SCHOOL OF INFORMATION MANAGEMENT
AND SYSTEMS

Module 1: Recordkeeping and Archiving: Storage

Topic 2: Information Representation

Objectives of Topic 2
Related Assessable Activities
Readings
Time Allocation

1. Introduction
2. Storage and Information Representation
3. Electronic Information: Digital Library perspective
4. Electronic Information: Data Archives perspective
5. Electronic Information: Recordkeeping perspective
6. Some Recordkeeping Strategies and Directions

Objective of Topic 2

• discuss information representation as an aspect of storage
• relate storage issues to migration and preservation issues
• demonstrate an understanding of strategic approaches to technological obsolescence
• discuss the strategies and techniques of preservation of electronic information from related information professions

Related Assessable Activities

Discussion: Recordkeeping perspectives on long term preservation of electronic information
Time Allocation: 2 hours
Due Date: Friday, October 23rd, 1998

Readings

Further reading
Digital Library approaches
Recordkeeping approaches

Web Browsing>
National Archives of Australia site
Public Record Office of Victoria, Victorian Electronic Records Strategy Project

Time Allocation

1. Introduction

Topic One pulled apart the notion of storage and suggested to you that there are various strands that we need to follow in order to better understand what we are doing. The strands are:

• Information representation,
• Recall, and
• Dissemination.

Traditionally we have lumped all of these aspects of storage together. The contention is that if we separate them out, we can see connections to other disciplinary concerns and also define our own concerns far more accurately.

Getting different perspectives on storage is particularly relevant to our professional endeavour relating to electronic records. Electronic records are technology-bound records. They are dependent upon software and hardware and the issues of managing electronic records over time are challenging many of our traditional archival practices.

Within considerations of electronic records (and electronic information communities more generally), issues of persistence and migration are fundamental. Issues of custody and distributed custody are also intertwined with discussions about what formats for storage should be adopted. How these issues are conceptualised affects the strategies and methods that are proposed for implementing both recordkeeping and 'archival' strategies.

This Topic, then, will focus on information representation – how the information is recorded, in what form, what media. While we are focussing on our own disciplinary area of recordkeeping, we draw connections to approaches from our library colleagues in an attempt to distinguish differences and to find commonality.

2. Storage and Information Representation

Storage decisions made at the point of capture reverberate through the lifespan of the record. Some of those decisions which are about information representation, determine:

• the form in which the record is captured
• the storage media on which the record is captured
• how it is re-presented for access (including decisions on migration of media, transformation into other media through processes such as digitisation etc).

and once the information has been captured:

• where it is physically or logically located
• the physical conditions in which it is placed

Many very familiar things that we have always done in relation to records storage are related to these issues. In the past we had paper based records as the norm and we routinely transferred records into archival custody. There is an extensive body of literature on archival preservation, all of which is relevant to this topic, and which is just too large to incorporate into our consideration here.

Archival authorities have been proactive in establishing standards to guide records programs located within their jurisdiction on issues of storage. Such guidelines include:

• Advice on paper quality
• Advice on physical care and records handling
• Advice on microfilming as a preservation technique
• Advice and standards for records storage facilities

The recently launched book by Ted Ling Solid, Safe, Secure. Building Archives Repositories in Australia, National Archives of Australia, 1998 is a fine example of the kind of detailed advice and statement of best practice available to practitioners seeking to physically secure their archival holdings.

In the records management world, the AS 4390, Australian Standard on Records Management, Part 6, Storage, provides detailed guidance on the physical aspects of records storage which need to accompany any good recordkeeping program.

In addition to the standards and guideline issuing roles of the archival institutions in this area, there are a number of well established best practices to ensure the physical conditions of storage are well maintained. Such best practices include:

• Fumigation
• Condition monitoring
• Disaster preparedness plans.

Browse

Browse the National Archives of Australia advices to identify the range and extent of the guidance available to agencies within the NAA jurisdiction on these issues. http://www.naa.gov.au

Discussions of storage have a tendency to become domain specific. Yet, the same issues and concerns arise in all contexts of recordkeeping. In agencies, records storage issues are of equal concern. From a records creating body's perspective, the issues are more likely to be focussed on:

• the requirements for storing or capturing records (functional analysis),
• balancing the requirements to capture records vs the risks of not having them,
• cost efficiencies,
• convenience, and
• speed of access.

Activity

Storage ( Media and Retrieval
Prepared by Anne Picot

Scenario Across Australia, people who drive or own motor vehicles are taxed for the privilege of doing so. The many offices (motor registries) which manage the business of licensing drivers and registering vehicles accept identification documents and other supporting documents from drivers and vehicle owners. The documents support the declaration of identity, the fitness to drive, the proof of ownership and of roadworthiness of the vehicle and the state of its insurance. There is a centralised computer system in each State/Territory which registers the vehicles and records the licensing of the drivers and the electronic record is generally regarded as the primary record. The other documents are supportive, they are evidence of the basis of the decision to license a driver or register a vehicle. In all state and territory jurisdictions recordkeepers in the relevant agency must make arrangements for managing these documents which generally become part of the record of registered vehicles and licensed drivers. In all of the jurisdictions, these supporting records may be recalled quite soon after the initial transaction is completed, or at any time for some years later. When the documents are retrieved within days of the initial transaction, it is generally required immediately, at the motor registry counter where a mistake of some kind has been discovered and needs to be corrected. The choice of storage medium, storage and retrieval arrangements and retention practices varies quite widely. The documents may be accepted as paper, photocopies generally, and subsequently held as paper, microfilmed or scanned and stored digitally. The scanned image and microfilmed versions of the supporting documents are generally kept for an indeterminate period ( not "permanent retention" but the records have not been destroyed to date. In the states/territories where the records are maintained as paper, the retention period varies from 7 to 10 years. One state is investigating the costs and benefits of the three identified storage arrangements(paper, microfilm and digital storage. The accumulation of documents has been measured at 4 shelf metres/per week (or 20 type 1 archival boxes which fit 5 to a metre of standard metal shelving). The rate of retrieval is approximately 30 items per week, 20% of which are required "immediately" to deal with a motor registry customer waiting at the counter. The other requests come from insurance companies, solicitors and the police for a variety of purposes. The first two are charged for access to the documents on a cost recovery basis, the police are not. The police often require the most complex searches for vehicle histories, changes of addresses of drivers going back several years and comparison of signatures. The access points for retrieval are: date motor registry location (name of suburb) motor registry officer ID (barcode) and the means of pinpointing those access points is the computer system which logs the Motor Registry officer's ID, linked to the motor registry, to every transaction she makes. Whether the documents are retained in paper or migrated to another medium, the computer system will continue to hold the principal record of the transaction (an electronic registration) and the means of accessing the supporting records will continue to be that computer system. Because these records, electronic and supporting documentation, hold current personal information, the control of access to them is of premium importance so all access transactions must be recorded.

Task

1. List the advantages and disadvantages of each storage media – paper, microfilm, scanned image – for the supporting documentation
2. Identify the issues that will need to be addressed in order to cost the best storage solution (eg what procedural steps will be needed to check image quality etc)
3. Propose your own recommendation for the agency.

Email: to Frank Upward

Due Date: Friday 23rd October, 1998

This case study brings our considerations of storage and recall down into a very concrete framework, one frequently facing recordkeepers in all agencies. It also reminds us that there are important distinctions to be made based on the reasons we take storage decisions:

• Is it to do with business activities?
• Is it to do with costs?
• Is it to do with accessibility?
• Is it to do with extending the life-span of a record?
• Is it to do with protecting a fragile original record?
• Is the decision one of converting from one media to another?
• Is it about the media of record capture?

The case study clearly locates itself as addressing the issues associated with converting paper into another media. The other media can be analogue (microfilm) or digital (scanned image). Even here we could further distinguish between an imaging application where the original is converted into a bit-map file of the original, kept as one entity and only manipulable as an entity, and a scanning process which digitised the information through optical character recognition techniques which enable the information content to be searched and manipulated.

3. Electronic Information: Digital Library Perspectives

The digital library community, then, has been very active in recent years thinking through issues to do with long term preservation of digital resources. Indeed, many of the discussions in that community have appropriated the term 'digital archives' to encompass their discussions. How similar or dissimilar are these conceptualisations from that of the recordkeeping community?

The discussion within the digital library community has, until recently, focussed on 'refreshing' of digital resources. 'Refreshing' is the notion of copying digital information periodically to new media.

'Refreshing digital information by copying will work as an effective preservation technique only as long as the information is encoded in a format that is independent of the particular hardware and software needed to use it and as long as there exists software to manipulate the format in current use. Otherwise, copying depends either on the compatibility of present and past versions of software and generations of hardware or the ability of competing hardware and software product lines to interoperate. In respect of these factors -- backward compatibility and interoperability -- the rate of technological change exacts a serious toll on efforts to ensure the longevity of digital information.'

Preserving Digital Information Report of Task Force on Archiving of Digital Information. Commissioned by the Commission on Preservations and Access and the Research Libraries Group, May 1996, p11

With the release of the 1996 report on digital archiving, the notion of migration became more prevalent within the digital library community.

'Migration is the periodic transfer of digital materials from one hardware/software configuration to another, or from one generation of computer technology to a subsequent generation. The purpose of migration is to preserve the integrity of digital objects and to retain the ability for clients to retrieve, display, and otherwise use them in the face of constantly changing technology. Migration includes refreshing as a means of digital preservation but differs from it in the sense that it is not always possible to make an exact digital copy or replica of a database or other information object as hardware and software change and still maintain the compatibility of the object with the new generation of technology. Even for information that is encoded in a contemporary standard form (e.g., a bibliographic database in USMARC or a corporate financial database in SQL relational tables), forward migration of the information to a new standard or application program is, as anyone knows who has witnessed or participated in such a process, time-consuming, costly and much more complex than simple refreshing.'

Preserving Digital Information Report of Task Force on Archiving of Digital Information. Commissioned by the Commission on Preservations and Access and the Research Libraries Group, May 1996, p12

'Stewards of digital material have a range of options when faced with the need to preserve digital information. One might preserve an exact replica of a digital object with complete display, retrieval and computational functionality, or a representation of it with only partial computation capabilities, or a surrogate such as an abstract, summary or aggregation. Detail or background noise might be dropped out intentionally through successive generations of migration, and custodians might change the form, format or media of the information. Enhancements are technologically possible through clean-up, mark-up, and linkage, or by adding indexing and other features. These technological possibilities in turn impose serious new responsibilities for presenting digital material to users in a way that allows them to determine the authenticity of the information and its relationship to the original object.'

Preserving Digital Information Report of Task Force on Archiving of Digital Information. Commissioned by the Commission on Preservations and Access and the Research Libraries Group, May 1996, p13 'http://www.rlg.org/ArchTF

Margaret Hedstrom, prominent North American archivist and educator, was active on the Task Force on Archiving Digital Information and in particular was responsible for the shift in focus from refreshing to migration.

Reading

Margaret Hedstrom, 'Research Issues in Migration and Long-Term Preservation' Electronic Records Meeting, Pittsburgh, PA, May 1997, http://www.sis.pitt.edu/~cerar

Note particularly the strategic approaches to migration identified in this paper. Many of them will be touched upon in the following discussion.

Further Reading: Digital Library Approaches

A large number of reports and issues papers are available from the digital library addressing these issues. Many of them can be accessed through links established on the National Library of Australia's Preserving Access to Digital Information page: http://www.nla.gov.au/padi/case.html

'Digital Collections: A strategic policy framework for creating and preserving digital resources' Arts and Humanities Data Service, First Public Consultation and Review Draft, Version 3.1, 24 April 1998 http://ahds.ac.uk/manage/framework.htm See particularly, section 5.4 The Institutional Archives (in my print pp34-40 of a 69 page report)

This recent report from the UK identifies the distributed nature of responsibilities involved in creating and managing digital resources. This is an interesting report as it embraces the notion of different types of resources within its frameworks, moving towards an understanding that there is a multiplicity of different requirements for managing digital resources. It also places some emphasis on the legal and economic environment in managing digital resources. However, it does these things from within a life cycle approach to digital resources (creation, management/preservation and use) and its title alerts us to the fact that there is an emphasis on collections. While this report seems to be breaking new ground, it still does not embrace the concept of a record – it places its emphasis at both a broader 'collection' level and at a narrower 'data' level. It ends up being about 'data archiving'. The understanding of records in this report is pretty shallow.

4. Electronic Information: Data Archives Perspectives

The data archive approach was an early strategy for management of electronic records. Terry Cook characterises strategies based on principles of data archiving and the transfer of data into 'software neutral' formats as the 'first generation' response to electronic records. This approach is exemplified by NARA who has been one of the key players in establishing 'machine-readable' archives. It continues to be a leader in managing considerable quantities of records in electronic formats.

The Center for Electronic Records at NARA 'appraises, accessions, preserves and provides access to Federal Records that were transferred to NARA by agencies of the Federal Government in a machine-readable format. Generally, the Center for Electronic Records does not preserve digitized versions of textual documents, or records in a software-dependent form' ( http://www.nara.gov/nara/electronic/faq.html). It (and its predecessors) have been undertaking such transfers and accessions for the past 25 years, making it one of the institutions with longest experience in dealing with custodial issues to do with electronic records.

The Center works on the premise that all electronic records accepted for transfer into the archives should be in a standard format. In the 1970s and 1980s this was a flat file structure in ASCII text, as suited most of the applications of computers in government. Since the 1990s relational database systems have been able to be accepted. The Center uses the Archival Preservation System (APS) to transfer electronic records into standard format, writes archival copies of physical files, automatically tracks all of the media volumes it writes and facilitates the eventual migration of files into new media. A second component of the system is the AERIC (Archival Electronic Records Inspection and Control) system. It is a processing system which provides for the preservation of logical and conceptual structures of databases.

There has been considerable controversy over the years about whether or not the Center is dealing with records or datasets. In general, it is a service to government which is dealing with whole datasets and databases etc. Most of those systems are information systems, as they were designed without the requirements of recordkeeping built into them, in common with every other database system written in the past (and still true today).

Almost every other archives institution which accepted deposit of electronic records during the 1970s and 1980s followed similar strategies for rendering 'machine-readable' records into 'neutral' formats.

This data archiving approach is still a strong feature of much thinking about the electronic records conundrum. Recently the National Archives of Ireland and the PRO London have both considered electronic records issues and both recommended the establishment of data archives in their respective countries. The PRO has established a program which seeks to preserve government datasets in the project NDAD (National Digital Archive of Datasets). The management of the project and the 'home' of the data archive has recently been outsourced to the University of London.

5. Electronic Information: the Recordkeeping Perspective

Persistence is a vogue word in the digital and electronic information communities at the moment. We have persistent URLs, and an increasing understanding that digital objects and information resources are inherently non-persistent. The word is constantly cited in the literature. But what does it mean? I assume in most cases it is used as an umbrella term to cover issues about maintaining digital resources in useable ways over time.

Most of the discussions of digital preservation and persistence seem to indicate that the object or resource being acted upon is a static and passive thing, a physical thing, which needs serious attention to extend its life span. The recordkeeping stance takes a somewhat different view of the record-object and imbues it with activity. The record is in constant process of creation. The content is only one of the three major planks of a record – the others being structure and context. Maintaining the contextual connections and links within recordkeeping systems is part of the emerging notion of records as constantly dynamic objects. This poses a different view to the passive, static view of record as finished thing, which is most common in discussions of digital persistence.

'Persistence to recordkeepers is not only an issue associated with being able to consistently access a digital object. It also includes the capacity to consistently interpret that object over time. Record objects need to be managed in ways that address complex notions relating to their meaning being dependent on context, and the fact that meaning is changing over time as society and organisations experience massive and interrelated cultural, functional and structural change. Links to the original contexts of creation must be carried with record objects for them to remain meaningful in future contexts. Such dependent metadata cannot be attributed once to stand firm for the whole of the existence of the object. [We must] construct mechanisms that enable record objects to be enriched by metadata throughout their lifespan, so that they continue to be meaningful over time and beyond their context of creation. Doing this involves deconstructing the nature of the dependent relationships inherent in the metadata elements attributed at the time of creation.

Records, in common with other types of networked information resources, also contain metadata which is contingent upon the exercise of rights or the permissions associated with particular actions. Rights to access records of business transactions may depend on whether individuals are party to the transaction. Many transactions are private and confidential. Issues of confidentiality, eg, are linked to a timeframe. A record may be confidential while it is in the process of development, while an action is being discussed, or while a person is alive. These sensitivities decline with age. Hard and fast rules on accessibility, or even the existence of the resource, cannot be 'hardwired' into the metadata framework, but must be resolved according to the status, time and nature of the subsequent request for access.'

1999 Large Research Grant Application, 'Metadata architecture to support persistence of essential evidence of business, social and cultural activity in distributed networked environments', February 1998 (Chief Investigators Barbara Reed and Sue McKemmish)

Such tentative beginnings to re-thinking some of our recordkeeping perspectives provide a different scenario in which to think about what we need to do to ensure that records remain active over time. This is not an approach which supports the notion of passive treatment of records as 'dead' objects.

These understandings affect the types of systems we design (or at this stage envisage) to meet our recordkeeping requirements to keep digital records over the long term. And, an approach along the lines above will look very different from a passive repository of 'things' approach.

6. Some Recordkeeping Strategies and Directions

At a conceptual level it seems that we are moving towards a better formulation and standardisation of the documentary forms (exemplified through the University of British Columbia's research project, particularly their document templates ) and through more routine capture of records (through initiatives and experiments with models such as the Business Acceptable Communications model emerging from the University of Pittsburgh research). As we are more able to clearly specify these things, we are able to specify better corporate records and then issue rules at a wider jurisdictional level to assist implementers in making these things happen. We will be discussing the practical implications of these projects and how they relate to implementation issues in Module 2 of this course.

What will work in one situation will not work in another. It is extremely likely that a variety of strategies will be needed into the future to deal with issues of software and hardware dependencies. Strategies now out of favour may well turn out to be successful, because they often have the benefit of simplicity. We are also stuck in our current understandings of a technological environment which is moving so rapidly that few can predict with any accuracy beyond 3 or 4 years.

6.1 Transfer to paper

This type of response has now almost completely disappeared from recordkeeping 'solutions', but until the early 1990s this was recognised as an achievable, if not ideal response. The Australian Archives slogan 'want it, keep it, print it, file it', was an example of the widespread practical reaction to early electronic document creation and communication systems.

The PROFs case in America did substantial damage to even the practical nature of this response.

'It [the Court] rejected the government argument that electronic copies are convenience copies if the primary organisational records are maintained in paper format. The court sided with the plaintiffs who argued that if anything is to be considered a convenience copy in an electronic environment it would have to be the paper copies because more can be done with the electronic record, not all records are copied to paper, and more information is present about the structure and context of the record in its electronic form'

David Bearman, 'The Implications of Armstrong v. the Executive Office of the President for the Archival Management of Electronic Records' Electronic Evidence Archives and Museum Informatics, 1994, p.124 (first published in American Archivist Vol 56, Fall 1993)

In a recent court challenge in America this notion was again upheld. In the judicial ruling against NARA in this case, one of the strategies that was rejected was that printing out electronic records and placing them in paper-based electronic recordkeeping systems. This was ruled as not a viable option because it allows agencies too much latitude to destroy electronic records.

6.2 Retain in native software environment

This strategy implies that documents created under proprietary software are maintained in that environment and migrated at appropriate times into the next appropriate generation of that native software. This is, in effect, what happens by default in most current workplaces, although the reality is that often the migration does not take place.

This is the reality that we all face, where workplaces are by default saving things in their native software environment. But, it doesn't have to be so. If records 'fall out' of the business into recordkeeping systems then those systems can impose different structural conditions on the record. This can be done at the time of capture. There is nothing inherently custodial or non custodial in this approach. So, we need to set up some guidelines of rules about this at point of capture.

This is also the arena where the distributed custody policies of the National Archives of Australian and the Archives Office of NSW are most immediately applicable.

This type of strategy is heavily dependent on constant monitoring for the demise of software able to process, extract or read the particular versions of software involved. It will also involve a huge diversity of software types and platforms. The most promising and optimistic outlook is if we can reach the 'self-migrating' record object that is outlined in David Bearman's 'Item level control' article. Whichever mechanism is employed, this strategy will involve migration as a continuous process. With the shelf-life of software averaging 2-3 years, there are many many migrations to consider during the life span of one record.

Given that migration almost inevitably involves the loss of some functionality, one of the research questions that Margaret Hedstrom poses becomes critical: 'to define acceptable levels of information loss during migration and to identify a set of minimal record attributes, which if not retained, would make investments in preservation pointless'. (Hedstrom, op cit)

6.3 Migrate to a system that is compliant with open standards

This strategy could also embrace the 'software-independent' strategy of data archives outlined above, but I am going to explore the notion of representational formats, standard formats or archival storage formats here.

Representation Formats or standard formats for capturing records have been part of a variety of electronic archives projects. In the early days, ASCII flat files were the representation format of choice. Recently there has been a move to discussing the Adobe Portable Document Format as a likely archival format. This has been adopted in the VERS project of the Public Records Office of Victoria (discussed below). Rich Lysakowski's CENSA (Collaborative Electronic Notebook Systems Association) is looking to identify functional requirements for recordkeeping of scientific research records and to identify long term formats. It is significant that Adobe has just been anounced as a CENSA member, with the statement that 'Issues like document fidelity, robust security, long term access and reusability of data are the cornerstones of Adobe's Acrobat technology and these capabilities coincide perfectly with CENSA's own long term needs and product development objectives.' (Electronic Records listserv E-Recs posting from Rich Lysakowski 30 May 1998.)

The Public Record Office of Victoria's Electronic Records Project has adopted Adobe PDF as its preferred storage format along with XML (extensible markup language - an emerging metadata formatting standard) for its pilot projects currently being developed.

The VERS project commenced with a report issued at the end of 1996, from a Project Team consisting primarily of Ernst and Young and CSIRO with input from a variety of archival and agency recordkeeping personnel. The report 'Keeping Electronic Records Forever, Records Management Vision Development' outlined the approach to be taken in the project.

The report recommended a solution that 'records be 'frozen' into a static view/print only format at the time of their creation. The 'frozen' image of the document is captured in a to-be developed or determined Representation Format through which the same record will be retrieved in years to come with satisfactory and evidentiary quality replicability.' (p3) The Representation format was not defined in the report which compared the performace of Tiff, Postscript, PDF and SGML. The report recommended 'An interim Representation Format comprising Postscript for the approximate form combined with a text dump for word indexing represents a combined format with desirable properties that can be readily generated from all common applications provided compression and authentication technology is added' and 'a migration to a presentation format (eg SGML or PDF) can be made should one be widely adopted in the future'.(p60). The report also recommends the use of CD ROM technology as the most likely storage format.

The report proposed this strategic direction and proposed that the PROV proceed with a Prototype Project rather than the usual development of a business case.

In many ways, this report was a preliminary document. In sheer practical terms it is often necessary to conduct such 'expert' reports in order to gain funding to do more work and that has been the outcome of the report. In March this year a tender document was issued which proposed further development of the prototype project. The tender document, development of a prototype Archival Representation Format' makes it clear that the VERS(Victorian Electronic Records Strategy) is seeking to develop an product independent representation format which will work in conjunction with electronic document management systems and records management systems. The VERS strategy seeks to establish a mechanism to enable 'the direct deposit of electronic records in an electronic records management system, separating the record from the algorithm with which it was created'.

As the project has developed the issue of archival format seems to have become less of a hard and fast thing. The project briefing of Dr Ross Wilkinson to the Electronic Records Special Interest Group meeting in Fremantle in August 1998 appeared to accept that all that can be specified is the most stable format that is available at the moment, but that there is no guarantee that this will remain fixed into the future. The strategy outlined by Dr Wilkinson is not dependent upon custody, which can take place anywhere - either in an agency records management system or in an accumulation located at the PROV. The models which form the background to this project present the record as an object wrapped around with layers of metadata.

The PROV site includes a page of information on the project, which is in prototype development phase. Major vendors of records management software are involved in the prototype project which is being conducted in one of the Victorian government agencies. This is an exiting project to watch and it is very likely that a working prototype will be operational within the next 12 months.

Adopting a representation standard seems to be a solid common sense strategy which is suited to a volatile climate. However, it does need to be placed into an understanding of the incredibly volatile electronic developments happening around us and be regarded as something that we can use as a present strategy which doesn't become a millstone around our necks as we force compliance to an aging standard. This is no more the 'solution' to electronic records storage and migration issues than transferring to paper is.

Browse

PROV VERS page http://www.vicnet.net.au/~provic/vers

6.4 Save the software needed for access and retrieval

Some of the museum of technology arguments might be re-run under this heading, but the costs and likelihood of this being an acceptable or cost effective strategy has been uniformly dismissed as a recordkeeping strategy.

However, the Public Record Office, UK has proposed an intriguing variant of this notion in their consideration of 'bundles'.

Over the last two or three years the PRO has begun to address electronic records seriously. They have identified a two pronged approach to this, with one arm addressing datasets – things like statistical databases and information systems (NDAD discussed in section 4), and another arm addressing electronic records from office systems (EROS).

EROS (Electronic Records in Office Systems) is an umbrella name for a project that has three major areas of action:

1. Records management guidelines for departments
2. The records processes of appraisal, accessioning and preservation
3. Long term transfer and access strategy.

The Public Record Office is proposing to accept custody of electronic records from office systems, but is investigating an approach which seeks to 'bundle' records with sufficient functionality to view, browse and extract and represent documents in their original form. The PRO has taken a strategic role in the formulation of this strategy. The bundle specification is being developed under the auspices of a British Standards Institute project, aiming to produce a British Standard for bundles.

6.5 Develop Software Emulators

There is little discussion on this issue. It is an idea floated by Jeff Rothenberg in a 1995 Scientific American article:

'The alternative to translating a digital document is to view it by using the program that produced it. In theory, we might not actually have to run this software. If we could describe its behaviour in a way that does not depend on any particular computer system, future generations could re-create the bahaviour of the software and thereby read the document. But information science cannot yet describe the behaviour of software in sufficient depth for this approach to work, nor is it likely to be able to do so in the near future. To replicate the behaviour of a program, there is currently little choice but to run it.'

Jeff Rotherberg, 'Ensuring the Longevity of Digital Documents' Scientific American, January 1995, Vol 272, No 1

'Fortunately, software engineers can write programs called 'emulators', which mimic the behaviour of hardware. Assuming that computers will become far more powerful than they are today, they should be able to emulate obsolete systems on demand. The main drawback of emulation is that it requires detailed specifications for the outdated hardware. To be readable for posterity, these specifications must be saved in a digital form independent of any particular software, to prevent having to emulate one system to read the specifications needed to emulate another.'

Jeff Rotherberg, 'Ensuring the Longevity of Digital Documents' Scientific American, January 1995, Vol 272, No 1

There is nothing we have seen in the archival literature relating to any specific work which has taken these ideas beyond Rothenberg's suggestions.

Email Discussion

The Task Force on Archiving Digital Information report mentioned in section 3 above has been criticised quite vocally in the archival literature as having a mind set relevant to life-cycle notions of records and information resources and putting in place only post-factum strategies, fail safe mechanisms and failing to address any pro-active action to keep resources current. (Adrian Cunningham, Review, Archives and Manuscripts Vol 24, No 2, November 1996 and David Bearman, 'Preserving Ditigal Information:A Review' Archives and Museum Informatics Vol 10 No 2, 1996).

These reviews from a recordkeeping perspective have great validity, because the digital library community are adopting the phrase 'digital archives' to encompass their strategies. The continuum based recordkeepers are posing a critique which is based upon that appropriation of the term and implying that this is not their view of what digital archives are or should be. However, as Margaret Hedstrom has pointed out:

'Both Bearman and Cunningham in their reviews of the digital archiving task force report stressed the need to differentiate records from other types of digital information. While this point is well taken, it does not really move the agenda forward unless the archival community can be more specific about why such a distinction is important and where and how the requirements for preserving electronic records differs from the requirements of other types of digital information.'

Margaret Hedstrom raised three questions coming out of the recordkeeping criticisms of the Task Force Report. These were:

• Are the requirements for long-term preservation of records fundamentally different from the requirements for other types of digital information? If so, how are they different?
• Do electronic records require fundamentally different preservation strategies from other types of digital objects, or are they a subset of digital information that can be maintained in common systems provided that a few additional requirements are met?
• Is it feasible, useful, and cost-effective to apply the requirements for preserving electronic records (metadata standards, process controls etc) to other types of digital objects?

From your understandings developed through this topic and your experience, post a response to these questions.

Email: a response to all of us

Due Date: Friday, October 23rd 1998

(Don't try to overdo this, these are huge questions and without substantial research can't be answered properly – all that is sought for are your impressions).

Further Reading: Recordkeeping Approaches

Ian Macfarlane, 'Overview of the EROS (Electronic records from Office Systems) Programme, (London Public Record Office and Central Computers and Telecommunications Agency), October 1996 http://www.sis.pitt.edu/~cerar

'Overview of EROS Project' from the PRO web site, http://www.pro.gov.uk/government/eros/default.htm

Tony McKinley, 'Why PDF is Everywhere' Inform Magazine September 1997 http://www.aiim.org/inform/Sept97/docpdf1`.htm
(mind you, note that the author has just written a book published by Adobe Press, so there may be a bit of bias here).

Kenneth Thibodeau, 'Electronic Records Activities at the National Archives and Records Administration (USA)' in Playing for Keeps, The Proceedings of an Electronic Management Conference, Australian Archives, 1994 http://www.naa.gov.au/govserv/techpub/keeps/p4k.html

School of Information Management and Systems