Theme b: Research data – Whose problem is it?

The notes from the event are available. This is an introduction to the notes, including a participant list, and this is the notes themselves.

Sessions in this theme are being co-ordinated by Neil Jacobs.

This page summarises some relevant JISC and related work on ‘data’:

You can use your comments on these pages to raise questions for Neil or the presenters or to raise issues you wish to discuss with your colleagues in the JISC community. For those of you attending these sessions there will also be materials that you will need to download in advance. For these sessions to work effectively, we invite you to share some information and opinions before the meeting. The details of this are given on the pages relating to each session:

Session 1 – Legal and policy issues [Live Blog Post]
Session 2 – Capacity and skills issues [Live Blog Post]
Session 3 – Technical and infrastructure issues [Live Blog Post]

Please be sure to look at these and share the information and views asked for there, at the latest by Thursday 10th July. Many thanks, I look forward to seeing you in Keele.


4 thoughts on “Theme b: Research data – Whose problem is it?

  1. Owen Stephens

    Just a couple of brief points to get us going:

    It is really important to differentiate between ‘Big Research’ and (with acknowledgements to Jim Downing) ‘Long tail research’. This differentiation exists in Science – not so sure about other areas? Anyway, basic idea is that ‘big research’ produces large datasets, but it is pretty economic to capture these as they come out of a small number of projects.

    Long tail research is much more expensive as this is lots of small research projects producing small(ish) amounts of data. The costs of capturing the data are high as each project has different requirements etc.

    I’m concerned at the moment that the costs of capturing the outputs from long tail research are very high, and it isn’t easy to see how we manage this economically. At a recent talk Peter Murray Rust said if you plan capture into your research it is cheap, if you try to add capture on, it is extremely expensive – I’m sure he is right, but we need to effect a cultural change here. However, alongside this we also need tools that help with the capture I think – at the moment the path of least resistance is not to capture the data in a structured/standard way.

  2. Joy Davidson

    Hi Owen,

    I agree that targeting those undertaking ‘long tail’ research is a much bigger challenge at the moment. My main concern is that we are not currently able to provide these smaller research projects with adequate incentives to convince them to invest a portion of their already limited staff time and project resources to data curation activities. Without much clearer incentives, it will remain difficult to get this group on board.

  3. Sarah Jones

    I wonder about this too Joy. From our experiences on the Data Audit Framework project there’s often very little money available for curation and it seems quite a low priority for organisations. In addition, the approaches to creating and managing data we’ve come across are very ad hoc. Where there is some consistency it’s generally limited to individual researchers and their own practices rather than stemming from a departmental / institutional data policy, so it’s going to be a challenge to manage the data being created. The data creators I’ve spoken with have tended to see their role as very distinct from that of curators too.

  4. Mark Hedges

    Hi Owen,

    This is true in other areas as well (perhaps even more so). Many of the research outputs in the humanities are on the small side, and looking at the access statistics for the former AHDS I note that, while some collections were pretty popular, a large proportion of them were not being accessed at all. This isn’t a reflection of their quality, nor of the importance of preserving them. It’s just that they are of very specialised utility and may be of use many years after they were created. I have myself used articles that were > 100 years old (recording inscriptions that have since disappeared); similar considerations will apply to digital material. I was recently talking with some classicists, who mentioned a multitude of small databases containing information that is in many cases irreplaceable (this was outside the UK so didn’t fall under the AHDS remit).

Comments are closed.