/font> Enterprise Information Systems, Inc.
White Paper No. Twelve
Introduction
An
Enterprise Knowledge Management (EKM) Model is a hierarchical network of rules that enables an agent to explain, anticipate and predict events and interaction patterns: (a) in the enterprise's Knowledge (Kn) Processes, or Knowledge Management (KM) Processes; and (b) in the enterprise's environment. An EKM model represents or models the Natural Knowledge Management System (NKMS) [1] of an enterprise.An Enterprise NKMS is an on-going, conceptually distinct unit composed of enterprise organizational and human components and their persistent interactions, both having properties. These interaction properties are not determined by design, but instead emerge from the dynamics of the interaction process itself. Moreover, this interaction process determines the outcomes of the enterprise's knowledge processes. An Enterprise NKMS includes mechanical and electrical organizational components produced by it, such as computers and computer networks, as well as human and organizational agents.
An enterprise
Artificial Knowledge Management System (AKMS) is an enterprise wide conceptually distinct integrated component produced by its NKMS:
A key aspect in defining the AKMS is that both its components and interactions must be fully designed. The idea of being fully designed vs. being partly designed, or not-designed is essential in distinguishing the artificial from the natural. Thus, in an enterprise or any other organization, even though we may try to design its processes, our capacity to design is limited by the fact that it is a Complex Adaptive System (cas) [2] [3] [4]. If we understand a cas, we expect to be able to correctly design its components and certain aspects of its processes, but in the end we expect its behavior to be emergent and to not be a precise consequence of our design. Once the cas starts operating, it is self-organizing and controls its own behavior. At best, we can only interact with it and influence it.
On the other hand, with an AKMS, we design both its components and their interactions. The connection between the design and the final result is determinate and not emergent. When we interact with the AKMS, we can precisely predict what its response will be.
A
Distributed Knowledge Management System (DKMS) [5] [6] [7] [8] [9] of an enterprise is a specific type of AKMS designed to manage the integration of distributed computer hardware, software, and networking objects/components into a functioning whole supporting enterprise knowledge production, acquisition, and transfer processes. The DKMS, in other words supports producing the enterprise's knowledge base. The relationship of the DKMS to the EKM model is one of mutual feedback and virtuous circularity over time. And it is also one of complexity and detail touching on many aspects of both the EKM model and the DKMS. This paper examines this relationship and articulates its many facets.
Each rule in an EM's rule network relates antecedent attribute values to consequent attribute values, concepts, or rule sequences. The attributes involved belong to a number of concepts that represent the components of the model. Declarative Rule networks are those whose rules fire in parallel to determine an outcome. Procedural Rule networks are those whose rules fire in sequence. An EM is composed of both declarative and procedural rule networks. Figure One provides an illustration of declarative and procedural rule networks.
[Image Figure 1 - Procedural Rule Network; inputs seem to be processed by a
Declarative Rule Network comprised of 5 levels, that feed possibly a
"Communication Limitation Rule" process, and then pass through a
"Transformation Rule" process, which yields a "Decision" as output.]
An EM's rule network is made up of rule patterns that an agent can use to
describe, understand, and explain enterprise interaction patterns, in order to
anticipate and predict future interaction patterns. The agent learns to
anticipate and predict better by adding, removing, and changing these rule
patterns based on it's beliefs about the enterprise's experience, its knowledge
validation criteria, and other competing models emerging from the enterprise's
knowledge creation, or knowledge retrieval processes. When an EM is confirmed
by testing it against competing EMs, in the context of validation criteria, and
by evaluating it favorably against the alternatives, it then becomes an
instance of knowledge for the agent or agents who accept both the validation
criteria, and the factual claims about the results of the testing process.
Validation criteria for EM's and other rule networks will vary across enterprises. Even within the scientific community there is no commonly agreed upon, precisely specified, and prioritized set of validation criteria [10]. So, clearly there will generally be no consensus across enterprises on validation criteria. As a result, there will also be no consensus on knowledge claims. What falls into the mere "information," and what falls into the "knowledge" categories is likely to differ across similar enterprises, and even across groups within the same enterprise.
Enterprise models are produced by the knowledge production process. In turn, the knowledge in EMs, as well as additional knowledge produced by the knowledge production process, is continuously used as the basis for action in implementing business processes other than knowledge processes, such as Sales, Marketing, Financial Management, Customer Care, etc. Knowledge use is not a knowledge process itself. Instead, it is an activity that is part of every knowledge process and every other business process as well.
There are three high-level knowledge processes that may be modeled in EMs and that also may be decomposed further [11]. These are: Knowledge Production; Knowledge Acquisition; and Knowledge Transmission. Before discussing them though, it's time to provide a definition of enterprise level knowledge.
Enterprise-level Knowledge
The previous definition of an EM goes a long a way toward providing a definition of an enterprise's knowledge. The hierarchical network of the enterprise's validated rules is the knowledge base of the enterprise. Such knowledge enables it to explain, anticipate, and predict events and interaction patterns in the enterprise and in its environment. The knowledge base formed by this rule network of the enterprise, contains: its set of remembered data; its validated propositions and models (along with metadata related to their testing); its refuted propositions and models (along with metadata related to their refutation); its metamodels; and (if the system produces such an artifact) the software it uses for manipulating these.
The enterprise's knowledge base is an abstract phenomenon. And it is one that emerges from the cas-style interaction of the various agents comprising the enterprise. Measures of an enterprise's knowledge base may be found in its cultural artifacts [12], including its linguistic products, its electronic artifacts, and its artistic expressions, if any.
The enterprise knowledge base is frequently viewed as being composed of explicit knowledge and tacit knowledge. In analogy to individuals, explicit knowledge is often thought of as knowledge that is verbally communicated, sharable, and a matter of public record, while tacit knowledge is seen as hidden, non-verbal, and tightly held. This distinction will not quite hold for the enterprise. Enterprise's don't keep secrets in quite the same way as individuals. Barring security classification of knowledge, all other knowledge is reflected in some way by the enterprise's artifacts, because the knowledge could not emerge without producing it through communication that leaves a trail of artifacts.
Some knowledge is highly explicit in that it is stated in documents or lingui stic formulations that are identified as enterprise knowledge sources whose contents has been previously validated. Formal, automated knowledge bases are examples of this type of explicit knowledge, as is the organization chart of the enterprise, the enterprise logo, etc. Other types of knowledge though, are not gleaned or measured so easily from cultural artifacts, and may not be explicitly acknowledged by the enterprise. Nevertheless, such "tacit" knowledge is not hidden, and non-verbal as is the case for the individual, it just requires more careful measurement modeling and inference from cultural artifacts to tease it out of them.
This distinction between explicit and tacit enterprise level knowledge talks past the currently important issue in KM circles of safeguarding or adding to enterprise knowledge assets by capturing the tacit knowledge of employees. This is an understandable continuing concern of KM. But the conceptual distinction implicit in this issue is that between the tacit knowledge (non-verbal and hidden) of individuals and the explicit knowledge of the enterprise. It has nothing to with the distinction between the explicit and tacit knowledge components of the enterprise knowledge base.
The process of producing knowledge is depicted in Figure Two. The core of knowledge production is the activity of formulating knowledge. I use the term "formulating" in a general sense. It encompasses (a) formulating new knowledge, (b) revising and refining previously produced knowledge, and (c) reformulating previously produced knowledge.
[Image Figure 2 - Knowledge Production Process - seems to show that knowledge use entails a continual process of searching, storing, retrieving, formulating, storing, testing, storing, concluding.]
In order to formulate knowledge we need to use previous knowledge about how to formulate knowledge. So knowledge use is a necessary requirement for formulating knowledge.
Formulating new knowledge is preceded by retrieving knowledge previously stored as the result of searching and receiving. Once knowledge is formulated, it is again stored pending testing of the knowledge claims produced by the activity of formulating knowledge. And following testing, knowledge may again be stored pending assessment of the results of testing and the activity of arriving at a conclusion rating competing knowledge claims against one another. This notion of rating competing knowledge claims does not, in general, imply rigorous rating or comparison activities. Rigorous validation methodology is not excluded as either an empirical possibility or a normative preference. But in the real world of enterprises, and even in scientific activity, the activity of concluding or validating is most often carried out with fuzzy comparisons and ratings.
In knowledge production, the activity of concluding is again followed by the activity of storing. In fact, each stage of activity within the knowledge production process is followed by knowledge storing activity. This reflects the difficulty of storing any appreciable amount of knowledge in human memory.
The knowledge production process is highly dependent for its effectiveness on knowledge storing. It is also highly dependent on using previous knowledge to implement all of the activities comprising the knowledge production process. And from this it is clear that both knowledge storing, and knowledge use, are not independent knowledge processes, but broad ranging activities that occur within knowledge production, knowledge acquisition and knowledge transmission as well.
Part of the activity of concluding is selling the knowledge that has been formulated and validated by groups and teams within an enterprise. By "selling," in this context, I mean persuading others that the knowledge claims that have been formulated and validated by one's team or group have, in fact, been justified. Selling, in other words, is persuading the knowledge production political system in an enterprise, that the products of the knowledge formulation process in teams or groups, their validated knowledge claims, are indeed valid, relevant, and useful at the level of the enterprise.
I distinguish five types of knowledge production according to differences in the linguistic form of the outcome of knowledge production. These are: Planning Knowledge, Descriptive Knowledge of Fact, Measurement and Non-causal Modeling, Causal Modeling Knowledge, Predictive Knowledge, and Assessment (cost/benefit) Knowledge. The general outlines of the knowledge production process are not different among these types, but there are differences in the details of formulating and testing knowledge of different types, because the different types of knowledge exhibit linguistic differences in the form and semantics of their characteristic rules.
Knowledge Acquisition
[Image Figure 3 - Knowledge Acquisition Process, seems similar to Image 2.]
Search activity is followed by the activities of data, information and external "knowledge" gathering. The activity of gathering may or may not be accompanied by purchase activity, depending on whether the external content carries a price. The last activity in knowledge acquisition is filtering and integrating the new acquisition into the enterprise's knowledge base.
Filtering includes cleaning and transforming data, information, and external knowledge, and staging it for loading into the enterprise's own stores [13]. All external knowledge is initially stored as information, unless filtering is supplemented by using the knowledge production activities of testing and validating to validate external knowledge as knowledge from the viewpoint of the enterprise.
Of course, gathering, purchasing, and filtering data, information, and knowledge, presupposes using previously produced knowledge to gather, purchase, and filter. You need to know how to get data from external sources, whether gathering requires writing and mailing a letter, faxing a document, or downloading a data set from the Internet.
Knowledge Transmission
Knowledge sharing means making knowledge available for retrieval by consumers. This can be done by placing documents in a library, or by placing them on the enterprise Intranet. Knowledge sharing is often preceded by "pushing" information or knowledge about the availability of knowledge that may be shared.
Searching/retrieving refers to the activities of searching the enterprise knowledge base, and retrieving certain knowledge from it. These activities can occur through using an electronic or a personal network to do the searching and/or the retrieving. While searching/retrieving through the enterprise electronic network is very powerful, and becoming more so as technology advances, one's personal network will remain a primary source of knowledge that is unobtainable from more formal sources.
Face-to-face knowledge transfer refers to both casual face-to-face knowledge transfer within informal groups, and also formal education and training contexts in which knowledge is transferred from teacher to student. The first provides the platform for transferring tacit knowledge by making it explicit. The second provides the means of transferring explicit and enterprise sanctioned information and knowledge in a highly organized and disciplined way.
All of the above knowledge transmission activities involve using previously produced knowledge. To "push" knowledge you need to know how to use techniques for pushing it. Those receiving it need to have minimal knowledge to use what they have received. To share knowledge you need to know how to transfer a document to the enterprise library, or how to get your document posted on the enterprise's web site, and so on. Similar examples could be given for searching/retrieving, and knowledge transfer. Knowledge use, again, is an inextricable part of knowledge transmission.
Keeping Levels Straight
Each intelligent agent above the level of the individual has its own knowledge production, knowledge acquisition and knowledge transmission processes. These need to be distinguished from the processes of every other agent. In particular, what is knowledge for one agent, may be only information for another. Knowledge validation for one agent doesn't automatically carry over to lower level, higher level, or agents at the same level of analysis. What is knowledge transmission for one may be only information transmission for another, and so on.
To keep straight what is data, what is information, and what is knowledge, we need to keep straight the level of our current discourse, and be constantly aware of when we are shifting levels. If we do not, we will be constantly confusing enterprise knowledge processes with the knowledge processes of lower level agents, and the portion of our EM dealing with knowledge processes and their impact on other business processes will become confused and unworkable.
KM is handling, directing, governing, or controlling knowledge processes within an organization in order to achieve the goals and objectives set by the organization for these knowledge processes. An Enterprise Knowledge Management Model (EKM) has the same general form as an EM model, but it enables an agent to explain, anticipate and predict events and interaction patterns either in the enterprise's knowledge processes, or in its NKMS.
The Knowledge Management Process (KMP) is an on-going persistent interaction among human-based agents within the NKMS who aim at integrating all of the various agents, components, and activities participating in the three knowledge processes into a planned, directed, unified whole, producing, maintaining, enhancing, acquiring, and transmitting the enterprise's knowledge base. Knowledge Management is human activity that is part of the interaction constituting the KMP. In modeling the NKMS, EKMs model KM activity and its consequences. But what kind of activity is it, and what is the subject matter of such EKM Models? The next two sections on levels of knowledge management, and knowledge management activities will address these questions.
Levels of Knowledge Management
Knowledge processes occur at the same level of agent interaction as other business processes. Let's call this business process level of interaction level zero of enterprise cas interaction. At this level pre-existing knowledge is used by business processes and by knowledge processes to implement activity. And, in addition, knowledge processes produce, acquire and transmit knowledge about business processes, using (a) previously produced knowledge about how to implement these knowledge processes, (b) infrastructure, (c) staff, and (d) technology, whose purpose is to provide the foundation for knowledge production, knowledge acquisition, and knowledge transmission at level zero. But where does this infrastructure, staff, knowledge, and technology come from, who manages it, and how is it changed?
They don't come from, by and through the level zero knowledge processes. These knowledge processes only produce, transfer, and acquire knowledge about business processes. So, this is where level one of cas interaction, the lowest level of knowledge management comes in. This level one KM process interaction is responsible for producing, acquiring, and transmitting knowledge about level zero knowledge production, acquisition, and transmission processes to knowledge workers at level zero. It is this knowledge, developed in a level one EKM model, which is used at level zero to implement its knowledge processes.
The KM process and EKM model at level one are also responsible for providing the knowledge infrastructure, staff, and technology necessary for implementing knowledge processes at level zero. In turn, knowledge processes at level zero use this infrastructure, staff, and technology to produce, acquire, and transmit the knowledge used by the business processes. The relationships between level one KM and level zero knowledge and business processes are illustrated in Figure Four.
[Image Figure 4 - Level Zero/Level One KM Process Relationships, shows interaction between different levels of knowledge processing in an organization, which support "business processes."]
Knowledge about level zero knowledge processes, as well as infrastructure, staff, and technology change when first level KMP interactions introduce changes. That is changes occur: when the level one KMP produces, acquires, and transmits new knowledge about how to implement level zero knowledge processes; and when it adds or subtracts from the existing infrastructure, staff, and technology based on new knowledge it produces. There are two possible sources of these changes.
First, knowledge production or acquisition at level one can change the EKM model, which, in turn, impacts on (a) knowledge about how to produce, acquire, or transmit knowledge about (level zero) business processes, (b) knowledge about how to acquire or transmit knowledge about level one knowledge acquisition or transmission processes (c) staffing, (d) infrastructure, and (e) technology. This type of change then, originates in the level one KM process interaction, itself.
Second, knowledge about how to produce the level one knowledge in the EKM model may change. This knowledge however, is only used in arriving at the level one EKM model. It is not explained or accounted for by it. It is determined, instead by a level two KM interaction process and is accounted for in a level two EKM model produced by this interaction. Figure Five adds the level two KM process to the process relationships previously shown in Figure Four.
[Image Figure 5 - Level Zero - Level Two KM Process Relationships, shows additional knowledge processing level added to Figure 4.]
Instead of labeling the three levels of processes discussed so far as level zero, level one, and level two, it is more descriptive to think of them as the knowledge process level, the KM level and the meta-KM level of process interaction. There is no end, in principle, to the hierarchy of levels of process interaction and accompanying EKM models. The number of levels we choose to model and to describe, will be determined by how complete an explanation of knowledge management activity we need to accomplish our purposes.
The knowledge process level gives us knowledge about business processes. The KM level gives us knowledge about how to arrive at knowledge about business processes. The meta-KM level gives us knowledge about how to arrive at knowledge about knowledge processes. Level four, the meta-meta-KM process level of interaction would give us knowledge about how to arrive at knowledge about the KM process. Level four then, seems to be the minimum number of levels needed for a fairly complete view of KM. And in some situations, where we need an explanation of our knowledge about how to arrive at knowledge about the KM process we may even need to go to a fifth, meta-meta-meta-KM level.
Enterprise Knowledge Management Activities
This hierarchy, ranging from activities to processes, applies to knowledge and KM processes as well as to operational business processes. Enterprise KM activities may be usefully categorized according to a scheme of task clusters which, with some additions and changes, generally follows Mintzberg [17]. There are three types of KM task clusters: interpersonal behavior, information (and knowledge) processing behavior, and decision making. Each type of task cluster is broken down further into more specific types of task pattern activities below.
Interpersonal Behavior
Knowledge and Information Processing
Decision Making Task Clusters
The DKMS
Two ways to look at the DKMS are in terms of its architecture, and its use cases. Let's proceed to examine its architecture, and then its use cases.
Architecture
Top - Down and Bottom-Up data warehousing architectures may be viewed as two-tier architectures utilizing clients and local or remote databases [18]. Enterprise Data Mart Architecture (EDMA), Data Stage/Data Mart Architecture (DS/DMA), and Distributed Data Warehouse/Data Mart Architecture (DDW/DMA) may be viewed as adding middleware and tuple layers to earlier architectures to provide the capability to manage data warehouse systems integration through unified logical views, monitoring, reporting, and intentional DBA maintenance activity [19]. But tuple-layer based management still doesnt provide automatic feedback of changes in one component of a data warehousing system to others.
DKM architecture may be viewed as adding an object layer to EDMA or to DDW/DMA to provide integration through automated change capture and management. Figure Six depicts DKM architecture. The object layer contains process distribution services, an in-memory object model, and connectivity to a variety of data store and application types. The layer requires an architectural component called an Active Knowledge Manager (AKM) [20].
[Image Figure 6 - DKM Architecture - seems to show overall hardware and software components and flow diagram for an enterprise-wide KM system, with Data Warehouse, Unified Data Model, and Data Marts.]
The Active Knowledge Manager
An AKM provides process control and distribution services, an object model of the DKMS, and connectivity to all enterprise information, data stores, and applications.
Process Control and Distribution Services include:
The in-memory Active Object Model and Persistent Object Store Model components of the AKM include:
Connectivity Services of the AKM include:
Only the DKMA among the preceding architectures, supports distributed, proactive monitoring and management of change in the web of data warehouse, data mart, web information servers, component transaction servers, data mining servers, ETML servers, other application servers, and front-end applications comprising today's Enterprise DSS/Data Warehousing System. Figure Seven shows the relationship of the AKM to front-end tools, application servers, and data stores.
[Image Figure 7 - .....]
Use Cases
[Image Figure 8 - .....]
Here are three side-by-side lists. A list of knowledge processes, one of associated Kn and KM activities, and a list of DKMS use cases that might be implemented in a comprehensive KM enterprise wide application. These also illustrate the point of partial support of the activities and the processes by the DKMS use cases.
Table One Kn and KM Processes and DKMS Use Cases
Knowledge & KM Processes |
Activities Within Processes |
DKMS Use Cases |
||
Knowledge Production | Searching (within the enterprise for data, information, or knowledge) |
|
||
Receiving (transmitted data, information, or knowledge) |
|
|||
Storing (and loading data, information and knowledge) |
|
|||
|
Retrieving (data, information or knowledge) |
|
||
Formulating new knowledge claims |
|
|||
Testing knowledge models and claims |
|
|||
Concluding (about knowledge models and claims) |
|
|||
|
Using Previously available Knowledge |
|
||
Knowledge Acquisition |
Searching (for data, information, or claimed knowledge external to the enterprise) |
|
||
Gathering (externally located data, information or claimed knowledge) |
|
|||
Purchasing (data, information, or claimed knowledge) |
|
|||
Filtering (including cleaning, transforming and staging data, information, or knowledge claims) |
|
|||
|
Testing knowledge models and claims from external sources |
|
||
Concluding (about knowledge models and claims from external sources) |
|
|||
Storing (and loading data, information, and knowledge) |
|
|||
Using Previously available Knowledge |
||||
Knowledge Transmission |
"Pushing," data, information and knowledge |
|
||
Sharing knowledge (by making it available) |
|
|||
Searching/retrieving data, information, and knowledge using an electronic network |
|
|||
Searching/retrieving data, information, and knowledge using personal networks
|
|
|||
Face-to-face knowledge transmission within small groups |
|
|||
|
Face-to-face knowledge transmission in formal training and education |
|
||
|
Storing (and loading data, information, and knowledge) |
|
||
|
Using Previously available Knowledge |
|||
Representing (at ceremonies) |
Signing Contracts |
|||
Attending Public Functions for KM |
||||
Meeting and relating to dignitaries |
||||
Leadership |
Hiring KM Staff |
|
||
Providing for KM Staff Training |
|
|||
Motivating KM Staff |
|
|||
Monitoring KM Staff |
|
|||
Evaluating KM Staff |
|
|||
Building relationships with individuals and organizations external to the enterprise |
|
|||
KM Knowledge Production | Activities are the same as those specified for Knowledge Production |
|
||
KM Knowledge Acquisition | Activities are the same as those specified for Knowledge Acquisition |
|
||
KM Knowledge Transmission | Activities are the same as those specified for Knowledge Transmission |
|
||
Changing Knowledge Process Rules |
Deciding to change Knowledge Process Rules |
|
||
Directing use of KM Knowledge Transmission to transmit new rules and mandate for using them |
|
|||
Crisis Handling | Various activities associated with other processes implemented in a "time box" mode |
|
||
Forecasting developing crises or crisis potential |
|
|||
|
Monitoring developing crises |
|
||
Evaluating developing crises |
|
|||
Contingency Planning for crisis |
|
|||
|
Evaluating crisis contingency plans against crisis management performance |
|
||
Allocating KM Resources and mandating implementa-tion for: |
KM infrastructure |
|
||
Training |
|
|||
Professional Conferences |
|
|||
Compensation for KM staff |
|
|||
Funds for new KM programs |
|
|||
Negotiating with business process representa-tives about: | Levels of effort for KM |
|
||
|
The scope and content of KM programs |
|
||
KM ROI targets |
|
|||
KM infrastructure support for business processes |
|
|||
KM staff support for business processes |
|
The breadth of DKM architecture discussed earlier, and the range of identifiable DKMS use cases just presented, illustrate the comprehensiveness of the DKMS as an integrative enterprise application supporting knowledge processes and KM across the board regardless of whether the necessary support requires batch, OLTP, or DSS processing. The comprehensiveness of the DKMS, and its integrative capability, far exceed that of either Data Warehousing solutions or Enterprise Resource Planning (ERP) solutions. The data warehouse as a conceptual construct provides only limited decision support capability in the form of querying and reporting against integrated relational data, while ERP solutions are focused on OLTP processing and fall far short in the DSS area.
Both classes of enterprise solutions are broadening in scope as time goes on, and as the demands of the market place force extensions of conventional data warehousing and ERP solutions. But, increasingly, extended data warehousing and ERP solutions are beginning to require the kind of functionality available from the DKMS. Thus both types of enterprise wide solutions are beginning to converge on DKMS architecture and use case functionality.
EKM Models and the DKMS
The relationship of the DKMS to an EKM model is a complex one of mutual feedback and virtuous circularity over time. The EKM model and the DKMS such an extensive and complex relationship because both operate at different hierarchical levels and impact each other both within and across levels. Figure Nine illustrates this relationship for levels zero to four of the Kn/KM levels hierarchy.
The EKM model is one of the continuously improved outcomes of the DKMS. Yet, the DKMS is also, at any level of analysis (n), both an outcome of a higher level EKM model at level (n+1), and partly an outcome of the EKM model at level (n). This means, that if the EKM model at a specific level is to be developed, it must be done hand-in-hand with the DKMS, and similarly development of an effective DKMS is equally reliant on development of an effective EKM model at the level above the specific DKMS level in question. An EKM model is thus the intelligence behind the DKMS. But the DKMS, in turn, is the tool for developing the intelligence behind an EKM model.
References
[1] Compare Edward Swanstrom, "What is Knowledge Management?" Discussion Rough Draft of a Chapter undergoing editing by John Wiley & sons, available at http://www.km.org/introkm.html.
[2] John H. Holland, Hidden Order (Reading, Mass.: Addison-Wesley, 1995)
[3] John H. Holland, Emergence (Reading, Mass.: Addison-Wesley, 1998)
[4] M. Mitchell Waldrop, Complexity (New York: Simon and Schuster, 1992)
[5] I introduced the DKMS concept in "Object-Oriented Data Warehousing," available at http://www.dkms.com/White_Papers.htm.
[6] Joseph M. Firestone, "Distributed Knowledge Management Systems: The Next Wave in DSS," available at http://www.dkms.com/White_Papers.htm.
[7] Joseph M. Firestone,"Architectural Evolution in Data Warehousing," available at http://www.dkms.com/White_Papers.htm.
[8] Joseph M. Firestone, "Knowledge Management Metrics Development: A Technical
Approach," at http://www.dkms.com/White_Papers.htm.
[9] Joseph M. Firestone, "DKMS Brief No. Four: Business Process Engines in Distributed Knowledge Management Systems," available at http://www.dkms.com/White_Papers.htm.
[10] Usama M. Fayyad, Gregory Piatetsky-Shapiro, and Padhraic Smyth, "From Data Mining to Knowledge Discovery: An Overview," in Usama M. Fayyad, Gregory Piatetsky-Shapiro, Padhraic Smyth, and Ramasmy Uthurusamy (eds.), Advances in Knowledge Discovery and Data Mining (Cambridge, MA: M.I.T. Press, 1996).
[11] My thinking on knowledge processes was developed in extensive conversational and written exchanges with Edward Swanstrom, Executive Director of the Knowledge Management Consortium (KMC) to whom I am very grateful. Of course, the views expressed here are solely my own, and may not agree either wholly or in part with those currently being developed by Mr. Swanstrom, or by the KMC.
[12] Joseph M. Firestone," The Development of Social Indicators from Content Analysis of Social Documents," Policy Sciences, 3 (1972), 249-263.
[13] Ralph Kimball, Laura Reeves, Margy Ross, and Warren Thornthwaite, The Data Warehouse Life Cycle Toolkit (New York: John Wiley & Sons, 1998).
[14] The application of the notion of a levels hierarchy to KM was suggested to me by presentations and unpublished drafts of Edward Swanstrom's, and by conversations with him, though again the specifics of my application will most probably differ from his treatment of the subject.
[15] Alfred North Whitehead and Bertrand Russell, Principia Mathematica (London: Cambridge University Press, 1913)
[16] Gregory Bateson, "The Logical Categories of Learning and Communication," in Bateson, Steps to an Ecology of Mind (New York: Chandler Publishing Company, 1972)
[17] Henry Mintzberg, "A New Look at the Chief Executive's Job," Organizational Dynamics," (AMACOM, Winter, 1973)
[18] Joseph M. Firestone,"Architectural Evolution . . ."
[19] Ibid.
[20] Ibid.
Biography
Joseph M. Firestone is an independent Information Technology consultant working in the areas of Decision Support (especially Data Marts and Data Mining), Business Process Reengineering and Database Marketing. He is developing an integrated data mining approach incorporating a fair comparison methodology for evaluating data mining results. In addition, he is formulating the concept of Distributed Knowledge Management Systems (DKMS) as an organizing framework for the next business
"killer app." Dr. Firestone is one of the founding members of the Knowledge Management Consortium (KMC), a collaborator in its Enterprise Knowledge Management Modeling Project, and the Chairperson of KMC's Artificial Knowledge Management System Committee. You can e-mail Joe at eisai@moon.jic.com.