20 June 1995
The Changing Workplace and the Nature of the Record
Richard E. Barry
Unpublished paper written in preparation for presentation made at ACA Conference, Regina, Canada, June 16, 1995
For a number of macro- and micro-level reasons, mainly of a global political and economic nature, private and public sector organizations have been transforming themselves or risking obsolescence or extinction if they don’t. Accompanied with these developments are significant changes in the workpatterns and the ways in which information, documents and records are created and used. Workplace changes show no signs of letting up, but appear to have become a way of life, at least for the foreseeable future, to survive in post-Cold War economic times. Accordingly, archives and records management (ARM) and information management and technology (IM&T) programs and professionals will increasingly be influenced by workplace changes over which they will have little control but with which they will have to carry out their mandates. Work pattern changes frequently involve new innovative uses of technology -- and in some cases result from rather than in such uses. Because they very often bring with them changes in the ways that documents are created or represented, ARM and IM&T organizations will have to join forces to keep pace with and adapt to these changes.
In this paper, I will attempt to develop these ideas by drawing from some of my own experiences in the planning and implementation of IM&T projects, related business systems analysis projects and, more recently, ARM experiences as well as on my observations of what is going on in the field of information technology and other related research areas. I will summarize those experiences chronologically to highlight some of the ways in which the workplace has been transformed in terms of work patterns, technology, interests and analytical tools during the past three decades, drawing lessons along the way, most of them with the clearer vision of hindsight and, I hope, not too much euphoric recall. Rather than simply leaving it there with another list of unanswered questions and unrequited concerns, I will take the plunge and offer both some conclusions for archives and records management (ARM) and some suggestions about what we might do about it all. These are not offered in any prescriptive way but rather as a way of provoking reactions, debate and better suggestions that, hopefully, will help all of us to move more rapidly out of the debating mode and into the “doing” mode. Before getting into the individual projects, however, I will begin by stating “for the record” my point of departure with respect to the meaning of ‘recordness’ and make some observations about technology in a broader sense.
An important part of the discussion in this paper has to do with the rapidly emerging area of multimedia documents and records. What used to be strictly in the domain of children’s games is now beginning to enter the board room. While preparing this paper for presentation at the ACA 1995 meeting in Regina, I found it difficult to make the necessary points regarding multimedia without actually demonstrating such forms of documents and records. That could not be done by simply reading this paper and, in any case, it was too long to present that way. To facilitate making several points concerning the impact of different presentation and storage media on ARM practices, I presented a considerably abbreviated form of this paper an on-line computer presentation where the various representation media could be demonstrated. Now, retrospectively, I have attempted to deal with the challenge of discussing multimedia forms of records, at least in part, by incorporating in this paper a few graphics taken from the multimedia presentation.
II. “Recordworthiness” and “Recordness”
There are others in the room far better qualified than I am to speak to the subject of the nature of records, including the next speaker, Trevor Livelton. Moreover, as will be noted later, in my opinion the nature of records has not yet changed because of the introduction of IM&T. However, as my task is to reflect on the impact of IM&T on ARM, I feel that I should begin by indicating my understanding of the term “record” and some prejudices I harbor about that. I will refer to more than one definition, in part to signal some of my own views early on in this presentation. I first drew from a traditional ARM definition given in the UN ACCIS report of 1990 which cited an earlier definition of the International Council on Archives. In this definition, the word record means:
Any recorded information, regardless of form or medium created, received and maintained by an agency, institution, organization or individual in pursuance of its legal obligations or in the transaction of business.
What I now find missing from that definition is the concept of use of records -- the notion that records are kept because there is a presumption of future use. This made me realize that the ICA/ACCIS definition is incomplete, if it hasn’t since been updated. I then turned to the Australian literature for another definition that was basically the same, but which added the phrase:
and subsequently kept as evidence of such activity though incorporation into the recordkeeping system of the organisation or person.
We owe a vote of thanks to Clive Smith and Glenda Acland for the improved Australian definition which, by drawing from the Jenkensonian definition, has the concept of subsequent usage of records (as evidence), if not explicit, at least implicit. I would like to suggest a further enhancement that would make usage both more explicit and broader, perhaps characterized by a further variation, such as:
and kept to be used for purposes of operational continuity, evidence, accountability, institutional memory, historical legacy and research.
Part of this may be semantic -- no doubt, the idea of use was undoubtedly intended to be implied in both definitions. Part of it may be a question of image. That is not a trivial matter for us to be concerned with, however. In this day and age of reengineering and reinventing government -- often code words for downsizing -- people who are seen simply to be keeping things with no explicit concept of serving fairly specific client communities are people who probably should be dusting off their CVs. In the electronic environment, I believe that records should be captured to the extent possible at the time of creation and only subsequently when necessary to reflect subsequent actions. Thus, in the electronic age, I submit that it is more appropriate to drop the word “subsequent”. Records are kept to be used and we should emphasize that -- used immediately and subsequently for the reasons listed. We might argue that in archival theory all of the reasons for usage noted above are subsumed in one word: evidence. Yet, it may be important to highlight and differentiate the uses of records because they reach out and speak to different client communities who should be advocating the case for ARM. Archivists and ARM programs need all the allies and advocates we can get these days. People who make significant use of records probably would become advocates if they felt well served in their own work by IRM programs. I question how successful we have been in culturing advocates outside of the ARM community with such terms as records or evidence (instead of, for example, mission-critical information), recordkeeping (instead of, say, information management), repositories (instead of information stores), central registry (instead of information intermediary or information service unit), etc. How we define things for our own purposes is one thing. How we project what we do to the client world perhaps should be another.
Archival science and diplomatics helps enrich our understanding of recordness with properties of impartiality, naturalness, interrelatedness, authenticity and uniqueness. Archival records also have content, form and medium. And they have context, which is a particularly important characteristic in electronic systems where part of the context may be supplied by the system software or hardware. The word ‘medium’ as typically used by authors in the field of ARM and diplomatics implies physical or storage medium. Now we begin to get into the area of how the use of words between disciplines can get us into trouble. Particularly in the electronic environment, the term “medium” should not ordinarily be used without a modifier since it may be used to describe quite different characteristics: perceptual media (sound, sight), presentation media (speech, paper, microfilm reader, video display terminal), recording media (analog, digital), processing media (text, data, image, sound, video), or storage media (paper, microform, floppy disc, videotape, hard drive, CD-ROM, DAT, etc.)
As one of my observations in this paper is that there is a need to bring the ARM and IM&T communities much closer together, I think it is important to mention these distinctions because I notice that quite different priorities are placed on these characteristics between ARM and IM&T professionals and even within those groups, between archivists and records managers and between information managers and information technology professional. This is, in part, why IT specialists and electronic document management systems traditionally give little attention to the issues of information value, appraisal, disposition management and long-term preservation; or to linking digital and paper-based or microform-based information and systems.
III. Let’s Stop Whining About Technology
In our understandable desire to focus on issues and remedies, we tend in many of our discussions of electronic records to lament the scourge of technology. I do it myself. Look at the problems that digital technology presents for records acquisition, appraisal, preservation, access and security. We focus much less on how technology might help us to overcome some equally or even more intractable problems with paper-based records management systems with respect to -- yes -- acquisition, appraisal, preservation, access and security.
To illustrate, we commonly use the issue of how easy it is to change electronic documents in an undetectable way as an example of the kinds of intractable problems we face with electronic records in comparison to paper records. Not long ago, the author had the occasion to print and sign a letter to an airline company to make a claim for lost luggage. While the original file copy was saved in electronic form, another paper copy of the letter had to be made to include in another letter to American Express where the author had baggage insurance to cover the difference between any loss and what the airline would pay. A copy of the signed letter was made on a personal copier machine (no high-tech office machinery). During the distraction of a phone call, the author got the original signed paper version and the copy mixed up. After spending about 10 minutes holding the two up to the light and trying to detect which was the original, I finally said to myself: What am I doing wasting my time here trying to detect what is for all intents and purposes an undetectable difference? I don’t know who got the “original”.
This experience with a $650 copier made me realize that with even inexpensive modern copying technology, it would be very easy to change a paper document -- particularly for the author with access to the original word processor, fonts and printer -- and to make a copy that would stand up well as the “original”, at least without the benefits of very expensive forensic lab tests. It is noticeably easier to do this today than it was 5-10 years ago, possibly when some of our notions about this subject were formed. Today, a well-designed electronic records system could make it easier, not necessarily more difficult, to prevent or detect document tampering. A relatively new technology called electronic time-stamping or digital time-stamping, offers an excellent way in which to prevent or detect changes to digital documents.
It can also be much more feasible in electronic than paper-based systems to maintain duplicate stores of the same information in different physical locations and even under different administrative control. While this solution to detecting changes in electronic documents might not always be feasible for the entire corpus of an organization’s electronic records, it might very well be feasible for 5-10% of those records, the typical amount representing archival documents or even a lower percentage of records considered most important. Or it might be feasible in larger quantities for records that are considered to be at great risk of manipulation by interested parties but needed only for a few years -- say the output of a computer-based procurement applications system. Thus it is possible to use a design strategy to achieve the objective of ensuring genuine records and of discovering attempts to alter or delete electronic records that is superior to what can be done in the paper environment.
Does this mean that technology does not bring with it a great many adverse or potentially adverse effects? No, it do not. Apart from the many well known archives and records management problems, there are many serious human issues that technology has given rise to -- but for which the solutions are not likely to be found in technological innovations -- at both the individual level (e.g., hardware ergonomics, human interface with software systems, potential isolation of people from the “real world” , etc.) and at the societal level (underemployment and unemployment, protection of personal privacy, access to public information, creation of a class -- even whole developing countries -- of information deprived “info-nots”, etc.) What I do mean is well expressed by an internationally acclaimed Renaissance scholar Walter Ong, S. J., who observed in 1968 in his book, Knowledge and the Future of Man:
It is not the inhuman effects of technological living -- our being “dominated” by machines, whatever that may mean -- but the human effects that pose our problem. The science that underlies technological living has given a new shape to the contents of the human mind....Opposition between technology and the humanities is more imaginary than real. The printing press, a technological device, was developed largely under Renaissance humanist auspices, and the use of computers for textual study and other humanistic purposes is already becoming commonplace.
Archivists and records managers love the written word. We would not be in the businesses we are in if that were not the case. As we wrestle with ever changing advancements in information technology, it might help all of us to pause and think about writing as a technology. As Ong later put it in Orality & Literacy: The Technologizing of the Word:
Because we have by today so deeply interiorized writing, made it so much a part of ourselves...we find it difficult to consider writing to be a technology as we commonly assume printing and the computer to be. Yet writing (and especially alphabetic writing) is a technology...in a way the most drastic of the three technologies. It initiated what print and computers only continue, the reduction of dynamic sound to quiescent space, the separation of the word from the living present.
In their days, wood, wax and parchment technologies were very popular and well accepted for the conduct of business in the middle ages -- wood for tally stick accounting records, wax for drafting other documents and parchment for the final version of those documents. Can we not imagine the headaches we would endure trying to store organizational accounting records on tally stick or carrying out document version control with drafts written in wax plates? Could CD-ROM possibly be worse than that? But wait! It would be more than a preservation problem. In the case of wax plates, it would also be a capture problem because, as with some modern forms of electronic records, people didn’t keep the drafts. They reused the wax for the next document much as today we use floppy disks, C-drives, PCMCIA cards and videotapes. Writing is indeed a technology, and one that has made use of many technologies. It is not, however, the only one. There are also speech, other sounds and video. Shouldn’t we be prepared to deal with whatever technologies are deemed effective or necessary by those most competent to create records of the actions for which they are responsible?
That many people hate to contemplate a change in what has become the favored technology of writing for producing records over other technologies, also is not a new phenomenon. It has always been contentious and a certain topic to raise emotions. In the fourth century BC, Plato ridiculed writing in any form as inhuman. Triethemius, the Renaissance humanist lamented the emergence of what today we might refer to as the ‘parchmentless abbey’ much as today some people lament the specter of the paperless office. He said:
“The printed word is on paper...The most you can expect of a book of paper to survive is two hundred years. Yet, there are many who think they can entrust their words to paper. Only time will tell.”
The fifteenth century promoter of the printing of the Latin classics, Squarciafico, wrote that the “abundance of books makes men less studious”. Drawing from Lowry, Ong carries the thought through to its modern day conclusion:
[the printed word] destroys memory and enfeebles the mind by relieving it of too much work ...downgrading the wise man and wise woman in favor of the pocket compendium.
Now, in a different way, the ‘printed’ word is under attack again, this time for largely economic reasons. Technical and professional journals and even newspapers are concerned about the significant threat posed by the increasing popularity of electronic publishing. While probably a large population of people seriously doubt that this trend should or will cause the demise of printed journals or newspapers, perhaps with the exception of some of the smaller journals that rely heavily on advertising receipts to sustain themselves, it is nonetheless seen by a number of publishers as something that will reduce their revenues and cause them to consider ways in which they will become more competitive in the future.
What will be next? Can we imagine our successors struggling to protect the paperless office from the onslaught of the next redefining technology? It could happen, and sooner than we may think. In November 1994, Leonard M. Adleman proposed a totally new approach to large scale computations in the form of a primitive DNA computer -- a biological computer -- to solve computer science problems. More recently, Eric B. Baum suggests the use of this model for purposes much nearer and dearer to the hearts and minds of the people in this room. He theorizes the use of Adleman’s model:
to produce an associative, or content addressable, memory of immense capabilities...one where a stored word may be retrieved from sufficient, partial, knowledge of its content, rather than needing to know a specific address as in standard computer memories. Content addressable memories are useful in a number of computer contexts and are widely thought to be an important component of human intelligence.”
CA memories may also be very useful in addressing some of the most taxing problems with electronic records -- navigation through extremely large document stores consisting of many, many terabytes, perhaps the magnitude of an entire nation’s archive. Some combination of “hardware” architecture, such as suggested by the DNA computer, and navigation software will be required.
We are comfortable and have learned to deal with the written word, but this does not mean that we should favor it over other technologies for creating and recording information. That is a matter more properly made (and that in all likelihood will be made) by document creators, not by the ARM or IM&T community. The emergence of multi-media (systems that manage separate stores for documents in various presentation and storage media, mixed-media (systems that manage documents with containing more than one presentation medium, hypertext (non linear text) and hypermedia (same as ‘hypertext’ but not limited to the textual medium) records in the form of ‘composite content objects’ (a.k.a. complex documents) will be a test of that proposition.
Finally, we should recognize that technical issues, albeit the easier to raise and discuss, are not the biggest issues. It is the social issues (privacy, preservation of democratic institutions, etc.), organizational issues (information ownership, service delivery organization, policy, etc.) and individual issues (underemployment, unemployment, ergonomic problems, etc.) arising out of the use of modern information technology that are the truly big issues. If we want to debate the pros and cons of technology, those are the issues needing most attention.
IV. Thinking “Out of the Box” About Electronic Records
One of the trappings of current business systems analysis and reengineering methodologies is the use of so-called “out-of-box” thinking. The term is used to contrast our thinking when we are so immersed in our problem world (the box) that we limit all possible future by the norms, constraints and tools that we have in the box. By contrast, when we engage in out-of-box thinking, we can visualize the box as though we were detached from it as we might be in a helicopter looking down on the box and trying to see surrounding landscape and previously unconsidered possibilities. In the former case, we might look at a barrier and consider how to climb over it. In the latter, we might smash through the wall or simply walk around it -- define it out of the problem. I find it helpful to apply out-of-box thinking to electronic records issues, e.g., to think about them in terms other than the usual paper-based paradigm. Let us explore some points about recordness using voice technology to help avoid getting trapped in the paper records box.
We know that an unrecorded telephone conversation in which I signal my agreement with your earlier proposal is not a record because, even if it met all other requirements of recordness, it remains information that is not recorded. We have chosen traditionally not to record telephone conversations for a number of reasons -- mainly because it is seen as a practice which would cross a strong cultural line of personal privacy and because of our preference for the written word. If something that takes place over the phone is of such importance as to be regarded as “recordworthy”, we expect one or both parties to the conversation to record the results in written form, e.g., a memo to files. Despite our almost universal distaste for answering machines and voice mail (or vmail), we put up with them because we dislike the alternative even more if it means that we won’t get the service we are after or the return call we anxiously await.
Now suppose that I fail to reach you when I call, and instead leave a message on your voicemail (vmail) system confirming my agreement with your proposal. Now this is a recorded voice document. It would become a voice record if other requirements of recordness are subsequently met, i.e., if you forward the voice document to others for action, as you might an email message, or otherwise use it as part of the business process of which it is a residue, and its context is established through linkage to the action. This requires that the vmail system be functionally able to facilitate this kind of linkage. Today’s vmail systems lack this kind of functionality even more than electronic document management systems.
Like most other information technologies, voice technology is moving beyond the state of simply providing a means for improving individual productivity and into the arena of organizational productivity and client services. The example given above is isolated and trivial and presented only to make a point that recordness may not be changing but the way in which acts are being recorded is. However, when organizations begin to use similar technology to perform business operations, then archivists and records managers do have to think at least twice. To illustrate, one very large insurance company, USAA, routinely records telephone calls from its policy holders because it does its business with customers directly by phone without sales brokers in the middle. As part of a business process re-design a few years ago, this company now permits its customers to make claims through a human interview process over an 800 telephone line. Conversations are recorded with the knowledge of the caller. They are linked to image and data files maintained on the customer for use as inputs to a work assignment and workflow system that parcels out claims and other requests for timely action and for later reference. They are voice records.
This practice is becoming more common as organizations turn more and more to the use of telephone services to provide more direct “just-in-time” services to their clients and less expensive means for delivering services, especially when those services are information based. Among the more common examples are the use of telephone touch-tone commands to transfer funds from ones savings to checking account without ever leaving home, or to charge a theater ticket over the phone for later pickup at the theater Will Call desk, and the telephone Help lines that are operated by most information technology hardware and software firms. In the not distant future, we will be able to get money from our banks without leaving home. An electronic box, perhaps the size of a cable box will allow us to insert a credit card with an embedded computer chip which will be used to download an electronic fund transfer to the card which we will subsequently be able to use at home or in stores to purchase goods and services and the card will be debited accordingly. It is not unusual these days to hear a recorded message when dialing into one of these services that says words to the effect: “This call is being monitored for training and quality control purposes.” This could mean that calls are monitored in real time and are not recorded at all, in which case they would not qualify as records; or that they are randomly or always recorded for later review, in which case they could be records. They could be disposed of immediately after such reviews or they could be filed for later use as evidence, should they become linked to some service-delivery or product liability action. In short, voice mail messages may very well qualify as records.
With speech generation and speech recognition capabilities now available in PC consumer products, the distinctions between the written and the spoken word will soon become blurred. Speech generation systems simply create computer-generated words and sentences, converting one digital form to another -- text to sound. With software that was preloaded on the notebook computer I am presently using, I could demonstrate this capability with this paragraph in a few different ways. It is not possible to do so through the medium of print; however, I will do my best to describe what happens. One easy way, using the particular software that I have on my system, would be to highlight this paragraph and simply do a COPY command off the Edit Menu. Then by simply clicking the right mouse key over the active speech generation software icon, we would hear this paragraph converted to sound. Pauses would happen at the end of sentences and you would find it amazingly understandable -- lacking in emotional intonations to be sure, but quite understandable. Of course there are shortcomings. For example, if the paragraph included a phone number, say my fax number -- (703) 241-7968, it would read that as: SEVEN HUNDRED AND THREE...TWO HUNDRED AND FORTY ONE MINUS 7 THOUSAND, NINE HUNDRED AND SIXTY EIGHT. But keep in mind that this technology in the hands of ordinary consumers 10 years ago would have been considered little short of a miracle. Even the fact that we are amused by or ridicule such seemingly obvious shortcomings is itself evidence of how much we have come to expect from information technology. Then, what is so newsworthy about this technology? After all, it has been around for a long time in such common applications as those used by telephone Information Operators when they say “Please hold for your number.” What is newsworthy is that this capability is now available in PC consumer software products and is included among the pre-loaded software packages provided on some multimedia computers. This should be taken as a wake-up call for ARM professionals to anticipate the interchange of email and vmail, once the email vendors catch on to the potential for their own products, probably before the end of this decade -- not a great deal of time to prepare for.
Speech recognition is a wholly different thing. It satisfies the opposite functionality and is vastly more complicated than voice generation, as it must ‘read’ widely varying voice sound signals created by individuals with different speech patterns and different accents -- even within the same vocal box when a person has the common cold -- and convert them to the written word form or to other outputs such as computer commands. Unlike voice generation, it has taken literally decades to bring this technology to fruition, mainly through a few leading computational linguistics laboratories, and mainly using simultaneous dictation applications. This very same technology were it to be applied to the field of automatic document classification could break some great barriers for the field of archives and records management. Like speech generation, speech recognition has appeared on the PC consumer market only in the past few months. For example, the simplest (and cheapest) of these systems (maybe $100) allows the user to easily navigate around a graphic user interface (or GUI) by speaking into a microphone incorporated in, or attached to, a PC (as all multimedia PCs now have) with a string of commands -- words -- such as: “Open Microsoft Windows,...Open Microsoft Word for Windows,...Open file name: “A C A PAPER PERIOD DOC”. In so doing users can change their GUI to a VUI or voice user interface. Other such systems, much more sophisticated and costly (say $1000), and not yet elegantly, may be used for the same purposes and more -- to dictate a memorandum or report without using a keyboard. To be sure, vocabularies are limited to a few thousand words that must be spoken by the system owner and dictation must be discontinuous and therefore slow. They also have difficulty distinguishing the various legitimate spellings of words with similar sounds such as the words “to”, “too” and “two”. However, the individual can build up his or her own dictionary by ‘teaching’ the system new words. For the hunt-and-peck typist, or the person physically unable to type or with a cultural objection to typing, it may sell some PCs to people who wouldn’t otherwise use them. More importantly, this is a rapidly emerging technology that has really turned the corner now. Ironically perhaps, but speech recognition may turn out to be a great friend to those in the archives community and historians who have fought to preserve records through the special medium of oral history. Consider the enormous possibilities of being able to do ‘full-text’ word searches on large voice record stores.
The author has been following speech recognition research for about 15 years in the artificial intelligence and natural language processing research and development communities. Every year, I have heard the same thing. The response usually went something like this: “We’re about three years away from the necessary breakthroughs needed to develop a product.” Many people didn’t believe it would ever happen. It was too difficult to achieve. Those who did, thought it would be possible to deliver such products only on large mainframes, including massively parallel platforms, because that was the level of power needed for complicated artificial intelligence applications during the research stage. Moreover, research in this area preceded the advent of personal computers. No longer is this the case however and, except for some special kinds of intelligence applications, most vendors involved in this field are going after the PC market -- now the most common instrument for document and record creation. Moreover, current research is underway to overcome the limitations noted above. To avoid getting blindsided, we should anticipate wide availability of a superior continuous speech recognition technology well before the New Millennium. If you are operating in a field where there is a fairly limited or specialized vocabulary, such as medicine, it is already here. Many physicians are already using speech recognition systems to dictate patient reports. What will be kept for the long term? The secondary text version created from the voice dictation, or the original voice records themselves? As with speech generation, the paper version of this presentation does not permit demonstration of this exciting and innovative technology. Seeing and hearing the use of these technologies for business purposes in the PowerPoint™ version of this presentation brings the point home in a way that is not possible to do on paper. Alas, that is at the heart of the multimedia records issue.
The point that email and vmail may be interchanged is a point that needs to be made more generally about information and records management in the digital environment. This is perhaps better described using the ‘slide build’ from the PowerPoint™ version of this paper. Through the use of image scanning technology, paper documents may be converted into digital form.
Paper Text 4 F Digital Image >
= scanning systems = paper F digital storage
Once the paper document is in digital form, it may be further converted from an image base to a character base, significantly facilitating information retrieval.
Digital Image > F Digital Text 4
= OCR systems = key-word F full-text retrieval
Now in character form, it is now possible to convert the original paper form into voice form through speech generation technology:
Digital Text 4 F Digital Voice (
= speech generation systems = email F vmail
A similar route may be taken going in the opposite direction beginning with the use of speech recognition technology to digitize a human voice document or record:
Human Voice J F Digital Text 4
= speech recognition systems = vmail F email; also = voice record creation
What this means is that, at least to a certain extent when the developers of GUIs catch up with the document creation and conversion technology, people will be able to choose their preferred way of operating by simply throwing a readily reversible switch or clicking an icon, or speaking a voice command, to have their email converted to vmail or vice versa. Obviously this could be quite important to the person who is hearing or sight challenged. But it could be very convenient also for any individual traveling without a computer who wishes to be able to pick up email by dialing into the vmail system back home. This poses an interesting question. Let us say that we decide as an organization that voice mail records will not be retained out of consideration of privacy; but that email records will be. But now if we have some users converting vmail to email, we wind up retaining some but not all records for reasons other than their relationship to a business activity. This is one reason why we should not make appraisal decisions on the basis of technology but rather on the basis of the acts or business processes that create these records. Reflecting on Charles Dollar’s Macerata paper, Luciana Duranti writes:
[I]ntense preoccupation to demonstrate the common nature of all records, regardless of physical form, and the need for management decisions (appraisal included) that treat all the records of one creator as an integrated whole, is certainly shared by many, but it is by no means the consequence of a generally accepted idea. The qualifier that Charles Dollar added to his statement that appraisal should be based on the functions or competences generating the records, rather than on the technical applications from which they resulted, that is, that this might not be so with electronic records having no paper analog, still stands unchallenged and is therefore quite disturbing.”
If we agree with the proposition that acts, not technology, should be the governing factor in appraising records, it is difficult to justify ignoring the appraisal of vmail records despite privacy considerations, even though retaining them in organizational multi-media record systems would be disturbing to many, including this author. The key here is the same as for email: to have an organizational policy on the subject that is clearly understood by all employees.
In the discussion of speech recognition above, it was noted that it is a rapidly emerging technology. This is a point that needs to be made more generally about information technology. It is a very fast moving field in which there is enormous competition. Thus, while we should of course recognize the limitations of IT where they exist, we should also realize that it is risky to base our future plans on the existence of the same limitations three to five years hence -- the typical lead time for introduction of an electronic document/records management system. As noted earlier, one of the major reasons that archivists expressed a few years ago for not maintaining records in electronic form was that electronic records could not be trusted to maintain their genuineness -- i.e., that one could not be assured that, over time, the record had not been tampered with or altered in some way. Research was carried out in the early 1990s that, although it was done for intellectual property rights purposes rather than for electronic records management more generally, addressed the document ‘genuineness’ issue that so concerns archivists. Another common concern of archivists is how email can be used as records in the absence of signatures. Considerable effort has been going on in the research community to tackle this problem that has resulted both in products and legislative changes to provide for certification of email messages.
V. Technology and the Transformation of the Workplace
We often observe that information technology is too often procured without an intended result or transformation in mind other than in some general terms of improved productivity, which organizational analysts and industry economist have found difficult to scientifically demonstrate. It is fashionable to state, as I have myself on many other occasions and as I do again here today, that desired organizational end results should govern organizational aims and objectives that, in turn, should be the basis for the creation of business processes, which should drive information and information management needs (including records) and architecture and this, finally, should drive information technology decisions. We say that it should never be the other way around. Despite my own theoretical convictions and incantations to this effect, my observations and experience tell me that there must be a more complicated arithmetic at work -- that there is a greater symbiosis between work and technology than I am always ready to admit. People do sometimes do it in reverse and sometimes with good results. Sometimes creative people can see in a new technology the potential for changing the way that some business process is currently carried out. The invention of the PC followed this route. Most people in the workplace didn’t cry for that invention; however, many immediately saw ways in which it could improve their work patterns. Emerging multimedia technology may very well prove to be the latest example of this phenomenon. When things are not done according to some prearranged analysis and plan, it is not always easy to discern or say which it was that led to workplace transformation -- a problem in search of a solution or a solution in search of a problem. Perhaps that is not the important distinction to be made. Perhaps it is more realistic to see work and technology as parts of a continuous feedback loop where work needs spawn technological requirements that may be only partly satisfied by technological innovation that is then reacted to in the workplace and refined in later innovations; and sometimes technology results in unexpected or unintended innovations in work patterns and the cycle begins again.
To the extent that technological innovation has an impact on document creation and use, these cycles carry with them significant implications for records management. The emergence of more complex document forms illustrates this point:
Simple documents are typically text-only documents such as telephone logs, call slips, calendars, text-only letters, etc. Compound documents are documents containing graphics (e.g., a logo or signature) or data (e.g., a statistical table). Complex documents are documents containing multimedia objects (e.g., sound, animation, video). Traditional records systems are based upon a simple document architecture in paper and/or microform storage media. They have been stretched to accommodate some compound documents, though often only by printing out such documents and filing them with simple documents in paper or microform storage systems. The emergence of complex documents will not be so easily accommodated by traditional records systems because multimedia documents cannot always be fully represented in paper form. Unfortunately, the printed form of this paper (as distinct from its multimedia form in a PowerPoint™ presentation) does not permit illustrating this point.
Figure 1 is an example of a simple document that will be followed through with more complex examples using the same case to illustrate why more complex forms of documents are beginning to be seen as providing a competitive advantage, either in terms of time, productivity, ease of understanding or attractiveness to the reader. The case is a person who is sending another person a video clip for a TV commercial spot that promotes travel to Alaska. The recipient’s comments are requested urgently because time is money in the advertising business. It is sent at considerable expense, relative to regular mail services, by overnight courier with a request that the recipient provide comments and a voice over to use with the spot immediately. We can imagine that this exercise would take about 3-4 days to complete.
Figure 1: Simple Document: text only
Figure 2 illustrates a compound document simply because it has a graphic logo on the letterhead -- something that archivists may find important, like signatures, to capture -- a reason why archivists typically favor image rather than ASCII representations of documents for electronic archives -- but it adds little if any substance or content to the document. Perhaps a bit of context. More interesting examples of compound documents that would add more by way of content and substance might be ones containing a statistical table or picture. In the case outlined above, there would be no difference in the manner of delivery or handling of the task. The introduction of the graphic figures in this paper make it a compound document.
Figure 2: Compound Document: text and graphic
Complex documents include embedded objects in addition to text that are linked from the document to some other independent computer file where the object is separately maintained and may be separately updated (and by so doing automatically update what is contained in the document). Such objects may be attached and mailed electronically with the document. Figure 2 could also be a complex document if, for example, the logo became an active icon that, when selected, would activate the embedded object. Objects so linked are said to be hyperlinked and may be hypertext, hypersound, hypervideo, etc. A document (or composite content object) with more than one such representation medium may also be called hypermedia. If the logo in Figure 2 were hyperlinked to the Alaska promotional video of the case in question, it would be illustrative of a composite content object containing both textual and video objects. The version of this paper that exists on my hard drive is a complex document because the graphics are hyperlinked to the graphics that exist in the PowerPoint™ version of this presentation. In its paper medium version, however, it is not so linked and is therefore a compound document.
Figure 3: Composite content object (CCO) containing text and embedded video/sound objects
Figure 3 is such an example. (The Asymetrix icon calls up a video capture system and is a trademark of Asymetrix, Corp. The microphone icon calls up the Microsoft Windows Recorder system.) Taking the example the next step, the document creator sends this document by electronic mail to the recipient with the attached linked object, in this case an Asymetrix “*.avi” or video-type file or video clip. The instructions indicate to the recipient that by ‘clicking’ on the video object the recipient may view the embedded video clip. The recipient receives the email request within minutes of its being sent and, after reviewing the video clip, attaches a voice annotation comment on the document and returns it to the sender. The whole transaction could be concluded in an hour or so.
In Figure 3 we see that the email message has been received and marked up with a return arrow to the sender ad dated 10/24. Now the CCO contain an embedded sound object (the microphone icon), which if clicked on will reveal a voice commentary on the video object (the Asymetrix icon). Rather than days having passed sending these communications back and forth in the mail, the whole transaction takes place in a matter of hours electronically. This kind of document may be stored in a normal digital storage medium, e.g., hard disk, but may be perceived in all of its presentation media (text, video and sound) only if stored in a multimedia information store and retrieved through a multimedia workstation. Until the latter conditions are met, the document would not qualify as a record, even if all other conditions of recordness were met, because it could not otherwise be perceived by human senses.
Imagine other examples of this kind of business use of multimedia, e.g., in the form of a university capital budget request that includes an opening “paragraph” in the form of a videoclip of a proposed new research center or student center. Think of the personnel file with the employee’s picture that has embedded in it a videoclip of the person giving a brief sketch of his or her professional background and current professional interests. Or the police report of a highway arrest which combines a written report with the picture of the person arrested which is hyperlinked to a videoclip taken of the arrest from the officer’s dashboard video camera. Many police departments already have computer terminals in their police cars where data is obtained from central data stores and where textual reports are submitted by the policeman in the field. Recently, some have installed video cameras. How long do we expect it to be before these record capturing media are combined into multimedia, hypermedia, documents or composite content objects? Not very long, I submit. Or consider the typical “back-to-office” report in most development assistance organizations. Traditionally such reports are fairly highly structured with statistical tables showing the financial status of the project and, if applicable, progress on the civil works schedule. How much more informative and meaningful such a report could be if it included a photograph of the developing country project manager or of a national archives building under construction. Beyond that, how much more informative it would be if, when one ‘clicked’ on the project manager’s picture, it revealed a video clip interview with the PM on the status of the project. One could imagine many such applications. In the German Government, one need not imagine. One of the most advanced operational multimedia systems anywhere is one presently being used by the German Government as part of the transition of the capitol from Bonn to Berlin, during which time portions of the Government are in both cities. As more real people with real business applications discover the powerful capability and multiplier effect of mixing the written form of language with other forms of information, it will not take long for hypermedia documents to become commonplace.
VI. Lessons Learned from Past Experience
In the following sections, I will attempt to illustrate some of these points with personal experiences and observations. They are not presented in isolation of the experiences of others as a way of ascribing greater importance to them than they deserve. They simply reflect the kinds of things that were going on in many quarters of the North America and elsewhere that, together, have had some impact on how information technology is used, how work is done and what it might mean for ARM.
Early Command and Control System Project
Information technology in the 1960s was largely confined to centralized, mainframe computer systems with data processing applications, as distinct from the text-based office systems of more recent years. Most of these applications were transaction-oriented and mostly in the financial sector -- accounting systems, payroll systems, etc. There was a certain likeness between the status of management information systems in the 60s and electronic records systems in the early 90s. That is, the subjects were very much topics of discussion and debate at professional conferences, but there was little by way of implemented, operational, systems. It was my good fortune, as a young naval aviator in 1960, to be assigned to a newly created “command, control and communications” group (or C3 as it was known, and subsequently C3I when “intelligence” was added to the function) in the Office of the Chief of Naval Operations in the Pentagon under the leadership of then-Commander Arthur K. Bennett, USN and to become involved in some of the first computer-based command and control systems that were military versions of management information systems, and more.
Prior to 1958, the Chief of Naval Operations exercised command and control of naval forces the world over. With the Defense Reorganization Act of 1958, however, operational command and control of all military forces was transferred from the service chiefs, under their individual civilian cabinet secretaries, to the Joint Chiefs of Staff reporting directly to the newly established Secretary of Defense.
Because of the new law, the Joint Chiefs faced the enormous problem of aggregating and digesting the status of all U.S. Armed Forces daily and set out to develop a computer-based operational readiness system for the Joint Chiefs of Staff,. It involved the creation of a reporting system consisting of about 30 highly structured reporting formats. These messages were transmitted daily to the Pentagon shortly after midnight Washington time so that they could be processed and summarized in a computer printout and used for preparation of the Joint Chiefs’ regular morning operational briefing. No such summary of all forces with such up-to-date information had been possible before that time.
The punched paper tape that was created by the teletype machines containing the digital representation of the messages were fed into paper tape readers that read the data into a large IBM 7090 -- what was at that time regarded as a scientific computer because of its computational power. It had all of 64K memory, a room full of tape drives and three shifts of about eight enlisted men and officers to maintain it. Remarkably, the punched-paper tape reader was connected directly between the teletype machine to the 7090 and messages were processed into the computer automatically as they arrived. The information was then interpreted by a software programs coded in machine-language. It was an early generation precursor to the modern data base management system. It was my first involvement in the design and implementation of computer-based information technology. The Joint OPerational Reporting System (JOPREP), consisting of about 30 separate structured reports, was fully digitized so that the system for reporting the conventional and nuclear readiness of U. S. Armed Forces world-wide was fully automated. The result was that, at 5 a.m. each morning, a highly classified stack of computer pages representing the state of U.S. Armed Forces individually and in the aggregate was in the hands of Pentagon briefing officers for use in morning operational/intelligence briefings of the Joint Chiefs in the National Military Command Center on the second floor of the Pentagon.
This project was an early lesson in how macro-level forces (in this case legislation in the form of the Defense Reorganization Act of 1958) can have enormous impact on local work needs and patterns, technology and organizational behavior that changed the ways in which information and records were created and used. In a similar manner, the demise of the Cold War has created global and national economic trends that are reverberating down to the smallest public and private sector organizations today in the form of competitive pressures, reinvention of government, etc. These forces are driving the rethinking of work patterns and technologies in ways that have a direct bearing on recordkeeping practices.
The experience reinforced the idea that if one focuses on the operational needs first, rather than the technology, results are more likely to be successful and enduring. The JOPREP system has undergone many evolutionary changes since it was developed in the 60s to reflect changing operational needs and technological improvements; but it is basically still in place today. We didn’t know them as business processes then and had no business systems analysis tool other than what was provided by old fashioned systems analysis, but it was the defining experience for me in the use of information management and technology tools to address business needs and was instrumental in my decision to leave the service and make a career in this field.
It was an early example of ARM and IM&T organizations missing each other’s boat.
Paper printouts were regarded as the residue of the system. Computer tapes were kept purely for backup reasons, not because anyone thought of them as “the record”. As such, they probably would not have been useful as an information store for selectively retrieving or presenting the component messages or summary reports. No military archivist, records manager or historian showed up, and we didn’t know enough to ask. That is not to say that there was no records management system in the Joint Staff. There were, for example, military historians whose job it was to reconstruct crisis situations retrospectively to learn lessons for the future -- a classical example of where records are needed for institutional memory purposes. It was simply a case in which the two communities of interest didn’t conceive that they had any common interests.
Early “Roomware” Project
Shortly after implementation of the JOPREP project, I had the good fortune to be involved (albeit as probably the most junior officer on the team) in the design of a new National Military Command and Control Center or NMCC. This was a room that was to replace the old fashioned “War Room”. It was one of the earliest attempts (the earliest for me) to integrate physical facilities with information management and technology. The task was to design a suite of rooms for the processing and presentation of high-level information with facilities for instantaneous world-wide communications at each seat around a large conference table in the central room where the Joint Chiefs met, as well as in adjacent rooms designed for the use of ad hoc battle teams convened to provide staff support to specific crisis situations. Just as the military was a leader in the development of information systems, including some of the earliest text-based systems that were developed for the processing of textual information in support of intelligence operations, it also played a very early leadership role in the development of decision-support rooms.
This project also opened up the thinking for many of us to look beyond the more traditional forms of data processing and presentation (mainly in the form of massive computer printouts) and to linking information processing with real-time communications -- the two principal elements that subsequently became integrated in what we now call office technology.
Unlike today’s “roomware” systems, the advanced version of this kind of technology, the NMCC did not automatically record decisions taken..
It is very easy to lose the record (or never make it) of ad hoc situations that are not a part of the daily routine of any particular office, such as during military crises or very common business project oriented activities. Yet, the associated actions may be among the more important records. With current and foreseeable trends in organizational design being toward flatter, less structured organizations, this is likely to be a problem that modern organizations will have to deal with more often.
Facility managers can be great allies of ARM programs when they see the potential of electronic records for reducing office space pressures and costs that in large urban centers can easily run $30-50 per ft. P.a.
Coordination of Information Sciences Research and Development Projects
Following a pause of a few years to return to an operational squadron and carry out postgraduate studies, I had the opportunity in the late 60s to return to the IM&T field in a very different capacity as Executive Secretary to an interagency information sciences technology group under the President’s Science Adviser’s Committee on Scientific and Technical Information (COSATI) whose purpose was to coordinate information sciences and technology research work among the agencies of the Federal Government. Here I first learned about some extremely interesting research that was going on in such fields as computational linguistics, large text-based systems and geographical information systems (GIS). Some of this was still highly classified at the time. GIS systems are now figuring significantly in modern electronic records. Presently, they constitute a relatively small portion of electronic records, because they are used principally in specialized applications dealing with mapping and building drawings. Nonetheless, even those applications often create important organizational records that ARM professionals have to deal with, usually through the use of very expensive and space consuming map filing cabinets. The storage of such records can be much more economically achieved electronically than in paper where the records are usually oversized, of different sizes and are sometimes on chemically treated paper. In the future, we may anticipate that the use of GIS technology will not be limited to mapping and building maintenance applications. Information scientists are using this technology to help tackle difficult information retrieval problems, especially as a retrieval interface for large data bases, e.g., any data base that has geographical components such as medical research data, real estate data bases, etc. Even applications that have no geographic components, such as very large directory systems are quite amenable to representation through the use of ‘information trees’ and ‘information maps’. GIS applications can, on the one hand, be used in ARM electronic directory applications. On the other hand, GIS represents one of the most complex forms of electronic records that modern archivists must learn to deal with, or to get others to deal with.
From this experience, I learned for the first time the great importance of the balance between research, development and implementation projects, including basic research with no specific application in mind at the start, development projects aimed at a particular problem set, or the functional needs of a particular user group, and good implementation projects to make the desired results happen. The Chairperson of the group, its members and invited guests from the research community were extremely knowledgeable in the field of information sciences and provided a most challenging intellectual experience. It also gave all of us the opportunity to see some state-of-the-art implementations. I learned the importance of networking with people in the research community, keeping up with their literature and to bringing them into operational settings and discussions when, in later management positions, I was involved with various information systems and services. I learned that researchers were just as wanting to have contacts in the real systems world to hear about current business needs and related gaps in the technology, as I was to know what their research might have to offer in the problem sets in which I was working.
Early Centralized Office System Project
The COSATI assignment led to another toward the end of the 60s, now as a civil servant back in the Pentagon, managing the information science and technology division in the Office of the Secretary of the Navy. There I had my first experience with a rudimentary office technology system -- conceiving and implementing a centralized office services system that provided dictation and word processing services for Secretariat staff. This was a ‘Rube Goldberg’ attempt if I ever saw one to put together some technologies that weren’t designed to work together to perform typical office services. It consisted of two secretaries located in a room with two single-tape IBM Magnetic Tape Selectric Typewriter (MTST) machines (at the cost of $7,500 each), one $10,000 two-tape MTST, a telephone line and an answering machine. Secretariat staff would call the service from their office telephones and would dictate the desired document over the phone. A secretary in the office services unit would transcribe the voice tape into its first typed form using one of the single-tape MTST machines. The document would be sent by special messenger to the author, usually the same day for markup and return to the office services unit for further update. The original MTST tape would be mounted on the two-tape MTST machine. The second tape drive would be used for merging the original tape information with the revisions and for creation of the new version for printout and return to the author.
No thought was given to keeping the electronic versions of the dictated documents beyond the time necessary to finalize those documents unless the author had reason to believe that it would be necessary to use large portions of the document in future documents. Any preservation in electronic form was for operational not for records management purposes which were seen to be totally under the purview of the principals creating those documents. The paper output of the service, not any electronic representation, was regarded as the record. Once again, the basic recordness of the paper residue of the actions that precipitated the documents was largely unchanged. Because the paper version was the record, neither content, context nor form were particularly effected. Only the process of document preparation and its related service levels had changed.
Private Sector Executive Involvement Project
For me, the end of the 60s and beginning of the 70s was marked by several projects in the private sector that involved interactions with executives on the design of computer-based management information systems, including organizing special seminars for managers in information management, interviewing managers and marketing systems to them, writing MIS proposals, etc.
Senior managers typically view information systems projects as computer projects or information technology projects, not as projects that support strategic aims and not as projects that manage valuable organizational assets in the form of information. This is changing nowadays because of the threat of competition and the greater computer literacy of people today than a generation ago. Even today, however, managers do not equate records to information. Whereas they are very anxious to talk about the strategic value of information, they will refer anyone wishing to talk about records to their secretaries. It is therefore important for ARM professionals to project their offerings and needs in ways that executives will relate to in their own terms and language.
Distributed Office System Participative Design Project
Word processing was first invented as a mainframe computer application in the 1950s by one of the great information scientists of all time, Douglas Englebart. A few years ago, I had the pleasure of meeting with Doug and having a demonstration of that early system and of the first (now ubiquitous) computer mouse (albeit today’s versions look quite different from Englebart’s hand-fashioned mouse) that he invented in 1964 to be used with a mainframe computer terminal. However, just as the mouse didn’t really gain wide usage until the latter part of the 1980s after something called Microsoft Windows was invented, word processing didn’t begin to catch on until the 1970s when dedicated word processing equipment made it possible to create documents locally rather than on mainframes and when equipment prices came down significantly at the same time that secretarial costs were on the rise. The type of office service described in the Pentagon experience was one of the precursors of what became commonplace during the decade of the 1970s -- the so-called “word-processing pool” in which the whole organization was served in its typing needs by a single central pool of secretarial staff working in what would today be regarded as an electronic sweat shop. This became feasible as the cost of word processors decreased and became more functional.
In 1979, while serving in an operational position of an international financial institution, I led a study of a regional vice presidential office consisting of several departments that serviced all projects in about 20 Eastern African countries. Since it was clear that document creation and preparation were tasks that were subject to differing individual work preferences and behavior patterns, we undertook this study in a participative manner involving managers, economic and financial analysts, sector specialists and support staff of various disciplines and areas of responsibility. The conclusion was that office technology should be fully decentralized to the unit level. Within a couple of years, the other regional offices followed suit and the centralized WP unit was disbanded and its resources were distributed to help seed an enterprise-wide decentralized office support system. For a short period, control of document preparation reverted back to the unit level support staff and the management of records improved at that level.
Among the noteworthy observations on this project are that:
Innovations may seem trivial or even absurd when they are first revealed. Sometimes they are; but other times they simply reflect that their creator is years ahead of his or her time and that they will become widely used when the time is right. It may be difficult for archivist to always sort out correctly which is which. The main lesson for the author was simply not to dismiss possible futures and to give thought as to how one might accommodate them if they appear on the scene.
The study was an early example of a participative design project in which considerable importance was placed on identifying and involving all key stakeholders as members of the team.
The notion that there might be stakeholders external to the regional office was not readily accepted because staff did not see the relevance of changes in their own work patterns to the work of other external offices. It didn’t occur to anyone that the records management division and archivist might also be stakeholders.
The study was very team oriented. On the other hand, it was a bottom-up approach to how work was done locally and therefore did not address the underlying business processes that produced the reports and other records. Thus even with the team orientation, the focus was more on individual productivity tools in a local organizational setting, rather than group productivity in an enterprise setting.
Individuals bring not only their own national and ethnic cultures to a project team, but the culture and value system of the organization in which they work. If that culture places a value on the importance of information and information sharing, it can make the archivists job much easier. If it treats information simply as a power lever, and not as an organizational asset, the archivist’s job will be much more difficult, and conceivably threatened by political pressures to dispose of records in inappropriate ways; and it will be necessary to consider strategies for bringing about changes at the cultural level.
No one was prepared for what would happen at the beginning of the decade of the 80s when IBM began marketing what it called a “personal computer” or “PC”. It was reportedly called that because it was seen as something that could be entertaining or useful in the home, not as something with any particular business application. Nor did most business organizations invest much in that technology, beyond supplying a few technology-oriented industrial-strength analysts with such tools, until about 1983/4. However, it took little time for analysts and authors to see how much better they could perform their jobs when they had control of the tools of document creation in their own hands. This realization fueled an even greater decentralization of office technology than had been dreamed even following the 1979 study.
Decentralization of Information Processing Power
Probably the single force exerting the greatest influence on the workplace of the early and mid-eighties was rapid diffusion in the control of information technology, information technology tools and all aspects of document creation. In my own organization, even before the introduction of PCs, a handful of daring economists and other operational officers requested word processors of their own and offered to make sacrifices elsewhere in their budgets to pay for the equipment. This quickly caught on. Whereas, in 1977, on average there was one terminal (pre-PC) for every 100 staff members, by 1980 it had changed to 1:25, and by 1983 the ratio had become to 1:4. By 1987, the ratio was 1:1 roughly where it stabilized until the generation of PCs that followed the IBM AT created a backlog of obsolescent PCs with little street value and they began to plow the older technology back to use in the homes of staff members, largely to give them after-hours access to email.
These events of the 1980s noted above mirror what has happened in many other North American organizations, with variances in the timing of full decentralization. It is what has happened or is happening now in other countries. What it meant was that support staff -- the original gatekeepers of office filing -- were increasingly dealt out of the document creation phase and the maintenance of local and organizational files, except for what was given to them in hard copy by their supervisors. As document creators were often analysts in their professional work, many began to make use of spreadsheet systems wherein they could define formulas for specified cells, fill in the data cells and let the program work out the arithmetic. It was easier than some of the more complicated earlier mainframe based systems, and was immediately available on ones own PC. Like WP, it had the great merit that it was easy to change the data and update the tables. Where reports involved the integration of statistical tables into the document texts, secretaries were typically involved in document creation, as typically this task was performed by secretaries using the more traditional literal “cut-and-paste” operations of the earlier years. Following the advent of Microsoft Windows, however, the invention of systems with “object linking and embedding” or “OLE”, made it possible for authors to import tables directly into their text documents and more recently to create them without appearing to even leave the document. This could be done nicely by the author with automatic changes and pagination -- without anyone else’s help. The 80s was the decade in which the author became master of his or her creations, making it also the decade in which the quality and completeness of records and recordkeeping systems began to slip.
Individual, Workgroup and Organizational Productivity
As noted earlier, in a broader way, it was a decade marked by tools for improving individual productivity -- word processors and “dumb terminals” that gave way to “smart personal computers”, word processing systems, spreadsheet systems, individual organizers, etc. But it was also the beginning of a shift toward concerns about workgroup productivity with the introduction of email (mostly mainframe based systems) and local area networks where the focus was on providing technological infrastructure for information sharing at the level of a small organizational unit. This experience and technology provided the operational and research base from which enterprise-wide networks and wide area networks of the 90s emerged -- the essential underpinning for enterprise electronic document/records management systems and workflow systems and what is emerging as the period of organizational productivity tools.
Business Systems Analysis and Information Management Projects
The 80s marked some other sea changes in orientation that would become much more pronounced in the 1990s in the form of increasing use of business systems analysis and information engineering tools. Business systems analysis (BSA) involves identifying broad organizational goals and supporting business areas and processes, business process definition and decomposition, and the development of improved processes and information architectures. (A processes is “a set of activities that, taken together, produce a result of value to a customer -- developing a new product, for example.) It helps to rationally link all these things and to drive systems development of supporting information technology architectures.
Since we hear a good deal these days, including from this author, about the desirability of linking records to business processes (BPs) and conducting appraisal at the BP level, it is important to recognize the complexity and commitment involved in undertaking such a program. It is not something to be undertaken lightly or without senior level air cover. It typically involves project sponsorship by the senior executive who is the common manager for all managers of offices that are stakeholders in the business areas to be considered, the commitment of several managers and senior professionals on the project task force, some on a full-time basis for several months, and in-depth interviews of all senior managers and other staff key to the BPs in question. During the 80s and early 90s, the author had the fortune to lead or be otherwise involved in three BPA projects at various organizational levels. The methodologies used were precursors to the business process reengineering (BPR) methodologies and computer-based tools widely used today.
Information management skills come in at the end-game of these exercises using various information engineering and data administration tools to establish how information resources can be organized in such a manner as to promote optimal information sharing and usage. It involves such things as designing and implementing enterprise information directories that rationalize and make it easy for users to discover, access and use divergent multi-media information stores. The design of corporate filing scheme is one of the oldest forms of information management. In a subsequent case, the author managed a project that used the business process model created from the above exercises and linked record series to those processes. It demonstrated that records could be organized according to BP and to a related provenance data base. More recently, the Dutch Government, as part of its “Revolution in Records” project resulted in research led by Tora Bikson that concluded that electronic information is operationally and strategically important to government agencies but lacks policy definition. It offered suggestions for filling that void. Furthermore, it concluded that the development of tools to provide for technological support for records management in diverse mission environments should be undertaken on a priority basis. Finally, it stated:
Finally, context-relevant constructs and methods for the management of electronic records and archives need to be developed. Reliance on paper-based procedures, plus the assumption that electronic records material has a print equivalent that can be managed according to traditional rules has probably delayed progress toward the articulation of new approaches that better suit today’s interactive information environment. 
BSA offers excellent approaches to business modeling that, in turn, can provide an excellent basis for developing information architectures and information directories (including paper and electronic records holdings). However, it is a very complex methodology that involves considerable investment of senior management commitment and time and is not well understood by many people, including most ARM practitioners. As valuable as BSA can be, it also has its shortcomings that are seldom mentioned with the positives. It is a complex approach which might not be easily carried out by many smaller organizations with limited human and financial resources. BPR specialists can be extremely expensive (several thousand dollars per day), are often dogmatic and wedded to their own approach and often take considerable time to orient to the local context. (To be sure, part of the value of BPR consultants is that it is the local context that too often gets in the way of identifying improved ways of doing things.) The BPR approach also very often gives only cosmetic attention to the human factor; e.g., it does not ordinarily make use of social models of the workplace to accommodate both process improvement and human adaptation, and it often relies upon identifying ‘better’ ways of carrying out business processes through BPR teams that consist in part of people who are afraid to come forward with suggestions that they feel may place their own jobs at risk.
I believe that the admittedly attractive and desirable notion of records appraisal at the business process level is better understood in theory than it is in terms of practical application. In my discussions with ARM practitioners around the world, I notice what appears to be a lack of appreciation of the complexities of implementing the theory, and a certain polarization of belief systems regarding electronic document/records management. It is sometimes also framed as the top-down vs. bottom-up approaches. Hopefully the NHPRC funded Pittsburgh project and the Dutch Revolution in Records project will shed a good deal more light on this subject.
The top-down or business process (BP) approach is broadly as outlined above. The bottom-up or docu-centric approach (e.g., as suggested in some of the diplomatics literature) begins at the other end of the spectrum with the document. Simply stated, this approach says that every document, especially every record, is associated with a business transaction and has a document profile which contains essential contextual data about the document and is managed as part of its parent record series.
Somewhere in between, we have computer “application systems” that support specific business process or sub-process, especially where the application carries out specific, usually repetitive, tasks. They are usually highly structured, transaction based systems. They are especially common in the finance business area, such as the payroll, cash management and accounts receivable applications systems. As such they operate above the record series level but below the major business process level. Most such applications systems are amenable to system-level appraisal and automated disposition management. This is not the case for many business processes, since they may involve both structured and highly unstructured documents from sources not easily disposition managed such as email systems which are not oriented toward any specific business process or application as most are not. Moreover, even when operating at the BP level, it is necessary somehow to populate a BP with all of the relevant documents or composite objects as discussed earlier, which means that documents must be identified as to what BP they support and be marked accordingly.
Thus, while stressing one or the other approach may seem to make good sense in theory, in the practical world of systems development and implementation, it is likely that some combination of both approaches will be necessary, and it will vary from case to case, BP to BP. I doubt that theorists in electronic records management see themselves as at opposite ends of the spectrum. Nonetheless, I find that many ARM practitioners are interpreting the literature to mean that one or the other of these approaches must survive over the other. Just as linking records to BP or provenance are not mutually exclusive, the choices among BP-level, applications-systems-level and document-level records management are neither universally applicable nor mutually exclusive. A benefit of computer-based solutions is that you can have it both ways.
These experiences suggest to me that, since the use of technology is largely driven by changes in business processes and work patterns, archivists need to give increasing attention to business systems planning and analysis and end-user work habits. As part of our involvement in such activities, we might also do a little “image enhancement” of our own -- to change the image of the archivist from one purely as the “keeper” of records to one also as the “purveyor” of documents, as noted in my opening comments on the definition of records. The purveyor of documents can also be and be seen to be a keeper of documents; but the reverse is not always as easily seen to be the case. This may be more a question of perception than of reality. To her great credit, although the national archivist of the U. K. is known as the Keeper of Public Records is also very much a purveyor of information. in these hard times, we have to address the perceptions of the profession as well as its realities. Perhaps we should reinstate the idea of the “Remembrancer”, established in sixteenth century London to help preserve the city’s institutional memory. This point is likely to be seen as unimportant if the evidentiary value of information is stressed to such a point that records creators think ARM professionals interests are only interested in the evidentiary (which most users will read as ‘legal’) aspects of records and thus are largely irrelevant to their own interests. I understand the logic and importance of stressing evidence, but caution that it not be overdone and that, as noted earlier, we also ally archives with the use of records for purposes of operational continuity, information management and institutional memory.
I have also heard it said, in advancement of the focus on evidence, that nobody really cares any more about institutional memory, other than historians. In my consulting work, I have not found this to be the case. Much of my practice involves dealing with executives and it is not uncommon that they raise concerns about erosion in their own organizations’ institutional memory, and without prompting on my part. I predict that this concern will grow as we see more downsizing activity, and more use of consultants in place of regular staff. To illustrate, one financial institution with which I had a recent engagement to examine its information management and records management programs had in course of approximately one year replaced its CEO and had the positions of general counsel and financial vice president become vacated. Thus, in a very short period, it had lost a substantial portion of the senior, experienced knowledge of corporate operations. Upon learning about this, I made inquiries to the personnel office and discovered that 40% of its investment officers (the mid-level front line deal makers for investment banking institutions) had been in the organization for less than three years. Remaining executives and investment officers readily offered, in the course of interviews, the importance they placed on the need to provide some document/records system to help them rebuild the institutional memory of the organization. This was an organization that was at the same time contemplating budget and space cutbacks for the records management function. They are now reevaluating the potential for the ARM program in maintaining institutional memory and are reevaluating priorities for their ARM program.
The Turn of the Decade of the 90s
Little attention was paid to email and facsimile by archivists and records managers at the beginning for a number of reasons. Firstly, low budgetary policies and organizational clout resulted in ARM programs being denied PCs, fax machines and email accounts. Secondly, ARM professionals were slow to appreciate the potential for using computer-based systems for managing paper and microform records, slow to make their own needs known to IM&T managers, and slow to gain senior management understanding of the role of ARM in carrying out core business processes throughout their organizations. For their part, IM&T managers did little to try to understand the related systems requirement or to represent those needs to the system development community and thus the market was largely ignored. There were exceptions, of course, such as MINISIS and MICRO-ISIS, but they were few and largely home grown rather than constituting an important market segment where most of the broad-based systems development work takes place. Thirdly, the view was strongly held by many that preservation is a major issue with records that are use something other than a paper or microform storage medium, and this continues to be a concern of many professionals. In addition, it was widely believed that most of the traffic going over email systems was junk mail, mainly in the form of lunch date exchanges and substitutions for “telephone tag”; and junk mail was not worthy of serious records management attention.
Integrating IM&T and ARM, Document Management and Records Management
In 1986, I became chief of a central division responsible for office systems services an organization of about 8000 employees. This included information technology standards, central operations for mini-computer systems, and support services for distributed PCs, printers, local area networks, email, word processing, spreadsheets, etc. -- most information technology except mainframes, telephone systems and long line communications. It was a year of mainly infrastructure building. In the same year, the archivist and records management division were moved from the administrative department to the central department responsible for information management and technology, as a sister division to my own. A year later, as part of a major corporate restructuring, downsizing and consolidation of management portfolios, most of the office technology support services and staff were decentralized throughout the user community, standards concerns were shifted to a policy group and my division was eliminated. I became chief of information services, which picked up the functions of the erstwhile records management division.
I was shocked to discover the paucity of information technology tools assigned within the ARM group and the generally lower educational levels of the staff. Whereas, using an oversimplified measure, the ratio of PCs to staff was nearly 1:1 elsewhere in the organization, it was about 1:25 in the ARM group. I came to the early recognition that I had entirely ignored this function in my previous position, to its considerable detriment. I also realized how little I or anyone else in the IM&T organization understood about imaging or character-based technologies that, with appropriate network facilities, would form the underpinning of document management systems. Like myself, virtually all of my colleagues had been raised on mainframe applications. Many in the IM&T profession, by no means limited to my organization, were just beginning to accept the notion that maybe PCs were here to stay.
We did several things to try to lift ourselves up by the bootstraps and to gain first hand experience in the ARM group with text-based systems that would prepare us to deal more effectively with electronic records issues and opportunities, including:
It takes a multi-pronged attack to make up for lost ground in the ARM group:
obtaining its own IM&T human and technological resources that could be trained in and become totally dedicated to ARM concerns and functions; a number of national and state or regional archives organizations are now doing this, most recently the U. K. Public Record Office.
cross training the ARM and IM&T staff; and
beginning to get training and hands-on experience with the application of the technology to ARM functions and a more mature understanding of ARM functional requirements.
These experiences resulted in my involvement in a leadership role in a major inter-disciplinary study of electronic records by some 30 organizations of the United Nations. Because of my interests in both ARM and IM&T, I was seen as someone who might be least offensive to either group (or at least equally suspect by both). It was a rich learning experience and an opportunity to work with many people in these fields throughout the UN community and with excellent consultants such as David Bearman, Tora Bikson and Charles Dollar and ARM practitioners such as Gertrude Long and Alf Erlandsson and many other members of the Technical Panel. For many of the ARM participants, it was their first experience with PCs and email, their first personal encounter with an electronic record. For many of the IM&T participants, it was their first conscious engagement with a record of any kind. The result of our effort was the publication of the UN report on Managing Electronic Records: Issues and Guidelines, and additional reports of the follow-on group.
Though I cannot escape thinking about them in drawing lessons and thinking about things we can do, I will not go into detail on more recent projects, some of which are reported elsewhere and others of which are still underway. These include: evaluation of a prototype workplace that integrated office building design, office landscaping and furniture, information management and technology, and records management; working with interdisciplinary IM&T/ARM teams in the development of functional requirements for enterprise electronic document and records management systems; designing conferences to bring senior executives, IM&T and ARM managers together to confront emerging electronic document and records policies, issues and opportunities; working on an approach for transferring personal electronic mail files to archives; and helping establish an IM&T function in a national archives organization.
The central lesson for me in all of this has been to understand the essential importance of joining forces between ARM and IM&T professionals, and of course the user communities managing the organizational business processes that produce the records. It is my conviction that none of these groups can resolve electronic records issues by themselves. I learned the hard way, however, that achieving this alliance is something more easily said than done. I have found that what seems to work the best where and organization is just beginning in its efforts to increase collaboration between ARM and IM&T groups is a non-threatening workshop in which representatives from both communities participate together.
Archivists and records managers, reading about litigation over the U. S. Presidential papers in the famous White House email trials in the last days of the Bush Administration also saw living examples of what some had feared for many years -- that email could constitute documents of considerable evidentiary or historical value. The lack of a well considered National Archives and Records Administration policy on electronic mail, and the politicization of the National Archives in bending to pressure by the Bush Administration led to national headlines that read: “Neglect at the Archives,” and a black eye for the ARM profession.
Where is IM&T Heading
Without going into a great deal on emerging information technology, it may be worthwhile to mention a few defining technologies. Of particular interest to ARM professionals are developments in personal scanners, PC chips and server chips.
Full-page personal image scanners have become common on the market for as little as about $1000 (half that if you are willing to forego color), making it possible to improve the readability of documents through the incorporation of graphics, pictures, newspaper clippings, etc., and therefore their attractiveness and likelihood of being read. For the most part, the use of imaging technology has been largely limited to centrally managed, fairly specialized, organizational applications. The use of personal scanners, because of cost, also has been typically limited to such applications as newsletters, even though small, hand-held, partial-page scanners under $300 have been on the market for years. With the increasing availability of full-page scanners that are relatively inexpensive, organizations will be more likely to provide such resources, at first on a shared basis much as they did initially with personal computers, to units throughout the organization. The recent advent of machines that package copying, facsimile and image scanning capabilities into the same box will further advance the move toward decentralized imaging services. This will also remove one of the main natural barriers against moving toward a less-paperful, if not paperless, office. In the past, parallel paper systems were necessary even when electronic systems were embraced for maintaining externally generated documents. If the price gets right, there will be less reason to maintain parallel paper systems, except for legacy records systems, and less reason to maintain paper copies of current documents except for short-term convenience purposes.
The introduction of desktop systems using the new Pentium, 586 and (in April) DX4 100 megahertz desktop computer chip brings a level of computing power not dreamt of by most people five years ago. At the same time, delivery commenced on 6-7 pound notebook computers that package this chip and deliver 75 megahertz (three to four times the speed of what was generally available just a few years ago) and with 500 megabytes (one-half gigabit or several file cabinet equivalents) of storage. Not much after the ink was dry on the first draft of this paper, Intel announced a 90 MHZ Pentium chip for notebooks and IBM announced a Pentium overdrive and a 810 MB removable hard disk for its 755 Thinkpad™ Notebook series as well as a CD-ROM drive that can be swapped into the floppy disc drive of the same machine. Other manufacturers did and will follow through in kind. This puts extraordinary computing power in the hands of people on airplanes, in hotels, and at home -- of such a magnitude as to make it practical to seriously consider the use of a single computer for office, home and travel use. It also fuels the engine of location independent information processing that is needed to permit location independent work. Newer P6 chips are of particular interest to ARM professionals. They are not for the end-user PC, but rather for much more powerful servers that will provide enormously efficient task sharing in client server architectures, a sine qua non for advanced electronic document and records management systems. When ARM professionals see enterprise networks implemented in their organizations, as distinct from independent LANs, they will know that the planning time for electronic document and records systems has just become zero.
In the past, my work has involved attempting to project future workplace scenarios, projecting 10 years ahead for purposes of early identification of potential workplace changes, and 25 years ahead for purposes of office building strategies. That was difficult enough to do for an organization with which I was very familiar. I will not, in this paper, attempt anything as ambitious for a comprehensive view of workplace futures more generally, although elements of those futures may be drawn from many of the lessons that are reported in this paper and the suggested things that we might consider doing in the paragraphs below. However, I will associate myself with some of the projections made by one of the leaders in the field of information management, Paul Strassmannn. Ten years ago -- about the same time I was writing an internal paper for my own organization on “A Scenario for the Workplace in 1995” -- Strassmannn predicted, in an excellent chapter “The Paperless Office” (that I cannot do proper justice to in a short quote), his view of what the beginning of the Third Millennium will be like.
There will be a lot of paper in use in the year 2000. There will be more of it, per capita, than at present because there will be so many more originals from which copies can be made. The information workforce will be more than twice the present size...The quality of electronic printing -- incorporating color, graphic designs, and pictures -- will make this means of communication attractive to use. The “intelligence of printing and composing machines will be of a sufficiently high order to cope with the enormous variety of electronic forms in which originals will be represented. All of this assumes that the present sociopolitical hurdles preventing the exchange of electronically communicated text will be resolved through international standards...we should expect to see the same progress...which now permits home-to-home dialing around the globe.
Paper will not be used for archival storage of routine business records. Optical recording... provides a much better means for the filing of information. Paper will be used for reading, due to its greater human compatibility....VDUs will not replace reading. They will deal with the logic of information search, with composition of text, and with terse, highly structured messages....The “paperless office” will not be one of the outcomes of office automation. Large amounts of paper will continue to be used, even though paper’s archival role will diminish.
Had the likely impact of multimedia, including the inability to fully represent multimedia documents in paper form, been as obvious 10 years ago as it is today, Strassmann might have altered some of his predictions. I take this not as an indictment of Strassmann’s forecasting skills, which I salute, but more as simply an example of how much IT changes in a decade and how risky forecasting really is. In a project in which I was involved in 1990, helping to integrate technology and human factors into the design of a large office building, I concluded that it was hopeless to try to crystal-ball what information technology was going to be prevalent during the projected 50 year life of that building. In the end, my recommendation to the architectural design project team was that the building specifically not be designed to a specific technology, but rather that it be designed as an adaptive building -- design it in a manner such that, even if it adds a bit more to the ‘first costs’ of construction, it will be less costly to alter in the future and probably more than recapture any incremental first costs. As it turned out, because this kind of debate took place before the architects did their construction drawings, the additional costs for making the building more adaptable were negligible. Interestingly, the archivist faces much the same problem, trying to figure out what it will be like 50 years hence in terms of information technology. Again, my only advice would be to assume that there will be many changes in technology over that period. We had better design our information architectures, enterprise networks and electronic document and records management systems to recognize and facilitate future change, e.g., through the use of such strategies as open systems architectures, object oriented systems, portable document formats, application-independent multimedia data bases, etc.
Some of the main conclusions emerging from this collection of personal experiences and other observations are that:
In thinking about approaches to electronic records management, especially at the national levels, we should be careful to take account of differences in national heritage and culture and not simply to be swept up by what is regarded as the way to success somewhere else. Made-in-America solutions should not be rejected out of hand because they might not have all been invented here any more than made-in-Canada solutions should be rejected in the U. S.; neither should other people’s solutions be adopted here if they cannot be effectively made to be Canadian. The variances in the traditional European, U. S. And Canadian approaches to appraisal and the Canadian ‘total archives’ approach as contrasted with the ‘public archives’ approach of most other countries are illustrative of this point. The use of IT is even more subject to cultural and human factors. For example, as was noted in a recent Toronto Globe & Mail article:
“The second annual Gallup survey indicates that 69.9% of Canadians have heard about the info-highway but 61.8% fear it represents a threat to Canada's cultural identity and say they want the federal government to assume responsibility for protecting that identity. 3.5-million Canadians, or 11.9% of the population, have used the Internet.”
It would be a mistake not to realize that ARM functions themselves are very vulnerable in the current budgetary climate, and that they could become the subject of outsourcing. Will the private sector want archivists it doesn’t already have for other reasons? If it does, will archivists find themselves in the untenable position of trying to preserve records of continuing public value when doing so might jeopardize their jobs?
It might be comforting to conclude that to prepare for the next millennium, archivist must simply become conversant with modern information technology -- comforting but unwise. That will be a necessary starting point. However, the more complex demands will be to properly understand the objects of automation through business systems analysis.
VIII. What Can We Do?
Having the business insights, communications skills and courage to become aware of and alert their senior managers to proposals for organizational, technological or procedural changes in the name of business process reengineering or innovation that have important business continuity, evidentiary or social dimensions will be an even more important aspect of being an archivist than has been the case in the past. Becoming skilled in business systems analysis and business process innovation, and in evaluating the implications of change on work patterns and social issues, and being able to articulate related archives and records management considerations will make it possible and more likely that archivists will be invited to the table when such discussions take place, thus enabling them to ensure that the important values of archives and records management that are worthy of keeping are indeed kept. Some of the actions suggested below have already been implemented in some organizations. More particularly we should consider:
As Professional Associations:
As National Archives Organizations:
As Operating Organizations:
As ARM Business Units:
As Designers and Teachers of Archival Study Programs:
If the distinguishing feature of information management and technology in the decade of the 80s was chiefly one of innovating and exploiting individual productivity improvements through the use of technology mainly in the form of individual utility tools, the distinguishing feature of the first half of the 90s has been innovations aimed at work-group productivity: e.g., email, group authoring tools and “roomware”. Imagine for the second half of this decade and the first decade of the new millennium:
REBarry:msw60/ba/assn-ca/aca-pv15.doc: 20 June, 1995
 Rick Barry is an author who consults and carries out workshops on information management and technology and electronic records management issues, strategies and policies. He has been a keynote speaker at numerous conferences on electronic records internationally; email: rickbarry [at] aol [dot] com
 For an elaboration of this subject, see the videotape: Electronic Records in the New Millennium: Managing Documents for Business and Government, University College London, 1995, written and directed by R. E. Barry, the accompanying teaching and discussion guide, “Managing Documents for Business and Government’ by R. E. Barry and Anne Thurston and paper: “Electronic Objects circa 2001: Problems or Opportunities?...Yes”, by R. E. Barry.
 The term “ARM” is used in this paper as a generic abbreviation for organizations and programs carrying out life cycle administration of records.
 The term “IM&T” is used in this paper as a generic abbreviation for organizations and programs carrying out information management and information technology tasks and services. Information management (IM) refers to the exercise of intellectual control over corporate information assets to aid in their easy discovery and use, usually carried out by people using information engineering and possibly business systems analysis tools. Information technology (IT) refers to the hardware, software and communications infrastructure, standards and human technical support necessary for the effective use of information.
 Interval Research Corporation in Palo Alto, CA, (one of the co-founders of which is also, with Bill Gates, a co-founder of Microsoft Corporation) is one of the most advanced information science research centers in the world. The disciplinary mix of the group is very interesting, combining information scientists with physicists with an array of other disciplines. One person in the group comes from the academic field of drama, and spent much of her prior professional career in designing the ‘look’ of children’s electronic games.
 Dictionary of Archival Terminology, ICA Handbook Series Volume 3, Munich, K.G. Saur, 1984, as cited in Management of Electronic Records: Issues and Guidelines, a report of the Advisory Committee for the Co-ordination of Information Systems (ACCIS) Technical Panel on Electronic Records, Chaired by R. E. Barry; United Nations Sales No. GV.E.89.0.15, New York and Geneva, 1990. Whenever the term “business” is used, unless specified differently, its meaning is intended to be inclusive of both the private and public sector organizations.
 Dictionary of Archival Terminology, ICA Handbook Series Volume 3, Munich, K.G. Saur, 1994, as cited in Management of Electronic Records: Issues and Guidelines, Advisory Committee for the Co-ordination of Information Systems (ACCIS), United Nations Sales No. GV.E.89.0.15, New York and Geneva, 1990. Whenever the term “business” is used, unless specified differently, its meaning is intended to be inclusive of both the private and public sector organizations.
 Keeping Archives, Judith Ellis, Ed., Thorpe, a part of Reed Reference Publishing, Port Melbourne, Australia, 1993, p.477.
 The argument I am advancing here is for promoting the use of the information that is maintained in ARM systems. It is separate and independent from the issue of whether documents should be appraised on the basis of the context of documentation as it relates to the actions it reflects versus on the potential future uses of records by historians or others based on the current (possibly short lived) views of the user community. For an excellent discussion of this latter issue, see Jean-Pierre Wallot’s “Free Trade in Archival Ideas: The Canadian Perspective on North American Archival Development” in American Archivist, Vol. 57, Spring 1994, pp. 380-399.
 For a further explanation of the distinctions among these properties see Luciana Duranti’s six-part series on “Diplomatics: New Uses for an Old Science”, in Archivaria, (Numbers 28, Summer 1989 through 33, Winter 1991-92) and her unpublished paper, “Authenticity and Reliability: The Concepts and Their Applications”, presented on September 8, 1994 at the Society of American Archivists 58th Annual Meeting in Indianapolis, Indiana.
 Duranti, Luciana, “Diplomatics: New Uses for an Old Science”, Archivaria, 28 (Summer 1989), p.16, and subsequent paper by the same title (Part V), Archivaria 32 (Summer 1991), pp. 6-7.
 See references to articles by Stuart Haber and Scott Stornetta and by Cirpa below.
 Lacey, Julia S. How to Survive Your Computer Workstation, CRT Services, Inc., Kerryville Texas, 1-800 256-4379.
See the videotape Organization overviews and Role Management: Inspiration for future desktop environments, by Catherine Plaisant and Ben Shneiderman; produced by the Human Computer Interaction Laboratory (HCIL), Center for Automation Research, University of Maryland, (301) 405-2768, 1993. A video summary appears in ACM CHI ‘95 Companion, (Denver, Colorado, May 7-11, 1995), pp. 419-420. CHI 95 Technical Program Video available through ACM. In paper form, see the paper by the same authors and title in Proceedings of the 4th Workshop on Enabling Technologies: Infrastructure for Collaborative Enterprises, April, 1995, CAR-TR-771, CS-TR-4373.
 Stoll, Clifford, Silicon Snake Oil: Second Thoughts on the Information Highway, Doubleday, N.Y., 1995, pp. 46-59.
 Ong, Walter J., S. J., “Knowledge in Time” in Knowledge and the Future of Man, Walter J. Ong, S. J., Ed., Holt, Rinehart and Winston, New York, 1968, pp. 11-13.
 Ong, Walter J., Orality & Literacy: The Technologizing of the Word, Routledge Publishers, London and N. Y., 1993, p. 82.
 Clanchy, M. T., From Memory to the Written Word, Blackwell Publishers, Oxford, U. K. and Cambridge, U. S. A., 1993; see especially, “The Technology of Writing” pp. 114-144.
 One of the author’ clients has a rich collection of analog tapes of radio conversations with missions in the field during disaster operations. It is more than the spoken words of people reporting from the field that give evidence to their actions. Other ambient sounds provide a record of the events in ways that could not nearly as faithfully be represented by second-hand human accounts.
 Courtesy of Charles Dollar, University of British Columbia.
 Lowry, Martin, The World of Aldus Manutius: Business and Scholarship in Renaissance Venice, Ithica, N.Y., Cornell University Press,, 1979, pp. 29-32.
 Ong, Walter J., Ibid., p. 80.
 See two recent EDUPAGE clippings: Newspapers Face Stiff Competition In Online Classifieds: The newspaper industry relies heavily on the $12.5 billion generated through its classified ads last year -- and is finding itself challenged by online upstarts such as Electric Classifieds, Inc. which offers a classified service on the Web. Unlike traditional publishing companies, which have millions invested in physical plant, fleets of trucks, and tons of newsprint, electronic publishers can set up shop for next to nothing. To combat this growing gang of competing Davids, the newspaper Goliaths are launching their own online efforts, but they may be overlooking the obvious, according to the editor of an electronic journal on online media is: "Newspapers have a tremendous advantage, if they don't blow it, and that's the infrastructure to take the ads, run the ads and bill for the ads." (Forbes 7/3/95 p.80); and: SCIENTISTS LEAD THE WAY IN ONLINE PUBLISHING: Scientists who used to rely on print journals for research sharing and peer review increasingly are turning to the Net, and the $4-billion technical publications industry is worried. The venerable New England Journal of Medicine is sticking to its guns -- an editorial to be published June 22 says it plans to "apply the same rules to Internet that apply to publishing anywhere else." In other words, if the article's appeared on the Internet, it won't be considered for publication. But other journals are looking at the numbers and deciding they can't afford to be left out. "Costs are up, postage is up, and ad revenues are down," says the American Medical Association's president for publishing and multimedia. "You can't grow enough new revenue sources. We've got to look at electronics as the future." Some scientists worry that bypassing the rigorous vetting process used by the journals will result in "low credibility, instant regurgitation." But others contend the peer review process enabled by electronic publishing can be just as thorough, and far more efficient. "We've only begun to scratch the surface of how much more effectively we can communicate," says the editor of Science. (Business Week 6/26/95 p.44). Source: EDUPAGE 6/22/95. EDUPAGE is a free electronic news clipping service covering the IM&T field. To subscribe, send a message to: email@example.com and in the body of the message type: subscribe edupage and your name.
 Adleman, L. M., Science, Vol. 266, 11 Nov. 1994, p. 1021.
 Baum, Eric, “Building an Associative Memory Vastly Larger than the Brain,” Science, Vol. 268, 28 April 1995, p. 583.
 This technology is not likely to be suitable for quick turnaround information queries, but could be very useful for complex research searches over massive information stores where it will not be crucial if it takes a few days to receive the search results. It has the potential of dealing with large and complex data searches slowly that would not likely be possible to perform today in any length of time.
 Considerable research and development work related to this subject is being carried out under the Text REtrieval Conference (TREC). See, for example, “Information Retrieval Systems for Large Document Collections”, by Alistair Moffat and Justin Zobel (University of Melbourne, Australia) in Overview of the Third Text REtrieval Conference (TREC-3), Donna K. Harman, Ed., U. S. Department of Commerce, Technology Administration, National Institute of Standards and Technology (NIST), NIST Special Publication 500-225, April 1995, pp. 85-94.
 The term “composite content object” or CCO is a term used by Jon Stewart, a consultant to NIST in an unpublished paper on document architecture in which the document is characterized as the integration of it component content objects (CO) of text, sound, video, etc.
 “Business systems analysis” (BSA) refers to the use of information engineering tools to model organizations for purposes of developing information and information technology architectures and for carrying out business process engineering with the aim of improving organizational performance against business aims. A “business process” is a set of activities that, taken together, produce a result of value to a customer or carry out all or part of a specific business aim.
 “PowerPoint™” is a trademark of the Microsoft Corporation.
 Dollar, Charles M. Archival Theory and Information Technologies, Macerata; Universita degli Studi de Macerata, 1992.
 Duranti, Luciana in paper to be published in the forthcoming issue of Archivaria: “The Thinking on Appraisal of Electronic Records: Its Evolution, Focuses, and Future Directions”.
 Haber, Stuart and W. Scott Stornetta, “How to Time-Stamp a Digital Document,” in Journal of Cryptology, Vol; 3. 1991, International Association for Cryptologic Research, pp. 99-111. For a lay explanation, see “Electronic Time-Stamping: The Notary Public Goes Digital” by Barry Cirpa in Science, the journal of the American Association for the Advancement of Science, Vol. 261, 9 July 1993, pp. 162-3.
 See EDUPAGE clippings:
POST OFFICE TEAMS UP ON CRYPTOGRAPHY: The U.S. Postal Service is working with software firm Premenos Corp. To develop public key cryptography for verifying the identity of e-mail senders and the integrity of electronically transmitted documents. Meanwhile, Verisign Inc. is developing a rival system that not only checks the digital signature, but also scrambles the content of the message. A number of companies, including Visa International, Mitsubishi and Ameritech, have invested in Verisign's technology, and Apple Computer and Netscape Communications have signed on as customers. (Wall Street Journal 6/22/95 B7) Source: EDUPAGE: 6/22/95; and
DIGITAL SIGNATURE GAIN LEGITIMACY: A law recently passed in Utah recognizes digital signatures as legally binding, and legislators in California and Washington are considering following suit. The Utah law is based on public key encryption, where companies and individuals register their public keys with a certification authority, which then uses them to decode messages created with private keys, verifying the senders' identities. Computer security companies, banks and the U.S. Postal Service all are expected to offer certification services. (Information Week 5/8/95 p.24) EDUPAGE is a free electronic news clipping service covering the IM&T field. To subscribe, send a message to: firstname.lastname@example.org and in the body of the message type: subscribe edupage and your name.
The terms hypersound, hypervideo, etc. Are simply extensions of the terms hypertext and hypermedia, the coining of which is attributed to Ted Nelson. On April 17, 1991, I visited Nelson to learn more about the elusive Xanadu system that he had created and, with a team of followers, had been working on its development for years. We talked about these words and his vision of a world of Xanadu document kiosks. As is his custom, he recorded and kindly sent me a transcript of our conversation. On the subject of hypertext, he put it this way:
I coined the terms hypertext and hypermedia...in ‘65...Again, these terms weren’t there, but we would have non-sequential documents and many data types. A server would be holding them in some authenticatable form...And then there would be a network of these servers and the user would send for arbitrary fragments defined by these. You would send for whole documents because you might not necessarily want the whole document. ...And the whole idea -- what hypertext really becomes is simply the idea of connected data. Because it means that you can leave things and follow connections....And so a data service, a data storage, engine and service protocol that will provide -- that will allow us to cross between documents across different kinds of bridges of linkage and what we call transclusion. So the two primatives (sic.)...are linkage and transclusion...Linkage means any connection between something on the left and something on the right. So this can be a piece of text, this can be a section of data and this can be an illustration...we’ll have a registry, so you can register a new link any time you’d like....Transclusion means virtual inclusion by reference. So that your document can virtually include a piece of that, a piece of that, a piece of that and a piece of that without copying. And that has become -- that has immense powers. And so it becomes a fundamental mechanism....In other words, the document is entitled to see something that it can link to and then transclude it. That means that you now have a new principle basis for all data. Or what I like to call a landscape of data and the applications just become windows or lenses which manipulate the data and work at it. And look at it....Hypermedia is ways of looking at this. And the style of interaction which you want is simply chosen by the user.
 POLIKOM - management, a 10-minute videotape (available in English and German) on the use of multimedia systems in the conduct of German Government business, 1992, Institute for Applied Information Technology, GMD, Schlos Birlinghoven, Postfach 13 16, D-5205 Sankt Augustin 1, Germany, 1992. For further information contact Uta Pankoke-Babatz, email: <email@example.com>.
 The author led the project and the technical work was led by Mr. Herbert Goertzel of the G. C. Dewey Company of New York.
 “Roomware” refers to systems that involve the integrated design of hardware, software and physical facilities. Teleconferencing rooms are rudimentary examples of roomware. More advanced examples, often generically referred to as “Arizona Rooms”, named after the university where the technology was created, include a dozen or so workstations, usually in a U shape, each of which has a low profile PC, a large screen projection capability, and software that facilitates presentation of design options, issues, questions or other agenda items and permits anonymous messages to be sent by the participants and projected in real time. The software facilitates grouping of responses, polling, etc. A manager or facilitator leads the discussion. “Electronic flip charts” are automatically created that could be used as a record of the meeting.
 Very recently, in response to a court challenge, it has been revealed that the Joint Chiefs do not keep records of such actions beyond their initial usage.
 This COSATI Panel was under the leadership of Dr. Ruth Davis, then Director of the National Library of Medicine which housed one of the first and most successful large text-based systems, MEDLARS.
 For an excellent review of some current research projects that use graphical approaches to information management, see the following video documents all contained in the video tape HCIL Open House ‘93, Catherine Plaisant, Ed., Video by John Reesch: “Dynamaps: dynamic queries on a health statistics atlas,” by Catherine Plaisant and Vinit Jain; “Hierarchical visualization with tree maps: making sense of pro basketball data,” by Dave Turo; “Tree VIZ™: file directory browsing,” by Brian Johnson; produced by HCIL, 1993 produced by the Human Computer Interaction Laboratory, Center for Automation Research, University of Maryland, (301) 405-2768, 1993.
 Negroponte, Nicholas, Being Digital, Alfred Knopf, N.Y., 1995, pp. 130-131.
 Reported in: “Democratic Automation,” by Jonathan Schlefer in Technology Review, July 1983; "Staff Participation in Office Systems Design: Two Case Studies at the World Bank," by R. E. Barry in Office Automation, Jekyll or Hyde?, Working Women Education Fund, Cleveland OH, 1983; and in The Silicon Jungle, by David H. Rothman, Ballantine Books, N.Y., 1985, pp. 132-134.
 This is not an uncommon situation today in many organizations where information technology decisions have been decentralized to end-user units, making it extremely difficult in the absence of insightful and strong top management to properly design enterprise document/records management systems that are interoperable across organizational units and different generations of technology.
 This is similar to what has happened 10 years later with multimedia technologies which gained first consumer application in the form of children’s games such as PAC MAN. Except for marketing functions, only in the past year or two have businesses begun to take multi-media seriously as an opportunity for improved business communications.
 Hammer, Michael and James Champy, Reengineering the Corporation: A manifesto for business revolution, Harper Business, NY, 1993, p.3.
 For detailed discussions of business process reengineering and suggested methodologies, see Hammer and Champy; Daniel Morris and Joel Brandon, Re-engineering Your Business, McGraw-Hill, N.Y., 1993; James Donovan, Business Re-engineering with Information Technology, PTR Prentice Hall, N.J., 1994.
 For a detailed discussion of information engineering, including methodologies, see James Martin’s Information Engineering, Book 1, Introduction, Prentice Hall, N.J., 1989, and Book II, Planning and Analysis, and Book III, Design and Construction. For an excellent guide to the application of information engineering tools for public sector financial management, see Information Systems Strategies for Public Financial Management by Hywel M. Davies, Ali Hashim and Eduardo Talero, World Bank Discussion Paper # 193, 1993. For an example of the use of one specific tool, state transition analysis, as it might be applied to the management of electronic records, see R. E. Barry, “Electronic Document and Records Management Systems: Toward a Methodology for Requirements Definition” in Information Management & Technology, the journal of Cimtech and UKAIIM, Herts, U. K., Vol. 27, No. 6, November 1994, pp. 251-56.
 An extensive training package was subsequently developed by the Dutch Government, “Analysis of Business Processes and Records Management Requirements,” outlined in an unpublished manuscript by Jan Acterbergh, Ministry of the Interior of the Netherlands, Division for Coordination of Documentary Information, The Hague, 1994, Tel: as above; Fax: +31 70 302 7600.
 Reported in Preserving the Present: toward viable electronic records, by T. K. Bikson and E. J. Frinking, European-American Center for Policy Analysis (RAND), Delft, The Netherlands, Sdu Publishers, The Hague, 1993, pp. 16-17. For further information, contact: Administrative Coordination and Information Systems Department, Ministry of Home Affairs, Postbus 20011, 2500 EA Den Haag, The Netherlands, Tel: +31 70 302 7677; Fax: +31 70 3106937; Email X.400 address: <c=nl/a=400net/p=idn/o=min biza/s=biza ibi>.
 Cain, Piers, “Robert Smith and the Reform of the Archives of the City of London, 1580-1623”, London Journal, Vol. 13, No. 6, 1987-88, University College London.
 Bearman, David, Electronic Evidence: Strategies for managing records in contemporary organizations, Archives and Museum Informatics, Pittsburgh, PA, 1994, ,p.2.
 This argument was a strong element of President Bush’s unsuccessful defense against litigation to prevent destruction of White House email records in the last week of the outgoing Administration’s existence.
 Regrettably, this turned out not to be a remedy that could be counted on as most operational people chose not to designate most messages as “official”, even when they were of considerable substance.
 See unpublished paper by R. E. Barry, “The Case for a Senior Information Management and Technology Position in Government Archives and Records Management Offices for Electronic Records” August 8, 1994, reported on at the annual meeting of the Society of American Archivists in Indianapolis, Indiana, September, 1994.
 Managing Electronic Records: Issues and Guidelines, United Nations Administrative Committee for Coordination of Information Systems (ACCIS), New York and Geneva, 1990.
 Barry, Richard E., “Addressing Electronic Records Management in the World Bank” in Electronic Records Management Program Strategies, Margaret Hedstrom, Ed., Archives and Museum Informatics Technical Report No. 18, Pittsburgh, PA, 1993, pp. 19-29.
 Reported in “Real Problems, Real Solutions” by Bronwyn Fryer in PC World, September 1993, p.35.
 Barry, R. E., “Electronic Document and Records Management Systems: Toward a Methodology for Requirements Definition” , Information Management & Technology, journal of Cimtech and UKAIIM, Herts, U. K., Vol. 27, No. 6, November 1994, pp. 251-56.
 Barry, R. E., unpublished paper “The Case for a Senior Information Management and Technology Position/Function in Government Archives and Records Management Offices for Electronic Records, August 8, 1994 (Rev.),” presented at the September, 1994 meeting of the Society of American Archivists in Indianapolis, Indiana.
 “Neglect of the Archives”, Washington Post, May 8, 1995, p. A 20.
 Strassmann, Paul A. Information Payoff: The Transformation of work in the Electronic Age, The Free Press, A Division of Macmillan, N. Y., 1985, pp. 176-7.
 See, for example, Michael Hess, “An Incrementally Extensible Document Retrieval System Based on Linguistic and Logical Principles,” in the Proceedings of the 1st Annual International ACM SIGIR Conference on R & D in Information Retrieval, Copenhagen, June 1992, p. 111 and associated bibliography.
 See Jean-Pierre Wallot’s “Free Trade in Archival Ideas: The Canadian Perspective on North American Archival Development” in American Archivist, Vol. 57, Spring 1994, pp. 380-399.
 Toronto Globe & Mail, April 4 20, 1995, p. B1. Source: EDUPAGE (4/20/95); to subscribe: send message to <firstname.lastname@example.org>. In body of the message type: “subscribe edupage” followed by your name.
See Overview of the Third Text REtrieval Conference (TREC-3), Donna K. Harman, Ed., U. S. Department of Commerce, Technology Administration, National Institute of Standards and Technology (NIST), NIST Special Publication 500-225, April 1995.