This dissertation proposes a new way of conducting long-term, large-scale social science research.  In order to understand both the present and proposed social systems of research, we need to examine many issues.  These issues include the nature of the research process itself, the nature of the scholarly establishment, the supporting technology that served in the past and the technologies that are necessary to support research in the future.  While the success of large-scale long-term research in the physical sciences is quite clear, social science research is dominated by solo or small-scale efforts.  We need to know why; and when we do we may see how new research efforts can be conducted on a scale that matches our ignorance of social processes and problems.

Figure I below shows some of the shortcomings of current research methods.  This is to be contrasted with Figure IV that shows the conceptual framework of this dissertation.

Figure I
Existing Collaborative Environment

We start with the selection of scientific realism as an underlying philosophy.  The research process is currently driven by the team's understanding of the scientific method and by the customs of their discipline.  The participatory process, so necessary in large-scale research, is currently typically unstructured, dominated by unrecorded dialog between the participants -- often on a dyadic basis rather than on a team basis.  Participation is now dominated by the sharing of document drafts in a process where criticism is not shared, but is generally channeled to the team leader.  This concentration of power is antithetical to current concepts of social participation.

We hope to remediate many of the problems in current research environments with a new process that again begins with the selection of scientific realism to guide our research.  A methodological approach to research will guide the research team from its conception to maturity; we select a methodology that resonates with realism. We end with a proposed conceptual framework that allows us to relate the interacting elements that comprise a research process and technological environment suited to large-scale long-term social scientific research collaboration.

2.1  Scientific Realism and Models

"There are only real things and the real ways they behave. And these are represented by models, models constructed with the aid of all the knowledge and techniques and tricks and device we have."

"The Fact is the basis, the foundation; Imagination, the building material; the Hypothesis, the ground plan to be tested; Truth or Reality, the building."

---J.H. van't Hoff (1852-1911) Chemist, Musician, Mathematician, Positivist

A realist believes in an objective existence of the objects found in the research process. The scientific realist might be one who is persuaded by the arguments found in Realism Rescued: How Scientific Progress is Possible (Aronson, Harré, and Way 1995). The principal argument in that work is that theory is actually embodied in a suite of three models. These models are the source, descriptive, and the explanatory models. The source model is an unexpressed model of reality from which the descriptive and explanatory models are derived.  The source model is the union of the tacit knowledge of the team members, the literature that has been identified by the team, and the intellectual content of the current received wisdom.  The model may include myth, ideology, and dogma (Denzau and North 1994, 3). The descriptive model is a representation of the objects and relationships between the objects of the issue domain. The explanatory model is a representation of the causal power of objects and the mechanisms driving the processes (Sayer 1992, 104).

I have extended the philosophical argument to include a simulation model.  This model takes findings from the descriptive and explanatory models and creates an environment that can be used to investigate the effects of change in the system.  The simulation model, if provided the state conditions existing in reality at a past time, and the inputs subsequent to that time will produce a description of the system state at the end of the input actions.  If the output of the simulation model consistently produces results practically resembling the actual behavior of the system, then the entire chain of modeling from source to simulation is validated by correspondence (Ziegler 1990, 2, 27).

The support that scientific realism provides to research is a rationale for modeling practice.  The descriptive model is an organized representation of all the information acquired and developed about the objects and their relationships. The explanatory model provides information of an operationalizational nature to the simulation model: qualitative and quantitative parameters.  Both the descriptive and explanatory models may be represented by diagrams and supporting text derived from existing genres, especially from software engineering and artificial intelligence (DeMarco 1979), (McMinamin and Palmer 1984), (Rumbaugh 1991), (Ziegler 1990).  Simulation modeling requires much more formality in modeling and calls for the rigor of Unified Modeling Language (UML) (Booch, Rumbaugh and Jacobson 1999), (Fowler 1997), Object Role Modeling (ORM) (Halpin 1999) or other formal methods.  The simulation model is a computer program represented by algorithms, which are in turn supported by a variety of graphic models and textual documentation. The model representation, as a genre, is discussed at length below (see §3.2.4.2).

We will see that the Research Process, described below in Figure I, is composed of three interdependent domains that are supported by the modeling rationale of scientific realism. The research work within the substantive domain, the observed world, is represented in the descriptive model. The Explanatory model represents the relationships among elements of the conceptual domain, the causally driven world. The experiments undertaken in the methodological domain, the manipulable world, are reiterated in the Source (simulation) model.

2.1.1  The Source Model
The descriptive and explanatory models are based on the team's source model.  This source model is simply all the knowledge the team has about the issue domain, the perceptions and presumed processes that constitute what is known by the researchers.  It is the union of the tacit knowledge of all research team members, the scientific literature known to members, and any new knowledge produced by the team.  Note that tacit knowledge may contain knowledge that is in fact not true (Dunn 1982): knowledge that is false, myth, or error.  Philosophical myths and dogma may also be a part of each member's tacit knowledge (Harré 1970, Chapt.1).  To a large extent, the source model is the union of all the team member's mental models.  Mental models are cognitive categories, beliefs about the world, held by individuals (Denzau and North 1994, 13).  The source model is unexpressed; but as it becomes expressed its objects and its processes become part of the descriptive model.  Like any other model, the source model changes with new knowledge; errors are excised and new knowledge, beliefs and conjectures are incorporated.

Tacit knowledge exists on a continuum of codification of the mental model through culture and routines into expressed knowledge.  The qualities of codifiability and teachability influence the rate of knowledge transfer (Zander and Kogut 1995, 77).  Moving tacit knowledge to more explicit codified forms is one of the principal jobs of the knowledge network known in this research as the Research Web.  The means of transforming knowledge is communication, and communication of tacit knowledge is dependent on the strength of social ties: strong ties are needed to transfer tacit knowledge, while explicit knowledge is easily transferred through weak ties (Augier and Vendelø 1999, 255).

Since the source model is unexpressed, it cannot be criticized.  As elements of the source model become transformed to explicit knowledge and expressed in other models, the other members can subject the elements to critical review.  The Research Web is designed to support this criticism.

2.1.2  The Descriptive Model  
The Descriptive Model (DM) is the repository for knowledge related to the observed reality of the issue domain.  A description of every type of tangible entity (object) that exists within the issue domain has a place in the DM.  Every process that is observed to operate in the issue domain is described in temporal, spatial and/or social terms.  Processes relate objects within the issue domain and relate parts of the issue domain to reality outside the issue domain.

As the elements of the source model are expressed they become part of the descriptive model.  Objects are identified and refined by extension and criticism.  Relationships between objects are expressed and entered into diagrams that organize the knowledge about the objects.  Observed processes are described, showing the actual operation of some of the activities within the issue domain.  The Research Web, through hypertextual organization, takes its shape from the hierarchical organization of the objects into natural types.  Process models relate all the objects and show how they operate in time and space and perhaps socially as well.

The Descriptive Model is created, extended, criticized and refined by the entire team, but most especially by those members who work in the Substantive Domain of the Research Process (see §2.3.2.1, below).  As time passes, regardless of the stage of completeness, the Descriptive Model is continually revised and those changes cause ripples to perturb the other models of the system.  The DM is the model that the Explanatory Model is modeled on.

2.1.3  The Explanatory Model

The Explanatory Model (EM) is inextricably paired with the Descriptive Model (Aronson, Harré, and Way 1995, 51).  While the DM seeks to represent the issue domain, the EM seeks to provide causal mechanisms driving the issue domain.  The EM is built on the elements, objects and processes, from the descriptive model and the set of hypotheses that emerges from the research.  The EM provides a model of a hypothetical generative mechanism for the observations of the DM (Harré 1978, 275).  For each observed process from the DM there will ideally be a set of explanations based on testable hypothesis.  The testing is carried out by experiment to corroborate the developing theory.

The construction of the EM is the principal responsibility of team members working in the Conceptual Domain of the Research Process (see §2.3.2.2 below).  The EM becomes the basis for hypothesis generation.  As findings accumulate from experiment and research, the EM becomes more mature and comprehensive and eventually becomes a definitive representation of theory.

2.1.4  The Simulation Model
The simulation model (SM) is derived from the DM and the EM.  The SM is the model driving the work of those researchers operating in the Methodological Domain of the Research Process (see §2.3.2.3 below).  The SM can be used to develop describe hypotheses and develop experiments.  In this way the SM provides results for the further refinement of the DM and EM.  The SM will be implemented late in the research process, in stage 3 (see §2.3.3.3 below) or the corroborating phase.  In many research projects the SM will not be fully implemented due to both time and expense.  The simulation model will be a very major undertaking which can really only be justified in long-term, large-scale research projects -- just the project configuration proposed for the Research Web.

Of great importance to the development of the Simulation Model is another model, called the Auxiliary Model, which explains the operationalization of the elements of the SM from their origins in the DM and EM.  The auxiliary model concept was proposed by Blalock (Blalock 1968, 24).  A carefully designed auxiliary model insures that measurements of a given attribute of a DM variable are properly represented in the SM.  Blalock (Blalock 1990, 34) later suggested that auxiliary models are necessary "whenever our theories are about postulated properties (such as attitudes, values, and abilities) that are only indirectly inferred on the basis of directly observed behaviors."  He explores four specific situations calling for careful use of auxiliary models, including very common situations such as the need to use data from two or more levels of analysis, and situations where the measured variable is intrinsically fuzzy or boundaries are rather arbitrary.  The auxiliary model is also necessary to operationalize variables whenever an experiment is designed to explain any phenomenon (Hox and Mellenbergh 1990, 124).

The hypertextual organization of the RW is essential to demonstration of the operationalization rationale of every element of the simulation model.  Since the auxiliary model is diagrammatic and textual, it may be easily criticized.  In short, the auxiliary model provides validation of the SM, under the assumption that the DM and EM are correct.

...it seems wise to develop a hypercritical stance that brings many ... hidden assumptions into the open as possible, and then to examine very carefully their implications.

In practice the auxiliary model will not exist as an entity, but will be expressed as augmentations to the DM.  The nature of the augmentation is to explain how the attributes of objects are measured.  The auxiliary model can be thought of as the argumentation that validates every operationalization decision in the SM.  Each of these arguments can be independently criticized and refined.  With multiple operational definitions of objects multiple versions of the SM may be implemented each with different results to discuss.

The simulation model is based not only in scientific realism, but in general systems theory (von Bertalanffey 1968) as well. General system theory considers hierarchy to be fundamental in all systems (ibid. 27) just as scientific realism bases its models on a hierarchy of natural kinds. This philosophical conjunction supports the existence of a hierarchy of models at different scales or levels of abstractions. These nested hierarchical models would seem to fulfill Morrill's (Morrill 1987, 540) wish for methodology that would support "several levels of explanation for the same phenomena, depending on the question asked."

2.1.5  Grounding of the Use of Models
The models, when well developed, provide support for processes fundamental to the scientific investigation.  Charles Pierce argued that scientists engage in syllogistic reasoning in their work, and this results in a 'logic of discovery' of four steps (Eliasmith, 1999):

1. observation of an anomaly.
2. abduction of hypotheses for the purposes of explaining the anomaly.
3. inductive testing of the hypotheses in experiments.
4. deductive confirmation that the selected hypothesis predicts the original anomaly.

In a well-developed model, an anomalous qualitative or narrative datum will likely be noted and perhaps critically discussed.  Certainly numerical data may be subjected to statistical tests that will identify outliers and other instances, which beg explanation.  In the development of explanatory models these data represent problems that need to be disposed of, as they defy the current explanations.

The means of explaining anomalies begins with a process of abduction.  Abduction is not well understood philosophically, but is well enough known to be identified as a practical tool.  Shank (Shank 1998, 843) associates abduction with 'ground-state' or ordinary thinking.  Abduction is applied any time we find ourselves in an ordinary situation and know what to expect (ibid., 848).  Six modes of abductive reasoning (ibid., 848), (Shank and Cunningham 1996) are identified as:

• reasoning to the omen (or hunch)
• reasoning to the clue
• reasoning to the metaphor or analogy
• reasoning to the symptom
• reasoning to the pattern
• reasoning to the explanation

In the research environment, omens, hunches, clues, metaphors, analogies, symptoms, patterns and explanations are supplied in abundance through data, observations, ideas, proposed explanations, etc.  The primary repository of these data are: the Descriptive Model; the RW essays, data stores and e-mail archives; and, to a limited extent, the Explanatory Model.  Hypotheses developed through abduction to explain anomalies may be presented in documents that may be criticized.

Inductive reasoning is used to examine the formalized and operationalized hypothesis through experimentation.  The results of the experiment will then be used as evidence in the Explanatory Model.  These explanations should also then be linked from the observation of anomalies that started the abductive examination leading to the hypothesis.

Deductive reasoning is seldom employable in the social sciences.  Deduction is the examination of truth based on axioms and rules of logic, and rules and axioms are absent or weak in most social sciences; thus formal deduction is not seen there.  In the research environment, however, we do have axioms and rules.  Due to the impossibility of knowing everything about an open social system, we build the Simulation Model to summarize and operationalize our knowledge about the system.  The Simulation Model must be realized as a computer program, and computer programs operate with rules and axioms.

So, to the extent that our rules (logic and synthesized explanatory processes) and axioms (observations) are true, we can deduce or predict.  Now, if we submit a scenario to the Simulation Model, and the Simulation Model predicts an outcome that is compatible with the expected outcome, then we have a weak confirmation of the models.  That weak confirmation is, in effect, an observation made by the Simulation Model and is no more to be trusted than an abductive conclusion.  These conclusions are, however, usable in several ways: first they do, if in adequate quantity, provide a demonstration of robustness; second, they may be used to abduce the fuzzy boundaries of the issue domain without running very expensive experiments; and finally, they may be used to examine new hypotheses before subjecting those hypotheses to actual, and very expensive, experimentation.

2.1.6  Validation by Correspondence with Systems Analysis Practice
The computer entered our lives in the late 1940s and became a business tool in the 1950s.  Early software applications were simply automations of existing business applications.  These programs required very little analysis, just programming.  In the 1960s and 1970s software became ubiquitous throughout the business world and with it came the realization that something had to be done to control its cost and quality.  The answer that business found was software engineering, a professionalization of the trade of programming.  The basis of software engineering was not programming, but systems analysis.

In the late 1970s a number of techniques were developed for software engineering.  One of these was Structured Analysis and System Specification (SA), popularized by Tom DeMarco (DeMarco 1979).  The technique was designed for and applied to information systems development, but as we will see, that application is a very good analog to any research project managed as a Research Web.

SA starts with a study of the current environment.  That study results in the production of a model called the current physical model.  In our vocabulary, the current physical model is exactly equivalent to that part of the descriptive model that deals with the physical objects of the system.  The next process in SA produces another model called the current logical model.  That model is exactly equivalent to the part of the descriptive model that deals with the processes that are observed in the system.  As we will see later, the equivalents of the current physical and current logical models are built in the first stage of the research project.  In realist terms, we call these models the descriptive model.

The next SA model is the new logical model.  As DeMarco puts it, this is the model that represents the system, as it should operate as opposed to how it does operate.  Clearly that is a model of an artifact to be constructed; not, as we might hope, a model explaining why the system operates as it does.  Yet, in this step of the SA system modeling of the new logical system there are parallels to the realist explanatory model.  The new logical model applies business rules to the system -- rules that may not have been well implemented in the observed system.  In the realist explanatory model, we might discover causes that make it necessary to go back and do some more observation of the issue domain for some more subtle objects or processes that cause the system to act as it does.

When the new logical system is implemented we have a new physical system.  This new physical system is exactly equivalent to the simulation model I've proposed.  The simulation model runs scenarios with data based on the descriptive model.  The activities that are simulated are drawn from the explanatory model and results in an outcome that should be recognizable as some behavior that the real system would or has produced.  The simulation model in the realist system would be the equivalent of the SA information system in operation with human activities simulated as well as the information system.

There is little doubt that there has been more effort poured into modeling for software system design and database design than in modeling for all other purposes in the entire history of mankind.  Large portions of software engineering modeling practice are excellent analogs of the realist models, validating the realist model by correspondence.  In the 24 years since DeMarco's work was published, software engineering tools have been much improved.  The object-orientation paradigm and universal penetration of relational databases into information science have resulted in not only a rich literature, but also a great number of computer-aided tools to produce both the graphics and text needed for models.

While accepting that software engineering has shown that modeling is effective and an established methodology, we must take note of the fact that software engineering is focused on the building of information systems.  Modeling of reality, an open system, is a much more demanding task than modeling the closed system of a business information system.  In our proposed new environment, the Research Web, the descriptive models are more complete than those used to model information systems.  Our team members must model a wider range of attributes because they do not have a goal in mind.  In an open system we do not know when a relatively obscure attribute may become important.

2.2  Adapting the Existing Research Environment to Collaboration
The existing research environment is the result of several hundred years of evolution punctuated with occasional periods of turmoil when adapting to new technology.  It has taken fifty years to adapt to the computer, and now we are in the most chaotic period of adaptation to the Internet revolution.  We have, in the past, adopted many revolutionary technologies that remain with us today as integral parts of our research environment.  Most adaptation calls for coevolution: the tools of the past are modified to work with new tools that improve the conduct of research.  So it is with the environment proposed in this work.  The document remains with us, the journal remains, the academic environment remains, older communication technologies remain, and above all, culture and the human psyche remain.

This section examines documents, general systems theory, reward systems, organizational behavior, and communication technology in order to provide knowledge that serves the design of new tools and the organization of research teams to use them.  We will discuss the old along with the new and benefits as well as pathologies.

2.2.1  Research as a Literary Enterprise

Speech is the representation of the mind, and writing is the representation of speech. --- Aristotle, De interpertione I

It is abundantly clear that science cannot progress without the recorded artifacts of research.  Scientific research is framed in documents (Ziman 1976, Chapter 5), and scientific progress is made through critical dialog about those documents (Weimer 1979, 78). While a lecturer can influence the lives of students and change the direction of colleagues; unless the lecture materials are recorded, they will have a very short life.  Early scholars wrote books, an expensive medium that did not offer an outlet for short articles.  It is probably the scientific journal that is responsible for the industrial revolution and the flowering of science that followed.  In the latter seventeenth century the scientific journal began an expansion that is not yet spent (Price 1963), (Veaner 1985, 6).

Publication of research findings in scientific journals allows alignment of an author's work with previously published works supporting parts of the author's hypothesis.  In a scientific paper, propositions other than the hypothesis being investigated must be supported by reference to supporting documents (Latour 1987, 33+).  The references in a paper recognize the contributions of earlier scientists.  Citation analysis (Garfield 1979, chapt. 10) is the quantitative study of the scientific contribution of an author as measured by published works and by the appropriation of the author's work by peers.  Recognition by peers is the leading reward in science.

The scientific paper is clearly the dominant expression of scientific findings.  The current Internet revolution will not change that fact: the genre is embedded in the practice of science.  The Internet revolution will, however, change the form of the scientific paper.  While the physical representation of the research paper is currently inseparable from the printed page, that is rapidly changing.  Many prestigious journals are now publishing facsimiles of their printed articles on the WWW.  Physics is driven by electronic preprints, with journals serving more as archives than as the primary mode of article communication (Taubes 1993).  Usage of electronic journals shows a growing acceptance, but there are reports of a (probably transient) sharp age-based differentiation in usage with 56% of academics under 40 using e-journals, as opposed to only 14% of those over 40 (Tomney and Burton 1998).

The question of submission of scientific papers to e-journals is quite a different question (Holoviak 2001, 14).  Disciplinary and departmental norms frequently deprecate e-journal publication, for good reasons.  Several general problems with electronic documents served from the WWW contribute by association to the deprecation of e-journals (Cronin and McKim 1996, 169), (van Raan 1997, 448): web-based documents are so easy to copy that originality and authorship is uncertain; ephemerality of the document is unavoidable without long-term institutional support; and revision control methods to prevent multiple versions are not widely enforced.  The debate about the relative merits and problems of paper and electronic publication rages on, but the advantage seems to be with electronic publication.  Odlyzko (Odlyzko 1995, 84) suggests that on economic arguments alone, e-journals will become dominant.  From the reader's standpoint, cost, accessibility, and utility favor e-publication; while from the institutional standpoint, archival standards, the weight of tradition, and the multi-billion dollar influence of the scientific publication industry favor paper.    Lawrence (Lawrence 2001) reports that every academic year from 1989-1990 to 1999-2000 showed a substantial increase in the percentage of research articles available online.  While Lawrence studied articles in computer science, it seems unlikely that similar increases are not present in the social sciences and humanities.

This dissertation proposes a form of the scientific paper that is more useful than the current printed genre or its electronic facsimile.  While no less literary than the printed research paper, the HyperDocument (see §3.4.4.1) remediates many of the shortcomings of the current representations of the research paper.  The HyperDocument allows the reader to quickly access references and associated documents while avoiding some shortcomings of paper-based documents: expense, difficult access, linear organization, and its inability to display animation and sound.  Even the argument of portability fades because a HyperDocument printed from the WWW on a color printer contains all the features of a printed document.  The HyperDocument is annotatable online, permitting reader criticism of the scientific paper.

Within the RW, there are many other forms of documentation in use.  In the detailed work of research, contributions such as critical commentary, definitions of terms, experimental data, opinions, and position papers all serve to advance the argumentation surrounding the research.  The RW serves as a repository for all this knowledge.

2.2.2  Research as a Dynamic and Open System
The issue domain of the Research Web provides the focus or organizing principle for the RW, usually expressed as a system model.  System models in the social sciences and life sciences are almost always models of open systems (von Bertalanffy 1968, Chapter 6).  Closed systems exist mainly in controlled experiments, those largely in the physical sciences.  The RW must, then, have the flexibility to respond to the unexpected.  Even more importantly, the RW must be able to assimilate the unexpected fact or event into its models.

The scientific research most suitable for expression in a RW is long-term large-scale research.  This class of research is characterized by frequent changes and additions as research continues.  The frequency of revision is directly proportional to the size of the research team and inversely proportional to the amount of current knowledge about the issue domain.  Interdisciplinary research is also subject to changes due to discoveries and changes in the dominant paradigm in each discipline.  However we choose to look at our knowledge, it is likely to be full of numerous feedback loops (Brinberg and McGrath 1985, 160).

The RW is designed to facilitate the revision of content.  The RW Essays can have information appended to them by annotation in DocReview.  New literature can be posted to the Annotated HyperBibliography quickly by the facilitator.  While preparing a new edition of an RW Essay begins with authoring that is equally arduous in all media, the facilitator has software that makes the re-presentation of the web page quite easy.

With its emphasis on modeling, the RW makes modeling of open systems an interactive and participative design activity.  If any member recognizes an omission or error in any model, there is an immediate and obvious way to annotate the model so the changes can be incorporated.  Modifying the simulation model programming is a technical task to be sure, but those modifications must be preceded by changes in the graphic/textual descriptive and explanatory models.  It is very important to document the presence of influences from outside the current boundary of the modeled open system (Sterman 1991, 219).

2.2.3  Reward Systems in Research

Yes, Virginia, scientists do love recognition, but only since Pythagoras.

Rewards systems are social constructs.  The major influences have been the practitioners of science, followed closely by politicians, ethicists, philosophers and business people.  The rewards systems have legacies rooted in the past, and like any legacy has a dominant paradigm and an elite, both of which serve to give the reward systems great inertia.  There are two ways to look at reward systems:  normatively and realistically.  Both approaches uncover desiderata and practical fact. The normative approach gives us the public face of how scientists should be rewarded, and the realistic approach yields a somewhat more sanguine story.

According to the normative approach, the reward system should serve to encourage successful adherence to the four factors in the ethos of science expressed by Merton in 1942 (Merton 1973): universalism, organized skepticism, communism and disinterestedness.  In order to be published a scientific work must adhere to the ethos of science while introducing new knowledge.  Only through publication can a scientist gain peer recognition.  Merton, in 1957 (Merton 1973), establishes recognition as the jewel in the scientist's crown, with recognition of priority as its brightest facet.  Reward systems are central to the study of the sociology of science.

A realistic approach examines personal motivation as the builder of reward systems.  Cohen (Cohen 1995, 1706) says, "In science, as in so many other professions, the coin of the realm is not collaborative generosity but credit - credit for individuals."  The psychology of reward systems is at odds with Merton's idealistic ethos of science.  The deepest motivation to scientists may be the desire for power.  Power is achieved through recognition (Beaver and Rosen 1978, 68).  And recognition is achieved by publication of research, but enabled through the exclusive professional structure of science: degrees, awards, memberships and position.

Thus recognition is not merely a passive phenomenon.  During both revolutionary and normal periods of scientific growth, it is a rite of passage which confers the right to recognize others: it represents a source of power in the scientific community.

Recognition is based on intellectual achievements, mostly published research that is cited by others, leading to acknowledgement of the origination of ideas.  The most effective publications are in peer-reviewed journals of high reputation.  The citations must be from established scientists also writing in quality journals.  The ultimate reward is to have principles and ideas attributed to oneself permanently (van Raan 1997, 446).

No discussion of reward systems within the collaborative environment can be meaningful without referring to the dominant rewards systems operating today: professional recognition; academic tenure and promotion; and industrial career advancement.  The outstanding characteristic of all three systems is that authorship of research papers dominates all evaluation factors.  A scientist's reputation rests on credibility which waxes and wanes with the scientist's publications.  A "cycle of credibility" strengthens with the volume of quality publications.  This body of produced knowledge is the driving variable in obtaining funding for more research and equipment to produce even more publications (Latour and Woolgar 1979, 200 et. seq.).

While the professional reward system evolved by scientists operates at the highest institutional levels, much of a scholar's career is engaged in satisfying the requirements of the academic tenure and promotion systems.  Academia has two or more masters: educating the populace and serving science.  There is a great tension in serving masters with conflicting goals.  The usual factors for evaluating performance are: teaching, research and service.  All take time, but in practice teaching and research compete for the scholars time, while service is largely incidental.  Our interest is primarily with research, but the tools and concepts presented here can also serve in teaching and service.  Our research will focus on the impacts of rewards systems on research activities.

In defining the nature of scholarly faculty work, Rice (Rice 1991, 12) defines four forms of scholarship: discovery, integration, practice, and teaching.  The first three components are types of research: basic, synthetic and applied.  Basic research is new original research, usually intradisciplinary; synthetic research is integrative and interdisciplinary; and applied research is the application of research to the solution of problems.  Each of these types of research is treated differently in the reward systems.  In academia, basic research is favored, while synthetic research is acceptable though marginal, and applied research is disdained.  Boyer (Boyer 1990, 36) points to several forms of scholarship that are seldom recognized, including writing computer programs and writing for the public press and even popular television.  Tool building is simply not rewarded at all (Star and Ruhleder 1996, 126).  In the humanities, computing is sometime regarded with suspicion and scholars so involved may find themselves marginalized in their departments (Birchard 2002).  Since all three types of research can benefit from the application of asynchronous collaboration in Research Webs, we need to find how the contributions to the RW can be rewarded.

2.2.3.1  Rewards in Practice
The ultimate incentive for participation is reward.  Without rewards, participation in a Research Web (RW) will likely wither after the first flush of enthusiasm is spent. Each person on the research team is encouraged by rewards coming from internal satisfaction, peer approval or institutional rewards.  Research groups can receive encouragement from peers or from their sponsors or other interested institutions.  Rewards reinforce participation while the barriers operate to discourage participation.  The response by individuals or groups to the pros and cons is complex and idiosyncratic.

Since participation is central to collaboration and the concept of the Research Web, we must develop new means to reward the team members for contributions other than research papers.  We must also attempt to ameliorate the more corrosive effects of current practice. Reward driven behavior as well as the products of such behavior can be used to measure participation in a Research Web.

Contribution of content by authorship is without question the most important single category of participation.  The rewards attending authorship are highly developed, though not without problems.  Existing rewards for authorship are deficient in dealing with new forms of publication.  In the "pecking order" of professional literature, the single author peer-reviewed research paper in prestigious journals is premier.  Publications in peer-reviewed electronic journals are often heavily discounted due primarily to the newness of electronic journals (Langston 1996).  The most prestigious journals in many fields frequently disregard critical or controversial work and emphasize "institutional" research that extends the current dominant paradigm in the field (Lindsay 1978).

Unpublished position papers, working papers, essays and conjectural works that are published on the WWW as part of a RW or personal home page can be important contributions even though they might not be considered for publication by the journals due to length, style, content or prior publication strictures.  Writing software for research purposes, even if useful and freely shared with colleagues, is seldom rewarded.  Invocation of a scholar's name on the WWW in the RW's public partition or within the RW's working area may become a reward.  Study of invocations on the Web may eventually allow alteration of existing methods of performance evaluation by recalling the scholar's name from unusual genres of communication such as commentary, acknowledgments, reports of professional activity, e-mail messages and meeting minutes (Cronin, et.al. 1998).

Criticism is extremely important; it is the means by which content is progressively refined from a draft to a canonical document.  Criticism of a RW essay can eliminate "holes", pose new hypotheses, and contribute to quality.  Criticism of references in an annotated bibliography can point out errors and where those errors are corrected.  Criticism of the annotated glossary can sharpen meaning of a term or introduce new nuances of meanings.  A three-sentence note can change the direction of a paper.  Criticism must be rewarded.  The RW provides a permanent record of annotations that establishes a record of critical contribution.

Leadership provides the necessary social glue to coordinate the efforts of the research team, to deal with the administrative details, and to keep the work going forward by encouraging participation by the members.  While the scientific leadership is generally vested in the Principal Investigator (PI), in practice the PI often shares many of the functions of leadership.  Foote (Foote 1999, 115) suggests faculty who build or maintain collaborations do not often receive much credit for their efforts.  The members who shoulder the day-to-day burdens of leadership should be rewarded.

An analysis of reward systems can proceed from two directions: first is an analysis from the viewpoint of the recipient of the rewards-individual, research team, or cooperating agencies (psychological, sociological, political); second is an analysis of the rewards that may be granted from distinct sources.

2.2.3.1.1  Received Rewards  
Individuals are motivated by internal satisfaction, recognition by peers and by the team at large and by organizations cooperating with the research.  Internal satisfaction stems from numerous sources: pleasure at the reduction of cognitive dissonance, closure of tasks, learning, etc.  Individuals are stimulated by their peers, especially those on the research team; by the team itself, expressed as pride in the team's accomplishment; and to a certain extent by approval of cooperating agencies, often expressed in letters of commendation and in press releases.  In the RW, as in other academic groups, social academic rewards are received by two main mechanisms: potential enhancement of reputation, and by reciprocity expressed as permission to use the information contained in the group's knowledgebase (Matzat 2001, 220).

Membership in an active community of scientists is itself a reward. Senior scientists expect this reward, but it is carefully doled out to junior members of the team, post docs and graduate students. This reward is the principal mechanism for socialization of new scientists. Lave and Wenger have named this mechanism legitimate peripheral participation and have thoroughly explored the issue (Lave and Wenger 1991).

The research team receives rewards of recognition from professional peers in the form of mentions in the press, citation of reports published by the team members, and by communications directed to the team; and from cooperating agencies in the form of additional support in the form of encouragement, additional equipment or support personnel and ultimately the reward of financial grants.  Cooperating institutions receive rewards in the form of prestige due to their sponsorship of successful research, recognition of public service, recognition of fulfilling a mandate.

2.2.3.1.2  Granted Rewards
Major rewards offered to individuals by the research team are: inclusion of their essays in the RW, inclusion as author on papers submitted to journals, public acknowledgment of contributions (Cronin 1995), financial support from grant funds, invitation to join the research team, and employment as a staff member.  Another class of rewards is ceremonial support that lends official recognition for service and the passing of milestones (Trice and Beyer 1984).

The importance of ceremony in team dynamics is often minimized.  This minimization may be caused by intellectual hubris.  Ceremony does play a large role in our lives and to neglect it is to ask for suboptimal performance.  Ceremony is easily overdone, especially when it is obviously artificial.  Ceremonies should be sincere occasions that show realistic expectations of participation and the desired effect.  Since conventional ceremony always requires physical presence, the distributed research team ceremony faces a special problem: celebration in isolation.

Ceremony in our special environment suffers greatly from the lack of media richness.  Lack of personal presence prevents the use of tone of voice, gesture, touching and other forms of communication.  We have great difficulty in reinforcing the message of the ceremony by engaging the members emotionally.  On the other hand we do have some advantages: the ability to very carefully express one's self verbally, the ability to reflect on the message rather than react to it, the ability of the receiver to recapitulate each communication, and the suppression of intemperate reaction.  If we recognize our weaknesses and capitalize on our strengths, we can successfully incorporate some ceremony into our Research Webs.

Rites of passage, especially those encountered in professional socialization such as graduate school, may be made less traumatic by collaboration with peers (Glenwick and Burka 1978, 213).  Daniels (Daniels 1990) has teased elements of rites of passage from the ordinary group meeting: Birth (introduction of a new person or idea); Maturation (recognition that a product or project has reached a new stage of usefulness); Marriage (restructuring in the organization); Leadership (promotions, appointments or assignments); Thanksgiving (recognition of awards or acquisition of new resources); Discipline (identification of poor performance and determination of penalties); and Sickness and Death (the identification of problems and allocation of resources to remediation, project, product, or performer termination).  While Daniels' insights are useful, Trice and Beyer present a more academic analysis (see Table I below).  They have developed a typology of rites that is quite useful for design purposes.

Except for rites of integration, there is an asynchronous solution for any type of rite.  As befits the medium these solutions are all documents.  These documents are likely to be issued by the scientific leader, and should anyone else care to issue them, they should coordinate the release with the scientific leader.  Announcement of awards with potential professional value to the recipient should be forwarded to organizations, such as the recipient's academic department, that might find them important.

There are likely to be other types of rites emerge from the genre of the Research Web.  For instance, one of the likely events in a mature RW is to develop a spin-off RW to investigate a new idea.  Perhaps the RW will be incorporated into a RW of larger scope.  Leadership of new RWs is likely to be awarded to outstanding performers, either substantive or collaborative.

Cooperating institutions can grant rewards far more valuable than the offerings of the RW team.  Academic departments, colleges and universities may grant tenure, make promotions, award cash grants, change the scholar's duties, or many other coveted rewards.  Granting institutions can offer endowed chairs, financial grants or fellowships.  Professional organizations or private foundations may grant prestigious prizes.  It is important to realize that none of these rewards are likely to materialize without a great deal of persistent effort on the part of the RW leadership.

2.2.3.2  Determining and Distributing Rewards

Designing the incentive to collaborate is just as important as designing the technology for collaboration.

Collaboration is quantifiable in a RW.  Most contributions are documented as authored essays, comments and e-mail to the team.  Gone are the days when the ability to work in a team is measured by guessing at popularity around the water cooler.  The results of evaluation of participation in collaboration may be surprising as it becomes clear how each member behaves: who takes the time to review a document: who comes up with new ideas; who volunteers to do humble chores; and who offers little to others.

Rewards for collaboration are very poorly developed.  At a psychological level, most people feel self-satisfaction with their contributions.  The contributions also need to be appropriately rewarded at the sociological level, or at the professional level (Cronin 1995). In the RW, rewards may be thanks, an acknowledgment in a published paper, or authorship, or may be the first author.  Thanks and other expressions of appreciation should be offered consistently as a matter of human courtesy.  At a more formal level, the authors of papers should acknowledge everyone who contributes commentary or minor bits of substance to the paper.  These contributions are characterized as subauthorship collaboration (Austin and Baldwin 1991, 23).  The papers produced by the RW team will be part of the RW as essays.  There is plenty of room to acknowledge everyone who contributed, perhaps in order of value to the authors.  While journal editors may request that acknowledgments be abridged or eliminated, the team controls the RW essay (which may be made public).  The acknowledgment is a neglected genre that may, in the coming full-text world, lead to better recognition of collaborators.

As a standard procedure, an advanced draft of any paper that is published should be recast as a RW essay and placed in the public partition of the Web Site.  This will provide an annotatable version of the paper.  As time has passed, publishers have become more willing to grant permission to allow the placement of an annotatable draft of the paper on the WWW.  This will increase the exposure of the authors and those acknowledged.

Authorship is the most visible prize and should be (and almost always is) very carefully determined in advance of the first draft release for comments by DocReview.  Additions to the author list, the order of the author list and perhaps the deletion of an author can be made as the project moves forward.  Occasionally a valued commentator may be invited to join the authors.  Endersby (Endersby 1996) suggests that all individuals making an active contribution share authorship, while those whose labor could easily be replaced by others, be acknowledged.  The APA (American Psychological Association) recommends that only those contributing to the content be awarded authorship, thus eliminating complimentary authorship to heads of departments and laboratories.

The RW team is frequently very independent of academic departments, especially if the team is interdisciplinary.  Since the team members are not usually subject to oversight by their academic supervisors, the scientific coordinator should appropriately commend each academic on the team to his or her academic chair.  If this is not done, collaboration other than authored papers will likely not be rewarded in the academic department.  The prevalent failure of academic administrators to appreciate collaborative behavior is cited in a special study sponsored by the Association of American Geographers:

Finally, we recommend that academic administrators and faculty acknowledge that reward mechanisms are now based almost entirely on individualistic conceptions of faculty roles.  Without exception in the committee's experience, rewards accrue to individuals evaluated in isolation.  That view of faculty roles may prevail for some time.  We opine, however, that the kinds of collaborative and team efforts that have proved productive in other industries eventually will prove useful in geography, probably in the form of instructional teams using several complementary methods of instruction. Research in geography also may move beyond the artisan or craft scale that currently prevails, to projects that are addressed collaboratively by organized groups. [emphasis provided]

2.2.3.3 Penalties
There is a variety of counterproductive behavior that must be prevented.  Besides abiding by the normal mores of professional behavior, all participants must avoid or at least reduce certain behaviors peculiar to working in a group. Since most of the work will be done over the Internet, all members should be aware of the rules of "netiquette." Much more serious is the problem of non-participation or reluctance to participate publicly. Many individuals, especially those in the social sciences and humanities, have been socialized in a world that rewards individual effort.  Some people may be shy or embarrassed with their language skill; others may feel that their stature is too elevated to bother with details.  Non-participation is a problem that can be solved by three paths: example, training, and leadership. A well-functioning collaboration is an example to newcomers and to team members who are slow to join the party. The facilitator can tutor team members who need help in learning the tools of the environment. Leadership is just as important in a distributed on-line community as it is in person. If a PI lacks leadership skills, the job of scientific coordinator can be passed to someone who possesses the needed skills.

Collaborators who fail to participate can be removed from the team. This penalty may be necessary in order to prevent demoralization of the team.  Resentment is a natural reaction toward free riders. David Coleman has established online communities of managers roughly equivalent to Research Webs (Liff 1998). These groups have an annual subscription fee for members, and if a member is not contributing, Coleman suggests that these "checkbook members" be asked to resign.

Members who are on authorship teams may be removed if they fail to fulfill their responsibilities. Since authorship is the principal reward in science, such a strong penalty must be used cautiously, with the concurrence of not only the lead author, but the scientific coordinator as well. Another, more effective, class of penalties is admonishment. Admonishments should be considered whenever one of Pröpper's rules of engagement (see Table II in §2.2.4.6.2) are violated. If admonishments are not effective, then removal seems to be the only alternative.   Antisocial or disruptive behavior is unlikely unless the RW is open to the public.  The team must be able to marginalize or exclude such people (Etzioni and Etzioni 1999).

The penalties suggested above are very severe and are unlikely to be applied in an academic environment.  Recent research in game theory in economics suggests another control over the free-riding problem.  Fehr and Gächter (Fehr and Gächter 2000, 990) show that if team members are offered the opportunity to punish free riders, they do so, even when it comes at some cost to them.  Assume management was to provide a bonus pool to be publicly distributed among the team members based on their peers rating of their collaborative behavior.  If the team members were allowed to distribute an allowance to other team members as they chose, then the free riders would be publicly identified by their poor bonus, and of course the prime collaborators would also be identified by a generous bonus.

American Association of Geographers, Special Committee on Faculty Roles and Rewards. 1993. Toward a Reconsideration of Faculty Roles and Rewards in Geography. San Antonio: AAG.  

Aronson, J. L., Rom Harré, and E. C. Way. 1995. Realism Rescued: How Scientific Progress is Possible. Chicago: Open Court.    

Augier, Mie, and M. T. Vendelø. 1999. Networks, cognition and management of tacit knowledge. Journal of Knowledge Management vol. 3, no. 4: 252-61.  

Austin, Ann E., and Roger G. Baldwin. 1991. Faculty Collaboration: Enhancing the Quality of Scholarship and Teaching. ASHE-ERIC Higher Education Reports, Vol 20, No 7. Washington, DC: The George Washington University.  

Beaver, D. deB., and R. Rosen. 1978. Studies in Scientific Collaboration: Part I. The Professional Origins of Scientific Co-authorship. Scientometrics vol. 1, no. 1978: 65-84.    

Bertalanffy, Ludwig v. 1968. General System Theory: Foundations, Development, Applications. New York: Braziller.      

Birchard, Karen. 2002. Canadian Universities Will Create on Online Portal for Studying Texts. The Chronicle of Higher Education vol. 2002, no. 2:  

Blalock, H. M. 1990. Auxiliary Measurement Theory Revisited. in Operationalization and Research Strategy. Editors J. J. De Jong-Gierveld J. Hox. 33-48. Amsterdam: Swets and Zeitlinger.    

Blalock, Hubert M. 1968. The Measurement Problem: A Gap between the Languages of Theory and Research. In Methodology in Social Research. H. M. Blalock A. B. Blalock. 5-27. New York: McGraw-Hill.  

Booch, G., J. Rumbaugh, and I. Jacobson. 1999. The Unified Modeling Language User Guide. Reading, MA: Addison-Wesley.  

Boyer, Ernest L. 1990. Scholarship Reconsidered: Priorities of the Professoriate. Princeton, NJ: The Carnegie Foundation for the Advancement of Teaching.  

Brinberg, David, and Joseph E. McGrath. 1985. Validity and the Research Process. Beverly Hills: Sage.  

Cartwright, N., T. Shomar, and M. Suárez. 1995. The Toolbox of Science: Tools for the Building of Models with a Superconductivity Example. Poznan Studies in the Philosophy of the Sciences and the Humanities vol. 44, no. 137-49.  

Cohen, Jon. 1995. The Culture of Credit. Science vol. 268, no. 1706-18.  

Cronin, Blaise, H. W. Snyder, H. Rosenbaum, A. Martinson, and E. Callahan. 1998. Invoked on the Web. Journal of the American Society for Information Science vol. 49, no. 14: 1319-28.  

Cronin, Blaise, and Geoffrey McKim. 1996. Science and Scholarship on the World Wide Web: A North American Perspective. Journal of Documentation vol. 52, no. 2: 163-71.  

Cronin, Blaise. 1995. The Scholar´s Courtesy: The role of acknowledgment in the primary communication process. London: Taylor Graham.    

Daniels, William R. 1990. Group Power II: A manager´s Guide to Conducting Regular Meetings. San Diego: University Associates.  

DeMarco, Tom. 1979. Structured Analysis and System Specification. Englewood Cliffs, NJ: Prentice-Hall.    

Denzau, Arthur T, and Douglass C. North. 1994. Shared Mental Models: Ideologies and Institutions. Kyklos vol. 47, no. 1: 3-31.    

Dunn, William N. 1982. Reforms as Arguments. Knowledge: Creation, Diffusion, Utilization vol. 3, no. 3: 293-326.  

Eliasmith, Chris. 1999. abduction [definition]. The Dictionary of Philosophy of Mind. http://www.artsci.wustl.edu/~philos/MindDict/abduction.html.  

Endersby, James W. 1996. Collaborative Research in the Social Sciences: Multiple Authorship and Publication Credit. Social Science Quarterly vol. 77, no. 2: 375-92.  

Etzioni, Amatai, and Oren Etzioni. 1999. Face-to-Face and Computer-Mediated Communities, A Comparative Analysis. The Information Society vol. 15, no. 241-48.  

Fehr, Ernst, and S. Gächter. 2000. Cooperation and Punishment in Public Goods Experiments. The American Economic Review vol. 90, no. 4: 980-994.  

Foote, Kenneth E. 1999. Building Disciplinary Collaborations on the World Wide Web: Strategies and Barriers. Journal of Geography vol. 98, no. 108-17.  

Fowler, Martin, and K. Scott. 1997. UML distilled : applying the standard object modeling language. Reading, MA: Addison Wesley.  

Garfield, Eugene. 1979. Perspective on Citation Analysis of Scientists. in Citation Indexing: Its Theory and Application in Science, Technology, and Humanities. Eugene Garfield. 240-252. Philadelphia: ISI Press.  

Glenwick, David, and A. A. Burka. 1978. Research Collaboration Between Graduate Students: Facilitating the Rites of Passage. Teaching of Psychology vol. 5, no. 4: 213-14.  

Halpin, Terry. 2001. Object Role Modeling: An Overview. WWW Page, Microsoft Corp. http://www.orm.net/pdf/ORMwhitePaper.pdf.  

Hardin, Steve R. 1998. Michael Schrage and Collaboration. Bulletin of the American Society for Information Science vol. August/September: 6-8.  

Harré, Rom. 1970. The Principles of Scientific Thinking. Chicago: University of Chicago press.  

Harré, Rom. 1978. Models in Science. Physics Education vol. 13, no. 275-78.  

Holoviak, Judy C. 2001. Scientific e-journals: The ´e´ Doesn´t Mean ´easy´. Geotimes vol. 46, no. 1: 14-15+.  

Hox, J. J., and Mellenbergh. 1990. The Interplay of Substantive and Auxiliary Theory. in Operationalization and Research Strategy. eds. J. J. Hox, and J. De Jong-Gierveld. 123-36. Amsterdam: Swets and Zeitlinger.  

Langston, Lizabeth. 1996. Scholarly Communication and Electronic Publication: Implications for Research, Advancement, and Promotion. http://www.library.ucsb.edu/untangle/langston.html.  

Latour, Bruno, and S. Woolgar. 1986. Laboratory Life. Princeton: Princeton University Press.  

Latour, Bruno. 1987. Science in Action. Cambridge, MA: Harvard University Press.  

Lave, Jean, and Etienne Wenger. 1991. Situated Learning: Legitimate Peripheral Participation. New York: Cambridge University Press.  

Lawrence, Steve. 2001. Online or invisible? Nature vol. 411, no. 6837: 251+.  

Lederman, Leon. 1969. A Great Collaboration. Review of Robert Jungk´s "The Big Machine". Science vol. 164, no. 169.  

Liff, Allen. 1998. Fostering online collaboration and community. Association Management vol. 50, no. 9: 33(4).  

Lindsey, Duncan. 1978. The Scientific Publication System in Social Science: A Study of the Operation of Leading Professional Journals in Psychology, Sociology, and Social Work. San Francisco: Jossey-Bass.  

Matzat, Uwe. 2001. Social Networks and Cooperation in Electronic Communities: A theoretical-empirical Analysis of Academic Communication and Internet Discussion Groups.. Amsterdam: Thela Publishers.  

McMinamin, S. M., and J. F. Palmer. 1984. Essential Systems Analysis. Englewood Cliffs, NJ: Prentice-Hall.  

Merton, Robert K. 1973. Priorities in Scientific Study. In The Sociology of Science. Robert K. Merton. 286-324. Chicago: University of Chicago Press.  

Merton, Robert K. 1973. The Normative Structure of Science. In The Sociology of Science. Robert K. Merton. 265-78. Chicago: University of Chicago Press.  

Morrill, Richard L. 1987. A Theoretical Imperative. Annals of the Association of American Geographers vol. 77, no. 4: 535-41.  

Odlyzko, Andrew M. 1995. Tragic loss or good riddance? The impending demise of traditional scholarly journals. International Journal of Human-Computer Studies vol. 42, no. 71-122.  

Price, Derek J. de Solla. 1963. Little Science, Big Science. New York: Columbia University Press.  

Rice, R. Eugene. 1991. The New American Scholar: Scholarship and the Purposes of the University. Metropolitan Universities vol. 1, no. 4: 7-18.  

Rumbaugh, James. 1991. Object-Oriented Modeling and Design. Englewood Cliffs, NJ: Prentice-Hall.  

Sayer, Andrew. 1992. Method in Social Science: A Realist Approach. 2nd ed. New York: Routledge.  

Shank, G., and D. J. Cunningham. 1996. Modeling the six modes of Peircian abduction for educational purposes. Paper presented at The Annual Meeting of the Midwest AI and Cognititve Science Conference.  

Shank, Gary. 1998. The Extraordinary Ordinary Power of Abductive Reasoning. Theory and Psychology vol. 8, no. 6: 841-60.      

Star, Susan Leigh, and Karen Ruhleder. 1996. Steps Toward an Ecology of Infrastructure: Design and Access for Large Information Spaces. Information Systems Research vol. 7, no. 1: 111-34.  

Sterman, John D. 1991. A Skeptic´s Guide to Computer Models. in Managing a Nation: The Microcomputer Software Catalog. eds. G. O. et. al. Barney. 209-29. Boulder: Westview Press.  

Taubes, Gary. 1993. Publication by electronic mail takes physics by storm. Science vol. 259, no. 5099: 1246-48.  

Tomney, Hilary, and Paul F. Burton. 1998. Electronic journals: a study of usage and attitudes among academics. Journal of Information Science vol. 24, no. 6: 419-29.  

Trice, Harrison M, and Janice M. Beyer. 1984. Studying Organizational Cultures Through Rites and Ceremonials. Academy of Management Review vol. 9, no. 4: 653-69.    

Veaner, Allen B. 1985. Into the Fourth Century. Drexel Library Quarterly vol. 21, no. 1: 4-28.  

Weimer, Walter B. 1979. Notes on the Methodology of Scientific Research. Hillsdale, NJ: Lawrence Erlbaum.  

Zander, Udo, and Bruce Kogut. 1995. Knowledge and the Speed of the Transfer and Imitation of Organizational Capabilities: An Empirical Test. Organization Science vol. 6, no. 1: 76-92.  

Zeigler, Bernard P. 1990. Object-Oriented Simulation with Hierarchical Modular Models:Intelligent Agents and Endomorphic Systems. New York: Academic Press.    

Ziman, John M. 1976. The Force of Knowledge: The Scientific Dimension of Society. Cambridge: Cambridge University Press.  

van Raan, A. F. J. 1997. The future of the quality assurance system: its impact on the social and professional recognition of scientists in the era of electronic publishing. Journal of Information Science vol. 23, no. 6: 445-50.    

Auxiliary model -- A model designed to test a theoretical model. The assumptions of the auxiliary model operationalize the theoretical model. The incommensurability of the two models are bridged by explanation of the assumptions made in operationalization. --- (Hendricksen - from Blalock 1968 The Measurement Problem: A Gap between the Languages of Theory and Research )

Canonical document -- An authoritative document. In the context of Research Webs, the canonical document is an essay that incorporates or references the accumulated knowledge about a topic as interpreted or synthesized by the research team. --- (C. Hendricksen)

Collaboration -- v. intr. To work together, especially in a joint intellectual effort. In the context of the Research Web, the creation of new shared knowledge. --- (American Heritage Dictionary, Michael Schrage)

Conceptual domain -- contains ideas, concepts, and their relations as well as the philosophical assumptions underlying them. --- (Brinberg and McGrath, Validity and the Research Process)

Cooperation -- The act of enabling collaboration. Institutions cooperate, they cannot collaborate. Individuals can collaborate or cooperate. Cooperation usually involves the contribution of resources to a joint project. --- (Hendricksen)

Descriptive Model -- A model that organizes selected elements of information from the source model. The selected elements are filtered for an adequate truth value, that is, errors are removed and myths are investigated before inclusion, and the elements possess validity in the context of the issue domain. --- (Charlie Hendricksen)

Document -- Any permanent recorded file of information, text, graphic, or sound, usually electronic, that teaches, warns, or serves as an example. --- (C. Hendricksen)

Explanatory model -- A description of a hypothetical generative mechanism that produces the phenomenon seen in the corresponding descriptive model. --- (Hendricksen, from Aronson, Harre and Way)

Expressed model -- That version of a mental model which is expressed by an individual through action, speech or writing. --- (J.K. Gilbert et.al. 1998, Models in Explanation, Part I)

Free rider -- A person who benefits from commonly held knowledge, but never contributes to that body of knowledge. --- (Hendricksen)

General Systems Theory -- ... general systems theory investigates systems inductively, looking at structure, behavior involving energy transfer, boundaries, the environment, the state of the system, and characteristic parameters. --- ( Laura Laurencio)

Methodological domain -- contains the methods, designs, and research strategies used to examine concepts and phenomena. --- (Brinberg and McGrath, Validity and the Research Process)

Model -- Real or imagined representations and analogues of naturally occurring entities, structures and processes. --- (Aronson, Harré and Way, 1995, Realism Rescued, p3.)

Operationalization -- The process of creating procedures to measure real properties based on abstractions from a theoretical model. For example, how does one measure argumentativeness? --- (Hendricksen - See Blalock 1968 The Measurement Problem: A Gap between the Languages of Theory and Research )

Research Web -- A WWW site which is the electronic embodiment of the intellectual capital of the network of excellence assembled to investigate a phenomenon. It disseminates information, provides communication facilities, and an infrastructure for collaborative interaction. --- (C. Hendricksen)

Scientific Realism -- ... the common sense (or common science) conception that, subject to a recognition that scientific methods are fallible and that most scientific knowledge is approximate, we are justified in accepting the most secure findings of scientists "at face value." --- (Stanford Encyclopedia of Philosophy)

Scientific coordinator -- A researcher who is delegated to oversee the management of the content of the Research Web. Those duties include responsibility for defining the boundaries of the RW's issue domain, so as to maintain interdependence of the research carried out by the authoring teams. --- (Hendricksen)

Simulation model -- A computer program, usually very complex, that when given a set of initial conditions and a script of actions (scenario), will produce an outcome that a real system would produce given the same scenario. --- (Hendricksen)

Source model -- In general, that which a model is modeled on. Ultimately, an unarticulated abbreviation of the entire context of the phenomenon. The nature of the abbreviation is just an abbreviation to the extent of our knowledge, with filtering of irrelevancies and more risky "unimportances." --- (Charlie Hendricksen, with a nod to AHW)

Substantive Domain -- contains the phenomena, processes, or focal problems of interest. --- (Brinberg and McGrath. Validity and the Research Process)