DEFINING ELECTRONIC

RECORDS, DOCUMENTS
AND DATA
David Roberts

David Roberts is the Manager of the Records Management Office of

New South Wales, one of the operational arms of the Archives
Authority of New South Wales. From 1980 to 1993, he worked in a
variety of positions with Australian Archives and convened the ACA/
ASA Standing Committee on Machine-readable Records from 1986 to
1990. He currently represents the ASA on the Standards Australia
Committee (‘IT/21’) preparing an Australian standard for records
management.

Basic concepts have a profound effect on the approaches taken by

archivists and records managers to their work. They are particularly
important for the Records Management Office because our work
depends on communicating with our clients. We have adopted a concept
ofrecords based on their role as evidence ofbusiness transactions, which
is essential to support accountability in the New South Wales public
sector and which leads to a particular set of approaches to electronic
records management. We distinguish electronic documents from
electronic records by the latter’s transactional origins and evidential
qualities and explore the appropriate use of electronic document
management tools within this conceptual framework as part of broader
electronic records management strategies. We distinguish data
management and data administration from the management of
electronic records and electronic documents and recognise their role in
electronic records management strategies.
 ROBERTS: DEFINING ELECTRONIC RECORDS 15

The purpose of this article is to describe work being done in the

Records Management Office of New South Wales to refine some of the
key concepts involved in the management of electronic records. As one
of the operational arms of the Archives Authority of New South Wales,
the Records Management Office is taking part in the work of the
Authority to develop a range of policy and procedural approaches to
the management of electronic records in the New South Wales public
sector. The focus in this article, however, is on the concepts of records,
documents and data as they are used in the electronic environment.
Inevitably, we will also need to look at these and related concepts in
archives and records work generally.
Why is it necessary to examine these concepts? Surely archivists and
records managers know what they mean when they talk about records,
and everyone knows what documents and data are! However, there is
ample evidence to indicate that there is little common understanding
of these fundamental concepts in the Australian or international
archives and records management communities. In addition to the
obvious need to improve the theoretical basis of the work of our
professions, there are specific reasons arising in the context of
electronic records management for reviewing and refining our
concepts.
Archivists and records managers are increasingly aware of the way in
which the definitions which we use for our basic concepts affect how
we approach our work. For example, if we accept that it is the
evidential quality of records which distinguishes them from other
kinds of recorded information, then the focus on capturing evidence
which is a hallmark of the approach to electronic records management
being explored by David Bearman and his colleagues1 makes a great
deal of sense. If we regard records as merely an information resource,
our approach to electronic records management will take a different
path.
Moreover, records managers involved in electronic records
management are increasingly recognising the need to augment tools
developed for conventional records management with tools for
managing data and electronic documents which have been developed
entirely outside the records management industry. For example,
document management software is growing in popularity as a means
for individuals or organisations to manage their electronic documents:
word processing documents, spreadsheets, databases and so on. If our
concepts of records and documents are indistinguishable, as much of
the traditional records management literature would have us believe,
how do we explain the difference between document management and
records management software or decide how to use these different
kinds of tools in an integrated strategy for the management of
electronic records?
16 ARCHIVES AND MANUSCRIPTS Vol. 22, No. 1

Ensuring that these fundamental concepts are satisfactory and

complementary is particularly important for the Records Management
Office. In our consultancy, training, publications and standards work,
we need to be able to communicate these concepts clearly to people
who, for the most part, are understandably little interested in the
subtleties of archives and records management theory, but who need to
use a wide variety of tools to manage their conventional and electronic
records. If we cannot get our thinking straight on these fundamental
matters, we cannot expect our clients or readers to do so. In the
Records Management Office, our concern in this regard has been
partly driven by our involvement in the development of means of
managing electronic records, in company with many of our colleagues
in the archives and records management communities. It has also been
driven by our own need for a fundamental review of many aspects of
our doctrine, leading to such specific tasks as rewriting the glossaries
which we use in our training and other materials.

Electronic records
For the Records Management Office, our review of what we mean by
records and, consequently, electronic records started with such simple
questions as the then names of our training courses. Was Effective
In formation Management really a course on information management
or on records management? If the latter, why call it the former? The
problem cropped up throughout our publications, training and
promotional materials and in our consultancy work.
The work of the Standards Australia ‘IT/21’ Committee which is
developing an Australian standard on records management has
provided, among many benefits and challenges, an opportunity to
examine what is meant by records management and, of necessity, by
records in the Australian and international archives and records
management communities. A study of definitions of records in the
literature, prepared as part of the Committee’s preliminary work,
showed that there is little agreement about what characterises a record.
The Association of Records Managers and Administrators Inc.
(ARMA International) defines a record as ‘recorded information,
regardless of medium or characteristics, made or received by an
organization that is useful in the operation of the organization’.2 Many
of the American records management texts3 simply equate records and
recorded information or follow the ARMA definition. Other
definitions4 define records in terms of the physical formats in which
they appear, along with the information which they contain.
A further group of definitions identified in the study5 give
prominence to the role of records in providing evidence to document
the transaction of business. This prominence is characteristic of the
concept of the archival document, developed in Australia by Frank
 ROBERTS: DEFINING ELECTRONIC RECORDS 17

Upward and Sue McKemmish6 and of the definition of a record

developed in October 1993 by the Electronic Records Committee of
the International Council on Archives.7 This emphasis provided the
basis for the working definition of records adopted by the
Committee:
For the purposes of the Standard, transactional records are defined
as:
• recorded information
• in any form, including data in computer systems,
• created or received and maintained
• by an organisation or person
• in the transaction of business or the conduct of affairs
• and kept as evidence of such activity.
Records in this definition are described as transactional in reflection of
the Committee’s terms of reference, emphasising that it is records in
this sense which are the subject of the Committee’s work. The debate
about the relationship between archives and records management and
information management has been canvassed in many places
elsewhere.8 The ‘IT/21 ’ Committee has recognised the tension between
the ‘unitary’ and ‘pluralist’ views, to use Sue McKemmish’s
distinction, without attempting the daunting task of resolving it or
imposing a particular position on the issues of the relationship of
records and information.
The evidential quality of records is particularly important for the
Records Management Office. The crucial role of records in supporting
accountability has been capably demonstrated in recent Australian
writings.9 New South Wales has by no means been immune from the
crisis of accountability described in these places. The Independent
Commission Against Corruption’s examination of poor recordkeeping
practices in Corrective Services10 is only one of a series of instances in
this State, albeit the most spectacular, which reveal the nexus between
records and accountability. It is particularly in this context that we
have confirmed that the business of the Records Management Office is
the management of records, rather than of information generally. We
have adopted the ‘IT/21 ’ Committee’s working definition and use it as
a vehicle for placing a strong emphasis on evidence, accountability and
documenting the transaction of business in our consultancy and
training work in the management of conventional and electronic
records.11
Accepting this notion of a record has a significant impact on what we
understand as an electronic record. We can, of course, start with the
observation that an electronic record is a record, however we
understand it, in electronic form. Thus, for the ‘records equals
information’ school, electronic records are ‘records which contain
machine-readable, as opposed to human-readable, information’.12
Similarly, where the transactional and evidential qualities of records
are accepted, electronic records can be described as ‘Recorded
18 ARCHIVES AND MANUSCRIPTS Vol. 22, No. 1

information that is communicated and maintained by means of

electronic equipment in the course of conducting a transaction’.13
Much of the archival practice in the management of electronic
records until the last few years, and of the literature describing this
practice, has been concerned with managing databases as machine-
readable or electronic records. Applying the transaction/evidence test
to many databases indicates that they should not be regarded as
records in this sense, operating as electronic information systems
rather than as electronic recordkeeping systems, to use the neat
distinction developed by Bearman and his colleagues. This can be
expected to pose a dilemma for many archival institutions, which may
accept the logic of the transaction/evidence argument — and indeed
adopt the approaches to electronic records management which flow
from it — but which have custody of, or legislative or jurisdictional
responsibility for, databases and electronic information which do not
function as records.
The problem cannot be avoided but perhaps it can be
accommodated, not by diluting the concept of a record, but by
recognising the informational value of such databases as information
resources, not as records providing evidence of transactions. This
means exploring and adopting the techniques which have been
developed by the ‘data archive’ community, which better suit these
kinds of materials.
In the Records Management Office, the importance of
accountability in our jurisdiction, coupled with the role of records in
providing documentary evidence to support accountability, forces us
to take a strict line on what we mean by records and electronic records.
As noted above, we have adopted the TT/21’ working definition of a
record; we discuss the concepts of records, information, transactions
and evidence, and establish the distinction between records and
information, in our courses; and we are exploring the approaches
developed by Bearman and his colleagues for electronic records
management as providing the potential for meeting important aspects
of the accountability requirements of New South Wales public sector
organisations.

Electronic documents
If we distinguish records from recorded information by their evidential
quality and their role in documenting the transaction of business,
where do documents fit into the picture? Is there any difference
between uses of the term in the paper and electronic environments, or
within and outside the archives and records management
communities?
In the archives and records management literature, we can identify
 ROBERTS: DEFINING ELECTRONIC RECORDS 19

two strands of meaning. Firstly a document is defined as recorded

information, regardless of medium or form.14 In the context of the
discussion above, then, a document may be distinguished from a
record by the latter’s evidential quality in documenting the transaction
of business. This distinction is most clearly drawn in the Keeping
Archives definition of records, where records are recognised as a class
of documents distinguished from other kinds of documents by their
transactional origin and evidential qualities.15 Where the essentially
evidential quality of a record is not accepted, that is, where records are
simply equated with recorded information, the distinction between
records and documents tends to disappear.16
Secondly, a document is defined as a physical record item or unit,
perhaps in terms of a specific physical format.17 This conforms more
closely to the usage of the word in the world at large. It is also used in
some records management software products, usually as an alternative
level of control to the file, while other products use folio for the same
purpose.
In computing, a document is a ‘named, structural unit of text that
can be stored, retrieved, and exchanged among systems and users as a
separate unit’.18
Each of these strands of meaning is evident when we look at
electronic documents in recent records management literature.
Richard Jones has described electronic documents as: ‘analogues of
paper documents. In other words they are multi-media forms of
information composed of text, numbers, images, figures: perhaps
extended to include sound ... Electronic documents may be completely
representable on paper, but this will not always be the case.’19 For the
Australian Government’s Information Exchange Steering Committee
(IESC) an electronic document is:
A collection of electronic data which may be produced by the creation of
original data (typically a text document, small database, spreadsheet,
graphic created within the electronic office environment) or by the
combination of existing data (which may include data extracted from data
files or databases). It should be managed as a unique entity by means of a
standard set of descriptors.20
We can identify several characteristics of electronic documents in
these definitions and their implications. Firstly, an electronic
document is discrete and identifiable from other electronic
documents, including other versions of the document. This means that
it can be managed as a unit. Secondly, it possesses a logical structure of
relationships between the smaller collections of data — paragraphs,
tables, cells, images, headers — which comprise the document. This
distinguishes an electronic document from other kinds of data files.
Thirdly it is increasingly likely to be a multimedia or compound
document, perhaps incorporating sound or moving images. This
20 ARCHIVES AND MANUSCRIPTS Vol. 22, No. 1

means that it is likely to be stored as more than one data file21; it will
require the use of more than one application program to use or present
it, especially in object-oriented programming environments; and it is
increasingly unlikely to be capable of satisfactory representation on
paper. All these characteristics are shared by electronic documents and
electronic records.
Prima facie, the IESC definition conforms to a model which sees
electronic records as a class of electronic documents, distinguished
from other kinds of electronic documents by their transactional origin
and evidential qualities. However, the IESC Guidelines define records
as ‘Papers, correspondence, forms, books, photographs, films, sound
recordings, maps, drawings or other documents, regardless of physical
form or characteristics, either originated or received by an agency’.22
Again we see a definition of a record based on a list of physical formats
(which, unintentionally, one hopes, does not include data storage
media like disks or tapes) and distinguished from documents,
apparently, by being originated or received by an agency. Presumably
document is used here in the physical sense which we noted earlier. Yet
are not documents in this context also either originated or received by
an agency? What difference is there between a document and a record
on this basis?
Moreover, the Guidelines consistently refer to provisions of the
Commonwealth’s Archives Act (1983) as relating to documents, when
the Act itself talks of records, albeit in the sense defined by the Act.23
Electronic records appear nowhere in the Guidelines, despite the
extensive use of the term by the Australian Archives. Clearly, then, the
IESC either does not realise or does not accept the transaction/
evidence distinction between records and documents. That it does not
is not intended as a criticism of the Guidelines. Rather, the point is the
effect of inconsistent terminology here and elsewhere. Much of the
advice in the Guidelines is of great practical value, which we
recommend to our course participants and clients as one of a number
of sources for developing their electronic records management
strategies. However, we need to make it clear that the practices
advocated in the Guidelines are tools for the management of electronic
documents which can be applied to electronic records as part of a
broader strategy. The Guidelines do not, and were not intended to,
provide answers to the problems of keeping electronic records which
will serve as evidence of transactions.
Software tools for the management of electronic documents are an
increasingly important part of the office environment, their popularity
fuelled by the rapid growth of networks and distributed computing,
work organisation methods based on work groups, document image
processing, large data storage devices and a variety of related
developments. As such they form part of the armoury available to
 ROBERTS: DEFINING ELECTRONIC RECORDS 21

archivists and records managers for the management of electronic

records, in addition to their value for the management of an
organisation’s broader electronic information resources.
A growing number of electronic document management products
are available off-the-shelf. How closely do they fit with the conceptual
view of electronic documents developed above? In the world of IBM-
compatible personal computers and networks, a product such as
PageKeeper is distinguished by its strongly textual orientation: the
basis of the program is text-retrieval software, analysing the syntax of
each document to identify its relevance to a search, and supported by
optical character recognition, data compression and network sharing.
However valuable such a product is, it is worthwhile being aware that
the concept of electronic document on which it is based is that of a
discrete text-based document. As we have seen, this is only part of the
story. A product such as Keyfile is closer to an enhanced (computer) file
manager, but with a user interface based on a filing cabinet and folders
and in and out trays. It is designed to manage more broadly defined
electronic documents than the text-oriented products, and supports
the Microsoft WAV format, which enables sound, such as dictated
comments, to be embedded in documents. Similarly PC-DOCS Open
produces a SQL database of information about electronic documents
in any media. It combines retrieval tools for searching the database
with text-retrieval software.24
Despite some limitations in their current scope, all three products
are essentially consistent with the concept of an electronic document
described above. This is fortunate for us, because it makes it easier to
fit them as useful tools into a strategy for electronic records
management founded on a sound conceptual basis and without
expecting users to adopt different vocabularies for different tools.
In the Records Management Office, we now specifically distinguish
electronic documents and electronic records by the transactional
origins and evidential qualities of the latter. We encourage our training
course participants and clients to investigate the appropriate use of
electronic document management tools, whether software tools like
those noted above or practices such as those described in the IESC
Guidelines, as part of a broad electronic records management strategy.

Electronic data
If electronic records are a class of electronic documents possessing
certain characteristics, then we may regard electronic documents as a
class of electronic data files possessing certain characteristics. When
we come to consider the management of data files25 in electronic
information systems, we enter the world of data management and data
administration.
22 ARCHIVES AND MANUSCRIPTS Vol. 22, No. 1

By data management, we mean ‘the function of controlling the

acquisition, analysis, storage, retrieval, and distribution of data’.26
Consequently, it can involve a large range of functions, such as
protecting the physical security of data through adequate backup
procedures and recovery arrangements, protection of confidentiality
and privacy in data, establishing and enforcing users’ responsibility for
data accuracy, reducing data redundancy or duplication, organising
data in rational and consistent ways, and ensuring the retention of data
for the required periods of time. Beyond these bread and butter
functions, data management can involve promoting the use of
common and consistent data across the organisation, establishing a
corporate data architecture and promoting the design of electronic
information systems which are consistent with the data architecture
and strategic IT plans. Anyone who uses a personal computer regularly
performs data management tasks, such as backing up our data files and
rationalising our directory structures, whether we think of it in these
terms or not. In organisations, data management is normally regarded
as a part of the management of information technology resources,
although it is not necessarily recognised organisationally with a
specific position or unit in an IT branch responsible for the function.
By data administration, we mean ‘the corporate service which assists
the provision of information systems by controlling and/or
coordinating the definitions (format and characteristics) and usage of
reliable and relevant data’.27 Thus it is a more specific discipline which
primarily involves the management of metadata, using such tools as
data dictionaries, Information Resource Directory Systems and a
range of types of system documentation.
Data management and data administration are increasingly
important for archivists and records managers for two reasons. Firstly,
developments in the last few years have demonstrated that both
functions have an essential role to play in approaches to the
management of electronic records. Thus data management and data
administration techniques appear among the methods advocated by
David Bearman and his colleagues,28 while the data management
principles identified in Australian Government agencies are harnessed
and enhanced for the purposes of electronic records management by
the Australian Archives.29 If archivists and records managers expect to
exploit data management and data administration techniques and to
forge alliances with IT personnel as part of their electronic records
management strategies, it is essential to understand the nature and
scope of these functions and their relationship with the management of
electronic documents and electronic records.30
Secondly, the growth in end user computing, with the proliferation
of personal computers and small networks, coupled with the reduction
or break-up of many organisations’ IT branches, has meant that there
 ROBERTS: DEFINING ELECTRONIC RECORDS 23

are increasing numbers of computer systems without formal or

practical data management arrangements. In these circumstances,
records managers increasingly find themselves taking on a data
management role by default or perhaps by design, because a work
group or business unit’s electronic information systems contain data
files functioning as records.
Promoting and performing data management at a practical level is
not new to archivists and records managers and useful advice is to be
found in the archives and records management literature.31 However,
we need to recognise that data management and data administration
are disciplines in their own right, part of the larger discipline of
computing. This is important for two reasons. Firstly it leads us to
literature and other sources of advice about their practice which we
will not find if we restrict ourselves to what is available in the archives
and records management communities. Secondly it makes us realise
our limitations in this area: few archivists or records managers can
expect to develop the levels and range of expertise or responsibilities
expected of a professional data manager or administrator, and few
organisations would justify the cost of our developing such expertise.
Thus it is a pity that Saffady’s otherwise valuable practical advice does
not place the data management practices which he advocates into
context, show how the records manager’s role in data management
might relate to those of other people in the organisation with IT
responsibilities, or recognise data management for what it is.
In our training and other work at the Records Management Office,
we have found it useful to make a clear distinction between data
management functions — ensuring that our course participants and
clients understand the full meaning of the term — and the
management of electronic documents and electronic records. This
enables us to discuss options for a range of electronic records
management strategies involving data management and data
administration, ranging from teaching basic practical data
management for individuals and work groups to the tactical use of data
management and data administration as part of the design of
electronic recordkeeping systems.

Conclusion
For the Records Management Office, refining our concepts of
electronic records, documents and data seems a small thing (and
indeed it is only one part of our work to review our doctrine) but we
have found it to be an essential and stimulating exercise. It has helped
us to impose some structure on the confusing world of information; to
focus on what we need to achieve in electronic records management; to
communicate the fundamentals of records management to our clients,
laying a sound foundation for what follows; and even to define our very
24 ARCHIVES AND MANUSCRIPTS Vol. 22, No. 1

business. The issues take us well beyond electronic records

management and pervade more and more of our work.

ENDNOTES
1. Most recently articulated in this journal in ‘Archival Data Management to Achieve
Organisational Accountability for Electronic Records’, Archives and Manuscripts,
vol. 21, no. 1, May 1993, pp. 14-28 and, applied to the management of electronic
mail, in ‘Managing Electronic Mail’ elsewhere in this issue of Archives and
Manuscripts.
2. Glossary> ofRecords Management Terms, ARM A International, Prairie Village, KS,
1989, p. 16.
3. Irene Place and David Hyslop, Records Management: Controlling Business
Information, Reston Publishing Company, Inc. Reston, Va., 1982, p. 4; Katherine
Aschner, editor, Taking Control of Your Office Records: A Manager’s Guide, G.K.
Hall & Co., Boston, Ma., 1983, p. 4; Norman Kallaus and Mina Johnson, Records
Management, South-Western Publishing Co., Cincinnati, Oh., 1992, p. 3.
4. Records Management Association of Australia leaflet, undated; Archives Act, 1983
(Commonwealth), section 3.
5. Peter Mazakina, Archives and Records Management for Decision Makers: a RAMP
Study, UNESCO, Paris, 1990, p. 21; Glenda Acland, ‘Glossary’ in Judith Ellis,
editor, Keeping Archives, 2nd edn, Thorpe, Melbourne, 1993, p. 477; Frank Evans,
Donald Harrison, and Edwin Thompson, A Basic Glossary for Archivists,
Manuscript Curators, and Records Managers, reprinted by the American Society of
Archivists from The American Archivist, vol. 37, no. 3, July 1974, p. 428; P. Walne,
editor, Dictionary of Archival Terminology (ICA Handbook Series, 7), K.G.Saur,
Munich, 1988.
6. Frank Upward, ‘Records Management and Record Keeping: The Archival
Document’, Informaa Quarterly, vol. 7, no. 2, May 1991, pp. 48-53; Sue
McKemmish and Frank Upward, ‘The Archival Document: A Submission to the
Inquiry into Australia as an Information Society’, Archives and Manuscripts, vol.
19, no. 1, May 1991.
7. ‘A record is information created, collected or received in the initiation, conduct or
completion of an institutional or personal activity.
Records have the following requirements:
• provide evidence (which is a by-product)
• comprise content, context and structure
• have integrity and immutability
• are unique
• exist regardless of physical format
• lead to an outcome.
All these attributes exist with electronic records, but for these context is more
important, to establish a relationship in a system of records.’
The definition is unpublished at the time of writing. The author understands that
the Committee does not regard this definition as definitive and expects to refine it
further. The author’s thanks are due to Steve Stuckey of Australian Archives for
bringing this work to his attention.
8. For example in two papers presented at the Records Management Association of
Australia’s 1992 Convention. See Maire'ad Browne, ‘Professional Development
through Education’, pp. 7-25, and Sue McKemmish, ‘Core Knowledge and Skills
for Information Professionals — Converging or Diverging: the Implications of
Diverse World Views’, pp. 103-1 14, in Proceedings of the 9th National Convention
of the Records Management Association of Australia, RMAA, Sydney, 1992.
9. Particularly in Sue McKemmish and Frank Upward, editors, Archival Documents:
Providing Accountability Through Recordkeeping, Monash Occasional Papers in
 ROBERTS: DEFINING ELECTRONIC RECORDS 25

Librarianship, Recordkeeping and Bibliography, No. 3. Ancora Press, Melbourne,

1993 and in articles on the accountability theme in the May 1993 issue of Archives
and Manuscripts.
10. Chapter 7 ‘Files, and Lies’ in Volume 1 of Report On Investigation Into The Use of
Informers. Independent Commission Against Corruption, January 1993.
1 1. Our governing legislation, the Archives Act. 1960 (NSW), has a definition of a public
record which does not, of course, reflect these recent developments. At the time of
writing, we are working towards a definition which takes account of them for
inclusion in draft state records legislation which will replace the 1960 Act.
12. William Saffady, Managing Electronic Records, ARMA International, Prairie
Village, KS, 1993, p. 163.
1 3. Charles Dollar, Archival Theory and Information Technologies: The Impact of
Information Technologies on Archival Principles and Methods, Odo Bucci, editor,
Ancona, University of Macerata, 1992, p. 85. A variation is to describe electronic
records as records, defined elsewhere in evidence/transaction terms, but in
electronic form, as in Acland, op. cit., p. 469: ‘Records [i.e. defined as we noted
above] capable of being processed in a computer system and/or stored at any instant
in a medium which requires electronic or computer equipment to retrieve them’; or
in State Archives of Western Australia, Electronic Records: an investigation into
retention, storage and transfer options, LISWA Research Series No. 4, The Library
and Information Service of Western Australia, 1993, p. 67: ‘Electronic record:
Machine-readable information in electronic storage media judged to be an
organisational record in accordance with the definition of what constitutes a record
adopted as a records management policy’.
14. Acland, op. cit., p. 469; Evans, op. cit., p. 421.
15. Acland, op. cit., p. 477: ‘Records: Documents containing data or information of any
kind and in any form, created or received and accumulated by an organisation or
person in the transaction of business or the conduct of affairs and subsequently kept
as evidence of such activity through incorporation into the recordkeeping system of
the organisation or person. Records are the information by-products of
organisational activity.’
16. ARMA Glossary, p. 10: ‘Document: Recorded information regardless of medium or
characteristics. Frequently used interchangeably with the word record ..2 Compare
with the Glossary's definition of a record noted earlier in this article. Also Violet
Thomas, Dexter Schubert and Jo Ann Lee, Records Management: Systems and
Administration, John Wiley & Sons, New York, 1983, p. 393: ‘Record: Any
document valuable enough to be retained’.
1 7. ARMA Glossary, p. 10: ‘Document: ... A single record item (letter, memorandum,
form)’; Mary Robek, Gerald Brown and Wilmer Maedke, Information and Records
Management, 3rd edn, Glencoe Publishing Company, Encino, Ca., 1987, p. 565:
‘Document. The smallest unit of filing. Sometimes called a page'-, Acland, op. cit., p.
469: ‘Document 2) The smallest complete unit of record material, e.g. a letter,
photograph, report’.
18. George McDaniel, IBM Dictionary of Computing, 10th edn, McGraw-Hill, New
York, 1993, p. 212.
19. Richard Jones, ‘Directions in Intelligent Electronic Document Management’,
Informaa Quarterly, vol. 9, no. 1, May 1993, p. 4. Jones is also careful to distinguish
electronic documents from other kinds of multimedia: ‘A key distinguishing feature
of electronic documents compared with general multimedia is that text is the
predominant medium, acting metaphorically as the glue holding the other media
forms together. This may be in the form of intra or inter-document references.’
20. Information Exchange Steering Committee, Management of Electronic Documents
in the Australian Public Service — A Report prepared by the IESC’s Electronic Data
Management Subcommittee, Australian Government Publishing Service,
Canberra, 1993, p. 6.
26 ARCHIVES AND MANUSCRIPTS Vol. 22, No. 1
21. Compare with word processing programs, where a document has traditionally been
synonymous with a file.
22. ibid., p. 55.
23. Under Section 3, a record means ‘a document (including any written or printed
material) or object (including a sound recording, coded storage device, magneiic
tape or disc, microform, photograph, film, map, plan or model or a painting or other
pictorial or graphic work) that is, or has been, kept by reason of any information or
matter that it contains or can be obtained from it or by reason of its connection with
any event, person, circumstance or thing’.
24. This brief analysis is not intended to imply anything about the relative value of the
products. It is only concerned with the scope of the concept of electronic documents
underlying the products.
25. Most archivists and records managers are familiar with the exotic use in computing
of such terms as archive, record and file, which presents another set of
terminological challenges beyond the scope of this article.
26. McDaniel, op. cit., p. 173.
27. Simon Holloway, Data Administration, Gower, Aldershot, 1988, p. 7.
28. See, for example, the use of data administration as part of a generic framework for
managing electronic records in David Bearman’s article ‘Managing Electronic
Mail’ elsewhere in this issue of Archives and Manuscripts.
29. See Rob Smith-Roberts, ‘Saving the Important Bits for Later: Data Management
Principles & Metadata’, in Dagmar Parer and Ron Terry, editors, Managing
Electronic Records: Papers from a Workshop on Managing Electronic Records of
Archival Value, Australian Council of Archives Inc. and Australian Society of
Archivists Inc., Canberra, 1993, pp. 68-86.
30. We should note that the distinction between data management and data
administration is not necessarily well recognised in the computing industry or in its
literature, with the border between the two often ill defined and with significant
areas of overlap when the two functions are described. Thus a useful recent text
covering the full range of data management functions uses data administration in
its title (Brian Horrocks and Judy Moss, Practical Data Administration, Prentice
Hall, Hemel Hempstead, 1993). However, the distinction is important for
archivists and records managers because both provide useful, though different,
tools for electronic records management. Again we need to impose a degree of
terminological precision which does not necessarily exist au nature!.
31. Such as National Archives of Canada, Managing Your Computer Directories and
Files, Ottawa, 1993; Saffady, op. cit., Chapters 6 and 7. The author’s review of
Saffady’s book appears in New Zealand Archivist, vol. IV, no. 4 (Summer/
December) 1993 and is reprinted in this issue of Archives and Manuscripts.