eCrystals – Repository preservation objectives

A little bit about eCrystals:
eCrystals Southampton (http://ecrystals.chem.soton.ac.uk/) is a data repository based on the EPrints platform, but heavily reconfigured to manage data files from chemical crystallography structure determination experiments. The repository evolved out of 2 rounds of JISC funding of the eBank-UK project (http://www.ukoln.ac.uk/projects/ebank-uk/). The repository has a schema that represents the data files generated during an experiment (a crude representation of the workflow) so that a record can be managed and presented in an understandable fashion. The next two phases of eCrystals evolution were concerned with developing a ‘Federation’ of such repositories and analysing the requirements, problems & pitfalls of a distributed network of data repositories (http://wiki.ecrystals.chem.soton.ac.uk). One workpackage of this project was concerned with preservation issues and looked into preservation planning, metadata and representation information concerning the network and individual repositories (WP 4).

A number of matters arose from this project and in general they were not technical, but mostly stemmed from the fact that these repositories are not based on the Institutional model, but are operated, maintained, administered and populated by individuals or research groups…often a single person takes on all these roles, in addition to their ‘day job’ of doing the scientific research! It is our belief that the future repository landscape will be very heterogeneous indeed and there will be a large number of archives & resources that do not operate under the Institutional model – the preservation requirements of these will be very different from those currently recognised by the community.

eCrystals preservation objectives therefore generally go beyond the technical and are concerned with understanding the issues around every-day researchers performing preservation activities:

Objective 1

to explore the preservation training and actions required for small groups and non-archivists. If all crystallographically active institutions (virtually every Chemistry Department has a crystallography centre) signed up to the concept of eCrystals, 95% of the repositories would be user-administered within research groups run by 1,2 or 3 people. These people are researchers trained in their art and would have very little concept of any preservation issues whatsoever.

Objective 2

to investigate how performing preservation actions can be made easy! Learning the minimum requirements for the maximum return (the 80% rule). What can be automated and what technologies can be implemented, both unseen by the repository software and as ancillary tools.

Objective 3

to develop an exemplar non-onerous preservation regime for the researcher administering a repository. Following on from objectives 1 & 2, a schedule of preservation actions can be derived and an objective would be to understand how this can be ‘coded’ into the repository software (eg an EPrints plug-in) so that the repository can automatically perform these actions or alert the researcher at the appropriate time as to which actions need executing (eg accession, appraisal, deletion, etc).

Objective 4

to develop example costings for researchers and administrators. Exemplar costs on a crystal structure record level would enable individuals to quickly and easily include the appropriate amount for preservation and administration of a repository to be included in FEC grant applications or departmental / research group levels.