Selection of Authentic Electronic Records
Terry Eastwood
University of British Columbia Vancouver, B.C Canada V6T 1Z1 [email protected]
ABSTRACT. The InterPARES Project is a collaborative international research initiative bringing scholars in several disciplines together with representatives of archival institutions and private industry to develop the knowledge required for the long-term preservation of authentic electronic records. The information technology revolution has dramatically altered the way in which governments, private organizations, and individuals carry out their daily activities. Technological obsolescence, storage media fragility, and the capability of electronic systems to manipulate data challenge our capacity to guarantee the long-term preservation and authenticity of electronic records. This paper will report on the work of the Appraisal Task Force of InterPARES to characterize the concepts, principles, and methods for effective selection of electronic records.
RÉSUMÉ. Le Projet InterPARES est un projet de recherche international qui regroupe des chercheurs de plusieurs disciplines avec de représentants d’institutions d’archives et d’industries privées, en vue de développer les connaissances nécessaires à la conservation à long terme des documents d’archives électroniques authentiques. La révolution des technologies de l’information a fondamentalement modifié la façon dont les gouvernements, les entreprises et les individus conduisent leurs activités quotidiennes. L’obsolescence des technologies, la fragilité des supports de stockage et la facilité de manipulation des systèmes électroniques défient notre capacité à garantir la conservation à long terme et l’authenticité des documents d’archives électroniques. Cet article expose le travail du groupe de projet d’InterPARES sur l’évaluation pour caractériser les concepts, les principes et les méthodes d’une sélection efficace des documents d’archives électroniques.
KEY WORDS: electronic records, information technology long-term preservation, authenticity, appraisal, selection, InterPARES.
MOTS-CLÉS: documents d’archives électroniques, technologies de l’information, conservation à long terme, authenticité, évaluation, sélection, InterPARES.
1. Introduction
In contemporary times, the technology of computers and communication have been combined to make instantaneous access to information a routine matter in the conduct of affairs in every sector of society. People value the latest, supposedly best information, and continually obtain the latest technology to acquire and manage it.
Some of this information is closely connected with the actions taken during the conduct of affairs, but a great deal of it is simply part of the social ambience. In such circumstances, it is often difficult to distinguish records or archival documents, the two terms meaning the same, from other stores of data and information. It is also difficult to manage records so that they remain uncorrupted in an environment that puts a premium on the capability to communicate and manipulate information to increase productivity. Such an environment contrives in innumerable ways to defeat the archival goal of long-term preservation of authentic records.
As MacNeil [McN 00a] puts it, “the authenticity of a record is assessed in relation to its identity (i.e., was it written by the person who purports to have written it?) and its integrity (i.e., has it been altered in any way since it was created and, if so, has such alteration changed its essential character?).” In the traditional environment of stable physical media like paper, assessments of authenticity have relied on the enduring existence of physical objects. In the electronic environment, we lose the obvious assurance of authenticity that can be gleaned from examining the physical aspects of the record. In the digital world, as Ken Thibodeau [THI 00]
remarks, “strictly speaking, it is not possible to preserve an electronic record. It is only possible to preserve the ability to reproduce an electronic record. It is always necessary to retrieve from storage the binary digits that make up the record and process them through some software for delivery or presentation.” In every twist and turn of the technology, even in its ordinary operation, exact replication in every case is not guaranteed. In such circumstances, what needs to be done if we are to give electronic records a measure of trust such that they will serve us in the ways traditional records have?
2. The InterPARES project and the Appraisal Task Force
To address these vexing difficulties, the InterPARES project sets itself the goal of developing the theoretical and methodological knowledge essential to the permanent preservation of electronically generated records and, on the basis of this knowledge, formulating model strategies, policies, and standards capable of ensuring their preservation. InterPARES is an acronym standing for International Research on Permanent Authentic Records in Electronic Systems. The project, an international, interdisciplinary, collaborative research initiative, began work in January 1999, and is scheduled to complete its first phase at the end of 2001. Further information on the background, organization, researchers, objectives, and
methodology of the project is available at <www.interpares.org>. The project’s researchers are divided into three task forces investigating the conceptual requirements for preserving authentic electronic records (the Authenticity Task Force), appraisal criteria and methods for authentic electronic records (the Appraisal Task Force), and methods for preserving authentic electronic records (the Preservation Task Force.) In the latter stages of the project, the results of the work of the three task forces will be integrated in the work of the fourth task force, called the Strategies Task Force. It is important to note that the work on appraisal is dependent on the Template for Analysis developed from the perspective of contemporary diplomatics by the Authenticity Task Force1, and on its work to specify the conceptual requirements for authenticity of electronic records. Readers may also refer to the glossary of archival and diplomatic terminology available on the project’s website2. This article reports the preliminary findings of the Appraisal Task Force.
The goal of the Appraisal Task Force is to determine whether the evaluation of electronic records for permanent preservation should be based on theoretical criteria different from those for traditional records and how digital technologies affect the methodology of appraisal. Before the Task Force started its work, a number of research questions in its domain were set out. They were:
– What is the influence of digital technology on appraisal?
– What is the influence on appraisal of retrievability, intelligibility, functionality, and research needs?
– What are the influences of the medium and the physical form of the record on appraisal?
– When in the course of their existence should electronic records be appraised?
– Should electronic records be appraised more than once in the course of their existence, and, if so, when?
– Who should be responsible for appraising electronic records?
– What are the appraisal criteria and methods for authentic electronic records?
The author of this paper is Chair of the Task Force. The other members are Barbara Craig of the Faculty of Information Studies at the University of Toronto, Phil Eppard, of the Faculty of Information at The University at Albany of the State University of New York, Gigliola Fioravanti of the Italian Central Direction of Archives, Norman Fortier of the National Archives of Canada, Mark Giguerre of the National Archives and Records Administration of the United States, Ken Hannigan of the National Archives of Ireland, Peter Horsman of the School of Archival
1. See<www.interpares.org/documents/TemplateforAnalysis_071100.pdf>
2. http//:www.interpares.org/documents/InterPARESGlossary_2000_3.pdf
Science and Research in the University of Amsterdam, and Du Mei from the central archives administration in China.
We began our work by reviewing the literature in English on appraisal of electronic records [EAS 00]. Archivists have been concerned with the challenge of appraising electronic records for over twenty years, through several phases in the evolution of technology, from the period dominated by mainframe computers, to the introduction and spread of isolated desktop personal computers, to the present with its organizational intranets and the Internet. Much of the literature dwells on the difficulties of defining what a record is in the electronic environment and of identifying and extracting records of long-term value (or data or information for that matter) from live systems. In addition, celebrated court cases like Armstrong v.
Executive Office of the President and Public Citizen v. John Carlin in the United States implicitly raised questions about the trustworthiness of records maintained in systems subject to few procedural controls. MacNeil [McN 00b] summarizes the significance of these cases.
Rather than discussing appraisal as such, many writers in the 1980s and 1990s fell to exhorting organizations to make provision for effective electronic records keeping, including procedures for disposition, in the processes of systems design and implementation. For example, an American archivist (Kowlowitz, 1991) maintained that “the most pressing issues facing electronic records appraisal today are not narrowly technical and methodological but broad program development and information management issues.” Most writers of the time believed that archivists had to be involved in the design of systems to build into them procedures for the appraisal and disposition of records. A Canadian archivist [BAI 89] asserted that archivists cannot wait until inactive electronic records are offered to them for appraisal, as they might have for paper records; too many computer records have vanished by then, and the documentation necessary for their proper appraisal has been lost, destroyed, or is hopelessly outdated. The sheer volatility of electronic records should be a powerful inducement for archivists to accept increased involvement in the scheduling process, beginning at the systems design stage.
Again, however, this is not an issue of new or revised theory or principle, but merely one of timing and strategy.
Australian ([ACL 91] and [OSH 91]), Dutch ([HOF 94] and [HOR 97]), and American [DOL 92] writers, as well as a publication by the International Council on Archives [ICA 97], take a similar stance. These difficulties explain in part why so few archival services have actually appraised and taken electronic records into custody. In the circumstances, it is not surprising that few writers discuss the difficulty of preserving authentic electronic records, and how those difficulties can in part be addressed during the process of appraisal. The aim of our work, then, is to demonstrate how during appraisal actions can be taken to initiate the process of preservation of authentic electronic records.
To further its work, the Task Force decided to engage in the exercise of developing a function model of the various activities undertaken during appraisal as a way of isolating the various theoretical and methodological questions that arise. A function model represents the various activities of a functional process in a series of structured diagrams. The Task Force used IDEF (0) or Integration Definition for Function Modeling, which is derived from Structured Analysis and Design Technique. IDEF (0) is a United States Federal Information Processing Standard, which is detailed in Publication 183 of the National Institute of Standards of Technology (see <http://www.idef.com/idef0.html> for more information). The model and the related definitions of terms, which are still in the process of being refined, are reproduced in at the end of this article.
The purpose of the model is to define the activities involved in selection of authentic electronic records for long-term preservation. Note that the archival function at issue is selection, which the Task Force takes to encompass the activities performed when evaluating electronic records for the purpose of continuing preservation and when carrying out the disposition of records so evaluated. The model is produced from the viewpoint of the entity responsible for the long-term preservation of electronic records of an organization. That entity may be an archival institution like the archives of a national, provincial, or municipal government, or it may be an archival program of an organization such as a church, a university, or a corporate body like a business. The assumption is that the same activities occur in any context where selection is performed.
The master diagram called Select Electronic Records (see Node: A-0 and the definition of the term in Model Activities Definitions) shows the various controls or constraints on the function as a whole as downward pointing arrows. The mechanisms needed to carry out the function are shown as upward pointing arrows, and the inputs or things consumed or transformed during the process by arrows on the right and outputs or things resulting from the process by arrows on the left. The other diagrams represent the successive decompositions of the selection function into its component activities. In this text, terms shown on the diagrams are in boldface type to assist readers to locate them on the diagram or in the appropriate list of definitions.
Selection of electronic records (or non-electronic records for that matter) is controlled or constrained by the perceived interests served by continuing preservation of records of the entity creating the records (Creator’s Needs), on the one hand, and of other than the creator’s (Societal Needs), on the other hand. The concepts of creator’s needs and societal needs reflect the traditional distinction between primary value and secondary value propounded by Schellenberg (1956).
The evaluative dimension in selection calls on a measure of interpretation of the needs of the creator and of the needs of others in society for continuing preservation of the records. This much, the Task Force assumes, is uncontroversial. How
determinations of value are made in any case is another matter, but not one germane to the exercise of determining criteria and methods relevant to the maintenance of the authenticity of electronic records that it has been determined need to be preserved. That is to say, we have nothing to say about what should be preserved, only about what needs to be done during appraisal to ensure continuing preservation of authentic electronic records.
There are other constraints. There may be Legal Requirements that dictate either preservation or destruction of certain records, or that raise considerations, such as for the protection of privacy, which must be taken into account. Of course, the theoretical concepts and principles of Archival Science, such as the nature of the archival fonds and the principle of provenance, also guide the conduct of selection.
For the purposes of this article, archival science should be taken to be synonymous with the French word “l’archivistique,” encompassing theory, method, and practice.
Working out in both a conceptual and methodological way what the requirements for authenticity are is of course part of archival science. However, because authenticity of electronic records is the focus of the InterPARES project, we have separately distinguished Authenticity Requirements as a control or constraint on the selection function. As we shall see, the Appraisal Task Force is relying upon the ongoing work of the Authenticity Task Force to specify these requirements.
In the process of selection, Electronic Records are evaluated and their disposition determined, the result being Electronic Records Selected for Preservation and/or Electronic Records Not Selected for Preservation. Essentially, the evaluation and disposition sub-processes depend alike on information that can be gleaned either from the records themselves (Information From Electronic Records), such as by viewing them to determine the degree to which they are complete and effective, or from their context. Because selection is so sensitive to context it is worthwhile elaborating at some length the InterPARES view of the matter.
3. Record’s context
The Template for Analysis (all quotations refer to this document) distinguishes five aspects of an electronic record’s context, defined as “the framework in which the action in which the record participates takes place.” The five aspects are the juridical- administrative, provenancial, procedural, documentary, and technological contexts. The juridical-administrative context refers to “the legal and organizational system in which the creating body belongs.” Appraisers usually carry with them considerable actual and intuitive knowledge of the constitutional and legal system and its rules and practices for accountability and the like. The provenancial (the word is our neologism) context refers to “the creating body, its mandate, structure, and functions.” It is hardly necessary to say that every appraiser must thoroughly understand the mandate (or mission of a private corporate body), structure, and
functions of the body creating the records in question. Various inferences from this information are habitually drawn about the significance, importance, or value of the records. The procedural context refers to “the business procedure in the course of which the records are generated.” The documentary context refers to “the fonds to which the records belong and its internal structure,” that is to say, the manner in which the records are organized and interrelated. Increasingly, organizations are engaged in integrating business procedures with documentary procedures in the electronic environment. Understanding these two aspects of context also cultivates inferences about significance, importance, and value. These four aspects of context are indicated as an input to the selection function called Information About the Records’ Context. Because of its importance, Information About the Technological Context of Records is set apart.
The fifth aspect of context, the technological context, is obviously vital to the understanding and appraisal of electronic records. It refers to “the hardware and software environment in which the records exist.” Although it is important to understand the hardware environment of the record, it is widely recognized that it is impossible to perpetuate it into the distant future, short of making archival institutions and programs incredible museums of technology. In large measure, understanding electronic records and perpetuating them in authentic form comes down to understanding and developing a strategy to perpetuate certain aspects or capabilities of the originating software environment. The problem, of course, is that this environment is complex and ever changing. It is the single most daunting factor in long-term preservation. The Template identifies five kinds of software. The operating system “manages, controls, protects, and facilitates the use of hardware resources in the electronic system.” It has such functions as process management, memory management, secondary storage management, disk scheduling, virtual memory management, user interface, and device and network interface, and as such has little direct influence on appraisal. System software (also called system utilities or system tools “creates an environment for application programs to be created, executed, and maintained, typically through system calls to the operating system.”
Examples would be machine language, compilers, interpreters, translators, coding for compression or encryption, and virus detectors. It may be necessary to know the characteristics of software in this realm if records are compressed or encrypted, although, at this stage, the world has little experience of dealing with compressed or encrypted records to go on where long-term preservation is at issue. Network software “manages networks and their resources in order to meet communication requirements of one or more applications, and has almost no influence on appraisal.
Application software is “any type of program that is tailored to satisfy real-world needs and requirements.” Application software usually has built into it a set of parameters or characteristics whose values affect the way in which data is formatted and stored. A great deal of the difficulty in analyzing electronic records occurs in determining whether the manner of handling data dictated by the software produces anything that we can recognize as records. Two questions need to be answered.
First, how are records constituted as data by the software? Second, how can we deal with the data over the long-term to ensure that the records can continue to be presented to users uncorrupted or unaltered, with all their necessary characteristics intact?
The Template defines data as “numbers, characters, images or other methods of recording which represent values that can be stored, processed, and transmitted by electronic systems.” Several aspects of the way data is organized are vitally important to understand when confronting electronic records. File structure is “the relationship and organization of files [in the sense in which the term is used in information technology] within a system.” It includes the directory structure, and may include the physical structure and organization of files in a file system, and the mapping of files on disks. The data format or file format is “the organization of data within files. These are organizations that are usually designed to facilitate storage, retrieval, processing and presentation and/or transmission of the data by software.”
Data formats may be standardized, such as ASCII text, or proprietary, such as Microsoft’s various WORD versions and Adobe’s PDF file formats. The tabular format of data files in database management systems, and the format (using tags) of data files used by mark-up languages (like SGML and XML) are other examples.
Obviously, standardized formats are easier to deal with over time than those tied to proprietary software.
Two other aspects of the way in which data is managed are important. System models “are abstractions that represent the entities, activities and/or concepts in the system as well as their attributes and characteristics, and the functional relationships between them.” As the Template puts it, “system models contrast with data format and file structure in that they represent behavioral, procedural, and/or functional aspects of a system or software application.” Because most software is in fact database management software, including document and records management software, systems models are often a valuable source of information for the purposes of appraisal about how a system behaves and how people use it. The final aspect related to data is system administration, “a set of procedures that ensure the correct, secure, reliable, and persistent operation of the system.” As we shall see, assessing the way in which these procedures affect authenticity is one of the steps in appraisal of authentic electronic records.
The Template provides the terms of analysis for gathering Information About the Technological Context of Records during the selection process. The relevance of the final input to the process, Information About Preservation Capabilities, is needed when considering the feasibility of preserving records selected for preservation, as we shall discuss in more detail when treating the activity of appraisal. To complete the picture of the process of selection, we need only consider the remaining two outputs or products.
4. Appraisal and authenticity
The Appraisal Task Force assumes that the selection process is the first step in long-term preservation. As such, from among all the information gathered as input to the process certain Information About Electronic Records Selected for Preservation needs to issue from the process in order to facilitate their acquisition and maintenance continuously in authentic form. Much of it will be a distillation of the information gathered about the records’ contexts, including vital information about the technological context necessary to maintain the data and the capability to present it as records to users. Some indication of the conceptual and methodological basis for selecting this information will become evident as we proceed to description of other activities in the process. The other output is Information About Appraisal Decision, that is, documentation of the reasoning to determine the valuation and feasibility of long-term preservation.
The selection function breaks down into three realms of activity (see Node: A0).
The entity responsible for preservation must establish, implement, and maintain a framework for the selection function, that is, Manage Selection Function. The other two activities are Appraise Electronic Records and Carry Out Disposition of Electronic Records. Managing the selection function means assessing information about the context of records and about past appraisal decisions and dispositions together with consideration of the needs served by preservation and legal and archival requirements in order to develop an appraisal strategy and disposition rules and procedures. In practice, a strategy for appraisal and disposition procedures will not be for electronic records alone, but any comprehensive framework will have to contain rules and conventions for dealing with the peculiar characteristics and problems associated with appraisal of electronic records. As previously mentioned, the work of the Strategies Task Force will be synthesizing the work of the other three task forces into recommendations about the policies, strategies, procedures, criteria and standards to be applied to the various processes of long-term preservation of authentic electronic records. Nevertheless, certain assertions about those policies, procedures, criteria, and standards will emerge in the rest of this discussion.
The central activity and main focus of the selection function is of course appraisal. These activities of appraisal are depicted on Node A2. They are conducted within the framework of rules and conventions laid down in Appraisal Strategies.
The appraiser must first Gather Information About Electronic Records, the result being a distillation of Relevant Information About Electronic Records, which is then analyzed when the time comes to Assess the Value of Electronic Records and Determine Feasibility of Preserving Authentic Electronic Records. The results to these two analyses then allow the appraiser to Make Appraisal Decision. As a result, Information About Appraisal Decision, that is, about both the valuation and the determination of feasibility, is recorded. Similarly, Information About Appraised
Electronic Records is a recording of the information necessary to effect disposition of the appraised records, both those destined for permanent preservation and those consigned to destruction or some other resolution of their fate. In the last case, we recognize that it may in certain cases be necessary to leave continuing preservation with the creating body against the day when the question of the ultimate destination of the records can be revisited.
The activity of assessing the value of electronic records is decomposed in Node A22. There are two aspects of assessment. One, called Assess Continuing Value of Electronic Records, encompasses all the analysis necessary to reach a determination of the capacity of the records to serve the continuing interests of their creator and society. It results in some written Assessment of Long-term Value. The second aspect, called Assess Authenticity of Electronic Records, encompasses all the analysis necessary to determine the degree to which electronic records in any given case were created in an environment meeting the requirements for authenticity.
As mentioned previously, the Authenticity Task Force is working on a statement of the conceptual requirements for authenticity. The Appraisal Task Force is guided by this work of its fellow Task Force in coming to a conclusion about the nature of the activity of assessing the authenticity of electronic records.
In wrestling with this difficult question, we recognize that part of the appraisal process has always been, whether explicitly or implicitly, an assessment of the grounds on which one can presume the records that are determined to have continuing value are in fact authentic records of their creator. Because of the volatility and vulnerability of electronic records, this process must be an explicit part of appraisal. To assist in this assessment, the Authenticity Task Force has developed a draft statement of “Requirements that Support the Presumption of Authenticity of Electronic Records.” An early version of the statement of requirements succinctly stated its main premises:
Establishing requirements for authenticity of electronic records over the long-term amounts to establishing requirements for the production of authentic electronic copies of authentic electronic records. The authenticity of electronic records must be verifiable from elements of the records (that is, either on their face or linked to them) and contextual to the records (that is, belonging to their juridical-administrative, provenancial, procedural, documentary, and technological context), while the authenticity of electronic copies of authentic electronic records is attested by the preserver, who has taken responsibility for the process of reproduction…. In other words, any electronic copy of an authentic electronic record is authentic if declared to be so by an officer entrusted with that function, namely the official preserver.
The first stage in the verification of authenticity is, we propose, performed during the appraisal process in order to establish the degree to which the preserver can presume selected records to be authentic. Normally, in assessing authenticity, we rely on that fact that the creator relied on them to carry out its affairs. Even so,
for electronic records on which the creating body no longer relies for its business, it is necessary to demonstrate the records’ identity and integrity. Such demonstration also characterizes the grounds on which future users may presume authenticity or the converse, that records may have been accidentally corrupted or purposefully altered.
Thus, we say that to assess the authenticity of electronic records during the process of appraisal means to determine the degree to which electronic records were created and maintained in an environment meeting the requirements for authenticity.
Only records whose identity can be established can be treated as being authentic.
Therefore, the appraiser needs to verify and document how the creator has made provision to express on the face of the record or linked to it as metadata the provenance, author, addressee, writer, originator, date, action or matter, and archival bond, status of transmission (i.e., whether the records are originals, drafts or copies), and attachments of the records. In order to determine the integrity of the record, it is also necessary to establish that the creator has made provision on the record or in metadata to identify the office responsible for handling the record, the office of primary responsibility for the record (if different from the handling office), annotations, and technical modifications.
The degree to which one can presume authenticity is also dependent on the degree to which the creating body institutes procedural controls as part of systems administration. Therefore, it is necessary for the appraiser to establish and document the degree to which the creating body has:
– implemented and monitored access privileges in the electronic system,
– set rules regarding which records must be authenticated, by whom, and the means of authentication,
– established procedures, where multiple copies exist, to identify the authoritative record and the form of authoritative copies,
– established procedures ensuring the regular transition of records from active to semi-active status, and the tracking of their location,
– established procedures to prevent, discover, and correct intentional or inadvertent loss or corruption of records, and
– established procedures to guarantee the continuing identity and integrity of records across technological change.
It is worth noting that authentication is a declaration of authenticity that occurs by inserting or adding an element to the certified copy that allows one to verify that a record is what it purports to be at that point in time. Obviously, then, authentication of a record at a point in time does not establish its authenticity over time.
The activity of verifying the identity and integrity of electronic records results in an Assessment of Authenticity documenting the grounds for presumption of authenticity. This documentation will be vital to the preserver and future users when assessing the trustworthiness of records. The assessment of authenticity and the assessment of long-term value are inputs to Determine the Value of Electronic Records, which is the analytical activity that produces Valuation Information.
Documentation of the grounds for presuming authenticity will be part of the valuation. It must be stressed at this juncture that, if the assessment of authenticity establishes that there are very weak grounds for presuming authenticity, it does not mean that the records are to be deemed of no continuing value. In fact we have often accepted such records for long-term preservation, and it will be no different in the electronic environment.
If, on the one hand, assessing the value of authentic electronic records means assessing the grounds for presuming authenticity, on the other hand, it means determining the feasibility of preserving the elements of the records that confer authenticity. The activity to Determine Feasibility of Preserving Authentic Electronic Records is depicted on Node A23. At this stage in our work, this diagram may be taken to be a proposition because it depends on the completion of the work of the Authenticity Task Force to specify the conceptual requirements for authenticity and the work of the Preservation Task Force on preservation requirements. We suppose that the first step is to Determine the Record Elements to Be Preserved. To some extent, they will be the elements of extrinsic and intrinsic form identified in the assessment of authenticity made during valuation. For instance, is there a special sign such as a logo or crest that should be preserved to identify the creating body? If so, it is then necessary to Identify Where the Record Elements are Manifested in the Digital Components. The result of this analysis will be specification of Record Components that need to be Preserved. The Preservation Task Force has devised a tentative definition of the term digital component. It is “a digital object that is part of an electronic record, or that contains one or more electronic records, and that has specific methods for preservation and reproduction.”
Thus, once one knows the record components to be preserved and the current and anticipated preservation capabilities of the preserver, one is able to Reconcile Preservation Requirements with Preservation Capabilities. The result is information for the preserver about the record components to be preserved and information as to the feasibility of preservation given the preserver’s existing and expected capabilities to preserve the various digital components that need to be preserved.
5. Conclusion
What can be concluded, in a preliminary way, from this account of the process of selection? In general, it is evident that the extensive analysis and documentation done during appraisal constitutes the first, vital phase in preservation, upon which
later phases will depend. It is clear that this analysis is most effectively conducted while the records exist in the system in which they were created, although that will surely impose a burden to track changes in systems and, if necessary, revise appraisals accordingly. It is also becoming clear that, at this stage in our knowledge, it is necessary to approach appraisal of electronic records case by case with the conceptual requirements for authenticity outlined here in mind. Efforts to develop a typology of records with specific authenticity requirements for each type of electronic records is not yet, if it ever will be, possible. Moreover, it is probably the case that certain types of information system (usually databases of one kind or another) used regularly to support bureaucratic work are not amenable to the forms of diplomatic analysis on which we have based our work. We have met such systems in the case studies we have done as part of the project. Our analysis suggests that they do not contain records as understood in diplomatics and archival science. Some separate discussion of appraisal of such systems will eventually be in order. At this stage, however, we can say this about records that exist in electronic systems. In the simplest terms, the appraisal phase establishes whether the records are what they purport to be, what elements of them need to be preserved, and how those elements can realistically be preserved in the electronic environment. It supplies later phases of preservation and future users with the necessary documentation to understand the context of creation, the grounds for presuming the records to be authentic, and the manner in which the elements conferring authenticity were expressed in the technological environment of creation. The preserver can then go about maintaining authentic electronic records across technological change, and future users will be able to establish the validity of inferences they draw from their reading of the records.
6. Model Arrow Definitions Appraisal Decision:
A determination of the disposition of a given body of electronic records (e.g., as embodied in a records schedule).
Appraisal Strategies:
The rules and conventions of the entity responsible for continuing preservation that govern the appraisal of electronic records.
Archival Science:
The concepts, principles, and methodologies governing the treatment of electronic records, including the concepts, principles, and methodologies defined by
diplomatics.
Assessment of Authenticity:
A record or records stating the reasons for presuming electronic records to be authentic in terms of the baseline requirements for authenticity.
Assessment of Continuing Value:
A record or records stating the reasons for continuing preservation of electronic records.
Authenticity Requirements:
The specification of the elements of form and context that need to be preserved in order to maintain the authenticity of a given type of electronic record.
Computer Equipment and Software:
Hardware and software to maintain and access electronic records.
Creator’s Needs:
The perceived interests of the creator served by continuing preservation of electronic records.
Disposition Rules & Procedures:
The rules governing the process of effecting disposition of electronic records selected for preservation and electronic records not selected for preservation.
Electronic Records:
A record that is created (made or received and set aside) in electronic form.
Electronic Records Not Selected for Preservation:
Electronic records identified for destruction or disposition to an entity other than the one responsible for continuing preservation.
Electronic Records Selected for Preservation:
Electronic records identified for transfer to the entity responsible for continuing preservation.
Facilities:
Places where records are stored.
Feasibility Information:
Information about the cost and technical capability required for continuing preservation of a given body of electronic records.
Information About Appraisal Decision:
A record or records detailing the valuation and feasibility of continued preservation of electronic records.
Information About Appraised Electronic Records:
A record detailing the information necessary to carry out disposition of electronic records.
Information About Disposition:
Information about the quantity and quality of records selected for preservation and records not selected for preservation, and about the cost of disposition of electronic records, utilized for management purposes.
Information About Electronic Records Selected for Preservation:
A record or records providing the necessary information about electronic records to maintain them continuously in authentic form, including the terms and conditions of transfer.
Information about Preservation Capabilities:
Information about the preserver’s current and anticipated capability to preserve electronic records, including the state of preservation knowledge, hardware/software capabilities, staff expertise, and financial resources.
Information About Record Components to be Preserved:
Information about the way in which the record elements to be preserved are manifested in the electronic environment, construed for the purposes of instructing preservation activities.
Information About Records’ Contexts:
Information about the juridical-administrative, provenancial, procedural and documentary contexts of the records.
Information About the Technological Context of Records:
Information about the hardware and software environment in which electronic records were created and currently exist.
Information From Electronic Records:
Information drawn from reading the form and content of electronic records.
Legal Requirements:
The concepts, principles, and specific statements in law relevant to the selection of records.
List of Record Elements to be Preserved:
A list of the extrinsic and intrinsic elements of form that need to be preserved to maintain the authenticity of electronic records.
Persons:
People who perform the selection function.
Digital Components that need to be preserved:
Description of the components in the electronic environment that manifest the record elements that need to be preserved to maintain authenticity.
Relevant Information About Electronic Records:
Information that is needed to appraise electronic records.
Societal Needs:
The perceived interests other than the creator’s served by continuing preservation of records.
Valuation Information:
Information about the criteria used to assess the value of electronic records and their application to the case at hand
.
7. References
[ACL 91] ACLAND, Glenda. 1991. “Archivist – Keeper, Undertaker or Auditor: the Challenge for Traditional Archival Theory and Practice.” In Keeping Data: Papers from a Workshop on Appraising Computer-based Records. Sydney: The Australian Council of Archives and the Australian Society of Archivists.
[BAI 89] BAILEY, Catherine. 1989-90. “Archival Theory and Electronic Records.” Archivaria 29 (Winter): 180-196.
[DOL 92] DOLLAR Charles. 1992. Archival Theory and Information Technologies: the Impact of Information Technologies on Archival Principles and Methods. Macerata, Italy:
University of Macerata.
[EAS 00] EASTWOOD, Terry, Shadrack Katu, Jacque Killawee, and Jeff Whyte. 2000.
“Appraisal of Electronic Records: A Review of the Literature in English.”
(http://www.interpares.org/documents/ERAppraisalLiteratureReview.pdf)
[HOF 94] HOFMAN, Hans. 1994. “Off the Beaten Track: the Archivist Exploring the Outback of Electronic Records.” In Playing for Keeps: the Proceedings of an Electronic Records Management Conference Hosted by the Australian Archives.
(http://www.naa.gov.au/govserv/techpub/keeps/hofman.htm.)
[HOR 97] HORSMAN, Peter. 1997. “Appraisal on Wooden Shoes: the Netherlands PIVOT Project.” Janus (1997.2): 35-41.
[ICA 97] International Council on Archives. Committee on Electronic Records. 1997. Guide for Managing Electronic Records From an Archival Perspective. Paris: ICA.
[McN 00a] MACNEIL, Heather. 2000a. “Providing Grounds for Trust: Developing Conceptual Requirements for the Long-Term Preservation of Authentic Electronic Records.”
Archivaria 50 (Fall): 52-78.
[McN 00b] MACNEIL, Heather. 2000b. Trusting Records: Legal, Historical, and Diplomatic Perspectives. Dordrecht: Kluwer Academic Publishers.
[O’Sh 91] O’SHEA, Greg. 1991. “The Medium is not the Message.” In Keeping Data: Papers from a Workshop on Appraising Computer-based Records. Sydney: The Australian Council of Archives and the Australian Society of Archivists.
[THI 00] THIBODEAU, Ken. 2000. “Certifying Authenticity of Electronic Records: Interim Report of the Chair of the Preservation Task Force to the InterPARES International Team.
Unpublished report dated April 19.
