Record, pre-record, non-record?

Primitive peoples often believe that it is possible to control things, persons, or nature simply by calling their names and reciting an incantation. Relics of this way of thinking are doubtless only slightly disguised in our collective consciousness. Acronyms furnish a good example. There is something bewitchingly official about acronyms. Ever since the Romans coined SPQR--and probably before--the mere appearance of these combinations of bold capital letters has magically legitimized both public and private practices.

Perhaps because they have devoted their professional lives to the handling of words recorded on paper (and other media), records managers have not been slow to seek just the right words (preferably reducible to a catchy acronym) to express how they wished to be identified. Notably, in an effort to ensure a place for ourselves in the burgeoning information society, we records managers have searched over the past two decades to discover just the right acronym to associate ourselves with "information," as opposed to mere records. Absorbed in this search and jealous of the vogue of MIS (one of the most successful and most equivocal of twentieth-century acronyms), we have striven to find magical combinations of letters that would convince our companies--and ourselves--that we are information managers also.

One of the most promising acronyms has been "IRM," which has been used to convey some important statements about the relationship between records and information. However, it does suffer from ambiguity, since some use it to mean "Information Resources Management" while others, to mean "Information Risk Management." Far more popular has been "RIM," which is usually translated as "records and information management" and has become almost the official acronym of records management. Exactly what it means is another question. Are not "records," as by nearly all records management texts and legal documents state, already included in "information"? If so, how do we explain the phrase "records and information management," which would seem to imply that they have two separate, if related, identities. However, analyzing such magical acronyms by mundane rules of grammatical logic is futile. The real question is what message are we trying to convey. This is not really very mysterious, for clearly records managers use RIM to assert that records management has a unique contribution to make to the management not only of paper records, but also of non-paper records and even of information that has not been recorded. "RIM" is not so much a description of an existing discipline as an ambitious future agenda for that discipline. This interpretation is substantiated by the inclusion of "information" in many records management titles, such as Maedke, Robek and Brown's Information and Records Management, first published in 1974.

It is not difficult to understand why records managers would be planning this agenda. After all, if we are eventually moving towards a paperless office, how important can an inventory of paper records be? Is the life cycle of paper records really that important in such a situation? How significant are the justifications of our departments that rely upon the space and labor savings resulting from the inactivation of paper records? Thus, feeling threatened by the replacement of paper records by electronic "information systems," records managers have sought to protect their positions by associating themselves with "information." This desire reflects not only the defense of corporate turf and economic existence (though surely this is important), but also a yearning to be associated with the "intelligent content" (i.e., the "information") rather than just the "irrelevant media" that carries this content. Of course, there are currently some media that seem exciting, but paper is not one of them. Unfortunately for records management, the "records" associated in everyone's mind with records management are paper, and paper has not been glamorous since it supplanted clay tablets and papyrus. Even we professional paper managers eagerly escape pawing through the dusty contents of archive boxes to the excitement of E-Mail, window icons, and spreadsheets that convert automatically into colorful graphics.

It is my contention that our attempt to assert our qualifications and right to be involved with non-paper records and even with non-record "information" is understandable and at least partially justified. However, I do not believe that we can control people or things simply by calling their names or reciting a spell. Similarly, to think that just by adding the word "information" to our title or combining records and information in a clever acronym is sufficient to secure our place in the "information society" is foolish. The ranks of "information workers" and "information experts" are already crowded. For records managers to justify for themselves a role in information management, they must show the unique contributions that their training and experience in records management can make to the management of non-paper information.

Actually, this demonstration will not be as difficult as at first it might appear. In our haste to jump on the information bandwagon and emphasize the non-paper, MIS aspects of our functions, we have sometimes lost sight of the key records management concepts that have been closely associated with the management of paper records. Old-fashioned or not, these concepts still have considerable validity, not just because so much information is still in a paper format, but also because they can be applied profitably to non-paper information.

APPLYING RECORDS MANAGEMENT PRINCIPLES TO NON-PAPER RECORDS

The obvious place to begin in expanding the purview of records managers beyond moldy old boxes of inactive business paper is to point out the benefits of applying records management principles to records in other media. For example, in dealing with paper records, it has been easy to prove that the presence of inactive records in active files wastes valuable office space. Records managers have further learned that even more harm has resulted from the loss in efficiency in having to search through files filled with irrelevant records in order to retrieve a few potentially pertinent ones. When the focus is on non-paper records, the argument is still valid. Storing obsolete data in computer files is wasteful; however, the cost of buying a few hundred gigabytes more of electronic storage is insignificant when compared with the computer time required to search for relevant data in a huge data bank consisting primarily of junk.

All of us know of the contributions records managers have made to the management of non-paper records. For example, in the PC environment, I remember that I first learned from William Benedon, CRM, one of records management's giants, about the development of automatic inactivation routines for micro-computers, while my MIS friends were still wringing their hands over the problem. An example in the mainframe realm involved the IBM 3280 which was responsible for Los Angeles transit operations and which was bogging down under the weight of data that had never been archived. I was impressed that the problem was resolved by the leadership and expertise of the records manager, Judith Butler, rather than the computer programmers and systems analysts.

The application of records management principles to networked PCs and telecommunications is a more recent challenge, one that most organizations still face. Especially promising in this area is the application of the records management's record-life-cycle theory to such media as electronic mail, fax junk mail, and multiple copies produced on the LAN. All of these display the classic defect of mixing a few pieces of valuable information with a preponderance of useless data. To resolve the problem, we must have an approach to network records management that goes beyond just an automatic backup of everything on the LAN or an automatic purge of everything earlier than a certain date. Much more profitable is records management's admonition to consider the use and disposition of records from the context of their total life cycle so that a record's disposition is already considered when it is created. Without such analysis it is far more difficult to purge the communications channels of the irrelevant and insignificant while preserving what is still needed.

Such an analysis requires the determination of which media to use for which application and of how many copies should be preserved. These have always been critical questions. However, recently they have become more acute as the result of the development of inter-changeability, or the quick and easy transfer-primarily by digital technology --of information from one media to nearly any other. Thus, we can microfilm paper, scan the microfilm into magnetic storage, transfer the information to optical disk, output it onto microfiche, and print it back on paper. Audio signals, as well as graphic and photographic images, can be similarly converted. Finally, we can use our computer to manipulate all of these.

Records management theory may have little to offer regarding the technology of the interchanges themselves, but it does provide the best tools for determining which of these interchanges make good business sense. Along with their sister discipline, archives management, records managers have studied the relative permanence of the different media and which media are best suited for various purposes. Records management writings are filled with comparisons regarding the portability, legibility, permanence, storage density, and communicability of various media. Even more importantly, records management, with its experience in developing media-specific records retention schedules, is familiar with tracking the different media on which the same data is copied.

APPLYING: RECORDS MANAGEMENT PRINCIPLES TO PRE-RECORD INFORMATION

There are many ways of classifying information. For example, Mary Robek et al, Information and Records Management, pp. 4-6, has analyzed information as historical and future; action and non-action; and recurring and nonrecurring. They have also analyzed it into "record" (or "documentary") versus "non-record" information. For our current discussion, this latter distinction is the most important. While in terms of volume, the largest component of "non-records" is the plethora of duplicates, a more significant part is composed of "pre-records," i.e., information that has not yet attained the "record" stage. As we have seen above, records managers can already claim the "record" type of information--no matter what the specific media--as their turf. Moreover, no one is really interested in contesting their authority over surplus copies. It is the pre-record type of information for which we records managers must prove the merit of our influence, for this has traditionally been the sanctuary of MIS computer gurus. It is in this arena--not just in that of non-paper records--that we must concentrate if we are serious about being "information managers" as implied by our RIM acronym and by the addition of "information" to the titles of many of our publications.

What is this pre-record information that we find so important, and why do we care about it? Records management theorists have developed a model of the "record life cycle," which includes creation, use, and disposition stages. Pre-record information may be viewed as part of the creation stage, but it has a totally different character from the other stages. Without wishing to carry the biological analogy to extremes, we can say that the pre-recorded stage includes both the conception and pre-natal development of the record. However, unless we believe in reincarnation, we must note that information differs from this biological model in one important respect: The same documentary information may evolve through a multiplicity of pre-record and record stages which are usually called "revisions."

Pre-record information is fluid, as opposed to the "snapshot" character of records. The preliminary computer sketch sent via electronic mail, the word-processor draft, the initial spreadsheet, the handwritten outline of a paper to be fleshed out are all examples of this pre-natal developmental stage. This is the time when we distribute a draft, receive comments, and make major modifications. To encourage creativity, it must be an unrestricted, "brainstorming stage," in which freely conceived ideas can be bounced around without worry about how the idea might appear in a newspaper or a court. This is the creative period in which business interactively develops its dreams and the strategy for realizing them using whatever informational media is available and without concerning itself with record storage or disposition.

We must refrain from distinguishing these pre-records from records by explaining tautologically that the latter have been "recorded." In fact, most pre-record information has literally been "recorded." Nor can we profitably distinguish between the media on which pre-record and record are respectively stored. While the pre-record may be only an idea in my brain or part of a conversation, it may also be a piece of paper, a file in my PC, or a dictation tape. The difference between the record and non-record stages of information is not determined by the media, but by the tentative or hypothetical character of the pre-record. When we use the term "record," we have in mind a piece of information that has--or at some time in the past had--a defined legitimacy, that was "in force." When we think of an "official record" or a "record copy," we come closer to what we mean by the word "record" than merely information that has been recorded on some media. For example, if I jot down a rough idea on the back of an envelope, it is a mere pre-record; when Abraham Lincoln wrote the Gettysburg Address on the back of an envelope, it was already a record. To summarize, it is neither the fact of being recorded nor the media in which it is recorded, but rather the degree of development of the informational content, that differentiates the pre-record from the record stages in the life cycle of information.

So what has records management to offer to business information in the pre-record period? A great deal. For it is essential in answering the critical questions of (1) when does pre-record information become a record, and (2) how can we tell the difference between these two states of information. These questions have become increasingly difficult as a result of the ease of transferring between the two states and among media through a common digital interchange, as we have already discussed above. At the same time, this distinction is extremely important for several reasons:

1. Business lacks resources to store, manage, and control pre-records that are often just preliminary notes and drafts that have been superseded.

2. For the effective handling of litigation support, providing information for public relations, and responding to external requests for information, it is often very useful to distinguish informal notes and preliminary drafts from official records. To preserve such pre-records is often unwise not only from the standpoint of expensive storage space, but also because of the damage they can cause to an organization's legal position, public image, and competitive standing.

3. Obviously, it is crucial to know if a given document is just an "interesting suggestion," a current procedure, or a superseded method. We must identify the current "record copy" and have this clearly distinguished from suggested and unapproved procedural or design revisions. This has long been a stated requirement in engineering projects. The ISO 9000 quality assurance standard has recently expanded this requirement to quality control and other sections of our organizations.

Records management possesses interrelated tools to enable an organization to distinguish pre-record from record and to track the development of one into the other. These tools include (1) the concept of "record copy"; (2) the configuration management concepts of baseline and historical versions (or audit trails); (3) the techniques of records inventory and appraisal; and (4) the information life cycle.

THE RECORD COPY

The concept of the record copy is one of records management's most rudimentary, yet most powerful, concepts. It states simply that only one copy of a record--traditionally the original--will be considered "the official record." It is this unpretentious concept that enables us to destroy multitudes of useless copies by assigning a retention period to them of "while useful." Without it, we would find ourselves having to keep surplus copies for the entire retention period indicated on the retention schedule for the record copy. It is also this concept, when expanded by linking the record copy to a custodial office, that enables us to assign responsibility for maintenance and keep clear in our minds whether or not a certain departmental copy is the official copy. Given the popularity of the strange --not to say irrational--tendency to house multiple "executed originals" within the same organization, the record copy is critical to avoiding endless confusion. Without it, we would end up throwing away the last copy of a document because we assumed "someone else had a copy," and we would keep needless duplicates of others fearing that no one else had any.

Clearly the record copy is a key records management idea, but does it have any application to pre-record information? Actually, the principle is probably even more important for information in the pre-record stage than after it becomes a record which can be easily tracked. For distinguishing between a record and its multiple clones involves only the naming of one as the original. In the pre-information stage, we must distinguish not only between the original and its exact duplicates, but also between different versions of it. To envision this, we need think only about what happens when we circulate a draft to many readers for "comments." Without an effective method for distinguishing both copies and versions, it becomes impossible to keep track of which comments were supplied by different reviewers, which revision includes which comments, and what the original draft looked like before revision. I know that in my own personal project files, I often (unwisely) keep multiple copies of documents and pieces of documents that become almost impossible to sort out and distinguish. Modern word processing, which makes it possible to electronically cut and paste pieces of verbiage and graphics from one document to another, can make the tracing of "what came from where" a detective story that would challenge Colombo, Jessica, and even Agatha.

PROJECT RECORDS: THE BASELINE

To apply the record copy concept to pre-record information, it is necessary to place it in the framework of time, as well as space. In other words, we must identify the current version among many preliminary versions of the same basic document, as well as the master among many clones. Actually this dynamic variant of "record copy" has already been developed by a sister discipline --if not an offshoot--of records management. This discipline is called configuration management, which I discussed in a previous column. Configuration management's concept of "baseline," defined as "that set of approved documents which describes the current configuration of the project," is really just an elegant term for "current record copy."

To control this "current record copy" of an evolving document, configuration management has developed strict controls which include elaborate change authorization procedures, identification of all revisions, controlled distribution, and an audit trail of all changes. The key to the success of these procedures is the maintenance of two separate files of the same documents: first, an historical file (or audit trail) of record copies of each revision, and second, the baseline. The latter integrates old pages, which were never revised, with new pages, which may have been revised many times, into a unified current version. These procedures eliminate any question as to which is the current version; enable all concerned to have current documents available at all times; ensure that no changes are made to the project baseline without proper review and approval; and provide an audit trail of snapshot documents of all previous versions of the baseline. With the addition of configuration management's baseline and historical audit trail, we are able to distinguish not only between record copy and clones, but also between current record copy and earlier versions or preliminary revisions.

In general, the principles of configuration management have been limited to engineering projects--especially those in the aerospace and nuclear energy industries. However, as modern business in every field becomes more complex and more dependent upon electronically manipulated information, the advantage of broadening the application of these principles becomes obvious. The concept of configuration management's discrete baseline and historical files may be used for any business project or process documentation that retains its identity through multiple revisions. For example, when applied to contracts, we can understand the distinction between the currently active contract file on the one hand and, on the other, the sequence of request for proposal, contract, and amendment, each of which at different times represents the "contract baseline." Similarly, we have the methodological framework for maintaining procedures, policies, project records, and all evolving business project documentation. Of special importance in recent months is the application of this approach to the maintenance of quality assurance documentation. The ISO 9000 standard for quality assurance systems requires the documentation of a system that maintains the current version of quality control documents while securing an audit trail of all previous versions. ISO 9001.4.5 and 9004.8.6-8.8 make it very clear that a configuration management of the quality control system's current "baseline," as well as all historical versions of this documentation, is what the International Organization for Standardization has demanded.

TRANSACTIONAL INFORMATION

Of course, the majority of business information is not--or is only incidentally--related to a particular project or to a single documentation opus, such as an organization's procedures manual. Most business information is what I term "transactional." It consists of the information required by a myriad of transient, if related, processes. Probably the most familiar is the minor-purchase procurement cycle that includes requisitions, purchase orders, invoices, payments, and perhaps follow-up correspondence if there is a problem. Similar are the letters of inquiry, complaint, and occasionally even praise. Information regarding taxes, individual personnel, and administrative services also fits into this category.

The problem in handling transaction records has always been the volume involved. In response, records management developed useful, relevant tools. While configuration management was designed to handle large-scale, individual projects, the tools of classic records management were fashioned for dealing efficiently with voluminous transactional records in broad document classes; e.g., "accounts payable" or "procurement" records. These classic tools included first an inventory of an organization's records, which determined exactly which records existed, where and how long they were maintained, and what functions they fulfilled. Based upon this inventory, the records management staff would work with the concerned departments to appraise the records' operational, fiscal, and legal values, as well as to determine retention periods based upon those values. Using these tools, it was fairly easy to identify the official record copy, duplicates, and pre-record versions, as well as to schedule their retention. Such was the bread and butter of classic records management in the era of paper records.

Now all of that is rapidly changing. For transaction information is being transformed from discrete paper documents into electronic information flows. In some cases, as for example with some electronic fund transfers, information is never in any but an electronic form. In other cases, paper records are scanned into electronic files and destroyed as soon as they are received into an office. In either case, a document may be revised and sent to multiple locations either simultaneously or sequentially. In each of these locations, it may be acted upon, and the difficulty of identifying the "record copy" becomes potentially great. Similarly, when transactions involve faxes, voice mail messages, electronic transfers, and simple phone orders, determining the current status of the transactions, not to mention the audit trail of what has occurred before, is quite challenging. In such situations, the lines separating pre-record from record blur, and the "most current copy" seems as mythical as the Holy Grail.

Thus, transaction documentation in an electronic environment ceases to be a collection of discrete, if uniform, pieces to be filed for an approved period. Rather it becomes a flow of electronic information of both the record and non-record varieties. Consequently, what is needed is a mechanism similar to configuration management for identifying the historical versions and current baseline of each transaction. Yet we have seen that such information does not form part of the documentation of just a single continuing project. Rather transactions are so voluminous that a full-blown configuration management set of controls for each of them cannot be economically justified. Instead, we must somehow modify the classic records management tools that were designed to deal with voluminous transactional information in groups. However, we would be foolish not to incorporate somehow configuration management's idea to control evolving project records by revising the stationary "record copy" into a dynamic "baseline" and by developing an audit trail of past versions.

To bring all of these ingredients together for transactional records means we must modify the records inventory and appraisal processes to operate in a dynamic environment. In particular, it seems appropriate to transform the records management inventory of static records located in permanent departmental locations into an analysis of the information workflow between departments. Such an analysis must track the distinction between pre-record and record information, as well as the identity of record copies, through multiple changes. Equally important will be identifying and determining the retention requirements for the various versions of these evolving records.

We must remember that, unlike their configuration management cousins, records managers must deal with transaction records as classes. They cannot afford to focus on individual documents. Thus, although what the records managers of the future will produce may well look more like workflow diagrams than records inventories, the principles underlying these work products will be rooted in the classifications developed in records management inventory and appraisal activities. To be sure software tools that help achieve these objectives are included in many electronic workflow packages. However, the choice of tools and the evaluation of whether these objectives are being adequately met requires familiarity with record classification and the concept of a dynamic "record copy."

Once information flows electronically between departments and once multiple departments can access the same document directly from their PCs, the classic records management tools of inventory and appraisal provide still a further benefit. For they have familiarized the records manager with the fact that the same information performs multiple functions and has multiple values which must be considered. These principles have also made records managers aware that the same record may have both operational and evidentiary values that are quite different. These realizations enable records managers to fulfill the role of mediator between one department's need for information, another department's need for confidentiality, the public's right to know, the requirement of legal discovery, and the MIS need to keep the system running efficiently.

INFORMATION LIFE CYCLE

One of the tools most useful to records managers once they begin to view information flow is the "life cycle" concept. The records management theory of the record life cycle has already taught us how information evolves through a series of stages (e.g., creation, use, and disposition). It has oriented us towards the need to focus on the records creation stage in solving problems of record proliferation that only become obvious in later stages. As mentioned above, we learned that if we did not file non-records in the first place, we would not have to worry about purging them.

However, whether termed "record life cycle" or "information life cycle," this model has not previously addressed the way in which current record copies very often become the pre-record predecessors of revised versions of the same documents. To clarify this situation, configuration management added the concept of a "project life cycle," a "living baseline" in which one version of the baseline record becomes the preliminary design for the next. When this evolutionary perspective is applied to transactional information as well, we discover the implication that, while individual records each pass through a life cycle of creation, use, and disposition, they are also part of a larger business process life cycle. Future records managers would do well to base their workflow analyses on these larger business processes.

In this context--and in keeping with our biological analogy--we might view records as limbs that grow from a trunk of business process documentation. When they are no longer useful to the process, they are pruned away, but the process as a whole lives on. Using this model, we can develop a long-range evolutionary approach that would not deal with records as isolated, spontaneous generations. Rather we begin to view them as the organic outgrowths of business processes, in which pre-records evolve into records which in turn become pre-records of a later stage. Such a model emphasizes that information must also be dealt with in the context of "records"--not just technology.

CONCLUSION

Does all this imply that the records manager, rather than MIS, is to be the real information guru? Not necessarily. In a team-oriented organization, who runs the show is less important. The important point is that the records manager has important contributions to make not only in the area of paper records, but also with regard to non-paper records, and even (as contradictory as it might sound) in the area of non-record information. For just as determining the difference between record and non-record was critical to the management of records, this same determination is as important to the management of non-record information. Once reengineered for a dynamic information environment, the concepts developed by records management disciplines--record copy, inventory, appraisal, information life cycle, and baseline --are just as useful now as they ever were.

Copyright Association of Records Managers Administrators Inc. Jul 1994
Provided by ProQuest Information and Learning Company. All rights Reserved