Same as it ever was? Significant Properties and the preservation of meaning over time
Stephen Grace and Gareth Knight
Centre for e-Research
2
Why Significant Properties?
“The fundamental challenge of digital preservation is to preserve the accessibility and authenticity of digital objects over time and domains, and across changing technical environments”
Wilson, 2008
InSPECT Significant Properties Report
3
http://www.flickr.com/photos/marcn/2175740608/Attribution 2.0 Generic
4
Change we can believe in
“We want to be able to guarantee that for a given object the reformatted version is equivalent to the original version with regards to some specific set of object characteristics”
Clifford Lynch, DLib 1999
If we change something in order to keep it safe, how do we know we can trust the results? And how do we reassure others?
5
InSPECT definition of SPs
The characteristics of digital objects that must be preserved over time in order to ensure the continued accessibility, usability, and meaning of the objects
Wilson, 2008
InSPECT Significant Properties Report
6
Authenticity, integrity, viability
• Authenticity – is this what it purports to be?
• Integrity – is this complete and “unaltered”?
• Viability – is this suitable for its audience (the Designated Community in OAIS terms)?
7
What is significant about the digital object?• How do you distinguish between essential, useful and superfluous?
• Impractical to present a single, definitive interpretation of significance
• Many stakeholders may be associated with an object
• Stakeholders vary and change over time• Stakeholders have different needs and knowledge
8
InSPECT approach to determining significance
• Formal methodology required to guide process of identifying, analysing and recording elements of the Information Object that are essential/beneficial to maintain over time
• Assessment framework should be rational, consistent in its application, while offering sufficient flexibility for widespread applicability
• Previous work performed in field, such as Rothenberg & Bikson’s Needs Analysis, InterPARES1 use of Diplomatics and PLANETS Utility Analysis methodologies
9
Significance is human
• Need to adopt a relativistic approach to determine aspects that are essential/beneficial based upon an interpretation of acceptable loss
• InSPECT builds upon two philosophical approaches:
• Teleology: study of design and purpose of object – why was it created?
• Epistemology: Understand meaning and process by which knowledge is acquired
• In combination, these encourage evaluation of context of creation and information needed to communicate intrinsic knowledge to a new audience (designated community)
10
SP assessment framework in InSPECT• Builds on Gero’s Function-Behaviour-Structure framework• Three categories:
• Function: The design intention or purpose that is performed
• Behaviour: The epistemological outcome derived from the function & structure obtained by the stakeholder
• Structure: The structural elements of the Object that enables stakeholder to perform a behaviour
• Behaviour is result of Function and Structure interaction
11
Assessment framework stages
• Object analysis• Identify functions, behaviours to be achieved
and properties needed for their performance
• Stakeholder analysis• Analyse functions a particular user group wish
to perform
• Reformulation• Perform a revised set of functions or different
behaviours
12
Analyse structure
Identify purpose of technical properties
Determine expected behaviours
Classify behaviours into functions
Associate properties with each
function
Review & finalise
Select object type for analysis
1. Object Analysis
13
Select object type(s) for analysis
Determine actual behaviours
Classify behaviours into set of functions
Object type analysis
Stakeholder requirements analysis
Assign acceptablevalue boundaries
Review & finalise
Finalise & record
Identify stakeholder
Cross-match functions
2. Stakeholder Analysis
14
3. ReformulationRedevelop object to perform a revised set of functions
or enable different behaviours (e.g. view, use)
Object type properties as req. by functions
Cross-match
Functions required by stakeholder
ReformulateInformation Object for use by stakeholder A
Information Object for use by stakeholder B
15
Reformulation in practice
http://www.flickr.com/photos/huggerindustries/3885401876/
Attribution-Non-Commercial-Share Alike 2.0 Generic
http://www.flickr.com/photos/mvjantzen/4113615243/ Attribution-Non-Commercial 2.0 Generic
16
dismantle
Object Analysis
Re-formulation
Stakeholder 1 Analysis
Archiving
Re-formulation
Stakeholder 2 Analysis
Re-formulation
Stakeholder 1 Re-Analysis
Re-formulation
Stakeholder 3 Analysis
Construction
Time
dismantle
Analysis & reformulation over time
17
SPs in digital preservation
• Document technical properties
• Describe intellectual entities
• Determine preservation priorities
• Measure the success of transformations• Choices, outcomes, relation to original
18
SPs in repository workflows
•Availability and adequacy of characterisation tools• (Partly) manual activity for the foreseeable future
•Encoding for machine processing• Requires a metadata schema e.g. extensions to PREMIS
•Possibility of standard ‘profiles’ over time• Sharing results and best practice
•Always relate to institutional mission• Whose needs are being served?
19
Findings of InSPECT
•Appraisal process required to identify aspects of digital object that are essential
•Analysis of functional requirements is a pragmatic method for determining acceptable loss
•Significance is fluid – variable and subject to change
•Methodology provides a vocabulary and framework for understanding design process
20
In conclusion
Significant Properties can act as a bridge
across time to ensure the persistence of
what is important in digital objects through
any required transformations
Grace, Knight and Montague 2009
InSPECT Final Report
21
Acknowledgments
InSPECT was funded by JISC and was a
collaboration between the Centre for
e-Research at King’s College London and
The National Archives
22
Contact us
Stephen Grace
Centre for e-Research
King’s College London
020 7848 1972
Gareth Knight
Centre for e-Research
King’s College London
020 7848 1979
www.significantproperties.org.uk
www.kcl.ac.uk/iss/cerch