such a way that it reaches the wider community.