| Introduction This article will diverge from one of the first guidelines given to a student of research writing. You are told not to waste the reader’s time with a self-referential description of the author’s research activities for the paper. For most research subjects, this is a wise rule to follow. But in the case of the emerging practice known as "Knowledge Management," I hope you will agree using that approach has some relevance. Knowledge Management (herein KM) is a term with no fixed meaning. I thought I had a fairly good sense of what it meant from my standpoint as a worker in the information technology trade. But as I have read many of the key works among the related literature and searched for the background of this subject as well as the current ways the term is applied, the meanings have multiplied rather than crystallized into something clearer. While the phenomenon of KM is most often associated with corporations and the business world, I would like to try to describe its general heritage and how it fits into the broader world beyond commerce. First I will talk about the business world and my own experiences during the genesis of KM, then about the more general context around this emerging area of study. Finally, I hope to describe KM in context by relating the general back to specifics of current practice and research. Not the least of the problems in getting a fix on this chimera is the definition of the two words Knowledge and Management. The definition of "knowledge" is a basis for whole careers in philosophy and within that field the study of epistemology concerns mostly the knowledge of individuals. Group knowledge, which is the practical focus of knowledge management, is an even slipperier slope. Taking a sweeping look through scholarly works concerning group knowledge, it is impossible not to delve into social science, where in early works the rather un-businesslike ideas of Hobbes, Locke, Marx and Hegel are found. The closest thing we will find to practical research is a bit of social psychology and some anthropology. Some ideas relevant to learning can first be found in the social psychology work done by Piaget and Vygotsky, which have led to what is now called the constructivist theory of "situated learning." If you read much anthropology, it will dawn that one of the closest things to (if not the definition of) group knowledge is culture – a society’s system of tools and signs. Since the makeup of culture goes all the way back to the first hominids, it is in fact a collective memory. Indeed, looking back to the business world, a subculture, corporate culture will pop up quite often when you really try to apply KM, so this idea could be useful. A recent development in the literature concerning knowledge management is a realization that the mating of "knowledge" with the word "management" begets an oxymoron (Wah 26). True knowledge can not be "managed" in the traditional business management sense of command and control. And of course the definition of "Management" will send us on an expedition nearly as uncertain as the one that sought "Knowledge." We can be pretty safe, though, in assuming that the common-sense traditional definition of Management is about hierarchical command and control of people after the military model. Management also means control of certain classes of resources, like assets, data and human resources. KM is a phrase usage that follows in the tradition of "human resources management" and "information management," treating knowledge as a resource. So the management of organizational knowledge, if it is to happen, must be a new kind of management, not the traditional interpretation as applies to other phrases dealing with more static resources such as "money management", "time management" or "database management." Unfortunately, at the management-fad level, many organizations that initially sought to use KM may have done so in the same spirit as the more cynical incarnations of rightsizing and business process reengineering. KM held the promise that the organization could simply extract the intellectual capital held by its workers and protect itself from the loss of those workers by holding on to their "knowledge." While there are many cases where this very basic type of KM is needed, focusing on this approach was due to lack of deep thought about what knowledge is. A rudimentary example of this pattern is the babysitter’s list. There is a temporary change in personnel when parents need to go out and a babysitter is called in. Certain explicit knowledge must be passed on from the parents to the temporary parent, the sitter. Just enough key information to do the job and contend with emergencies will usually fit on a short list. The same pattern applies to more professional and complicated positions but the "list" becomes a document or knowledge base or more elaborate system of assuring that specialized knowledge gets shared. Many vendors of technical so-called KM tools are still using the management of this type of explicit, codified knowledge as an implicit part of their appeal to customers. White Rabbit or Sorcerer’s Broom? For the KM researcher there is also the problem that knowledge management still enjoys the special moving-target status of a fad (and one tailor-made for management consultants) rather than of a formal academic area of study. Everybody is talking about it but few can speak from deep experience. This is because of its newness as well as the rapid growth of both interest in it and of new ideas about it. It is "in" at some schools but barely recognized at others. Library indexing systems do not yet commonly use "Knowledge Management" as a keyword, though it may fall into a cross-reference category. Knowledge Management (herein KM) is also one of the first of such fads blessed to debut during the widespread adoption of the World Wide Web. Since it is both a business-oriented and an information-intensive subject, there is an explosion of Web sites, in addition to management books (many with associated web sites) on the subject of KM. The profusion and confusion of public documents that claim to address the subject of KM, then, is a prime example of one of its own areas of concern – the explosion of easily available and sometimes unwanted information of questionable quality. Many vendors have simply begun to use the word "knowledge" in their marketing materials and have repackaged or shifted emphasis to aspects of their existing products - such as database or communications systems - that relate to KM. There is a huge bandwagon loading up, and it is not full yet. So what does anyone really know about Knowledge Management? I would like to begin with a summary of my own take on the subject. A recent trend in the online literature from knowledge consultancies has introduced the word "ecology" alongside "knowledge," replacing KM with "Knowledge Ecology" for example. I think ecology is an apt metaphor, as Tom Davenport, who is arguably the top KM author, also felt when he titled his 1997 book "Information Ecology." I like to think that a successful KM effort at an organization would try to follow some of the same methods as a good wildlife management program. Find the bad things - kudzu and rats - to discourage. Find the good things to encourage - the so-called "native" flora and fauna. Expert trackers of knowledge already have some field guides. Valuable knowledge comprising many of an organization’s strategic talents is often found operating in what are now being called Communities of Practice (CoP). A CoP is a natural occurrence in any field of interest, profession or marketplace where people engage their skills. By sharing tips, encouragement and stories about their special interest or skill, they form a kind of community, however fragile and ephemeral. An organization cannot force a CoP to appear, but it can seed, foster, support and encourage one and it can align its organizational practices with a CoP that is strategically important. In both the technological and human resources sense, a CoP can be nurtured with the "if you build it they will come" approach to tools that support a community. If I had to sum up Learning and Knowledge in Organizations, I might simply turn to Micklethwaite in "Witch Doctors":
Sounds simple enough! It has been pointed out (Stewart) that CoPs are easier to kill than to grow and that many of the typical structures and practices of modern organizations are very good at stifling CoPs. CoPs are arenas where a touch of what is known as tacit knowledge gets shared – knowledge that is unwritten, unwritable and is only learned between the lines, by experience. Many of the best of the org learning and KM efforts are the ones that seek to remove barriers to fledgling or potential CoPs that have been suppressed or frustrated within organizations. This is where I think the "heart of KM" is to be found. I think Davenport, though he does not dwell on the CoP concept, is in accordance with this emphasis when he says "Firms need to shift their attention from documents to discussions" (Working K 106). There are many other kinds of KM projects and processes but the instances where true KM is alive and well for long times are bound to be those where at the root a CoP lives. The CoP concept is a cousin to other workgroup concepts such as high performance teams, special interest groups (SIGs), start-ups, skunk works (Bennis - Org Genius) and Sally Helgesens’s concept, Webs of Inclusion (Helgesen). Of course, Senge’s Learning Organization is a sort of generalized CoP in the large, especially when focused on "team learning." After giving a broad overview, I will give examples of how CoPs can be recognized and how they fit in the context of KM. There are many facets of KM, but for simplicity I think we can make a few categories of KM grouped according to the disciplinary specialty of the gurus in each camp. In fact, I feel that there are two main sides of the approach to the issue. My own jump into the lake called KM was from the shoreline of the computer technology encampment. Since I am a specialist in GroupWare, email and Intranet systems, I had always heard the KM phrase (mostly from technology vendors) in association with various types of collaboration systems, online communities and with shared knowledge bases. I also recognized that it probably had some ancestry in artificial intelligence, decision support systems and expert systems. And so indeed I found in my reading that one of the main camps in this field is the technology focused one. Knowledge Management as viewed from the technoid approach is simply the latest phase in the evolution of business automation concepts that started with data management and next moved to information management. Davenport explains this progression in terminology in the first chapter of "Working Knowledge."(2-4) But even those in the techno faction of KM practice must admit that the proposed KM systems, unlike the mainframes and supercomputers of yore or the hypercubes and drill-down crosstabs of last year, are people-focused systems. The KM system is not only for "retrieving" knowledge as a stored "thing," it is also for finding and communicating with knowledgeable people. At the "street level" if a person has heard of KM, they probably think of something related to knowledge bases or expert/knowledge-based systems. One of the main problems with the technoid approach, as is often emphasized by Davenport and other moderates, is that the tools can take on a "life of their own" and the Knowledge Project team assumes that the tool is the solution to the problem. This is Kaplan’s "Law of the Instrument" in action (Shera in Machlup 383), which is a generalized way of saying "if you have a hammer, every problem looks like a nail." On the other shore is the noble people-focused camp, who I call the socioids. I had my introduction to them when Peter Senge came to talk at a presentation (att CCS seminar xxx) hosted by my employer of the moment, AT&T, back in 1993. He is a visionary, eliciting the comment "This isn’t management, it’s abstract art" from a Swiss seminar customer (Mickelthwaite 133), but he has been doing consulting all along, so his vision is accompanied by practical experience conducting workshops and grass roots discussions groups. His approach in his first book, The Fifth Discipline was almost entirely aimed at interpersonal relations between executives of large corporations. It was in the end a spiritual book. Its focus was on the day to day practice of new ways of thinking and learning in teams. The point was for management to learn new ways to manage by changing their core assumptions about business and life in general. This sea change among managers, if taken to heart, would naturally fan out to the rest of the company. His only focus on technology tools was the use of MicroWorlds, a simulation tool developed at M.I.T., in training workshops held for managers. Senge’s large but well-written book had a mesmerizing effect on seekers of more fulfilling organizational lives everywhere. AT&T had just implemented Steven Covey’s "Principle-Centered Leadership" concept as a part of its management training strategy, so Senge’s approach in his talk came across as a slightly more intellectual or maybe Zen version of Covey’s holistic approach. It’s interesting how both Senge’s and Covey’s books used nice low numbers (5 and 7) in their titles, too, within George Miller’s oft-quoted "magical number 7" which estimates how many ideas most people can easily hold in their heads. When I saw Senge speak in the AT&T auditorium in New Jersey, there were admirers with well-worn and color-tabbed copies of his book to sign at the end of the talk. I had been inspired by his talk as well and considered asking him a question until I saw the throng of more ardent and obviously more well-read admirers already gathered there. The Learning Organization had been Senge’s catchphrase for his lecture and that was also his contribution to the management fad-phrase pool in that period. People started talking about building Learning Organizations. It was in the same year I heard Senge speak that I went to a multimedia demo of Lotus Notes version 3, which presented to a crowd of hundreds a very idealized and flashy vision of groups of people (knowledge workers) using software to work in teams across continents. The catchphrase GroupWare gained currency with the release of that year’s version of Notes. This was the technoid approach but with a people oriented flavor that was new to the software-marketing world. At that time, around 1992 or 93, the phrase "Knowledge Management" was found only in some articles by management gurus at the head of the pack. During the early 1990’s, many people were thinking and talking about the same general subjects that now fall under the category of KM, including the phrase "Intellectual Capital." Certainly the idea that an organization’s pool of knowledge needed managing was being discussed, often as an issue prompted by quality improvement programs, which also popularized the concept of "best practices," but the "branding" of that two-word phrase as a recognizable keyword had not taken place in most business circles. Many leading companies by the early 1990’s had added the title of CIO to their upper ranks and since then, in accord with the accepted data-information-knowledge hierarchy, many have added the title "Chief Knowledge Officer." Now, in 1999, on the World Wide Web there are many different large sites that offer themselves as KM portals or central clearinghouses for both current and seminal writings about KM. If a fad is "puppy love," then this trend has graduated in its romantic relation to the corporate imagination – I would say it has moved perhaps past marriage and the honeymoon phase into setting up housekeeping. Working with my admittedly simplistic "two camps" of KM conceit, I would like to offer some background on the intellectual heritage that these different flavors of practitioners might bring with them to a KM project. They are certainly not from totally different worlds, but people who have one or the other viewpoint will tend to have different basic assumptions about the problems to be solved. Those assumptions will be characteristic of their professional training and the body of knowledge that typically supports the training. A socioid KM expert has most likely majored in management with several courses that served up various aspects of management theory. Management theory has tapped in to several other disciplines such as industrial/organizational psychology, operations management, systems theory, economics, sociology, business philosophy and even anthropology. Any KM practitioner from the management area will likely have studied at least one of those specialties in addition to the standard "management core." In some cases, they have advanced degrees in social science and have been enlisted into the management world. For example, the Xerox Palo Alto Research Center (PARC) has a staff of anthropologists who aid in product design as well as designing systems that support organizational knowledge (Buderi, Brown). From the techno side, cybernetics, artificial intelligence, databases, expert systems, hypermedia, networks and collaborative systems are the most obvious predecessors. Systems theory, information theory, game theory and even biological/evolutionary computing are also sometimes mixed into a computer science or MIS curriculum that a KM expert may have followed. Notice that there is an overlap between the camps centering on systems theory, cousin of cybernetics. People who work on Operations Research and Game theory - rather mathematical pursuits - tend to also know Economics, a social science that also uses numbers. And the lot of them, if studying as graduates, would have taken at least one course that outlined systems theory. There is a third factor, the wild card in the academic background that may have influenced someone from either camp. Libraries became heavily computerized in the 1980’s and librarians, along with some computer scientists in many schools had begun studying "Information Science," which was first offered as a program at the Georgia Institute of Technology in 1963. Special applications of computer systems arose to support libraries and the problem domain they address is not far from the one that presents itself in the world of KM projects. In fact, after noticing the uncertainty I found about the term KM when starting my research, I have taken some comfort in observing the admitted confusion around the term "Information Science" that is evident in the textbooks and articles on that subject from the 1960’s through the 1980’s. One of the main orders of business in many such writings seems to be staking out the territory of the discipline and explaining how it is multidisciplinary and that a definition is not yet settled. One thing that is well agreed on is that this new phrase appeared just as computers made their appearance on the horizon of the library’s kingdom. Once the computer people realized this title had been staked out, a cold war ensued over it, which has yet to be settled. Fritz Machlup’s collection, "The Study of Information," in 1983 featured contributions from many of the leaders of each academic tribe that sought or had a hand on the flag. Machlup, in his introductory article, explains that "we have chosen to deal here with only thirty or forty cultures (3)" and then goes on with some amusement and frustration to explain the different way each area of study treats the key word information. For "Information Science", he finds four different groups using the phrase uniquely as their own. Sadly, I have found no sign of anyone of that caliber doing a similar survey of studies of "Knowledge." And so, stepping back a bit, how did "Science" - which was originally focused on the discovery of nature’s "laws" - come to turn back on the thoughts and writings of men rather than continuing to explore the outer world? How do you get from biology or physics, straightforward studies of things living or things in general to something called "Information Science," which along with "Management Science" is an ancestor of KM? That’s a pretty big question, but in short, this kind of stuff – the study of studies – of information and knowledge - was relegated to philosophy and the humanities up to some point. But then somehow there came to be an economic advantage to it, and so people paying the bills began to want some practicality applied to the study of man’s own way of organizing his learned knowledge. What were the milestones along this path? Of course a major one was the scripted language, which many cite as the beginning of civilization (McLuhan, Beniger) –where tribal, orally related wisdom began to be recorded outside the body. The code of Hammurabi is a key marker in the transition from "prehistory" to history and the burned library of Alexandria is still known as one of the disasters of human times. One even more primal example of a cultural loss is recounted by the explorer Robin Hanbury-Tennison in his memoir "A World Apart." He went on a trip on the Amazon in search of a plant with reputed blood-clotting properties. When his party found a group of a tribe said to use this plant, it turned out they had missed the shaman by a few months. The government's drug based disease fighting program had undermined the shaman’s authority and self-worth and as a result,
Contemporary hunter-gatherer groups are now being recognized as "walking encyclopedias" of regional biological knowledge (Diamond 143). This historical development is mentioned by McLuhan in describing that in early Greece the new way of learning was education by written, classified data, a shift from what Plato called the "tribal encyclopedia" (Understanding, vii). This demonstrates to me that KM is in fact the mediated extension (a la McLuhan’s "extensions of man") of a core human activity, the passing on of accumulated knowledge vital to a group. Certainly managers of today live in fear that the knowledge of their best workers will disappear or are frustrated because such knowledge is untapped. The project of KM is to extend this activity to support artificial tribes – modern, or perhaps I should say postmodern organizations. The extensions employed to solve these problems are both new and re-adapted skills along with the new types of tools that complement them. The compilation of tribal stories into single codified stores of wisdom and laws was also a step along the way to KM. Scribes are still with us in many flavors, having many titles but all descended from that original "scribe" and since they originally worked as official codifiers of knowledge, describing the evolution of their subspecies in the civilized world is a way to track the roots of the emergence of KM. As long ago as 2000 B.C. there was an official royal "tablet keeper" in the kingdom of Ur (Harris p24.) The Torah and the Bhagavad-Gita certainly carried a lot of weight with their readers. In a way, the scribe can be found incarnated today as the political speechwriter or the cultural feature writer, still trying to capture the stories and spirit of the age. And of course the bibliographer, historian and the ethnologist, trying to capture the knowledge of the past before it fades. Since the middle 1980’s, the personal computer is the hardest working scribe, recording nearly all text produced by humanity both past and future. This last scribe is our most powerful amplifier of infoglut, nemesis of KM. KM is a problem we assume we must use computers to solve and one that computers have virtually caused. Even before tablets, scrolls and books were common, people had formed specialized trades that used rituals, tools, oaths, relationships and language to impart extra meaning that helped to train apprentices. The unwritten part of what is learned in this and all similar types of situation is called tacit knowledge. The passing on of tacit knowledge probably began during man’s mimetic phase, even before language had progressed very far. As language developed and mimetic culture was extended with verbal culture, the oral tradition was born (Donald 2). Even today, most people are familiar with the effect of verse and rhythm on learning – children learn their nursery rhymes and the easiest thing to remember is the words to a catchy song. Long verbal passages were used to recount the exploits of ancestors. An apprentice to a trade in the neolithic and later periods also would have been subject to the prototype for another key KM activity, filtering – he would have been protected and restricted from certain activities and experiences in order to keep focused on learning his appointed trade. This specialization of thinking and communication is an efficiency that allows depth of skills just as the symbolic language of mathematics allows powerful concepts to be expressed quickly. We certainly all know about the scholarly genius who, overspecialized, couldn’t survive a day outside his lab or university niche. But specialization creates huge barriers between the experienced and the rest of the world that includes the novice and the layman. Even the written, codified part of what the specialist knows can be difficult for outsiders to apprehend since it is often coded in a special jargon, or as Foucalt (*cite*) called it, Discourse, oriented to "shop talk." After written script many other knowledge tools appeared; the list, musical and mathematical notations, the letter, the scroll, the reusable wax tablet, the journal, the reference table, the book, libraries, the codex, reference books, maps, the dictionary and encyclopedia, the printing press, the chalkboard, the newspaper, the digest, the scholarly journal and review. An example of how a body of knowledge became ossified during the middle ages, and probably because of the lack of easy access to or ability to create group knowledge, is the legacy of Galenism. Galen was a roman doctor who gathered together much of the medical knowledge of his day and synthesized his own version of that age’s system of medicine. Because of his leadership and because of the turn of events that led to the "dark ages," his system became the one way to practice medicine in the West for over 1000 years in spite of many discoveries that disproved his teachings. The continuation of this system in the face of obvious contradictions illustrates the power of many concepts, such as groupthink, mindlessness, memes and mental models, relevant to group knowledge. Howard Bloom, in an article attempting to explain many of these concepts, illustrates the hold of Galenism:
I might add that these handwritten volumes were all in Latin, which not all of the students could read. Not until the Reformation and the printing press would people begin to disseminate knowledge in their own tongue. The codified knowledge of western civilization was "managed" by an elite who could read latin and physically held the books. There was a divide between that kind of knowledge and common knowledge that was verbal and tacitly transmitted. Along with the printing press and the explosion of vernacular communications in print, came the first issuing of infoglut, so that along with the jostles and cries of street vendors, the city dweller became subject to posters, leaflets and handbills. Certainly as literacy spread with the influence of the press, codified knowledge began to move to the center of common concerns. McLuhan (*cite*) said that a concept of "the public" did not exist as now understood until the printing press quickened the spread of topical knowledge. The rule of law was aided by the publication of laws, rules and of the results of legal proceedings. The ability to make a good speech had to give some ground to the ability to cite precedents and decisions from the body of legal knowledge. Science and invention began to work a bit differently with mass publication of findings, with the academy coming to the fore. As the books and publications piled up, special skills were developed in the course of organizing and using them all for research. In 1930, American schools began to organize graduate studies in "Library Science" (Rayward 352 in Machlup ) Some years after that, by 1970, the words "Information Science" were often appended to that department’s hallway placard. And finally just a few years ago, somebody got the idea that this idea of getting a handle on knowledge wasn’t merely practical anymore, it offered a strategic advantage to commercial organizations. In his book "Libraries and the Organization of Knowledge," the eminent librarian Jesse Shera pointed out a similar genesis of the discipline now called Economics from pragmatic individual works to an ill-defined area of study to a discipline with its own schools. He points out in his article (27) that an economy of knowledge would be another frontier soon to be explored and he gave his envisioned new field the name "social epistemology." He outlined it rather clearly:
This was roughly contemporary with Peter Drucker’s first writings of the change in economies from physical to intellectual focus, not to mention McLuhan’s books. The ideas proposed were not exactly championed by his fellow librarians in any way that affected the rest of the world or even the academy, but they are exactly relevant to KM as it applies to codified knowledge. Somehow KM got started without much influence from the library crowd though they are recognized as natural participants in the game. The rise of research publications in tandem with the research university nudged the science librarians to split into a specialty called "Documentalists" before the First World War. This branch of the trade took on the official "Information Science" title in 1968 when the American Documentation Institute became the American Society for Information Science. Some advocates of Information Science such as Shera and Kochen had grand visions for the growth of the field in a direction that was very much similar to a sort of global knowledge management, but the majority still seems to treat it as a slightly more technical domain of library science. Kochen summarizes in his Machlup article:
Explorers and raiders were closing in on the territory of KM on other fronts as well. If you are to believe the textbooks of the late 1980’s, the discipline called "Knowledge Engineering," which is engaged in the building of Expert Systems and Knowledge-based systems, might have taken the world by storm by now. This subdiscipline of computer science was looking to apply Artificial Intelligence and database research to problems of capturing knowledge into somewhat intelligent, expert systems. A knowledge-based systems textbook published in 1991 ended its introduction with this advice:
Although progress has continued, said knowledge has turned out to be much harder to capture and use in a system than was hoped. Many expert systems are in use in special situations but the world of work was not transformed in most cases. KBE is now often a featured special in the menu of KM practices and tools. Some of the aforementioned vendors who have shined up their product lines with new "KM" labels are Knowledge Engineering specialists playing a new game. Back in 1945, one milestone along the way toward a technology to support KM was an article by Vannevar Bush, "As We Might Think" that imagined a tool he called the Memex. This Memory Extension was a desk that essentially contained a personal KM workshop. It could scan in any sort of printed material and then link it all together as the owner desired. What Bush described was very similar to hypertext as we know it as well as several other technologies that are available but have yet to be seamlessly integrated as he described it. His description did not elaborate any collaborative or networked aspect to the Memex other than the ability to automatically link its existing data store with related materials dropped in from another Memex. His idea would fall in to the "personal knowledge manager" side of the current crop of KM tools. Another interesting development that started the very next year was in a confluence of management theory and industrial psychology. In the late 1980’s, a discipline named "Organizational Development" (OD) sought to influence in a comprehensive way how organizations handled growth, training, diversity, reengineering, downsizing and reorganizing. Some of the same writers who influenced Senge, including Chris Argyris, were in this crowd or at least cited by them. This field seems to have been the descendant of the "sensitivity" movement started by Kurt Lewin in 1946 (Patten 4)with the use of "T-groups." Industrial psychologists seemed to be a very strong presence in its ranks and there were also some sociologists. Books about this discipline (Sikes) seem to fade from the scene (according to library records by subject keyword at least) after around 1990, coincident with the publication of Senge’s book. The viable OD ideas pertinent to upper management and strategy seem to have been taken up by Senge and the Organizational Learning movement while the other concerns seem to have been split off into other management/administration specialties such as HR and training. The term OD can still be found in use as a more elegant term for corporate training. Along with librarians, OD people are natural players in the KM game and some KM peddlers will have come to KM through that career path. One keyword topic I discovered in a book on OD was an area of research called "sociotechnical systems" (STS), an early specialty concerning workflow that seems to have been obviated by the introduction of the PC. This area was staked out by industrial psychologists, efficiency experts and systems analysts, and it appears that it was wholly based on shop floor automation type systems in industrial rather than "knowledge" type work situations. Shoshana Zuboff covered a lot of the worker experience side of this discipline in her much-cited "In the Age of the Smart Machine." Her book points out the depth of knowledge, mostly in the tacit dimension, that even factory jobs require and that is not captured in automation efforts. Some other work has been going on for decades in the social sciences that may begin to integrate with the work going on around KM. I already mentioned the trend that has brought anthropologists into the realm of the cubicle. There has also been some work in sociology and social psychology that could prove to be relevant, with all these disciplines perhaps learning to speak more similar dialects to each other. Said one academic in a recent collection on a 1998 conference on social cognition:
This same writer, along with observing that many social psychology concepts and terms have independently derived counterparts (same concept, different discourse) in the management research world, later notes that the psychologists "were introduced to a new title in many organizational charts: CKO, or chief knowledge officer."(Thompson, 2) Of course CKO has been in use at top companies since 1994 (Davenport 1), so it’s good that these people are now talking to each other! This pattern of two populations working on similar or related problems but hardly communicating with each other is often called the functional "silo" or "stovepipe" structure of organizational culture (Martin 88, Brown et al 99). A similar, more willful construct is the "not invented here" syndrome – a group form of solipsism (Wiig 3-15). At the level of comparing two different specialized academic disciplines, segmenting into specialties is considered normal, but this sort of ghetto of knowledge with learning isolated between organizations is one of the chief targets of KM. For the most part, though, KM is a practice that is only applied inside the business world and pretty much in the Fortune 500 part of it – a rather large silo, but a silo nevertheless. There are some signs of KM efforts being applied in the running of universities and governments but these are late-breaking news (*cite*). One application of KM technology that is definitely at work in colleges is the virtual classroom. Both vendor supplied environments and patched together ensembles of tools such as web sites, email and listservs are supporting courses all over the world. Aside from that, KM is an idea that has been thus far totally oriented toward competitive advantage in the business world - but that is only a result of the "spin" applied by the experts in the field to perk up audience interest. The advantage of shared knowledge to society is considerable, but since KM is a set of practices that requires effort and expense, those who want to make a career of it are concentrating on the appeal to the competitive advantage provided to paying customers rather than more general benefits. As we can see from glimpses at the social sciences and library science as they touched on the study of group knowledge, there is an older trend that has sought similar ends to what the gurus of KM seek. With the focus on human understanding in the large rather than competitive intelligence for individual organizations, many pertinent ideas, practices and findings have been noted, but not collected. This is where my interest in showing the larger context of KM is focused. I hope I can show how a widening of scope might point the way to improve how KM is approached and for KM practices and tools to be more of a boon to other communities outside of business. In short, to break through the silos between the different fields of academe and business would be a great thing. Admittedly this way lies utopia, but this tradition has often contributed to practical pursuits. The predecessors to KM, Organizational Development, Information Science and Knowledge Engineering as well as the Quality movement were formed while grand unification visions danced in the heads of their prophets, patrons and champions. As the new wave and its praxis would settle into a niche, its mark would be made on the world but not as near the center of attention as hoped. Organizational Development is a mature field, having its origins with Kurt Lewin’s "T-groups" in the 1940’s, and while consultants from the OD fold have made some forays into KM, the OD community as a whole is trying to cover its back from criticisms like Peter Drucker’s comment to Tom Davenport that more damage was done to business by the sensitivity movement than by the reengineering craze (Davenport "Meeting" 3). Nevertheless, OD has made some contributions in the direction of helping organizations manage their group knowledge, including its influence on the Org Learning field as exemplified by Senge. Both Information Science and Knowledge Engineering are adolescent fields that are still trying to range out of their niches, offering some insights and tricks ready to be shared in the practice of the latest Management craze, KM. Indeed some of the more technoid practitioners of KM are old knowledge engineering hands and it occasionally shows. The KM website of the "KM Consortium" (www.km.org) has a glossary proposal that betrays a deeply technical bias. Most of the terms are defined in the specialized language of knowledge engineering with no evident concession to lay people. In fact the terms are so theoretical that any references to learning and knowledge are worded so as to refer to either machines or people as "agents" that engage in managing knowledge. The entry for Mental Model leads off with "Mental Models are rule systems that are internal to humans" (Swanstrom 7) and this is one of the more plainly worded entries. One angle of KM that has been studied within the world of science and technology but has not been marketed very much as a business service or technology is the diffusion of innovations. This also relates to knowledge networks and the meritocracy of science – who knows what and how knowledge actually spreads or is hoarded. There have been quite a few books written and studies conducted in the attempt to understand how discoveries and new ideas move around. The concepts of "the invisible college" – essentially a CoP - as applied to scientific research communities (Price, D) and scientometrics – the surveying of citations in papers - have come out of these studies. The most notable output from that area of study is the overused and misunderstood paradigm that was gradually popularized after "The Structure of Scientific Revolutions"(Kuhn) appeared in 1962. Ironically, the point where the Kuhnian meaning of paradigm became corrupted was during the 1980’s when it found its way (especially in the term paradigm shift) into the language of social commentators, then business writers like Steven Covey, becoming essentially a synonym for mental model. A recent direction in these types of studies, especially prompted by the work of David Hull (Hull), has steered the discussion into the realm of memetics. There are now even groups trying to apply memetics to management science (Price, I). Another area of research that I think should be a much bigger part of serious KM studies is sociality and group cognition. As mentioned earlier, social psychologists are making some forays into management schools. The work of Linnda Caporael of Rensaeller on the structure of groups and how group relations affect thought could prove helpful. She has pointed out some very basic observations that have not been paid much attention to – that people tend to have certain ways of dealing with each other according to the size of the group involved in the interactions - these being the dyad, the workgroup, the deme and the macrodeme (Caporael). This certainly will have a bearing on the discovery of how knowledge is created and shared in organizations. Another researcher with some basic ideas about group relations is Fiske, who has outlined four different types of interactions that happen universally in group relations (Fiske RM web). Since group knowledge is culture and sociality is the study of culture in action, I hope that the works of these researchers will be brought to bear more often in the study of knowledge at work. On the management science side, the concept of "innovation" is typically only discussed in the sense of how to best encourage it in corporate team settings rather than defining it or tracking its diffusion. One writer who has made a personal quest of collaboration and innovation is Michael Schrage. His work on shared space in the process of prototyping has focused on what works well for high-performing innovators. His conclusion is that the most important part of the process is to have a compelling prototype for skilled people to get excited about with the sustaining motivation of a shared workspace to drive the project forward.
Some work has been done on how to track relationships in companies such as advice and trust networks (Krackhardt 42)and also the flow of technical knowledge through "gatekeeper" networks (Allen 160). In general, there has been very much work done on subjects that are clearly germane to KM but that are not widely recognized to be so. And the reason for this is partly that "silo" effect - the gap in jargon between the various disciplines where this work has taken place and partly because of all the smoke still being blown by vendors and consultants eager to be known as KM experts. They are contributing to a plume of information that continues to spread, obscuring some of the older and less "buzzwordy" works done on group knowledge by people who are mostly innocent of the business world. Finally there is the work being done in spite of the KM craze that is nevertheless knowledge related. Schrage and the Collaborative systems people come to mind. Many continue to work on systems that support group work, mostly in small groups but also there are certain tools that simply are picked up by groups and collaboration emerges in new ways. Many of these people were working on these problems before the KM fad and have not bothered to slap the "KM" shingle outside their office. Collaboration is knowledge management, only a little smaller and faster than the enterprise system types like to focus on. A seminal example is the emergence of the internet itself. The original ARPANET project workers came up with a method of gathering technical input that combined with the new electronic mail and document sharing tools they were building to pave the way for virtual communities of practice. This vehicle for creating technical value was the RFC – Request For Comments document. A proposed tech specification was circulated as a RFC and the comments that came across the wire were combined to refine the specification. The very way that the internet was built was collaborative and involved shared knowledge.
And today this virtual spirit of community can also be seen at play in the linux/open source software community. The linux movement started as a hacker’s free operating system project – supported by the users, who could contribute new features and bug fixes. It is now making a serious dent in the enterprise Operating System market and in 1998 drew fire from Microsoft (cite halloween) , which rallied the community as never before. Yet another field of endeavor converging on KM, perhaps without knowing it, is the work of a special part of the software engineering community. In a sense, software engineering itself is a KM tool, a way to reuse knowledge. The object-oriented software community latched on to Christopher Alexander’s "Pattern Language" idea in the late 1980’s. This way of thinking about designing things has gotten lots of interest and in 1997 and 98, the software developers started to apply that method to higher-level things – the social structures they saw affecting their software process. The "OrganizationalPatterns" community was born and a book that sought to catalog a few of these patterns came out (Brown, AntiPatterns). What these people are doing is a bottom-up sort of way of the same thing Senge has been working on with his more abstract "system archetypes" – perhaps the two will converge soon - if they reach across their silos and adopt some common terms it could happen. One researcher who has made a recent contribution to the vision of how humanity could manage its knowledge with technology is Pierre Levy. His book "Collective Intelligence" builds on the work of both Teilhard de Chardin and Vannevar Bush as well as incorporating recent developments. He is a computer scientist with some recent computer mediated knowledge innovations to his name, but his book is just as visionary as Teilhard and Bush’s writings were. His Cosmopedia is a sort of global knowledge manager. Where Bush’s Memex would be an assistant to any research that a single person had underway, the Cosmopedia is a globally networked meta-memex that is constantly making new associations between all research projects as they proceed and helping to show its users where the connections are. It translates so users can work in concert across cultures and continents. It would host intelligent agents which he calls angels that would help navigate and visualize the landscape of the cosmopedia for its users. It would show people how to find other people with the same needs or interests, in other words, eliciting the formation of communities. Some of the tools of this cybersystem are the knowledge tree, whereby individuals track and share their own interests and learning plan, and the Cinemap, a dynamic relational map of knowledge space(190). This Cosmopedia has some of the messianic flavor of both the Singularity as posited by the extropian community and the Omega point of Tielhard de Chardin. The free-flowing market of ideas and resources that this meta-tool would promote is a feature of what Levy calls "real-time democracy." Levy follows a theme similar to many cultural histories that categorize advances into evolutionary phases, with the next phase being the time when most of mankind will migrate from the commodity space - where we have been since the enlightenment - to the knowledge space, which is the way of thinking that will be enabled by real-time democracy (Levy 75). He characterizes the current time as the noolithic period – the prehistory of the knowledge age (140). But Levy’s envisioned tools have roots in research and there are prototypes of some of the cosmopedic components in play. Certainly there are even some experimental and commercial knowledge visualization browsers that are a step in the direction of his Cinemap and knowledge tree ideas. Several products have dynamic browsers that attempt to represent enterprise knowledge maps – both continuously scrollable and zoomable maps of knowledge bases and also of where people who have certain expertise can be found. These use ideas that are primitive forms of Levy’s ideal tools. A recent market development that has grafted a kind of collaborative feature, the chat window, onto the expert-finding knowledge map idea is the combination of instant messaging with expert maps. One such product is Lotus Sametime. People who participate in the network advertise their areas of expertise and then the system keeps track of whether they are available or reachable in real time. This results in a live map of experts in an organization that can be used to initiate live chat and whiteboarding sessions based on a person’s status on the network at any given time. This is a tool that could be thought of as collaborative and also knowledge-based in the sense of the knowledge of users being put "on tap." The Knowledge banner that is currently flying above the throngs of KM vendors and consultants is not really theirs. The KM craze is simply at the forefront of attention because all enterprise leaders have come to realize that Peter Drucker’s early focus on the knowledge worker was well aimed. There are several other fields of study that deal with organizational innovation and knowledge that simply don’t happen to be known by a phrase that includes "knowledge." What the best of the consultants have to say is not very reassuring, though. They say that knowledge work is hard work – new tools will not do the job of making an organization a knowledge-creating one. Simply focusing on knowledge is not going to create knowledge or conserve the knowledge already in the minds of an organization’s workers. KM tools and practices are a good investment for organizations that are ready for them, and being ready is a feature of good management. As Michael Schrage puts it, "knowledge management is a bullshit issue …I can give you perfect information, I can give you perfect knowledge and it won’t change your behavior one iota" (Schrage interview 4). Schrage is a tad bitter because his pet specialty, collaboration, got overshadowed by the KM craze. The two fields came out of the gate together and are not too far apart - and he has a point - since KM is often an excuse to simply move more static information around the enterprise instead of to enable active engagement, which is the focus of his favorite collaboration technologies. The prep work for KM is dynamic work that affects an organization’s core culture, starting with the leadership. Gurus like Senge and Helgesen, who look to the very basics at work between management team members, show the way to a place where KM tools can provide leverage. John Seeley Brown, who is found quoted in books about Collaboration, KM and CoPs, sums it up well:
KM, whether named as such or not, is simply one of the skills of an activated and self-aware organization. The best Chief Knowledge Officer is one who will strive for the elimination of that position – when for that organization, the transition to the knowledge space has been made and it is simply the norm for day to day work. I would like to take some words from Peter Drucker’s mouth as related by Davenport and ask the reader to apply them to KM, while remembering that an important part of managing knowledge is forgetting what needs to be forgot:
I have talked to peers in the Information Technology field who have expressed frustration with KM or claimed that it’s on the way out. I certainly don’t agree, but Drucker’s insight seems to apply. Since no one person "named" Knowledge Management, I doubt it can be renamed, but the same striving to improve organizational memory and intelligence will continue. I hope I have shown some of the names it has gone by and perhaps some of those it will be known by in the future. |
September 20, 1999
Charles
Ross
[email protected]