Managing Distinctions: Enterprise Information, Document, Records, Knowledge and Content Management
by Richard E. Barry
N.B. This paper was originally published in Records and Information Management Review, Vol. 18, No. 2 / February 2002 and is republished here with the kind permission of the publisher and Richard Cox, editor of RIMR.
These are words of one contributor a few years back on an Internet discussion list for professional records managers, archivists and other information management professionals when ‘knowledge management’ (KM) was beginning seriously to emerge as an important business concept and approach to organizational performance. Others, doubting the potential impact of knowledge management on recordkeeping, thought that it didn’t matter that much whether KM was real or just another buzzword:
· If it’s recorded, we’re interested, but then it’s just like any other kind of information…If it isn’t…we’re not interested.
But there were other voices, not so dismissive:
· Knowledge management (KM)…will change the way we manage records.
· With downsizing, the "knowledge" is walking out the door, so records become more than just dust catchers.
· What the heck is it really, and how do record managers fit in? It seems to me that records managers should be right in the middle of knowledge management, but I don't think it's so, Horatio!
This kind of exchange is not unique to ‘knowledge management’ discussions. Rather, it illustrates the professional confusion that arises with the introduction of most new concepts or terms. A recent discussion of ‘content management’ ensued on one list that served to highlight the fact that, until something like content management is understood and established in the workplace, it may be very difficult to find papers on the subject in the professional literature, especially ones with any consideration of potential recordkeeping implications. The absence of such dialogue can result in substantial setbacks for organizations and archives and records management professionals trying to sustain a trustworthy recordkeeping environment.
The purpose of this essay is to discuss some key distinctions among important enterprise knowledge, content, information, document and records management concepts and supporting technologies in current use, and to reflect on some implications for records managers and recordkeeping. It is also to reflect on how these concepts often grow out of and reinforce one another. Before discussing those matters, however, let us consider how we learn about new ideas, technologies and product lines, so that we might better prepare ourselves for the inevitable new ones yet to come.
Where does one look for early answers?
It seems that no sooner than one new business strategy, organizational ‘paradigm change’ or technological innovation is barely digested that a new one floods into the workplace and market. In many of these cases, archives and records management (ARM) and other information management (IM) professionals find themselves asking: is it just an old house with fresh paint? Will it last any longer than the latest skirt lengths or tie widths? In short, is it substance or fluff? And, even if it is real, does it particularly matter for recordkeeping? In the words of another discouraged list discussant:
· It appears that "knowledge management" is a buzzword about everything! …I’m not sure how to incorporate KM into my records strategies – or even if I should.
Who can one look to for answers that are not self-serving, particularly early in the game, before there are established standards? It is highly unlikely that answers will be found in professional literature, possibly with the exception of journals associated with such professional groups as the Society for Information Management (MIS Quarterly) and IEEE Computer Society (IT Professional). But do not expect to find recordkeeping impact statements in those places. Nor is one likely to find answers in conference presentations at ARM professional meetings in the early days of most business or technology innovations.
Vendors and technology ‘early adopters’ may be among the only sources of information on technological innovations at the early stages. Richard Cox notes, “[T]he web offers an amazing ability to discover information about vendors offering services in archives and records management.” He later cautions, however, that there are few places where one can go for comparative information on the strengths and weakness of such offerings. Examples of this are used here to illustrate differentiators. Whether and to what extent reality lives up to advertising, I will leave to the Doculabs, Gartners, IDCs, and Yankee Groups to evaluate – groups that do offer comparative product information once a market has begun to mature. This need not lessen, however, the insights advertising can offer regarding capability differential statements to help make the kinds of distinctions with which this essay is concerned.
Because of the critical role software developers play in the way records are created, professional ARM organizations might consider rethinking their relationships with the that community, including more frequently breaking tradition and inviting vendors to make presentations at ARM conferences as was done was the NAGARA 1997 Conference in Sacramento. Such presenters can be asked to speculate on how a particular innovation or new product type might impact on recordkeeping; or they might be coupled with an ARM professional who fills that role, as was done in the NAGARA conference keynote session, where Chauncey Bell, Senior Vice President of Business Design Associates, a newcomer to records management, challenged some of our ritual thinking about the role of ARM professional and this author reacted.
At the same conference, Raul Medina Mora, Senior Vice President and Chief Scientist for Action Technologies, Inc., offered some excellent views about recordkeeping considerations for workflow management (WfM) systems. Both speakers remained for most of the conference to listen and engage other participants, breaking a habit of many IT speakers of parachuting into conferences to give their presentations and then departing immediately thereafter or simply remaining in their exhibit booths and not partaking in the conference dialog. In the case cited above, Mora invited participants to communicate with him following the conference on recordkeeping concerns in his leadership role in the international WfM Coalition Standards Committee.
Similarly, at SAA’s 1999, Pittsburgh conference, William Hooton, then-President of Tower Software USA shared a keynote session on the development of records management applications with Steve Matsuura of DoD’s Joint Operability Test Command, who addressed the certification of records management application (RMA) software under the 5015.2 standard, and Authentica CEO Lance Urbas spoke at the Cohasset MER 2001 conference on access control of records through advanced encryption technology. Cohasset’s MER and ARMA Houston/Tarian’s e-solutions Conferences are considerably ahead of most professional associations in integrating private sector speakers into their programs and in attracting key developers to register exhibit booths; although even those conferences are still looking for the right balance in fairly providing time for exhibitors to demonstrate their wares.
However, vendors are not always welcomed in substantive conference sessions unless they purchase conference registrations. Users tend to think that vendors should be grateful to pay for whatever they do for us. Yet, just the cost of creating, transporting, installing an exhibit, paying the exhibit fee, possibly hosting a continental breakfast or coffee break and providing a demonstration and refreshments suite, will add up to several thousand dollars. Then they have to be tended by expensive people, limiting the number of staff that vendors can afford to send to conferences and the number of conferences they can afford to exhibit in each year. When, by necessity, limited vendor employees are on hand, it is usually necessary for them to remain at their booths to be present for visitors. If professional associations want to have vendors become more conversant and sympathetic with ARM issues, they should ensure that exhibits are closed for key sessions (as for example was done in the e-Records Solutions 2001 Conference) and that exhibitors are encouraged to attend those sessions without charge. Conversely, breakfasts and breaks should be held in exhibitor spaces and reasonable periods should be specifically scheduled without conference sessions to encourage conferees to visit exhibit booths.
To illustrate the potential value of industry sources, one would have found it difficult when enterprise information portals (EIPs) were emerging to find good resources in the professional literature on what they were. Even today, what exists on the subject is principally in the trade literature, although some of us have regularly used workshops and conference presentations to alert the ARM community of such upcoming technological innovations and their potential implications for recordkeeping. An excellent analysis of the Northern Light EIP was prepared by IDC, a firm that analyses IT industry trends and products. Even if one wished to challenge analyst Susan Feldman’s conclusions about the specific product (which this author does not, except possibly with respect to her less vigorous IT usage of the term ‘archives’), one would gain the insights of at least one knowledgeable industry analyst about the key features of the best EIPs. Such insights help in distinguishing EIPs from other information product classes, such as EDMS. Feldman says:
EIPs are high-end systems that …act as a central gateway to any information the enterprise generates or needs...The enterprise information system comprises all the functions that create and manage collections: acquiring information from diverse sources such as Web crawlers or authoring tools, indexing materials, organizing those materials into a structure that is pertinent and understandable to the organization, maintaining versions and archives, and, most importantly, making the information readily accessible through advanced, easy-to-use search and browsing tools. 
The questions always boil down to: what, if anything, does this mean for my profession, my organization and my job? What do I need to know about this? Who do I need to warn and what do we need to do? Unfortunately, the answers are rarely unambiguous. Often there is both substance and hype in the same term depending on who is using it and for what purposes.
Where else can one seek information that will shed light on some of the above questions? After the concept or product line matures, some will be found in the literature (explicit knowledge in the KM vernacular) and in the heads of the relatively small number of people (in the form of tacit or implicit knowledge) who are actually implementing such organizational and technological innovations in a recordkeeping context. For example, the Indiana University Electronic Records Project represents one of the first projects implementing an EIP system (MyEden) and an enterprise resource planning (ERP) system (PeopleSoft™) in a trustworthy recordkeeping environment. As ERPs are a common approach to the replacement of legacy financial and human resource applications, and represent one of the largest producers of records in an organization, capturing ERP records into a trustworthy recordkeeping environment should be a top priority for executives, program managers, CIOs and ARM professionals.
As valuable as they are, the views of ARM professionals reporting on the implementation of innovative technologies are not without their own issues. ARM professionals reporting on their own projects may be no less disinterested than vendors promoting their technology and thus may also experience conflicts of interest. Are implementers as likely to be as forthcoming about the down side of a project experience as they are about its advantages? Not if they fear losing the next installment of grant money or budget, or if it might expose poor design or implementation decisions for which they were responsible. Are they any more likely to openly profess project failure? Probably not if they are in an organization that penalizes people for making ‘mistakes’ in ways that might be career limiting.
Even if they are willing to candidly share valuable case information, which obviously other information professionals need to know, will they be allowed to publicly say so in a conference presentation or professional journal? Maybe not if their papers require prior management clearance. Still, objective case studies from managers of implementation projects can be excellent sources of information. These questions give rise to the question of whether conference papers are placed on websites or in CD/printed proceedings and the question: would speakers be more candid if their presentations were not published? Obviously, as one who electronically publishes my own papers and several others, I favor publication where possible. Perhaps there is room for more critical analysis of projects in a conference panel setting that would not be published with the other conference papers.
Although non-ARM professionals are rarely invited to make conference presentations or write papers for ARM journals, the users of implemented projects can constitute one of the best sources of information on how the projects really went, if allowed to speak in public about it. In the end, reporting from independent consultants about their experiences with various technologies in an ARM setting, without revealing client identities can yield some of the most informative insights. This is why a relatively small cadre of senior and well-respected consultants regularly appears on conference agenda in the US (Rae Cougar, Julie Gable, Bob Williams), Australasia (Connie Christensen, Mike Steemson), Canada (Terry Cook, John McDonald) and the UK (Marc Fresko, Tony Hendley, Peter Emmerson).
In light of the above discussion, and since most technologies with a heavy recordkeeping impact are emerging from technology firms, ARM professionals must find better ways to relate to such organizations. This will require such steps as:
· incorporating into ARM organizations IT professionals who can become specialists in recordkeeping technologies;
· integrating the ARM function with the broader IM functions at the policy, systems and/or organizational levels;
· keeping tuned into developments in relevant advanced technology research communities and in current business technologies and practices, such as those that are the subject of this essay; including professionals from other disciplines in conference programs;
· providing (and taking) the time necessary at conferences to visit vendor exhibits; and
· giving more prominence to business special interest groups, such as the SAA’s Business Archives Section (BAS). In recent years, there have been some outstanding presentations at BAS pre-conference meetings that, in this author’s opinion, were important enough to have been put forward as conference sessions.
· IM vs Information Technology (IT)
· Information management (IM) vs Records Management (RM)
· Records vs electronic records
· Enterprise management vs Enterprise Systems
· Knowledge, Information, Documents and Records
· Content vs Information and Knowledge Management
We should recognize up front that these concepts are not mutually exclusive or always a complete departure from traditional approaches. Often they represent improvements along a continuum of ideas and technologies. Knowledge management makes use of information management and portal technologies – and more. Content management (CM, or more often ECM for enterprise CM) can be implemented separately but can greatly support knowledge management. Advanced portals try to put it all together.
Information Management vs Information Technology
Let us begin with the more traditional and better-understood concepts of information management and information technology before moving on to records management, knowledge management and content management. There are important distinctions to be made even here. In my experience, a great many people with the title ‘Chief Information Officer’ (CIO) are by education and experience really ‘Chief Technology Officers’ (CTOs). And most related organizational units with which I have come into contact are well endowed with information technology experts but have few if any information management specialists. Perhaps this is in part why many CIOs consider that the balance between IT and information management in their own organizations is out of whack – too oriented to IT and too little to information management. This was always an important distinction to make; but it is much more so the case today as more organizations move to knowledge management, content management and electronic records mandates, especially when the KM, CM and ARM functions are not collocated with the CIO function.
The term information management (IM) is often wrongly used synonymously with, or subsumed within, the term information technology (IT). It is important to distinguish between these terms, because most organizations are deficient in the former, potentially leading to a situation in which technology interests rather than management and operational information needs dominate management attention, resource allocation and related design decisions.
Information management is the means whereby the existence of needed information can be discovered by managers, action officers and support staff. Unlike later content management systems, vanilla IM systems require the user to know what s/he is looking for and to go after it. IM fosters the effective use of information for specific business purposes and information conservation, sharing and recycling throughout an organization and among organizations with mutually supportive aims. The practice of information management in modern enterprises, whether in the public or private sector, may also involves data administration, information engineering and business systems analysis to link organizational, processes and information models. This requires understanding and analysis of the mission and objectives of the organization (or business system) and the development of families of information categories that are related to one another through their inherent or organic connections with business aims and processes. The terms information engineering and information architecture are becoming more frequently used to embrace IM, as new tools have developed to serve IM needs in recent years. These distinctions are more than semantics though they are that too. I once was associated with a department called “Information, Technology and Facilities” that emphasized the importance of integrating the technological and physical workplace. I always made a point of placing the comma between ‘Information’ and ‘Technology,’ probably a nuance that probably went unnoticed by most people. Beyond semantics, the kinds of people and skill sets are different, with IT specialists more likely coming from computer science educations and technical work experience and IM specialists more likely coming from a liberal arts, information or library science/studies background, including many ARM professionals.
When articulated in graphic form with explanatory text, the information resulting from the above analyses may be used to describe the organization's information architecture. Information architecture shows what information is needed for, and is produced by, its key business processes and how the sources of internal and external information relate to one another in the context of this particular organization, which may differ from how similar information is used in other organizations. It articulates information standards and depicts directories or maps.
Directories may be of information stores managed by the organization, whether internally or externally generated. They may also be in the form of tools for the effective navigation, use and disposition management of an organization's information assets. Directories may be of people with needed expertise in fields of importance to the organization. They may also be of physical assets – buildings, analog files, etc. The most effective directories embrace, and where appropriate link, all of these assets. Files schemes and records series familiar to ARM professionals are special purpose directories that should be an organic part of the organization’s information directory system, although it may be argued that the days of records series are limited in light of other alternatives for recordkeeping, such as business models and rich description and metadata.
Because of the manner in which business systems analysis (BSA) is carried out to model various aspects of the organization and develop its information architecture, it is referred to as a ‘top-down’ approach. It uses various BSA techniques to ‘decompose’ the organization from the top, first examining:
§ the organization’s ultimate purpose/mission; then its:
§ aims and objectives,
· major business areas or functions (operations, human resources, finance), including but not limited to organizations by the same names
· supporting business processes (hire staff) and sub-processes (check references) and related workflows that typically involve multiple units,
· information needed and produced by processes and related
· information categories and other metadata
Those may, in turn, assist in: linking information assets (including records) to corporate aims; developing directories and schemas; and determining appropriate information technology. This approach also has the potential for assisting in making records appraisal and disposition management decisions based on macro-level value assessments and possibly value-chain like analyses at the process level. Information management is important because it facilitates the intellectual or logical control over information assets that is necessary for the effective management of information toward strategic ends, including electronic records.
More often, however, a ‘bottom-up’ approach is used – inventorying existing records in individual offices and focusing on record content and creating organization rather than the importance of the business aims, processes/functions and context in which they are created. This approach concentrates more on the physical control of records and, in my opinion, inadequately on their intellectual control. This is partly because of the fact that up until recently, most recorded business information has been created in the form of paper documents. Digital creation systems can facilitate the management of records in much more powerful, efficient and effective ways if appropriate provisions are made for their trustworthy control, integrity, preservation and long-term access, in the face of rapidly changing technology. Such systems also facilitate the ‘macroappraisal’ approach outlined above.
Information technology is about hardware and software infrastructure. In the Internet age it often also embraces communication technology, described with terms taken from the transport sector – information super-highway, info-bahn, info-strada. IT makes it possible to do a much better job of information management today than was possible even a few years ago, but it is not to be confused with information management any more than information is to be confused with technology. Just as organizations must develop information architectures to understand how they can most effectively organize and use their information resources, they must also develop technology architectures specifically geared to serve related business needs and to articulate technological/communications standards..
The higher the level of organization to be served, the more generic the technology architecture must be. Thus, at a national or international level, the technology must be such as to support the most diverse needs in terms of kinds of information to be transported and common protocols for access to the network. For example, the Transmission Control Protocol/Internet Protocol, better known simply as the "TCP/IP" standard, at the transport layer of the Internet technology architecture was created to permit access to, and use of, the global Internet worldwide. At lower levels, such as the individual enterprise level, technology standards may be required at the application level, such as word processing, email, document management, recordkeeping, etc. (Ironically, in the standards literature, the application layer (interoperability level) is seen as at the top of the standards reference model and the transport layer (interconnectivity level) at the bottom end.)
The traditional relationship between information and technology – that business needs should dictate information requirements that in turn should drive technology decisions – is being challenged. This age-honored axiom has been giving way to a new reality in recent years as managers see technology as a way of streamlining operations. Although large companies may be unwilling to openly discuss this subject, many are using technology to drive their businesses and redesign their organizations. This is true because in a free market economy, there are tremendous pressures to exchange capital (technology) for labor (people).
This is happening especially where front-office, customer-facing systems (a.k.a ‘business-to-customer (B2C)’ systems such as customer interface management (CIM) or relationships management (CRM) systems) are integrated with back-office (ERP) systems (a.k.a. ‘business-to-business (B2B) or business to enterprise (B2E) systems). In the old days, in-house systems programmers became experts in one of their organization’s major functional area systems (finance, personnel, operations), and maintained such systems in ways that supported changing organizational needs. This was done often at high cost in terms of patchwork systems and turf protection on separate mainframe systems that resulted in ‘stovepipe’ information systems, unable to ‘speak’ to one another, usually with little current documentation and an enormous and risky dependence on one or two system programmers who have it in their heads.
Integrated enterprise systems are changing that and now such systems are often integrated and cohabitate on single mainframes or networked servers. Now, many large companies (some well known to all of us) are taking the position that these new major enterprise systems will not be maintained in house. (Not a bad approach, in principle, since most user organizations do not have the level of expertise needed to fool with that kind of software.) Not only will such systems not be massaged to adapt to the internal organization, but just the opposite: the internal environment, organizational structure and staffing will bow where necessary to accommodate upgrades in these large systems. If an upgrade means that some sub-process should be changed and automated in ways that might reduce the need for organizational entities and related people, policies or procedures, then so be it. It is just one more way that organizations can gain a competitive edge even for a short period. Management says: why not? Often, no one has a good reason to the contrary.
Many aspects of recordkeeping make prime candidates for automation, in some cases enhancing recordkeeping. Archivists and records managers need to be discerning in making those distinctions honestly for themselves, their profession and their particular organizational settings and either make the case against such automation or lead the way. If ARM professionals are seen simply as Luddites, their influence will be diminished when there is a good case against automation or outsourcing of ARM functions.
The emergence of ubiquitous digital recordkeeping
In the recent five years or so, we have seen the emergence of electronic recordkeeping systems such as those that were originally certified under the US DoD 5015.2 Records Management Applications (RMA) standard or that were short listed as a result of RFI/RFP evaluation exercises by national archives (e.g., in Australia and the UK) as being trustworthy RMAs. If you had an EDMS, you needed to couple it with a recordkeeping system to get it right. The 5015.2 standard did a great deal to get the attention of software developers concerning the importance of recordkeeping and recordkeeping functionality in systems. Now, some of those RMAs have become full-fledged EDMS in their own rights with full traditional document management capabilities and industrial-strength recordkeeping functionality, as well as possibly workflow and knowledge management supporting capabilities. In the remainder of this paper I refer to such systems as EDMS Plus or EDMS+. So what we now have are: EDMS without serious recordkeeping functionality; RMAs that are lacking in rich EDMS capabilities; and EDMS+ that have both. Unless an organization has some very unusual needs, they shouldn’t require more than one EDMS+ to do it all.
Integration of recordkeeping functionality is further developing as EDMS+ vendors make deals with developers of noncompliant recordmaking systems. There are a number of ways that recordmaking systems can be matched up with recordkeeping systems. First, we are all familiar with the old “OEM” (original equipment manufacturer) arrangement whereby one company purchased hardware from another and packaged it under its own name brand – e.g., PCs. This approach has begun to appear with recordkeeping software. Something with the look, feel and functionality of Tower Software’s TRIM Captura™ (because it is Captura™) is repackaged by a former competitor such as Tower Technologies’ (no relation) in its T2E™ system. It works seamlessly with Tower Technologies’ Web Capture™ software and the client sees it as the product of a single company, and deals with Tower Technologies for its support, not Tower Software that produces TRIM Captura™.
A second way of converting recordmaking systems into recordkeeping systems is through the use of an application program interface (API). The Defense Advanced Research Projects Agency (DARPA) uses an AMS financial management system called Momentum™, which is an important financial recordmaking system, but not a trustworthy recordkeeping system. DARPA uses Tower Software’s API with Momentum™ to capture financial records into a trustworthy recordkeeping environment, effectively making Momentum™ a recordkeeping system. Tower Technologies has also partnered with Authentica for industrial-strength security and access control with encryption, making for a most powerful web recordkeeping solution. (ARM professionals commonly take the view that they see no relevance for security systems in unclassified records systems, because they are promoting the use of records. However, even in unclassified systems, one of the strongest hallmarks of a trustworthy recordkeeping system is its ability to properly control access and prevent accidental or malicious alteration or deletion of even open records.)
A third way is the ‘commercial gateway’ approach whereby the recordkeeping system is integrated with the recordmaking system at the source code level, as exemplified in TrueArc’s (Foremost™) recent alliance with Filenet. Depending on the method chosen, end users will be working with one company or two companies for support. In each of the above cases, an EDMS or EDMS+ is involved. However, all recordmaking systems are not EDMS. A fourth approach is one in which recordkeeping software is integrated directly into the business application software. This is the approach taken by Tarian Software.
Finally, there are other experimental approaches to recordkeeping being undertaken for different reasons, including in some cases budgetary necessity. For example, the Indiana University project cited above is exploring the use of simpler WfM instead of EDMS+ tools to manage electronic records, simply because of budgetary decisions taken at higher levels that pulled the rug out from under the latter possibility. On the other hand, the Army Information Warehouse (AIW) proposed by Edward Arnold, Deputy Director for Army Records, would make use of warehouse-based technologies. Other strategies focusing on long-term (hundreds of years) preservation and access to electronic records, including software emulation and IM techniques in which information is preserved in application-independent form, as in the joint NARA/ARPA/San Diego Supercomputer project.
Together, these strategies constitute the beginning of ubiquitous recordkeeping. One way or the other, recordkeeping functionality will become integrated with business processes and practices. More of the now troublesome big-volume recordmaking systems, such as ERPs, that are not currently trustworthy recordkeeping systems, will become so. The good news is that the trend has definitely begun. The bad news is that it will take some years to fully come about. We can’t just wait for that to happen.
Information Management vs Records Management
Senior line managers and information technology managers often suffer from misconceptions about the relationships between information management and records management, which many regard as very different activities. Yet, well-implemented records management is simply a rigorous, efficacious form of IM to serve broader and more demanding personal, organizational, legal, regulatory and societal requirements.
Misconceptions about this are chiefly due to historical accident, because of the fact that usually hard organizational lines have been drawn that should not exist between computer-based and paper-based operations of most organizations. Further, IT professionals traditionally have been more interested in the medium than the message. ARM practitioners have typically been more interested in the message than the medium – traditionally paper and microform. The paucity of IM professionals in IT organizations, and IT professionals in ARM organizations, has made it difficult for ARM-based IM concerns to be properly addressed.
For some reason, managers are usually very interested in the use of information in the workplace, but are disdainful of records. This observation has caused the author to ask senior executives in interviews: what is it that happens to information when it moves from paper form to electronic form that transforms it into something important? Confronted this way, executives usually acknowledge that there is no sense in such a rationale. For relevant information policies to be established and enterprise document management systems to be developed, policy makers and developers must see the highly interconnected nature of information, documents and records.
While all records constitute documentation of some kind or another, and may be referred to generically as "documents," irrespective of storage or presentation medium, the reverse is not true. Some documents do not qualify as records as they are purely reference documents. Dynamic database information and external economic intelligence information resources may be consulted in an organization through a computer-based information system or through printed reports; but that information do not usually constitute records until imported into internal document that are records. The reference material becomes part of that record it becomes part of something else that qualifies as a record. In like manner, documents in preparation or at some early draft stage are not normally regarded as records until such time as they leave that state and are communicated into the business unit or corporate (institutional) domain in support of some business process.
We should also distinguish records from evidence. Yes, all records constitute evidence; but the reverse is not the case. A scrap of paper in a person’s handwriting with a potentially incriminating telephone number on it, taken from her wastebasket, may become important evidence; but it wasn’t ever a record. And, or course, there are all manner and sorts of physical evidence that do not constitute records. Better making this distinction, in order to streamline recordkeeping so that records only truly essential for legal evidentiary purposes, has been central in the discussions of the Association for Information and Image Management (AIIM) C-22 Committee, chaired by Dan Schneider. Its Evidentiary Support Project, approved 11/01 by the AIIM Standards Board, covers Internet, intranet, and extranet web-based activities (e-mail, on-line transactions, web advertising, voice-mail, videoconferencing, etc.) and the evidentiary management issues in the other recorded information holdings related to litigation discovery judgments.
Likewise, in his e-Business Solutions 2001 keynote session, Edward Arnold characterized the traditionalist’s definition of ‘record’ as: “Everything created by anyone for any reason (e.g., Post-it notes, unsigned draft documents, etc.),” in contrast to the AIW’s definition:
“Information documenting agency operations.” I believe most ARM theorists
will agree with the proffered AIW definition. If so, this highlights continuing misunderstandings between professionals coming from a strong records management and archival backgrounds. With so many other electronic records challenges, there is little excuse for continued misunderstandings in our own community.
Records vs Electronic Records
There has been a great deal of discussion and debate in recent years about the difference, if any, between records and electronic records, and it is not the purpose of this paper to review that field of literature. While there may be nuances that theorists will debate about such distinctions, for all practical purposes electronic records are simply records that are created, captured or maintained in digital electronic forms rather than in paper or other analog forms. There is no compelling reason (at least not yet) to conclude other than: the technology or medium in which documentation is created, stored, used or presented should not be what decides whether it is a record or not. To say that we should destroy all email after a week or month or whenever, would be little different than saying that we should destroy all paper after a given period. What gives documentation recordness, or makes it recordworthy, is the fact that it represents the conduct of business.
Drawing again on Arnold’s excellent presentation, he understands the traditionalist definition of ‘electronic record’ as: “All information created electronically, e.g., email, all drafts, etc.,” compared to the AIW definition “Records readable only by computers (e.g., email and database out-put).” This again appears to suggest more by way of misunderstanding between archivists and records managers than disagreement.
Perhaps the underlying issue in both C-22 and AIW is not so much one of definition of recordness but more one of what records are worthy of retention and for how long, if at all. The choices of whether to keep such records, or for how long, are separate questions. That brings us back to the central importance of the records appraisal process. It also draws attention to negative effects of the bifurcation between national and state/provincial governments’ management of so-called “current records” at the department/agency level and the management of archives at the state/provincial or national levels. It further stresses the importance of ensuring greater integration of archives and recordkeeping skills and experience at all levels, including in private sector organizations.
Thus, while technology does not change
the nature of recordness, it substantially changes the manner in which
records must be managed, the related skill needs and the design, development
and implementation of information systems. It also raises complications
regarding the integration of paper and electronic records. Finally, whereas
recordkeeping was a matter of little interest to CIOs and IT managers when they
were all in paper, now the omnipresence of electronic records makes
recordkeeping a high risk area in which CIOs and IT directors must be deeply
involved and can ill afford to ignore.
Still, many CIOs are of the opinion that it is necessary to have both EDMS and records management applications to cover both general document management needs and recordkeeping requirements. That was true when RMAs were in their infancies. Now, however, at least some of the 5015.2-certified RMAs have are full EDMS+ offerings. The rationale for using both is disappearing, and the penalties for having both are also becoming very high in terms of total cost of ownership (TCO) – a term well known in the IT world to indicate the full costs of specific technology solutions – licensing, technical staff and user-training, testing, conversion, implementation, technical support, future upgrading, etc. CIOs and ARM managers should ask short-listed EDMS and EDMS+ vendors to demonstrate satisfaction of both traditional document management and records management requirements. Then decide if dual systems are really needed.
Enterprise, the Adjective
Like so many terms in business and technology today, the noun ‘enterprise’ has come to be overused. As recently as 10 years ago, some dictionary treatments of the term limited its meaning to business entities, undertakings, projects or initiatives, often with a connotation of risk. In its broadest usage, it was synonymous with ‘business’. The adjective form was ‘enterprising’. Today, we hear the term enterprise used more as an adjective than as a noun to emphasize the application of a concept or technology to the whole organization – ECM, enterprise content management; ERM, enterprise relationships management; ERP, enterprise resource planning system; and of course EDMS, enterprise document management system.
For many years, EDMS meant electronic document management system, and technically still does when used to reflect isolated systems in parts of the organization. As organizations came to realize the importance of instituting organization-wide rather a hodge-podge of fragmented, non-interoperable departmental systems, and as developers responded to ever increasing scalability requirements, EDMS began to mean enterprise document management system. It was not just fashion, but business bottom-lining that made the term a modifier for virtually any kind of business concept or technology. Yet, most RMAs and EDMS have not been implemented on an enterprise level, or have been for only very limited aspects – correspondence tracking, management of paper records, etc. Needed enterprise-level recordkeeping necessitates either upgrading recordmaking systems to become recordkeeping systems, or porting the records they products into a trustworthy enterprise recordkeeping system. Either way, it entails executive and CIO appreciation of the requirement for enterprise recordkeeping to be a central part of the organization’s IM and IT architectures.
The growing use of the term enterprise to reflect scope and scale of application has important connotations for ARM professionals and systems for two principal reasons. Firstly, enterprise systems are now frequently being introduced to replace legacy systems – especially financial and human resource systems – mainly in the form of ERP systems that integrate all of the business process transactions within and between the financial and/or HR functional areas and possibly other areas as well, such as CRM. Since elimination of paper and other analog forms is often one of the central objectives of introducing ERPs, the recordkeeping capturing mechanisms of the old legacy systems disappeared with many ERP implications – and in too many cases, so did the records.
Secondly, enterprise systems are major recordmaking systems. Recordmaking systems may be distinguished from recordkeeping systems because the former produce large quantities of transactions, with or without documentation, that meet professionally accepted definitions of records; however, they lack the recordkeeping functionality necessary to qualify them as trustworthy recordkeeping systems.
In the remainder of this paper, I refer to EDMS that also have robust recordkeeping functionality as EDM Systems Plus or EDMS+. Having recordmaking systems that are not also recordkeeping systems is a formula for serious trouble, because it means the organization is very much at risk creating evidence of its functioning outside of a sound recordkeeping environment.
Knowledge, information, documents and records
A Washington Post story about the placing of NASA’s Triana program in ‘cold storage’ for a few years, expresses part of the rationale behind the knowledge management movement in modern organizations.
Valero, the lead scientist, and Jim Watzin, the Triana program manager at Goddard, fear that the greatest loss during Triana's storage period will be intellectual. Engineers will drift away to other projects. Information could be misplaced, lost or forgotten. Even certain pieces of technology, currently assigned to Triana, might be carted away for other purposes. "That's another risk of going into long-term storage -- the predatory projects," Watzin said.
Management guru, Peter Drucker, stated it more generally:
[I]ndividual accountability is key, says Drucker, who scoffs at the notion of knowledge management. "You can't manage knowledge," he says. "Knowledge is between two ears, and only between two ears." To that extent, Drucker says it's really about what individual workers do with the knowledge they have. When employees leave a company, he says, their knowledge goes with them, no matter how much they've shared.
Few would disagree with Drucker’s proposition that knowledge is an inherent property of the human mind. Yet many corporations are investing heavily to build organizational climates and human and technology infrastructures, including Chief Knowledge Officers (CKOs) and supporting programs and organizations, that facilitate human collaboration, expertise discovery, conversion of tacit to explicit knowledge, timely deployment of experts and quick delivery of expert knowledge wherever needed. At the same time, they are also attempting to capture what can be captured in the form of documentation and to distill large quantities of information into knowledge kernels to ease the path for others.
Bain & Company, a Boston-based strategic management consulting firm, discovered a few years ago that it was having difficulty responding rapidly to client needs and that staff were often unknowingly plowing over old ground in addressing similar client problem environments. Bain found that some of the most important documentation of its client work was in the form of hundreds of PowerPoint™ slide presentations constituting in many cases the firm’s intellectual products for their clients. By definition, these were important corporate records and assets, as they probably are in most organizations. Yet, they were clearly not being managed as records or systematically in any other fashion. With the involvement of their records manager, Chris Bednar, Bain developed an extensive database of the PowerPoint™ files. It turned out to be among its most important records and one of the best knowledge sources in the organization, both in terms of providing models for how certain situations were handled and for identifying subject-area experts or skills, an idea that goes back at least as far as the Middle Ages guilds.
At about the same time, the World Bank saw several barriers in the management of its diverse intellectual assets that led it to making major organizational, work culture and technology changes needed to support a KM mandate. These included divergent information capture processes; much uncaptured information, divergent capture destinations, divergent processes to maintain, no single catalog for information, and divergent cost recovery models, access services and sourcing strategies. These factors were all exacerbated by the Bank’s size and physical presence all over the world. How could the organizations they serve in Harare, or even their own staff, hope to know that similar development issues were being addressed in Delhi? In the mid- and late-1990s, to begin to break down these barriers, the Bank introduced an enterprise-wide KM mandate and support system, including a Director of Knowledge Programs, first under the CIO and later moved to the Operational Core Services Network vice presidency, and cross-organizational communities of practice networks oriented to major thematic areas reflecting the Bank’s core business aims and led by ‘anchors.’
We will not go into a literary review or scholarly discussion of knowledge management here. Bruce Dearstyne has done a brilliant job of that already in these pages. Readers will also find an excellent series of KM papers in a dedicated issue of IEEE Intelligent Systems. In this paper, the discussion of knowledge management is focused principally on differentiating it from the other management modifiers – information, records, content, etc., and considering recordkeeping implications. We will spend more time on KM, because it may be thought of as the center of a web that has threads out to all of the others (IM, EIP, CM, RM).
Perhaps the most distinguishing feature of knowledge management that separates it from the traditional concepts of information, document, and records management, as suggested above, is that KM is more human-centric than object- or docu-centric. Of course, humans make use of documents and information in all forms and both information and technology play heavy supporting roles in KM. However, KM is more about establishing an organizational climate and culture that encourages identifying expertise and creating and establishing ‘communities of practice’ that gather both virtually and actually to share and update experience and bring that experience to bear on current issues facing the organization and its clients. KM isn’t easy. Perhaps that is why Bain & Company research discovered that only something more than 20% out of 631 worldwide interviewed companies apply knowledge-management. Because of its human-centric nature, ARM professionals may mistakenly conclude that KM is of little import to recordkeeping.
Foreign Affairs and International Trade Canada offers an excellent example of innovative use of records metadata to support the human-centric side of KM. The records database has been used to search for authors on particular topics and lead Department staff to living knowledge sources. In further support of one of the most challenging aspects of KM – knowledge capture – FAITC developed a scenario in which a metadata server would allow staff at missions and HQ to search across their combined collections.
Documents, like other forms of information, are basically inert objects, neutral as to their usage. It is largely up to the reader to discover their existence, find and access them and determine their relevance. They are amenable to the application of information management techniques and technologies such as EDMS, EDMS+, resource planning systems, databases, warehouses, etc. By contrast, and design, KM-based information and systems are not inert. At least in theory, they merge and extract and may imply or suggest a course of action based on experience or best practice.
As with other innovations, as soon as the KM concept caught on, many software developers declared their products as KM Systems. To the extent that they were information systems and KM requires information, a case could be made. But many contributed little to our understanding of KM technologies that support an organization with a KM mandate in ways beyond what any IM product does. By understanding the concepts behind KM, and the ways in which KM has been implemented in such organizations as Accenture, Dow Chemicals and the World Bank, we can surmise that the key differentiators between KM and traditional IM technologies are that KM support systems have specific functionality beyond simply something that requires writing custom APIs to the IM systems. KM systems must facilitate the identification of individuals with special expertise of interest to the organization and the collection, distillation, dissemination of, and access to, that expertise as well as intelligent discovery of not only the source of the expertise, but its target interest groups.
In 1998, InformationWeek Labs and Doculabs combined forces to evaluate five software products targeted at the KM market, including Wincite™, Intraspect™, ChannelManager™, Backweb™ and KnowledgeX™. KM differentiators included:
· Context relationship analysis and tools for augmentation by subject-area experts.
· Creation of group memory stores mapped to individual needs through context-sensitive user subscription profiling and automated alerting.
· Linking of historical, current and emerging information.
· Automated linking of email, attachments, word-processing files, web pages, by subject, topic, etc.
· Multidimensional data design allowing correlation of information from disparate platforms and formats and viewing from multiple perspectives or in different contexts.
Other products, e.g., Tacit's KnowledgeMail™  and Verity’s Knowledge-to-Enterprise (K2E) ™  support subject-expert discovery and directory building. This is done through continuous analysis of email, other documentation, and user’s search patterns and behavior. Experts are linked up with clients or others requiring such expertise. Automatic classification and document spidering technologies are also employed for these purposes. Unlike documents, which typically can stand on their own, records exist in the context of other records related to the same business process, action, transaction, policy or decision. Trustworthy recordkeeping systems thus also can use some of the same technologies that support KM to capture context and relate records to the business processes that produce them. Some system developers have or are working toward EDMS+ systems that incorporate functionality to support KM requirements.
KM and ARM are in important ways sisters, although there is scant evidence that CKOs and ARM managers have done much to recognize or build on their relationships. ARM and KM are both oriented to actions – KM’s whole thrust is to guide best future practice and actions. ARM reflects actions taken, but also provides a source of information regarding the contexts of prior actions to help guide future actions. An organization’s records constitute perhaps its best source of legacy information on which to build the explicit end of a KM system. Records provide excellent pointers to expertise in an organization that can be used to develop communities of practice. Oral history programs and records can be especially effective in this way, although this is an area that is often sadly neglected in organizational budgets and ARM programs.
KM knowledge-capture mechanisms sometimes provide mechanisms and technology to facilitate the expert’s preparation of what are sometimes called ‘knowledge kernels’ that summarize lessons learned and best practices. Kernels may be prepared purely from memory in the absence of documentation. A recent example of this, caused not by a KM mandate but rather by the 911 disaster, is taking place in the Department of the Army where staff have been asked to prepare such documents to substitute for many, many paper records lost in the Pentagon attack that were not being maintained in digital form. In other cases, knowledge kernels may be prepared by experts at the end of a project to distill and summarize key project decisions and lessons from project records. Such documentation may be thought of as ‘macrorecords’ that also could be seen and used as part of a high-level form of archival description.
From the above, it can be seen that ARM professionals need to have an excellent understanding of KM for more reasons than simply keeping abreast of trends in business practices. Rather, it is important because advanced business practices are organic to modern recordkeeping, and because they offer opportunities for more enhanced recordkeeping.
For these reasons, one of the best business cases for the implementation of advanced recordkeeping systems is the support that RM provides to the KM culture and knowledge base. But even sisters sometimes have disagreements. So it is with KM and ARM. Communities of practice anchors may have a different view on the value of legacy records and want to retain them beyond their established retention periods. If this is just related to context, to point to sources of expertise, it can be remedied without necessarily retaining the records beyond established destruction dates by simply retaining related metadata. If it is because the content of the records is seen as having enduring KM value, then it is a matter suggesting reappraisal. Since leaders of communities of practice will not ordinarily be program managers responsible for the business processes in which the records were created, their legitimate concerns can easily be overlooked. This suggests that those leaders are serious stakeholders in the appraisal process, and this is a topic that should be addressed in policy governing the appraisal and scheduling of records.
Intermediation is not dead! People are back at the Council!
KM and content management systems often make considerable use of intermediaries in both back office and front office operations – welcome news to librarians, archivists and other IM types. Some websites use ‘virtual humans’ to try to make us think we are still dealing with humans. Listen to perky, green haired Ananova deliver the world news, for example. However, a countertrend in the making is for sites to provide links to live chat help when visitors want true human interaction – old-fashioned intermediation. Accenture and Dow Chemical are examples of large corporations that make use of knowledge officers to work with experts to facilitate – cajole, urge – the capture and organization of knowledge and otherwise help serve knowledge needs of the user communities of interest served by communities of practice. We are likely to see some of these needs filled by ARM professionals. Records managers have historically been very familiar with the substance and context of paper records, because they had to physically work with the records and their creators to ensure proper disposition. That makes them an interesting commodity to CIOs and others looking for content managers. Bain & Company is a good example of where a records manager was integrated into the knowledge management team. Similarly, in Queensland Rail (Australia), senior archivist Del Cuddihy was made a part of the business systems analysis team where she has made leading contributions in modeling the whole organization and its information and records assets.
KMWorld, a KM technology trade magazine, regularly announces new products aimed at supporting KM. Some of them, in this author’s opinion, do not differentiate themselves. Others do, if you can trust but verify. A recent Smartlogik announcement purports to differentiate its product in ways that help us understand what KM technology ought to do in addition to traditional IM technology. Smartlogik has developed search and categorization technology that it says integrates human insight and expertise with corporate information stores.
Content vs. Information Management
The Wall Street Journal, earlier than most, introduced a rudimentary ‘Personal Journal’ portal that was limited to news – a service that continues today. Using key words, company names and stock symbols, the on-line subscriber could define, what subject-specific and company/stock-specific articles should appear in the user’s daily ‘personal edition’ of the Journal. Such providers as AOL and Yahoo deployed portals in the form of their home pages with broader capabilities in the public domain. Portals were seen as the door to the Internet and the big players wanted you to get to the Internet only through their door, hopefully pausing to read advertisements and perhaps purchase services and products along the way. Their focus was to supply the individual client with whatever resources that individual wanted each time they logged in – email, chat, news, weather, etc. In web lingo, well-designed portals created ‘stickiness’. Investors and advertisers like sticky websites. Yahoo subsequently invented MyYahoo, a welcoming page personally and easily designed from a menu by the user to choose news, weather in their favorite cities, local movie schedules, lowest airfares to their favorite destinations (and of course the ability to actually book tickets), etc., each time they enter the space. Now, the MyEden portal being created as part of the IU Electronic Records Project uses the same philosophy to help make record capture sticky.
As KM matured, it became clear that what was needed was that same kind of portal technology but internalized to meet KM and other needs within the organization. Internal portals developed into channels for delivering information of particular interest to each individual subscriber. Such portals could then be used for drawing together people with common expertise or interests. Thus, EIP technology was channeled into KM where it filled a large need. It was also seen as a means for channeling information to an organizations market via its own website. In other words, the organization would become its own Yahoo, serving its own employees and client needs.
Content management began chiefly as a way of managing dynamic information that organizations placed on their web sites. One might ask: what possible difference is there between content management and information management? Aren’t they the same thing? In some ways they are. Both deal with documentation of one or more kinds. In other ways they are not. Information management and information systems, as noted earlier, tend to be inert objects. They house the documents produced by the organization, and possibly externally, in some logical manner. Content management is much more dynamic and aggressive. Content managers do not wait for documents to appear on their desks for proper filing. They are informed of or see needs within the organization or on the part of its clients for information that may not exist. They hunt it down. They make use of good graphics and other marketing skills to keep their internal and external customers coming back. While webmasters originally handled content management as well as technical tasks for the website, nowadays at least larger organizations have separated those jobs because different skill sets are needed and because they are both full time jobs.
Unlike more traditional information managers and records managers, content managers are regularly changing the content that is available on the web, again to draw and keep people interested and happy with the usefulness of what they are finding on the website. Everyone who manages a website knows that the owners of the website must constantly be changing the content in ways that keep the attention of internal and external customers’ attention. Otherwise, the website turns into a candidate for the Yellow Pages. People don’t ordinarily visit the same Yellow Pages entry more than once or twice. One industry analyst group, the Aberdeen Group, states that content flow is as important as cash flow.
Interwoven, a supplier of CM products and major holder of market share in this field say:
Its content infrastructure product suite includes content aggregation, content collaboration, content management, content intelligence and content distribution.
AIIM has established this definition for content management:
The technologies used to create, capture, customize, deliver and manage enterprise content to support business processes.
Noticeably missing from these definitions are classification, preservation and disposition management.
As noted earlier, Filenet has described its own CM product (Panagon ECM™) in ways that help to understand how CM differentiates from other products. Some of those differentiators are listed below:
· Content interchange utilizing XML capabilities for extracting and exporting metadata
· Template-driven and HTML content authoring
· Web content lifecycle management to maintain accurate and time-sensitive content
· Process management and reporting
· Web site staging and publishing to make exporting content to Web application servers significantly easier
· Search feature enhancements to provide multi-library searching
Dynamic web content is one of the most significant ways in which organizations interact with clients and the public. This heightens the necessity of ensuring that such content/records are properly captured and preserved in a trustworthy recordkeeping environment unless they have already been reflected in such an environment prior to being electronically published to guard against there being more up-to- date information on the website than exists in the record systems. The NHPRC-funded Syracuse study of federal and state websites found that, “Many agencies use disclaimers to state that the information on their websites is not the ‘official version.’ Despite these disclaimers, in-depth interviewing revealed multiple cases where record-quality materials appear uniquely in agency web postings…because agencies continually explore…modalities that no one anticipated even a few months ago.
Conversely, as John Harney and others have noted, content managers are discovering the potential of organizational records and are ‘repurposing’ legacy information by filtering it and integrating it with other resources to meet new needs. Geoff Moore addresses migration concerns as entities become integrated through mergers and acquisitions or simply through reorganizations. Existing content must be migrated into new websites or other information resources. For these reasons, and as CM technology and our understanding of KM have improved, the concepts underlying CM have been extended beyond website management to embrace intranets, extranets and actually information in all formats and data types. CM is no longer limited simply to websites.
While identifying some important ways in which current business practices – IM, KM, CM, RM, etc. – may be distinguished from one another, it is also evident that these concepts are not isolated from one another. Each one is more of an outgrowth of what came before it than an independent movement of its own. New concepts and technologies are being integrated with current practices to take business to the next higher level of operations. Using the same calculus, perhaps we will become better at envisioning ways in which gaps in current practices will foster new developments and lessen their negative recordkeeping impact.
Most business practices and technologies have consequences for the manner in which records are managed. Unless practices such as information management, knowledge management, and content management specifically provide for legitimate enterprise recordkeeping needs, ARM professionals should assume that such functionality is not provided and perhaps isn’t even intended. They should act to ensure that CIOs and other senior executives: appreciate that most recordmaking systems that support current business practices are not trustworthy recordkeeping systems; understand the risks such systems give rise to; and foster remedial actions to integrate recordkeeping requirements in organizational information and technology architectures and systems. Conversely, I hope this discussion will stimulate the adoption of advanced business concepts and technologies in ways that will contribute to improved recordkeeping solutions and more trustworthy recordkeeping environments.
© 2004 Richard E. Barry
 The term ‘archives and records management (ARM)’ is not commonly used in Australia and Europe where there is purportedly not the same bifurcation we have in the U.S between the management of so-called ‘current records’ and ‘permanent’ or ‘archival records’ as personified by records managers and archivists. Underlying is also the debate between advocates of the life-cycle records management (which advances distinctions between “current” and “archival” records) and the continuum approach (which fosters integration of recordkeeping at a broader recordkeeping environment level). My use of the term and the shorthand ‘ARM’ is not religious here, but rather is simply to make it clear that I am referring to professionals and programs associated with the management of all public and private records.
 <http://www.misq.org/> URLs cited in this paper were accessible at time of writing in December, 2001.
 < http://computer.org/itpro/>
 “Records Professionals and the World Wide Web: Resources and Responsibilities: Part One: Basic Professional Information Sources,” by Richard J. Cox RIMR, Vol 16, No. 8, October 2000, p.7.
 National Association of Government Archives and Records Administrators <http://www.nagara.org/>.
 Bell’s presentation can be found at <www.rbarry.com> under HOT TOPICS/Organizational Change along with other conference papers (including Mora’s presentation) on that subject. Mora’s presentation and a subsequent version of Bell’s paper presented at the Society of California Archivists 1998 meeting are also in the Guest Authors section.
 Society of American Archivists <www.archivists.org>.
 See program at <http://www.erecordsconference.com/>.
 “Northern Light: New Model for the Enterprise Information Portal: an IDC White Paper,” (Analyst: Susan Feldman), pp. 1-2. This paper can be purchased at <http://www.infoworld.com/research/articles/01/05/11/01051124586.xml>.
 Indiana University Electronic Records Project, Phase II: 2000-2002, Phillip Bantin, IU Archivist, Project Director, <http://www.indiana.edu/~libarch/ER/NHPRC-2/index.html>. Also see three presentations at the September 2001 Society of American Archivists conference at <www.rbarry.com> under Guest Authors.
 One has to be careful these days with the use of the term IM. The rapidly emerging use of instant messaging for substantive business communications has resulted in X- and upcoming Y-generation IT professionals more often using the acronym ‘IM’ to mean ‘instant messaging’.
 For an excellent discussion of macroappraisal, see Terry Cook’s “Archival Appraisal and Collection: Issues, Challenges, New Approaches,” accessible at <www.rbarry.com> in the Guest Authors page.
 Unfortunately, technologists too often fail to understand or guard against unintended consequences due to human error in automated systems, a point well illustrated in “Crazy Clocks: Counterintuitive Consequences of ‘Intelligent Automation,’” by Kim J. Vicente, IEEE Intelligent Systems, in “Human-Centered Computing section, Vol. 16, Nov-Dec 2001, p 74. Automation that eliminates human involvement can introduce error instead of eliminating it: <http://www.computer.org/intelligent/ex2001/pdf/x6074.pdf?SMSESSION=NO>.
 There are many worrisome, often unintended, social consequences of modern technology – employment, privacy, intellectual property rights, ethics, accountability, quality of life, digital divide, etc. It is not the purpose of this paper to address the impact of technology on society or the ARM profession. Still, ARM professionals may have to face up to them in their organizations. Interested readers are invited to consult Computer Professionals for Social Responsibility resources <www.cpsr.org>, papers on ethical and social issues for IM professionals in the Recent Papers section of <www.rbarry.com> and the Forthcoming Book [Richard: please insert latest book title/info] by Richard Cox and David Wallace on these subjects.
 See Edward Arnold’s November 6, 2001 presentation on the AIW project at the e-Business Solutions 2001 conference in the Guest Author’s section of <www.rbarry.com>.
 “Knowledge-Based Persistent Archives,” by Reagan W. Moore, SDSC, January 18, 2001 <http://www.sdsc.edu/TR/TR-2001-07.doc.pdf>.
 See slide 8 in “Army Information Warehousing: Advanced Army Records Management,” by Edward Arnold. Ibid.
 In the niche but growing facility management and construction sectors, EDMS means engineering document management system. With growing use of such systems, they are becoming important organizational electronic recordmaking systems, but typically not trustworthy recordkeeping systems.
 For further discussion of enterprise document management, see this author’s review of Document Management for the Enterprise: Principles, Techniques and Applications by Michael J.D. Sutton at <www.rbarry.com> in the Other Papers section; and, better still, read the book.
 “For Gore Spacecraft, All Systems Aren't Go: Earth Observation Satellite Shelved,” By Joel Achenbach, Washington Post, Wednesday, August 8, 2001; Page A01.
 Kontzer, Tony, “Management Legend: Trust Never Goes Out Of Style,” Informationweek.Com News, June 4, 2001, accessed July 20, 2001 at <http://www.informationweek.com/story/IWK20010604S0011>.
 Dearstyne, Bruce, “Knowledge Management: Concepts, Strategies, and Prospects,” RIMR, Vol. 15, No. 7/September 1999. See also, “Knowledge Management and Records Management: Is There a Difference?” by Susan Mybrugh, RIMR, Vol 14, No. 7/September 1998.
 See IEEE Intelligent Systems, January/February 2001 <http://www.computer.org/intelligent/> for excellent papers on knowledge management, including use of advanced ontologies and technologies. These do not address recordkeeping implications.
 Formerly Andersen Consulting
 See < http://www.tacit.com/>.
 See <http://www.kmworld.com/news/index.cfm?action=readnews&news_id=2449>.
 It can be appreciated how some people resist KM because they regard analysis of their behavior as an invasion of their privacy.
 See “Managing Business Information – Content in Context,” by Geoff Moore, July 2000 <http://www.ustrim.com/product/files/context.pdf>.
 I played with this idea in a 1993 paper “EDMS and ERMS: Towards a Methodology for Requirements Definition,” at <www.rbarry.com> in the Other Papers section.
 “Analysis and Development of Model Quality Guidelines for Electronic Records Management on State and Federal Websites,” by Charles R. McClure and J. Timothy Sprehe, Ph.D., January 1998, Chapter 6. The report and guidelines may be accessed at <http://istweb.syr.edu/~mcclure/nhprc/nhprc_title.html> or, along with other related papers at <www.rbarry.com> under the HOT TOPICS/WWW vs. Records section.
 “Content Migration: The Race to Repurpose,” by John Harney, e-doc <www.edocmagazine.com>, Nov/Dec 2001, p. 24.