ADASS XVII: Trusted Data Repositories (D. Giaretta)

David Giaretta is the associate director of the UK Digital Curation Center

Need for trustaable repoitories.
  • A task force on archiving of digital information declared (1996)
It's relatively easy to do, given a lot of funds and time...
It's relatively easy to test reliability claims as long as repositories are long-lived

Key OAIS Concepts:
  • "This data is being preserved": how to we check this claim?
  • Carve them in stone, but we need more that this, we need understandability....
  • We need to define a concept of a "designated community"
  • Metadata is not a blanket term
Long term preservation of information
  • original bits?
  • look and feel?
  • behavior?
  • performance?
  • explicit/implicit/tacit?
  • "Ensure the information to be preserved is Independently understandable to (and usable by) the Designated Community
Representation Information (is this a proxy for metadata?)
Need to take into account the change in environment (software and hardware)

Idea: do we need a sort of "snapshot" for astronomical data archives, over time?

So far the community has produced TRAC (missed acronym...)
  1. Governance
  2. Organization structure
  3. Procedural accountabiluty
  4. ...
  5. ...
TRAC is linked to ISO 17799/27001 for security and ISO 9000
TRAC has started the process of becoming an ISO standard via CCSDS - same route as OAIS
Do we need another ISO standard? 27001 cannot be modified satisfactorily.
The market has some role to play into this... for example IBM is rather interested in this. In particular, transparency, the existence of trustable repository and the cost of an archive.

RAC: http://wiki.digitalrepositoryauditablecertification.org
TRAC: http://www.crl.edu/PDF/trac.pdf
Digital Curation Center: http://www.cdd.ac.uk
CASPAR project: http://www.casparpreserves.eu

Q. Is this process not really applicable (cost?)
A. This standard need to be procuded and the market will have to bear its cost