The demand from institutions to avoid each having to build their own
data repository (etc) was very clear when DCC surveyed institutions
earlier in the year. DCC colleagues are involved in developing the
national Research Data Registry and Discovery Service (RDRDS) though
that is about metadata of course, not about managing storage or the
policies relating to specific datasets.
We do however want to offer more relevant case studies and support in
this area, and to better understand the potential for efficiencies. I'm
interested in what the uses cases are for scaling data repository
services beyond the institution. By that I mean any services higher
level than the bit-level preservation service already provided through
the Janet framework agreement with Arkivum.
My assumption is that an institution should want to own and control its
data assets (sharing that responsibility with its data producers) but
place them where they will get the best 'return' in terms of providing
the enhanced description etc need to promote discovery and reuse. Some
of the high-level repository services needed to do that might be better
managed at the specialist sub-domain level; as Sebastian I think
suggests below, a long-tail of specialist niche repositories could offer
potential efficiencies to institutions for some functions, while others
such as persistent identification make more sense at institutional and
national level.
There are different economies of scale that could work if you break down
a 'data repository' into different services. Some might be more
economical at research group, institution, regional, national or broader
levels. What examples of repository functions that are above file
storage and 'bit-level' preservation, and might be provided by a
national or regional service are people looking for?
Angus
--
Dr Angus Whyte
Senior Institutional Support Officer
Digital Curation Centre
University of Edinburgh
On 15/10/2014 22:36, Sebastian Rahtz wrote:
> On 15 Oct 2014, at 22:28, John Milner<[log in to unmask]> wrote:
>
>> The big sciences and UKDA are excellent, but it's not practical to do this for every discipline group. Therefore we thought that taking lessons learned there and elsewhere into institutions who can then cope with the long tail of research groups is a better strategy.
> the two problems with this approach are that:
>
> a) the institutional repository has to adopt a one-size-fits-all approach and therefore doesn’t offer the specialised (e.g.) APIs and visualisations that the specialist repo might provide
>
> b) dividing up the data by institution doesn’t actually fit the needs of the researcher. most academics seem to be barely aware of what their
> institution is, but identify with a subject grouping instead.
>
> The answer, plainly, is that subject repositories become aggregating caches and not archival data holders; which is fine,
> _if_ the institutions also accept some of the responsibilities for hosting and maintaining subject repositories.
> --
> Sebastian Rahtz
> Director (Research) of Academic IT
> University of Oxford IT Services
> 13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431
>
> Não sou nada.
> Nunca serei nada.
> Não posso querer ser nada.
> À parte isso, tenho em mim todos os sonhos do mundo.
>
--
The University of Edinburgh is a charitable body, registered in
Scotland, with registration number SC005336.
|