(Please send comments to Ross Coleman by 14 June 1996.)
A Draft Statement of Principles for the Preservation of and Long Term Access to Australian Digital Objects has been circulated for comment to assist with the development of strategies to preserve such objects*
This statement, initiated by the National Preservation Office (NPO), was drafted following a workshop on the topic in December 1995, and several subsequent rounds of discussion. CAUL was one of thirty groups of producers, publishers, users, library and archive organisations represented at that workshop. The NPO has called for comments by a deadline of 1 July 1996.
As more information becomes available in digital form, along with a greater use for research and study, questions and issues of long term access become more critical. Digital medium is not an inherently stable form the physical longevity of magnetic tapes and CDROMs is limited or disputed; there is a high level of obsolescence of both hardware and software which is already leaving some digital information inaccessible; there is a multitude of software types and an increasing complexity in language and styles (eg HTML); there are few standards to enable strategies to update or migrate data; and, as a dynamic communication medium, responsibility for archiving and maintenance is far from clear.
Strategies and agreements on preservation will effect the way University Libraries will access and provide such resources to users in the future (be it from inhouse or remote sources), so some involvement in the development of strategies from members of CAUL is both appropriate and important.
The intention of this paper is to highlight some of the issues and questions around these Principles that relate to University Libraries as a means of initiating some discussion to develop strategies or suggestions for input to the NPO.
The full NPO document is available at http://www.nla.gov.au/3/npo/natco/princip.html
The seven Principles in the draft Statement are interrelated and need to be considered in conjunction with each other. However these principles are very generalised and raise as many questions as they do provide a context for strategies. This paper will draw out issues relevant to University Libraries in relation to these Principles.
* Digital objects generically include data stored in digital form and accessed using electronic equipment. Examples are databases, images, sound, video, documents, etc.
1 Problem of definition of "digital object" this is so broad it is difficult to get a handle on, especially as there is a vast and increasing array of such "objects" being created. (Use of the word "object" may also cause some difficulties).
A more fruitful approach to this issue would be to consider the method of access to the "object" something that is either held inhouse (eg multimedia product, CDROM databases), or that will be used remotely over the networks, such as the internet. Inhouse ownership puts the preservation issue to each library to determine whether and how it will maintain the object (this may not be too unlike issues in preserving audiovisual material). However, implicit in reliance on remote access is the assumption that someone else will be maintaining and archiving the information, and without any understandings or agreements this will become problematical for assured long term access.
2 Archiving and access the process of storing or archiving of digital material must be considered in conjunction with provision of continued access to that material. The two are inseparable, archiving also means continued long term access. Strategies for migration or refreshment of data also includes handling metadata, indexing, internal and external linkages etc etc a complex process
3 Australian digital objects the statement is only concerned with Australian product (local mirroring/archiving of significant sites or databases from overseas is a separate issue). Some consideration of what constitutes an "Australian" object may be necessary to clarify what it is understood to be, ie only objects created in Australia, related to Australia, by Australians.
Existing understandings on storing and preservation of Australian print publications are based around the national or state collecting institutions taking those prime responsibilities, while University Libraries take care of their own institutional product. Can these existing understandings be applied to digital objects?
If so, the implications for University Libraries could be significant, particularly if there is an upsurge of institutional digital production/creation, or a move to "claim back" distribution of intellectual product from publishers. This may also relate to broader institutional IT policies as well as any role the Library may accept
4 Costs preservation of digital objects involves new kind of longterm costs rather than capital costs this requires recurrent equipment and maintenance costs for ongoing refreshment or migration of data, or maintenance of older hardware to access obsolete data.
5 The Principles:
Principle 1
"The cooperation of all with a role in the generation, use,
management, distribution and preservation of digital objects is
essential. Continuing access to digital information will be dependent
on the cooperation of information creators, information system
designers, manufacturers, publishers/distributors and information
custodians and providers."
Issues:
The role and responsibilities of University Libraries for preserving
and providing access to the product of their own institution (probably
in conjunction the creators/producers or IT services) and any
unique digital objects held, needs to be addressed. Relevant to
this is the extent to which national and state collecting institutions
define their responsibilities, and assumptions they may have about
University Libraries
The key question here is whether University Libraries accept such a role ?
Principle 2
"Creators of original digital objects hold an initial, and
in many cases a continuing, responsibility for their preservation;
Creators have the power to facilitate or deny the continuing existence
of digital information."
Issues:
Underlying this is a change of relationship between library and
producer/publisher rather than being a storehouse of publication
output, libraries will become dependent on the ability of the
producer/creator to store and maintain long term access. Establishment
of contingencies by collecting institutions to take over these
responsibilities to ensure long term access may be necessary (eg
when creator/producer ceases to exist)
The primary role of creators of digital material is to create, and distribute, that material. Accepting a preservation role would be a major conceptual shift which many would be unable or unwilling to do nor could they be relied on to do so.
The role of University Library when University is creator of original objects.
Principle 3
"The finding, selection, identification/cataloguing and retention
of digital objects is best achieved using the principle of collective
responsibility. A collectively responsible alliance of information
custodians and providers which manages access through distributed
national networks will best serve to preserve and maintain long
term access to significant digital objects."
Issues:
Replication and extension of current cataloguing and processing
obligations or understandings for print material to cover digital
objects
Does this assume that the cooperative imperatives of traditional library relationships (such as resource sharing) remain the same will they ?
Principle 4
"Digital objects should be retained and preserved only for
as long as they are judged to have continuing value and significance;
Resources need to be expended on maintaining access to digital
information only for as long as it is valued."
Issues:
How is "value" defined, and by what criteria is "continuing
value and significance" determined. If creators of objects
have the initial responsibility for their preservation, they will
to a large extent determine what is kept is this adequate for
research and study purposes?
The issue is how to ensure creators understand importance of longterm access, especially when their interests are usually in shortterm sale or distribution. Where is the imperative/rationale to ensure longterm access for research and study when this is not the creators primary concern? What is the economics of longterm preservation for the creators of objects will it be economically viable for commercial producers ?
Selection (of object or version of object) for longterm access will become very pragmatic, and to some extent may become selfselective by dint of creator's interests
Digital material has private and public origins and purposes which will influence the attitude of creators, and the need for preservation
Principle 5
"Rights of creators, owners, providers and users of digital
objects must be balanced and protected; The preservation of significant
digital objects should not be jeopardised by censorship of form
or content. Rights of access and the right to receive benefits
arising from the digital object need to be carefully balanced
and protected."
Issues:
What will be the legal and procedural mechanism for recognition
and "balancing" of these rights, and how will they be
defined
Principle 6
"The adoption of best practices and standards is essential.
Migration and copying strategies and procedures must ensure the
integrity of preserved digital information. Continuing research
is essential."
Issues:
Who has responsibility for development of standards, should they
be international standards, and how will they be enforced
Should these standards and practices be initiated by libraries
Principle 7
"Appropriate Commonwealth and State government regulatory,
legislative and policy regimes are essential. The continuing and
equitable access to digital information is dependent on and must
be supported by appropriate regimes, including legal deposit legislation."
Issues:
Development of concept of legal deposit in a networked environment
and when dealing with a "liberating" medium