This dissertation is grounded in an essential conviction: Computer conferencing. represents an important new medium of human communication. The conviction is hardly a new one. For those who have attempted to create computer conferencing systems, the conviction dates back at least 20 years. It can be argued, moreover, that the vision goes back more than 40 years to the earliest days of computing and the Vannevar Bush's (1945) vision of the 'Memex' (effectively a terminal based database system).
It can be argued that starting a study with such conviction is inherently risky. Yet conviction, whether grounded in the intuition of phenomonological insight or the deduction of semiotic interpretation, is an essential cornerstone of scholarship. We observe. We record. We intuit. We analogize. We interpret. We classify. We deduce. We systematize. We test. When we believe we have found something interesting we assert. Having asserted we test our assertion with additional observation and record those observations so others can test them. The result should be a base from which others can test the truth or falsity of our assertions.
Hence it can also be argued that conviction is an almost inevitable precursor to any study. Where one starts a study in conviction, however, one must insulate the study from the conviction that spawned it. These steps are most commonly a rhetorical recasting of belief as hypothesis; of attempts to demonstrate as attempts to observe; of hypothesis as null hypothesis. This recasting takes the form of an agreed upon value neutral vocabulary and well exercised presentation formats. Instead of saying "I believe ..." we say "Prior literature suggests ...". Instead of saying "We demonstrated ..." we say "Results suggest ...".
There is nothing inherently wrong with such recasting. The content of a study remains the same with either vocabulary. There is nothing wrong with the time tested formats of many scholarly studies, in which one proceeds in an orderly fashion from overview to literature review to hypothesis and rationale to methodology to results to conclusions. It should be noted, however, that such recastings do not reduce the potential bias of conviction. They simply make conviction less obvious and allow us to pretend it isn't there. Such recasting cannot make objectivity a reality. Indeed, it has the potential of hiding conviction within a skin deep apparent objectivity.
A more effective counter to conviction is probably found in the simple act of acknowledging it. It puts the reader on the lookout for faults in one's evidence and argument. It alerts the writer to the potential for accepting evidence because it supports one's view; to the potential for ignoring evidence because it does not. It provides a base from which the observer can potentially study human interaction with an indifference to one's acknowledged preconceptions. Such studied indifference is, of course, a key element of the practice of at least one perspective, ethnomethodology, with which this study will share some common assumptive and methodological ground.
A more effective step, perhaps, can be found in the selection of a methodology that makes it possible and even probable that the conviction will not be supported. This kind of methodological selection is rarely treated as seriously as it ought to be. Indeed, many scholars have extremely limited methodological toolboxes, often using only one or a few narrowly defined methodologies over the course of a career. Hence it is often the case that a methodology is selected for its familiarity rather than the advantages it may have in addressing a particular research question.
Critical analysis, experiment, survey, interview, archival data, and participant observation represent only some of the very broadest categories of methodologies. Each offers unique sources of insight as we test conviction. The advantages and disadvantages of a wide range of research methodologies are well documented elsewhere. Similar comparisons in this study will be limited to the chapter on methodology.
It is the author's intention, in the face of acknowledged conviction, to test it with appropriately selected methodologies used in counterpoint to each other. Theory, participant observation, archival statistics, and survey data will be intermingled. Some chapters will be dedicated to only one of these forms of analysis, but many will freely intermingle theory, survey, observation, and/or archival data.
The effect, it is hoped, will be that of a "Jazz" dissertation: frequently syncopated; prepared at any moment to disappear into a seemingly tangential musical phrase; yet always targeted on central themes: the nature of computer conferencing, its relationship to other media, its prospects for success, and its lessons for the study of communication.
The conviction from which this study proceeds is ultimately little more than a starting point. It is possible and even likely that we will not be able to fully determine computer conferencing's prospects for success, its lessons for the field of communication, its relationship to other media, or even exactly what computer conferencing is. It is likely, however, that we can learn a great deal about computer conferencing, and perhaps something about human communication in general, by exploring computer conferencing playfully, open to both questions and answers that present themselves along the way.
This, in some sense, is exactly what has occurred over the course of
this study. The conviction that "computer conferencing represents an
important new medium of human communication" is not a hypothesis. It
is an assertion and a point of curiosity. Accident has created
opportunities to assert that curiosity, first in the editorship of a
industry newsletter which focused on
The submission for publication of any paper written by an IBM employee that deals with some aspect of IBM's business (e.g., this doctoral thesis) requires appropriate internal IBM clearances. This study and thesis has been supported by my immediate manager, Dr. Gerald Waldbaum, the manager responsible for IBM's first and largest computer conference, and a proponent of computer conferencing within the company. Waldbaum has reviewed this thesis for historical accuracy (he was involved for two years before my observation began) and appropriate IBM business controls. The primary impact of this review was on the discussions of rules and enforcement.