What would be the main benefit in performing an archiving operation on a computer system?

Archival systems, frameworks and scalability

The term ‘archival system’ has most often been used to refer to specific systems that exist to manage and document records of continuing value under the control of an archival institution. This formative view of archival systems focuses on the processes needed to ensure that records are accessible and that their meaning is available over time. Systems of this type are discussed in this chapter as background to a more extensive discussion of recordkeeping and archiving processes associated with description, appraisal, preservation and access that are applicable within a broader archival framework.

From a records continuum perspective, the concept of an archival system also encompasses a framework-setting role, involving the administration of ownership, custody, access rights and responsibility for all records within a particular jurisdiction, including those of continuing value. The framework is set at a high level by legislation, regulations, standards, rules and policy, but the implementation of the framework occurs simultaneously at different points, at all operational layers within the framework encompassing each of the layers of the archives, archive, records and document, that is in all dimensions of the records continuum. A key difference between life-cycle and records-continuum approaches is that each stage of the life cycle is conceptualized as a one-dimensional space in which recordkeeping or archiving processes take place from the perspective of that stage alone. The records continuum defines recordkeeping and archiving as multidimensional. When operating in any one dimension of the continuum, all the other dimensions are present, although the particular focus may be creation, capture, organization or pluralization. From this viewpoint, collective archivists focus on pluralization, operating in one dimension of a multidimensional space, The theoretical underpinning of this view of recordkeeping spaces is explored in chapter 8.

Archival frameworks also operate at a social level The network of individual archival systems makes up a community of practice that operates to preserve an aspect of the social or collective memory of society. This use of the concept of archival system does not just refer to specific descriptive systems or regulatory regimes, but also includes the archives/records profession with all its complexity − incorporating professional training, resourcing and the health of the profession as a whole − and interactions with other institutions serving similar long-term social expectations of preserving collective memory.

Within an increasingly complex Internet-enabled world, the records continuum concept of an archival system can be seen as a system that operates to impose controls that need to apply beyond the boundaries of one agency. From this viewpoint, an archival system manages an aggregated layer of records existing outside the physical boundaries of a single agency, and encompasses multiple individual agencies, often from disparate layers of government, private organizations and individuals. Applying archival techniques to managing records arising from cross-agency services or ‘joined-up’ services has yet to receive much professional attention, yet within the records continuum model, such transagency controls locate the techniques for managing records at the pluralized (collective archives) layer of operation.

In addition to multiple meanings, the notion of archival system is scalable. Thus it is perfectly reasonable to discuss an archival system that is implemented within one agency, in Australia often referred to as ‘in-house archives’, of which BHP Billiton or many of the university archives would be examples. Such systems are as legitimately archival systems as those that manage the same set of concerns for the whole of government or multiple agencies within broader constructs. Such organizational archival systems regard subelements of their organizations as agencies within their archival systems and apply concepts developed to manage across their defined agencies.

Often understanding ‘archival systems’ is made more complex by the fact that multiple interpretations may coexist within one institution. Teasing out some of these strands does not diminish any of the roles, but might help explain why some of our overseas colleagues consider that Australian recordkeeping approaches ignore (or in extreme cases by imputation actively reject) the cultural role. Positioning an archival institution or program as regulator for agencies within its domain involves it in setting up an archival framework in the broad sense described above for application at agency levels. The archival institution or program is a cross-organizational body, but its rules are applicable at lower levels of aggregation. When archival institutions and programs collaborate with other institutions and programs dealing with collective memory, they are also operating in a pluralizing dimension.

The term archival systems thus can be used to encompass multiple meanings, some of which have been identified here. These interpretations are not necessarily contradictory or incompatible. However, it is necessary to ‘place’ or locate discussions of archival systems so as to identify what the scope and effect of strategies adopted can be.

Much of our professional practice, both in Australia and overseas, currently locates discussions about archival systems in the sense of the rules and frameworks established to regulate agencies, as well as the pluralizing processes in place in those domains to identify, manage, describe, appraise, preserve and make accessible records of continuing value, within specific jurisdictional or domain boundaries. How these notions of archival systems may play out in future in a post-modern world and across domain boundaries is explored further in the final section of this chapter. In these early sections we explore the evolution of such archival systems to this point in time, focusing on how associated processes have been reengineered and repurposed in Australian records continuum management frameworks, with particular reference to the Australian series system and the advisory and regulatory role played by government archival authorities in Australia.

Website history research

In Chapter 12, we discussed using the Google and Bing’s cache feature. This provides a snapshot of a website created the last time a search engine crawled the website. Again, it can provide clues to the investigator about any changes that might have been made to the website since the last search engine crawl. As the created cache is subject to replacement, it must be documented and preserved at the time it is collected. These website caches are only a short-term snapshot, which is replaced as soon as the search engine crawls the site again.

However, there exists a more long-term website archival system on the Internet. Since 1996, the Internet Achieve (http://archive.org) has been collecting and cataloging websites, which to date exceeds 240 billion webpages or almost 2 petabytes of data. It is currently growing at a rate of 20 terabytes per month (archive.org). It is a US 501(c)(3) nonprofit organization, which is supported by donations but collaborates with institutions such as the Library of Congress and the Smithsonian. The Internet Archive’s website, the “Wayback Machine”,1 has an easy-to-use interface to search for website information. The site provides the date and times of when the site has been crawled, as well as a capture of the site, so the investigator can see how the site has changed over time. These achieved webpages may provide the investigator with additional useful information. This could include ownership information in the archived “About Us” section that may have been deleted or later changed to prevent the current webpage from disclosing website ownership.

Just like any other webpage, the investigator can also look through the HTML source code of the achieved page to look for possible usable information. Investigators should be aware that the site does not crawl and record everything found on a website or webpage. It does not record every page if the Robot.txt file is set to tell search engines not to crawl the page. Additionally, certain Java code and other newer active content scripting are not collected. The Internet Actives FAQ page lists circumstances when the site does not collect information on a particular website or page. Regardless of some limitations, this is still a hugely valuable tool for the investigator to identify past website information (Figure 13.5).

Backup of the Internet Archive

The data stored through the Wayback Machine project also has a mirrored site in Alexandria, Egypt. Bibliotheca Alexandrina maintains the only copy and external backup of the Internet Archive. The Internet Archive at the Bibliotheca Alexandrina includes the web collections from 1996 through 2007. It represents about 1.5 petabytes of data stored on 880 computers. The entire collection is available for free access to researchers, historians, scholars, and the general public.

The Bibliotheca Alexandrina Internet Archive is the first center of its kind established outside US borders. It is designed not only as a backup for the mother archive in San Francisco, but also as a hub for Africa and the Middle East.

Authentication and the Internet Archive

The Internet Archive is a nonprofit organization and as such is not in the business of responding to requests for affidavits, or authenticating pages or other information from their Wayback Machine. Accordingly, they ask, prior to requesting authentication and an affidavit on the results, investigators to consider the following:

Seek judicial notice or simply ask your opposing party to stipulate to the document’s authenticity.

Get the person who posted the information on the URLs to confirm it is authenticate.

Or get the person who actually accessed the historical URL versions to confirm that they collected and it is an accurate copy of what was accessed. (This is what this text has been stressing: proper collection, preservation, and documentation of the process, is a must in authenticating online evidence.)

However, if investigators are determined to obtain an affidavit and authenticating printouts, they provide procedures for doing so on their website (http://archive.org/legal/). Fees are $250 per request plus $20 for each extended URLS, except those which contain downloadable/printable files. Such URLs (e.g., .pdf, .doc, or .txt) cost instead $30 per extended URL. Copies are not automatically notarized. If the investigator wants the affidavit notarized, there is an additional $100 fee.

The French Revolution and the nineteenth century

The French Revolution provides perhaps the clearest example of the mutable nature and purpose of archives and their tendency to inspire extremes in human emotion. Between 1789 and 1793, much of the archives of the ancien régime were attacked and destroyed by mobs or in state-sponsored bonfires and paper recycling campaigns, with the aim of obliterating what the revolutionaries regarded as symbols of their erstwhile oppression. While such actions might sometimes have had the practical benefit of destroying the evidence of feudal debts and obligations, by and large they were cathartic acts of retribution and ritual cleansing of the body politic.

In the midst of this destruction of old archives there co-existed a desire to create new archives, out of which emerged a new archival system for the new society. A legislative repository was provided for by the new Assembly just two weeks after the fall of the Bastille. In September 1790 a law was passed establishing a new National Archives that was to be open to the public and which was to report to the Assembly. By 1794 the desire to destroy the documentary evidence of the Ancien Régime had been replaced by a desire to preserve and manage those records as nationalized public property, reinvented for the purpose of symbolically highlighting the glory of the new Republic in contrast to the sinful decadence and oppression of the old regime. A decree issued in June 1794 granted the National Archives jurisdiction over the records of government agencies, provinces, communes, churches, universities and noble families, thus creating the world’s first centrally controlled national archival system. The same decree also proclaimed the right of public access to these records, thus establishing the first modern instance of archives fulfilling a legal role as protectors of the rights and entitlements of the people and as instruments of accountability and transparency in government The creation of national archives as both symbols of nation building in the midst of turbulent change and ideological − indeed almost mythological − assertions of legitimacy by new orders is a pattern that has been repeated often since. The fate of the archives of the Ancien Régime testify to the fact that no archives can assume an eternal mandate − in the words of Judith Panitch, they are forever ʻsubject to the judgement of the society in which they exist’.23

Another aspect of the impact of the French Revolution on archives is worth exploring at this point. Luciana Duranti has argued that the 1794 decree created for the first time a dichotomy between administrative and historical archives − the distinction between the archives of the Republic and the archives of the Ancien Régime. Duranti considers this an unfortunate development in that it represents a usurpation of the administrative and legal functions of archives by social and cultural functions − a usurpation that has echoes in various places and times since the Revolution.24 Other commentators, however, beg to differ. Judith Panitch, for instance, argues that in the 1790s the notion of French archives as sites of ʻhistorical or cultural scholarship had yet to take hold’. While they had acquired the new function of public access for the new purpose of accountability, their essential role as legal, administrative and symbolic institutions remained unaltered.25

Nevertheless, Duranti is correct in highlighting the distinction between the administrative/ legal and cultural/historical roles of archives − a source of contestation that shall be explored in more detail later − even if the cultural role of French archives did not become apparent until some decades after the Revolution. Duranti’s portrayal of one role as being innately superior to another is, however, a position that is far more difficult to sustain, as we shall see. Nor, as we have already seen, is it true that the world had to wait until the late eighteenth century to witness an example of an archives that was established for cultural and historical purposes. While such phenomena were indeed unusual, they were not unprecedented − see for example the case of the Han Dynasty Bureau of Historiography referred to above.

The creation of a centralized national archives in France provided a model for archival development in a number of other countries such as Finland, Norway, the Netherlands and Belgium during the nineteenth century. Similarly, in Sweden, Denmark and Prussia central archives evolved out of pre-existing royal or administrative repositories. Forty-eight years after the creation of the French national archives, the English followed suit, but for very different reasons and in much less dramatic circumstances. Between 1800 and 1837 a variety of committees and commissions of inquiry had highlighted the scattered and poorly controlled and preserved state of public records in that country. These efforts culminated in the passage of the Public Records Act in 1838 and the eventual establishment of the Public Record Office during the 1850s by a government that was concerned to ensure the proper care and preservation of records that guaranteed the legal rights and entitlements of English people. Lawmakers in Westminster were no doubt aware of the fact that their counterparts in Scotland had beaten not only themselves but also the French in establishing a national archives, when their principal collection of public records had been assembled in Edinburgh’s General Register House as early as 1784.

By the middle of the nineteenth century the growth in historical scholarship based on the use of written sources was becoming an important factor in the evolution of European archival institutions. Selected series of historical documents were published, such as the ʻRoll Series’ and the ʻCalendars of State Papers’ in England. In 1869 the Historical Manuscripts Commission was established in the United Kingdom to identify, describe and promote the preservation and use of significant historical records that were not otherwise catered for under the Public Records Act. The Commission, which existed until April 2003 when it was amalgamated with the Public Record Office to form a rebranded National Archives, is probably the best example of a state-sponsored documentation program for the nationally distributed holdings of historically significant private records.

Digital repositories

Digital repositories are information systems that ingest, store, manage, preserve, and provide access to digital content. The OAIS model provides a conceptual foundation for designing standard-compliant repositories. Digital repositories are a relatively new phenomenon that emerged in the early 2000s. A concept of trusted digital repositories has been advanced to ensure high-level preservation services for all types of repositories. There are several repositories types, including institutional, disciplinary, government, and centralized repositories, which aggregate content from several subsidiary repositories. According to the Directory of Open Access Repositories, most of the content in open access (83.2%) is available through institutional repositories (OpenDOAR, 2015). As Lynch (2003) observes, institutional repositories offer an essential infrastructure for scholarship in the digital age and a potential to revolutionize scholarly communication. He also notes that a key part of the service is to manage technological change and the migration of digital content from one set of technologies to the next.

Institutional digital repositories serve multiple purposes. Their primary goal is to support scholarly communication and provide open access to articles, dissertations, and research data. In addition, they provide platforms for storing and preserving the digital master files created as a result of digitization projects. The boundaries between a repository and digital libraries are sometimes blurred, as repositories also host digital collections for access. The combination of access and preservation functions poses significant challenges. McGovern and McKay (2008) investigated the juncture of institutional repository implementation and digital preservation programs and provided a set of recommendations for leveraging the benefits of institutional repositories to strengthen long-term preservation. A number of research studies examine the current practices of digital preservation in the institutional repository environment (Banach and Li, 2011; Kunda and Anderson-Wilk, 2011; Neatrour et al., 2014; Oehlerts and Liu, 2013).

Trusted digital repositories perform preservation functions. This notion was first introduced in the seminal report, Preserving Digital Information (Waters and Garrett, 1996). The authors emphasize the role of trust in managing the identity, integrity, and quality of digital information in archival systems and recommend developing a process of certification. The concept of a trusted digital repository was fully articulated in another foundational report, Trusted Digital Repositories: Attributes and Responsibilities, prepared by a RLG/OCLC working group (Beagrie et al., 2002). A trusted digital repository is defined as one “whose mission is to provide reliable, long-term access to managed digital resources to its designated community, now and in the future” (Beagrie et al., 2002, p. 5). In order to gain recognition as “trusted,” a repository has to have certain attributes that ensure the reliability and authenticity of stored information. The RLG/OCLC group outlines the following characteristics of sustainable digital repositories:

Accept responsibility for the long-term maintenance of digital resources on behalf of its depositors and for the benefit of current and future users

Have an organizational system that supports not only long-term viability of the repository but also the digital information for which it has responsibility

Demonstrate fiscal responsibility and sustainability

Design its system(s) in accordance with commonly accepted conventions and standards to ensure the ongoing management, access, and security of materials deposited within it

Establish methodologies for system evaluation that meet community expectations of trustworthiness

Be depended upon to carry out its long-term responsibilities to depositors and users openly and explicitly

Have policies, practices, and performance that can be audited and measured (Beagrie et al., 2002, p. 5)

In addition, the RLG/OCL report discusses methods and strategies for the certification of trusted digital repositories. A regular cycle of certification and audit is recommended for digital repositories to remain trustworthy. The process of certification has gained considerable attention in the last decade, and a number of standards and checklists have emerged, such as TRAC (The Trustworthy Repositories Audit & Certification Checklist), superseded by the ISO 16363:2012—Audit and Certification of Trustworthy Digital Repositories international standard. A range of tools have been developed in Europe, including nestor, DRAMBORA, Platter, and Data Seal of Approval. The recent publications on digital preservation provide an overview of these tools (Brown, 2013; Corrado and Moulaison, 2014).

Securing Documentation

If documents relating to the target network architecture fell into the hands of malicious hackers, the customer would be at risk—if identified vulnerabilities and exploits were included in the compromised documents, the customer may be severely impacted, depending on the sensitivity of the data.

Any documentation and penetration test data that we collect and store needs to have the appropriate protection. We can either encrypt the data itself or encrypt the system the data reside on. If we want to encrypt the data, we could select either password encryption or certificate encryption. The other alternative is to encrypt the system that stores the data using full-disk encryption, which can also use both certificates and passwords to secure data at rest. The advantage of encrypting the system that stores the data is that once a user has validated himself or herself to the system, all documents stored on the data can be viewed without the need of additional passwords (assuming the files themselves do not have additional encryption mechanisms in place). Another advantage of full-disk encryption is that passwords can be easily changed, according to password policies. Changing passwords on large quantities of individually encrypted documents can be an enormous undertaking, especially if no change-control management process exists.

Trust and Technology

Indigenous Australians’ lives began to change nearly 220 years ago with the establishment of the first of six British colonies in Sydney in 1788. The contrast between the two cultures could not have been stronger. Whole clans and nations were dispersed, killed, moved to reserves and made the objects of misguided assimilation, conversion and welfare including the separation of children from their families. While attempts have now been made at reconciliation and fostering self-determination, one legacy of dispossession, exploitation and paternalism was, and remains, a distrust of white institutions, including religious, legal and government institutions. This inevitably extended to archival institutions and to the content of the files they and government agencies held.

The Trust and Technology project,4 begun in 2004 at Monash University, Melbourne with Australian Research Council funding, aimed:

to develop a way of preserving through recording orally transmitted indigenous knowledge, and controlling access to it; and

to develop specifications for an annotation system which would allow today’s Indigenous Australians to ‘answer back’, that is to respond to institutional archives by creating websites of collections comprising comments, digitised documents and even lists of other relevant collections.

Needless to say the research and development is entirely predicated on consultation, the project partners including a strong representation of indigenous organisations and others.5 The project is half completed and is currently assessing results and planning the scope and funding to develop the specification for the second objective. Already, however, several features are noteworthy from an appraisal viewpoint.

The first is the direct involvement of communities in the formation of collections and control of access to them. The system contemplated by the Trust and Technology project involves Indigenous Australian communities lodging their own records in their own virtual archives, the appraisal choices being shaped by the contested records about them in government archives and to which they claim, morally, a shared ownership.

The second feature concerns the question: what is to be appraised? Traditional appraisal assumes records exist, the primary aim being to determine their length of retention. By contrast, many Australian archivists assume that business functions and activities exist, the primary aim being to determine which should be evidenced through records captured in recordkeeping systems.6 Now the Monash project is pushing the threshold question even further back by assuming indigenous memory exists embedded in word, song, music, dance and ritual. Starting with storytelling, it is saying in effect that one aim of appraisal is to provide the means for indigenous people themselves to select and appropriately preserve them as records.

Risk Choices

Once we have assessed risk, we must decide what to do. Options include accepting the risk, mitigating or eliminating the risk, transferring the risk, and avoiding the risk.

Metadata

The above case study is in this chapter because nineteenth-century British and Northern European systems for the registration, aggregation, classification and tracking of documents (and even earlier systems in parts of Europe) were built around recordkeeping schemes that have many similarities with emerging recordkeeping metadata approaches. In such systems the document can be validly considered in isolation, although it does not exist independently. It has a set of information that describes it – the document-level application of metadata schemes discussed in the previous chapter.

Sometimes this document-level metadata is ‘point-of-capture’ metadata. Point-of-capture metadata in modern approaches establishes the initial picture of the document by making a record of that document (memorializing it). Links are made between the record, the creator and the business context that created it. The metadata typically describes such things as the creator of the document, the time or date of its creation or receipt, the business transaction or activity it forms part of, often expressed within a classification code. Various specifications for records metadata exist and all start with this point-of-capture level of information.

But for a document to be considered as an authentic, reliable and usable record (see Definitions 1, 2 and 3 above), it needs to be connected with information which describes in richer detail the tapestry of actions it has taken part in, such as what has happened to it, how it has been used, who did what and in relation to what business activities. We need to be able to ask (and often answer) a host of questions, including why a record was formed in the course of doing something (in a business activity sense), and whether the person forming it was authorized and able to undertake the action at the time the action took place.

Being able to link a document with attendant details of what is going on is a major part of the transformation of a document into a record. This type of metadata can be distinguished from point-of-capture metadata by calling it process metadata, although, of course, in recordkeeping they are brought together within the one scheme, a consistent pattern for the application of both types of data elements.

These process-based layers of information about the document make them records by providing continuing contextual links, as we saw in the nineteenth-century example above. In effect, what the record process metadata does is to connect the document to a particular set of actions on an ongoing basis, establishing relationships with other documents. This involves embedding and disembedding actions as the document moves into different contexts. By the use of process metadata documenting and describing the actions taking place a document (or any object about which metadata is being kept) is disembedded from its immediate contexts and can be re-embedded (captured) into new contexts within files, dossiers, series, the archive of an organization and in archives.

The establishment of connections to actions and relationships with other documents are what is referred to in the phrase ‘capture of records’ in this part of the book. The capture of records is the process essential to bringing objects into a recordkeeping system of control. Once captured (also referred to in other traditions also as being ‘set aside’11), ongoing systemic recordkeeping processes can be invoked to ensure the characteristics of records are maintained.

Depending on the formality of the records system, capture can be as simple as the deliberate assignment of a record to a folder (either physically or virtually on a computer). At this most simple of capture levels, the deliberate act is of significance – it indicates intent to treat the record as a record. Evidence of intent can be as informal as assembling emails in folders, putting children’s drawings, school reports, or other documentation together in one place over time, or physically putting papers into a labeled manila folder thus creating a proximity association. In simple systems such as these there is no guarantee that necessary recordkeeping metadata will be available. There is still a metadata scheme present involving such consistently applied data elements as file or folder name. It is, however, a weak form of capture because the bonding of the documents within the folder is easily disrupted by altering the contents and their order within the receptacle without any maintenance of data recording this fact.

To ensure more formal capture and the possibility of a fuller range of recordkeeping processes that make the links between items more or less indelible, registration usually occurs as described above, creating a unique identifier which provides a fixed reference for connecting both point-of-capture and process metadata to the registered object. The registered object itself varies. It might be a draft document, a document that has been created, a document that has been received, a file, or, in archival systems, a registered series of files or other objects. In systems that use registration processes, no further recordkeeping actions can be taken on the object until it is appropriately registered, and registration can only take place once for each object. While the contents of the object can be altered or changed, the metadata can record this along with links to other registered objects of which it is part. Thus a document, as a draft, as a sent document, as part of a file, or as part of a series of records, could involve a number of unique, but linked, registration numbers depending on how the system is configured.12

The establishment of initial point-of-capture and process metadata with registration identifiers that can connect data to an object is, then, a beginning, not an end, to metadata application processes. The application should also be directed at the carriage of documents through time and not be a once only occurrence. The record-making metadata is perpetually accumulating around an object.

Multiple threads of action take place and each one provides a potentially different way of reading the record. Using the example introduced in the first chapter of the photograph of children in the water, central to the Children Overboard case, the photo, when located within the context of HMAS Adelaide’s immediate transmission, is evidence of the bravery of individual serving personnel. If placed in the photograph album of an individual sailor, it becomes a personal record of the event. If the same photo, now without the metadata containing the captions contextualizing it, is a record in the Minister’s office, it tells a very different story purporting to prove that asylum seekers were throwing their children overboard. In the published versions, the photo becomes part of the collective memory through dissemination of the official ‘untruth’. It’s the same photo, but the contextual metadata provided about the photo − either explicit or provided by location – enables us to tell many different stories about it, each one occurring at different points in time, each one providing its own sequential yet recursive reading of events.

It is not enough, then, for a record to be an object created in the course of doing something, a characteristic of records identified in the introduction of this chapter. The capturing process needs to take the object, which in a recordkeeping system can be any object, even a desk in an inventorying system, as we saw in the introduction, more or less indelibly linking it to the actions and other objects that surround it. In the process of managing records something new is likely to be formed involving new registration processes. The draft registered document can become the registered document, then part of the registered file, then part of a series of files, then part of the company’s archive, and then part of archives elsewhere. In the introduction to this chapter it was pointed out that records are far more than simply physical things. They are the physical thing bound to information which provides all the contextual detail surrounding their creation and ongoing management. In technical terms the record is the object and its associated metadata and the scheme for recording information on all the contextual detail is known as recordkeeping metadata.

If this logic is applied, the distinction between data, documents, information and records becomes easier to explain. Data, documents and information can be created, received and maintained in the transaction of business. But they may not be managed as evidence of actions in the same formative way that those same elements would be if they were managed as records – they may not record every change to the data with the time, date and name of the person making the change, and they may not record all instances of access to the data. That does not mean that some evidence of action elements cannot be reconstructed if required using for example audit logs, but where the possibility of reconstruction is not integral to the management of the information, document or data, they are not being managed as records.

In relation to data, documents and information, the process of making something a record (establishing its evidential nature) by retrospectively tracing all these things is often undertaken where they are required to prove something. Retrospectively identifying all these elements by tracing what has happened to data or a document is, however, a very expensive exercise. What is preferable, if analyses reveal that the need may well arise to use documents or data as evidence, is to enable these processes to be put in place up front, to identify which resources need this layer of protection to enable them to act as evidence of action, and to manage them in this ‘record’ way from the time of their creation.

Part of the complexity of formative records management is that records can reach and cross, or not cross, thresholds of formality at different times. This is particularly observable in personal papers, where the formalizing layer to create systems to facilitate contemporary systems may be completely lacking, with records rather moving directly to a higher level of aggregation as the whole of the records of an individual or family. At a later date the degree of formalization that has been established might disintegrate as the person moves house or dies and relatives split the records up amongst themselves. This does not make personal papers any less records, but it means that the logical elements of contextualization are often hard or impossible to determine. Nor does this make them different in type from government records, as the U.S Department of Defense example earlier illustrates, and the issues of relocation and inheritance are common phenomena in all circumstances where records exist. Electronic recordkeeping approaches in fact may be expected to, in time, minimize this difference when personal records systems, modeled on specifications of recordkeeping requirements, become available.13

Recordkeeping requirement analyses allow flexibility in finding ways to implement appropriate recordkeeping. Using appropriate specification of the structural elements of what is required and the stringing of one or multiple technologies to deliver the functionality frees us from the physical view of records. It also allows degrees of flexibility to be introduced to suit particular organizational, industry or individual needs. In keeping with these ideas, different ways of defining and articulating recordkeeping metadata are being articulated. These are being expressed in terms of recordkeeping metadata standards, and are structured in XML schema. By expressing the dynamism of records and recordkeeping processes in ways that are accessible to technologists and systems designers, the prospect of integration of recordkeeping into business systems of all types and into vital areas such as electronic transactions and service delivery becomes more realistic. But not every recordkeeping metadata standard takes the process-oriented approach; many often stop at defining metadata for the record at point of capture. The ability of recordkeepers to understand requirements and to rethink how to apply records capture and management in diverse and ever-changing technologies will shape the record of the future.