KeepIt course 3: significant characteristics

KeepIt course module 3, London, 2 March 2010
Principal topic this module: significant properties
Tools this module: PREMIS, Open Provenance Model (OPM)
Tags Find out more about: this module KeepIt course 3, the full KeepIt course
Presentations and tutorial exercises course 3 (source files)

Significant characteristics are “ the characteristics of digital objects that must be preserved over time in order to ensure the continued accessibility, usability, and meaning of the objects”.

To continue introducing the more technical aspects of preservation workflow in course module 3, we welcome Stephen Grace and Gareth Knight from the Centre for e-Research at King’s College London (KCL), to explore the practical implications and impact of significant characteristics. Alternating presentation duties, Steve and Gareth split their coverage into six self-contained presentations interspersed with practicals and discussion.

[slideshare id=3364267&doc=1sp-introducingsignificantproperties-d2-100308064856-phpapp02]

Presentation 1

One of the difficulties of digital preservation is that although we as users hope to be able see what we want in a digital object, this may not be the same as another user, or the author or creator. In other words, there is scope for interpretation. In addition, machines interpret and present the object depending on the technology framework available at a given time. As we know, that framework can change. So if we allow the possibility of creating a new version of an object – e.g. by format migration, which by definition must change something in the object, however small – to enable presentation in a contemporary technical framework, how do we decide what changes to the object are to be permitted? Such an analysis is the focus of what has become known variously as significant properties or significant characteristics (although Dappert and Farquhar argued for ‘characteristics’ and against interchangeability of these terms).

Perhaps surprisingly, given its wide scope and multi-faceted nature, the topic of significant characteristics has received far less attention than, say, file formats, as the timeline in presentation 1 recognises.

[slideshare id=3364378&doc=2sp-inspectframework-100308070838-phpapp01]

Presentation 2

A recent substantive investigation of significant characteristics was carried out by the InSPECT project, led by KCL. This project provided a framework for determining significance based on Function-Behaviour-Structure (FBS), originally developed to assist engineers and designers to create and redesign artefacts. The FBS approach is used throughout subsequent presentations in this set and in group practicals, and presentation 2 introduces the method and adapts it for data objects:

1. Analyse structure for technical properties
2. identify the purpose of properties and categorise
3. determine expected behaviours
4. and classify into functions

However, the behaviour of an object may vary depending on the user, or ‘stakeholder’ in the analysis presented. Which activities might the user wish to perform on the object? In this presentation we learn how to cross-match object functions with stakeholder functions.

Group exercises

[slideshare id=3364808&doc=3sp-practicalobjectanalysis-100308081920-phpapp02]

Presentation 3

Now we are ready to test our understanding of this framework with our first group exercise, designed to analyse the content of a familiar type of digital object, an email, and consider how it will be used. Presentation 3 invites participants to start (slide 3) with a blank FBS structure (or rather an SBF structure, since we are performing the analysis on this direction), and shows a generic example (slide 4) of what this might look like at the finish. (Note, some of the detail of these slides may be better viewed on the originals – use View on Slideshare button, lower right-hand corner of the slide viewer – or the source MS Powerpoint, see link at the top of this blog entry.) A handout sheet to support the exercise tabulates and defines the elements of an email, also providing a functional description and examples. Users were also given a sheet with a blank SBF framework to complete. Among the source materials you will find a version of the framework filled out with some possible answers.

[slideshare id=3364939&doc=4sp-practicalstakeholderanalysis-1-100308083910-phpapp02]

Presentation 4

The second part of the exercise, presentation 4, links the needs of different stakeholders with the actual behaviours required of an email. The stakeholders considered in this exercise are the creator, a recipient, and a custodian (e.g. an archivist). Groups performing the exercise are asked to identify 2-5 behaviours for each stakeholder. You will notice some parallels with the first exercise. In the final slide participants are asked to review the correlation between object and stakeholder properties.

A participant’s perspective of these exercises and of this course module has been reported.

[slideshare id=3365063&doc=5sps-in-archive-100308085155-phpapp02]

Presentation 5

By this stage we are becoming aware that it is not feasible to manually maintain a large number of digital objects. Presentation 5 introduces a selection of tools available to perform anaysis of digital objects, including DROID for format identification, JHOVE for format validation, and XCL (eXtensible Characterisation Language) to extract and document object properties.

[slideshare id=3365130&doc=6sp-summary-100308085925-phpapp02]

Presentation 6

Concluding, presentation 6 considers the role of significant characteristics in repository workflows.

Next, we consider how to describe these characteristics, and how to record actions taken on the digital object that might alter them.