Embed Size (px)
Citation preview
Establishing Trust in Data Curation: OAIS and TRAC applied to a Data Staging Repository (DataStaR)
Gail SteinhartCornell University Library
Ann GreenDigital Life Cycle Research & Consulting
Dianne DietrichCornell University Library
IASSIST 2009
Image courtesy of the Cornell Biological Field Station
What exactly is a data staging repository?
DataStaRPermanent Repository(domain,
institutional)
metadata
upload publish
publishcreate
data set institutional)
colleague
upload publish
disseminate
data set
share
user
IASSIST 2009
Where does it fit in the life cycle?
IASSIST 2009
Where does it fit in the life cycle?
IASSIST 2009
But DataStaR isn’t a preservation repository...
“...if repository developers and administrators are guided by a reference model, they are more likely to consider the right issues.”
Allinson 2006: OAIS as a Reference Model for Repositories: An Evaluation
“A repository is Trusted if it can demonstrate its capacity to fulfill its specified functions, and if and if those (...) functions satisfy (...) if and if those (...) functions satisfy (...) minimal criteria which all trusted repositories are assumed to require.”
DigitalPreservationEurope 2008: Repository Planning
IASSIST 2009
Checklist and Guidance
An OAIS view of DataStaR
DataStaRPermanent Repository(domain,
institutional)
metadata
upload publish
publishcreate
data set institutional)
colleague
upload publish
disseminate
data set
share
user
IASSIST 2009
An OAIS view of DataStaR
DataStaR“pre”-SIP
Permanent Repository(domain,
institutional)
metadata
upload publish
publishcreate
data set institutional)
colleague
upload publish
disseminate
data set
share
user
IASSIST 2009
An OAIS view of DataStaR
DataStaR“pre”-SIPAIP
Permanent Repository(domain,
institutional)
metadata
upload publish
publishcreate
data set institutional)
colleague
upload publish
disseminate
data set
share
user
DIP
IASSIST 2009
An OAIS view of DataStaR
DataStaR“pre”-SIPAIP SIP
Permanent Repository(domain,
institutional)
metadata
upload publish
publishcreate
data set institutional)
colleague
upload publish
disseminate
data set
share
user
DIP
IASSIST 2009
OAISAn OAIS view of DataStaR
DataStaR“pre”-SIPAIP SIP
AIP
Permanent Repository(domain,
institutional)
metadata
upload publish
publishcreate
data set institutional)
colleague
upload publish
disseminate
data set
shareDIP
user
DIP
IASSIST 2009
Look at other approaches to implementation
IASSIST 2009
Put TRAC into context
IASSIST 2009
What did DataStaR need...
Three things:Three things:
• Data depositor agreement• Set of repository policieso po o y po• System documentation
IASSIST 2009
Data deposit agreement
IASSIST 2009
Repository policies
IASSIST 2009
System documentation
IASSIST 2009
System documentation
IASSIST 2009
How did we do?
IASSIST 2009
How did we do?
Number and percentage of TRAC criteria addressed by Number and percentage of TRAC criteria addressed by (agreement, policies, system)
TRAC Depositor Repository System doc/ SECTION
pagreement
p ypolicies
y /requirements
A (24 criteria) 6 (25%) 9 (38%) 3 (13%)
B (44 criteria) 4 (9%) 14 (32%) 30 (68%)B (44 criteria) 4 (9%) 14 (32%) 30 (68%)
C (16 criteria) 0 0 6 (38%)
IASSIST 2009
How did we do?
Number and percentage of TRAC criteria addressed by
TRAC Depositor Repository System doc/
Number and percentage of TRAC criteria addressed by (agreement, policies, system)
SECTIONp
agreementp ypolicies
y /requirements
A (24 criteria) 6 (25%) 9 (38%) 3 (13%)
B (44 criteria) 4 (9%) 14 (32%) 30 (68%)B (44 criteria) 4 (9%) 14 (32%) 30 (68%)
C (16 criteria) 0 0 6 (38%)
S ti A it i ( i ti l i f t t ) dd d i l b li• Section A criteria (organizational infrastructure) addressed mainly by policy
IASSIST 2009
How did we do?
Number and percentage of TRAC criteria addressed by
TRAC Depositor Repository System doc/
Number and percentage of TRAC criteria addressed by (agreement, policies, system)
SECTIONp
agreementp ypolicies
y /requirements
A (24 criteria) 6 (25%) 9 (38%) 3 (13%)
B (44 criteria) 4 (9%) 14 (32%) 30 (68%)B (44 criteria) 4 (9%) 14 (32%) 30 (68%)
C (16 criteria) 0 0 6 (38%)
S ti A it i ( i ti l i f t t ) dd d i l b li• Section A criteria (organizational infrastructure) addressed mainly by policy
• Section B and C criteria (digital object management and technologies, technical infrastructure and security addressed mainly (but not exclusively) b t
IASSIST 2009
by system
What didn’t we do?
TRAC SECTION Address at transition to production system
Not relevant to DataStaR
A (24 criteria) 11 (46%) 1 (4%)A (24 criteria) 11 (46%) 1 (4%)
B (44 criteria) 0 8 (18%)
C (16 criteria) 10 (63%) 0
TOTAL (84 criteria) 21 (25%) 9 (11%)
We are making an effort to address 64% of the TRAC criteria, in the pilot phase.
IASSIST 2009
Some observations
• Understanding/interpreting the criteria is a lot of work• Understanding/interpreting the criteria is a lot of work.• The right tools might simplify policy development.• The right software might simplify system specification.• Compiling /presenting evidence: for auditors, or for users?o p g /p g d o aud o , o o u• Picking your partners...
TRAC h l t t ff if l t TRAC has a lot to offer, even if long-term preservation isn’t your focus.
IASSIST 2009
datastar.mannlib.cornell.edu
Thank you. Gail [email protected]
Dianne [email protected]
This material is based upon work supported by the National Science Foundation under Grant
IASSIST 2009
Image courtesy of the Cornell Biological Field Station
No. III‐0712989. Any opinions, findings, and conclusions or recommendations expressed in this material are those of the author(s) and do not necessarily reflect the views of the National Science Foundation.
![A Risk Register for Digital Curation aligned with the OAIS ... · Digital Curation is an ongoing process, alongside with the process of long-term digital preservation. [22] To maintain,](https://img.pdfslide.us/doc/110x75/5f75d736f084d267c21afba7/a-risk-register-for-digital-curation-aligned-with-the-oais-digital-curation.jpg)
