As a genre of computer conferencing, a computer-mediated query represents a fairly specialized event. It is characterized by many short events, each involving one person asking a brief question and one or several people offering brief answers. Answers usually follow quickly, and no CMQ event is likely to last longer than a few hours or a day. One almost never sees debate in these events. Several people may offer alternatives, but it is rare for anyone to dispute any of the advice given or even ruminate on which of several alternatives might be preferred.
One should not, at least in the case of CMQ, interpret specialized as synonymous with limited use. ISTHERE FORUM is consistently one of the fastest growing forums on IBMPC, and it is hardly alone in its dedication to this single genre of communication. Its companion query forums, including WHEREIS FORUM, WHICH FORUM, WHATIS FORUM and a variety of topic specialized variants including PRINTER QUERY, BEGINNER QUERY, OS2BEGIN FORUM and others, are also fast growing forums. CMQ is hardly limited to such "query forums", moreover. Instances of computer mediated query can be found on many forums, and the occasional use of computer mediated query is a defining characteristic of several other genres of communication on IBMPC.
It is possible to identify an array of such genres of computer conferencing event, each serving a particular need in the IBMPC computer conferencing community. The definition of these genres is hardly clear-cut. Just as a dramatist can combine elements of a comedy with elements of a tragedy in building a play that is not quite either, so computer conferencing genres can be combined. Such combinations can create both problems and opportunities.
The problems come in identifying distinctive processes of communication that rarely occur in isolation. ISTHERE FORUM makes it easy to identify CMQ as a distinctive genre of computer conferencing, but the creation of this generalized (to a variety of PC topics) yet dedicated (to questions and answers) query forum can be considered to be something of a freak event. Before the emergence of ISTHERE FORUM, queries were either intermingled with other interaction on more topical forums, or posted to a strongly specialized query event in which a user would typically start a forum called XXX QUERY (where XXX described the topic of the QUERY) and answers would be posted to it.
The former behavior was a problem to users in that questions were often posted to inappropriate forums. The latter behavior was a problem for users because it led to a proliferation of small forums that were hard to find and often were never used again.
The emergence of dedicated and generalized query forums is, in a real sense, a size-induced innovation. Computer mediated query was already an important genre of communication on IBMPC before ISTHERE FORUM was created. The size of IBMPC (and concomitant overload of information) made some of the existing patterns of CMQ use problematic, however, and query forums provided an answer to that problem. What would be off-topic queries in other forums were the main purpose of ISTHERE FORUM. What would isolated queries if posed in a specialized query event were now packaged together into a single large query event.
Computer-mediated query represents a bare introduction to the varieties of generic interaction which occurs repeatedly on IBMPC. Most are probably fairly representative of processes that occur, or eventually will occur, in many computer conferencing environments. These variations of generic interaction in computer conferencing include:
The most common application of these "voluntary newsletters" on IBMPC is the distribution of program source code, usually in forums with the filetype PROCS (short for PROCedureS). An exception to this usual application is found in one of the fastest growing forums on IBMPC, however. This forum, NEWSCLIP, a repository for abstracts of IBM and Personal Computer related news as reported on other media, will be discussed in a coming chapter.
Content is not always focused on a single product. There are numerous examples of generalized product forums on which any product within a given class can be discussed or compared. WORDPROC FORUM discusses word processors. PRINTER QUERY discusses printers. Other forums discuss spreadsheets, hard disks, plotters, laptop computers, and other classes of products.
These forums can be regarded as on-line user groups in which users of specific equipment and software share information. Indeed, discussion on these forums is similar, in many ways, to discussion in live user group meetings. It includes computer mediated queries, voluntary newsletter contributions, general discussions of product related news, electronic focus group surveys of who's using the product for what purpose, and even occasional polar debates. The content of these forums, which exists as a sort of prototypic superset of all discussion on IBMPC, is perhaps the most varied on the facility.
The most significant change in structure is found in the nature of bug reporting and the practice of suggesting improvements to the program. Because the author(s) of a program can be directly addressed on-line, contributors to ISD forums come to expect that many of their suggestions will eventually be incorporated into the program and that the bugs they report will be fixed fairly rapidly. Interactive software development will not be discussed in detail in this study.
These seminars generally convene around a single generalized topic, which may be a problem that requires a solution, a solution that requires a problem, or a general concept. FEAR FORUM, for instance, discusses fear of computers, the reasons for it, and the things that can be done to humanize computers, their uses, and their effects. Another electronic seminar, USRFREND FORUM, discusses techniques for creating user friendly software. There are dozens of such forums on IBMPC, including CREATIVE FORUM (discussions of creativity in the workplace), QUALITY FORUM (discussions of what constitutes quality work and products and how it can be achieved), OOPS FORUM (discussions of object oriented programming techniques), SMARTCAR FORUM (discussions of the potential use of computers to improve the automobile of the future) and many others.
The criteria used in assigning the name "genre" to each of the above vary somewhat. Some are recognized primarily on the basis of prototypic patterns of interaction that occur repeatedly in these genres. Others are recognized primarily on the basis of the way the forums are used; their generic purpose. Both sets of criteria matter to the existence of a genre. Generic patterns of interaction reflect the application of a medium to a given purpose. Where one finds patterns of interaction one should find an application of the medium. Where one finds an application of the medium one should find generic patterns of interaction.
At least one specific purpose can be associated with each of the genres outlined above. These purposes, including education, general discussion, problem resolution, obtaining current information and news about PC's and PC Software, asking questions and getting answers, reporting bugs and improving software, getting support, participating in technical discussions, getting new ideas, meta-discussion, and others can be reduced to, as it turns out, two primary applications, discussion and information.
A series of questions in the 1988 survey of IBMPC participants asked respondents to estimate the value of various applications of the of the IBMPC Computer Conferencing facility, including most of the purposes of IBMPC noted above. The scale for these questions (numbers 83 to 95; see the 1988 survey) runs from "Extremely Valuable" to "Not at all valuable", with three intermediate points. These questions assess the relative value of features of IBMPC that people found particularly valuable in 1986 (based on responses to open ended questions). Only one of these questions, a holdover from the 1986 survey, is not concerned with the applications of IBMPC.
A principal factor analysis of the responses to these thirteen questions yields two factors which, after oblique rotation, account for 31% of the total variance in the thirteen questions. One question fails to load on either factor. This question, concerning shadows of IBMPC, is unrelated to the content of IBMPC and has already been discussed. The two factors (shown in the table below) can be regarded as broad descriptions of the two major uses of the IBMPC Conferencing facility: information and discussion.
| Value Variables | Factor 1 | Factor 2 | Means | Standard Deviations | N | |
| PCTOOLS Software | 0.81* | 0.00 | 1.68 | 0.96 | 176 | |
| Ask Questions | 0.13 | 0.53* | 1.70 | 0.92 | 176 | |
| Local Shadows | 0.24 | 0.17 | 1.80 | 0.91 | 177 | |
| Current Information | 0.51* | 0.32 | 1.84 | 0.80 | 176 | |
| IBM IUO PC Info | 0.54* | 0.29 | 1.84 | 0.86 | 177 | |
| Search IBMPC | 0.39* | 0.19 | 1.86 | 0.92 | 176 | |
| IUO Software Support | 0.78* | -0.11 | 1.91 | 0.99 | 176 | |
| Problem Resolution | 0.25 | 0.58* | 1.96 | 0.96 | 175 | |
| Technical Discussions | -0.09 | 0.68* | 2.13 | 0.90 | 176 | |
| Bug Reports | 0.27 | 0.46* | 2.16 | 1.06 | 176 | |
| New Ideas | -0.11 | 0.63* | 2.52 | 1.03 | 176 | |
| Product Descriptions | 0.52* | 0. 8 | 2.61 | 1.04 | 176 | |
| General Discussions | -0.04 | 0.67* | 2.75 | 1.06 | 177 | |
| Interfactor Correlation | ||||||
| Factor 2 | .43 | |||||
The two factors seem to be equally important in explaining variation in the questions. After rotation, each factor accounts for roughly 15% of the total variance. Still, an examination of the means for the various questions appears to support the notion that the information applications of IBMPC are somewhat more valuable to IBMPC participants than the discussion applications of IBMPC. Five of the top seven means are associated with the information factor. Five of the bottom six means are associated with the discussion factor.
In fairness, the spread of these means is not large. The most valued application is rated, on average, somewhat better than "very valuable". The least valued application is rated, on average, somewhat better than "valuable". Among the two top applications, moreover, one (PCTOOLS software) is a member of the information factor and the other (Asking Questions) is a member of the discussion factor.
Even with all of these applications rated as at least "valuable", however, there are significant differences. Indeed, t-test comparisons yield a continuum of five overlapping clusters of applications. Five features from the Information factor are among the eight features in the two more highly valued clusters. Four features from the Discussion factor are among the five features in the three less valued clusters. One concludes then that, on average, IBMPC participants value the facility more for the information they get from it than for the discussion that surrounds, creates, and sometimes obscures that information.
However much more valuable the information function of IBMPC may be to its participants, it is ultimately very difficult to separate information from discussion. This difficulty is apparent in the factor analysis, where the two factors are themselves fairly highly correlated (R=.43, df=1/160, p<.001). It is also apparent in the structure of the forums on IBMPC, where discussion and information functions are freely intermingled.
This intermingling is apparent in computer mediated queries, where an information requirement is satisfied through interaction. The comparisons with conversation and citizen's band radio are indicative of the discussion qualities inherent to the genre. The comparisons to classified advertising and advice columns are symptomatic of the information qualities inherent in the genre. The discussion probably dominates in the acting out of the query event. The information dominates in any subsequent reading of or search through the forum which is its host.
A similar mixture of information function and discussion function is inherent to several of the variants of computer conferencing that have been identified in this chapter, including on-line users groups, interactive software development, electronic seminars, meta-forums, lightning rods, polar debate, and electronic focus groups. Each of these genres of computer conferencing mixes discussion and information in a fairly interesting way.
On-line user groups, for instance, form around a general need to obtain information about a given product or group of products. The nature of the information needed will vary with the sophistication of the individual participant, but can vary from a need to understand the basic features of a piece of software to the need to push a piece of software into applications for which it was not designed. Intermediate levels of information can include understanding what a new user is doing wrong in trying to start the package, details of a new release, or learning the tricks that allow operations to be performed more quickly and easily.
This information is exposed in what is otherwise a discussion process, however. Sometimes the discussion proceeds in the form of a computer mediated query, with brief question followed by brief answer. On other occasions it proceeds in the manner of an electronic focus group, with a series of longer and fairly detailed appends building on each other, each offering an answer that is a little bit better than the ones before. Sometimes the discussion begins with an announcement of a new release or a related support program and is followed by a dialogue in which different participants reveal their discoveries about the availability, cost, ordering process, and features of the product. The forum will sometimes detail a series of neat tricks that can be performed in the program and sometimes detail bugs and procedures for avoiding them.
In more active on-line user groups, these discussion threads will be hopelessly mixed together, with several of these events proceeding in parallel. In less active groups these threads may be episodic, with one or another of these themes keeping the forum active for a few days at a time. Information may, in the end, be the more important to on-line user groups. It is, after all, the information that is the focus of both current interaction and subsequent reference. It is in discussion, however, that this information is uncovered and distributed.
Many of these same forces are at work in interactive software development. The participation of the program's creators or maintainers personalizes the interaction, however, inevitably changing it in the process. A bug that is reported in an on-line user group is directed to the programmer in interactive software development, often in a "good news, bad news" format (e.g. this program is really great, but this feature doesn't work right). A desire for an additional feature might, in a user group, be expressed in a "Does anybody know how to ..." question. In interactive software development, it is more likely to be expressed as a request for the feature, often prefaced with the frequently invoked term "WIBNI" (Wouldn't it be nice if), and sometimes accompanied with specific suggestions about how the new feature might be implemented.
Discussion, in interactive software development, centers around the program's authors. Where an on-line user group expects to find answers in other members of the group, participants in interactive software development frequently expect answers from the authors. Other individuals may know the answers to questions, but will frequently give a program author a chance to handle them before volunteering information. This dynamic can make discussion in interactive software development highly dependent on the author's participation. When the author participates, discussion blossoms. When the author is reticent, discussion can whither to nothing.
The purpose of discussion in interactive software development is ultimately information, but the information is most often strongly tied to fairly distinctive patterns of discussion, not the least of which is the distribution of software in reply to previous requests for new features.
Other combinations of discussion and information can be found in electronic seminars and electronic focus groups. Electronic focus groups are more strongly oriented to the collection of information, with all appends responding to a single initial question. Many of these appends will assume the content of prior appends, but they will rarely respond directly to any but the first. The result is substantially less than a discussion, yet still discernibly interactive.
Electronic seminars, by contrast, are more strongly interactive. There is generally a strong information component in such seminars, but that information is nearly always mixed with a substantial amount of personal opinion and disagreement. Interaction sequences in electronic seminars on IBMPC often extend for hundreds of appends, with each responding directly to one or more preceding contributions. Electronic seminars are, perhaps, the most discussion-oriented forums on IBMPC, but they also entail a substantial information component. It is often information that attracts interactants to electronic seminars, but discussion remains the core of the genre.
Meta-forums, lightning rods, and polar debates can be considered even more strongly discussion-oriented variants of electronic seminars. Meta-forums are, in some sense, seminars devoted to the question of how interaction should be conducted on IBMPC. It is these forums focus on meta-discussion that distinguishes them from other genres on IBMPC. There is a continuing information value in these forums in the knowledge of proper conferencing behavior that they instill, but this information is not distributed through the forums so much as it is distributed through the public and private interaction of IBMPC participants.
Lightning rods, by contrast, are generally loosely-focused seminars which are intended to pick up random discussions, hence drawing their thunder away from forums where they would be considered off topic. IBMPC has tried a variety of lightning rod forums over the years, including SOAPBOX FORUM, TEMPMISC FORUM, NEWSTALK FORUM, and most recently NEWSCHAT FORUM. Many IBMPC participants ignore these forums in their quest for specifically PC related information. Many others consider these forums a highlight of their IBMPC participation. They are nearly always among the fastest growing forums on IBMPC. The information value of these forums is real, if diffuse, and there are numerous occasions where discussions on lightning rods have seeded new electronic seminars.
Polar debate (discussed in more detail later in this chapter), when isolated to its own forum, entails the same function as a lightning rod forum, that of drawing what many consider to be off topic discussion away from forums that have other goals. Polar debate forums are generally much more tightly focused on a single issue than any other genre of computer conferencing on IBMPC, however. Polar debate can occur, moreover, in almost any forum. All that is required is an issue that invokes strongly polarized beliefs. Computer, operating system, and computer language choices are often subject to this kind of strong belief, and it is no accident that dedicated polar debate forums have names like APLDEBAT FORUM, C-DEBATE FORUM, and OS2-AIX FORUM. Each came into existence because debates on these language and operating system choices are frequent and intense, and IBMPC participants felt they deserved special, and non-interruptive, treatment.
The information value of polar debate can be high for the casual observer who can read through the rhetoric. The process of marshalling arguments for one perspective and against another inevitably points up strengths and weaknesses that might not have been recognized otherwise. It may be nonexistent for at least some debate participants, who often appear to read opposing appends primarily for the purpose of ripping them to pieces in a subsequent append. This kind of behavior, whether intended as such or not, is the major problem with polar debate. The administrators of IBMPC try to encourage participants to look for common ground in their discussions, and this sometimes means closing down a debate that appears to be on the verge of violating the formal IBMPC RULES or which is otherwise out of control.
The trend to discussion in computer conferencing is not universal, however. At least two of the computer conferencing genres identified here are strictly informational, entailing no discussion: answers files and voluntary newsletters. Both of these genres are essentially presentational. In the case of voluntary newsletters, individuals make personal contributions of source code, news abstracts, or other discontinuous material. In the case of answers files, a discussion (usually a computer mediated query) is abstracted to its essential question and best (in the judgement of the contributor) answers. The material is, once again, discontinuous, and were it not for the highly specific question-answer structure of every contribution and the way in which answers entries are used, answers files might not be distinguishable from voluntary newsletters.
The intermingling of some measure of both interaction and information is probably inherent to any medium. The interaction may be minimal; restricted, as it frequently is for mass media, to letters to the editor and other media complaints. The information may be minimal; restricted, as it frequently is in interpersonal media, to selectively presented "facts" that bear on whatever issue in under discussion. Books and newspapers are primarily information oriented (even when the information is presented to entertain). Telephone calls and face to face interaction are primarily interaction-oriented.
Computer conferencing, by contrast to these and other media, entails a strong intermingling of interaction and information. Some genres of computer conferencing are entirely oriented to information. Other genres are almost entirely oriented to interaction. This strong intermingling, and the strongly contrasting genres of computer conferencing that it entails, are probably largely responsible for the typological middle ground that computer conferencing enjoyed in both the formal typology and the user assessment of media using that typology.
The higher value users placed on the information oriented applications of computer conferencing in the survey suggests a desire to treat the medium as something closer to a mass medium. But so long as the medium remains oriented to minimally edited discussion, it will continue to straddle the middle ground as an interpersonal mass medium. The two genre reviewed in the remainder of this chapter should continue to illustrate indirectly the intermingling of these dimensions of computer conferencing value.
Just as it is possible, in computer mediated query, for a brief question to receive a small set of answers in a short period of time, so it is possible for such a question to receive a large number of answers, many of them related to prior answers, over an extended period of time. This variation on interaction is, in some sense, inherent to most forums on IBMPC, as each has a topic that can be expressed as one or a few general questions. There are specific forums on IBMPC, including IBMPC BENEFITS, TELAPART FORUM, WHATS4ME FORUM and CUSTOMER BENEFITS, which are entirely focused on this specific mode of interaction, and on which one sees nothing but new and improved answers to a single question.
The general form of this genre, which might be labeled an electronic focus group, is illustrated in IBMPC BENEFITS, a computer conference whose content will be of considerable interest in coming chapters for reasons that should be apparent from its opening append:
Please use this public file for describing how the IBMPC disk has been of benefit to you or others at your location. If you have described such a benefit elsewhere, please simply reference the appropriate append. Thank you.
This statement was posted by the IBMPC owner/manager at 2:18 PM on Saturday, February 2, 1985. The statement poses a simple question: "How has the IBMPC computer conferencing facility been of benefit to you and to IBM?". On the surface, it is simply a question, not unlike the simple query that was made in ISTHERE FORUM. The form of the question is a bit different than the typical computer mediated query, however. It is a request for information, but the request is fairly nonspecific about the kind of information needed. There is, moreover, an implicit expectation that there will be many varied responses. An entire file will be devoted to answers to the question. Appenders are expected to simply reference descriptions of benefit that have been made elsewhere.
Although the event started by this question is not computer mediated query, the difference isn't obvious from the speed with which its first append appears. Indeed, the first answer appears remarkably quickly, especially given its initiation on a Saturday, a non-work day for most IBM employees and always a comparatively slow day on IBMPC. The first response is at 4:00 that same afternoon, barely an hour and forty minutes after the forum opens. The difference from CMQ is immediately obvious, however, in the content of this first append, which sets the tone for many of the appends that follow:
At the Programming Center at Myers Corners Road, DSD Poughkeepsie, there are 100s of PCs installed with very little information available. For a non-technical person such as me (a planner) this presented a problem when I received a 3270PC to replace my 3278. IBMPC has been a life-saver. By browsing certain files I have educated myself to the point that I purchased a PC/AT to use at home under the Home Terminal program. IBM has benefited greatly by this as I now do a great deal of IBM work at night and on weekends. The Conference disk has explained how things work and has given me a ready source of help. If I have a problem a query will usually get a prompt response without anyone making me feel silly for asking a stupid question.Bugs in programs are found and fixed promptly. The library of internal programs has been of tremendous value. I am presently using many including PCTERM, MAILMAN, SWAP PRINTERS, Upload and Download programs, SCRIPT, etc. To my surprise I find that I am becoming somewhat of an authority in my immediate area on what's available - only because I keep abreast of what's new on IBMPC.
Although I am not yet able to make significant technical contributions I have been able to answer some queries so I am able to start "paying back" in kind for some of the help I have received.
The speed of this response is about the only thing it has in common with a response to a computer-mediated query. CMQ responses tend to be short and pointed. This response, by contrast, is nearly twice as long as all of the three forum responses to the simple query taken together. The append is also rather less pointed than any of those appends. Indeed, the appender takes the opportunity to tell us a little story of how the use of IBMPC has made him a more effective IBM PC user and IBM employee.
The appender exhibits somewhat atypical IBMPC conference behavior. There are people, including both the appender and the observer, who regularly access IBMPC on weekends, but they are a clear minority. Contributions to IBMPC shrink to a mere trickle on weekends. Oddly, however, the append is highly representative of the kind of append that occur throughout IBMPC BENEFITS in most other respects. Consider the append's:
The above are just a sampling of the kind of enthusiastic statements made concerning IBMPC over the course of the forum. Although it is difficult to project that this enthusiasm is typical of electronic focus groups, similar levels of enthusiasm has been observed in other electronic focus groups on IBMPC. Electronic focus groups appear to attract strong opinions, including both highly enthusiastic appends and highly critical appends. This is certainly the case in WHATS4ME FORUM, whose central question asked respondents to name the best and worst software they had used and explain what was good and bad in each. WHATS4ME shows little middle ground. Appends are either highly enthusiastic about products or highly critical of them.
It may be that this kind of extreme feeling in responses to electronic focus group questions is a reflection of the questions asked, and more particularly the circumstances that lead to the question. This is almost certainly a factor in IBMPC BENEFITS (see the next section), REXXBEN FORUM, and WHATS4ME FORUM. Hence it may be that the right question could lead to much more moderate answers. The observer has yet to observe an electronic focus group in which this kind of moderation has prevailed, however, and one cannot discount the possibility that the expression of extreme feeling is an integral element of electronic focus groups.
Based on these observations, the first reply to the question that opens IBMPC BENEFITS is be a fairly representative one. Its length, content, and style are fairly representative of most appends on the forum, and, at least by comparison to other appends, its enthusiasm for conferencing is subdued.
The opening of IBMPC BENEFITS by the owner/manager of IBMPC provoked a certain amount of concern among IBMPC participants, with people looking for hidden motives that are not expressed directly in the question. Numerous appends to the forum, especially in its early days, express concern that the forum has been started in response to a threat to IBMPC's existence. The third append to IBMPC BENEFITS states this concern directly:
Just out of pure curiosity, is this forum here because somebody is saying that we should limit, curtail, or abolish this excellent communication vehicle???????
The initial question is stated simply. There is no hint of problems in the request that people describe "how the IBMPC disk has been of benefit to you or others at your location." But motives are assigned anyway. This may well account for the rapid growth of IBMPC BENEFITS over its first few days. By the end of its first weekend, IBMPC BENEFITS contained four responses totaling 115 lines, and all but one of its major themes (areas of benefit) had already been introduced. Four responses in a single weekend is a torrid pace for any forum on IBMPC, and was a clear portent of things to come. On Monday there would be 16 new responses. On Tuesday there were 12 more. By Thursday it was clear that things were slowing down -- appends were both less frequent (5) and shorter. Forty-six appends were made to the forum during its first week; sixty-two in its first month. After this initial flurry, the forum's pace slowed considerably. A total of seventy-eight appends would be made during its first year. Another 77 appends were made to the forum during 1986; 55 in 1987; 37 in 1988; 24 in 1989.
The story of IBMPC BENEFITS outlined above is very different from CMQ. The differences already apparent include the form of the question, the content, length and number of responses, and the duration of the event. The key difference, however, is the question.
The simple query on ISTHERE FORUM was both specialized and highly specific. It required specific expertise, and left little room for the vagaries of opinion. There are many users of IBM PC's that would not have understood the question, and could not have answered it. It demanded, and received, highly specific answers. As a result, the answers to the query were interchangeable. Almost any one of the answers received would have fully satisfied the need.
The same cannot be said of the question posed on IBMPC BENEFITS. It may be that only one or two benefits can be identified, but the question carries no such expectation. Indeed, the request that people "simply reference the appropriate append" if they "have described such a benefit elsewhere" clearly anticipates that a large range of benefits might be identified.
These differences can be overstressed. Computer mediated query carries no requirement that a question be highly specialized or even highly specific. Still, both characteristics are generally found in the questions that open Computer-Mediated Query events. Neither characteristic is found in the question that opens IBMPC BENEFITS.
As it turns out, the fact that this communication event opens with a question is about the only distinct similarity it will have to a Computer Mediated Query. Differences include:
There is no such right answer to the central question of IBMPC BENEFITS. Indeed, there are potentially a huge number of answers, at least one for every person who ever derived benefit from computer conferencing.
These characteristics pose a strong contrast to the behavior observed in computer mediated query. Hence we must conclude that, whatever IBMPC BENEFITS represents, it is not the CMQ genre. Like CMQ, however, the forum it is an intriguing event that is difficult to describe in terms of non-computer-based communications media, including single question, multiple answer genres on other media. Consider:
The mix of characteristics described here is clearly highly distinctive. IBMPC BENEFITS, as an event, has much in common with a survey or a focus group in its goal of open ended answers to a fairly general question. The nature of the event is clearly very different from either, however. The interaction, in the form of appends that draw on, even when they don't reply to, other appends, is clearly distinct from the circumstances of either conventional alternative. A survey would generally be more highly structured and less interactive. A focus group would generally be more interactive, but with less breadth of participation. The event is ultimately closer to a focus group. Hence its naming as an electronic focus group.
The strongly enthusiastic and strongly critical responses that seem to characterize electronic focus groups can also be found in other genres of computer conferencing interaction, notably polar debate. Polarization of opinion seems to be a common occurrence on IBMPC. Such polarization, when it occurs, rarely seems to be resolvable simply through discussions on the conferencing facility. There are certainly hundreds of examples of protracted polarized discussions that have arisen on IBMPC over the years. Some of these discussions are almost routine, and have occurred enough times in enough forums to earn the designation "deceased equine" (an issue that has been discussed to death in the minds of participants who have seen it before) and, in some cases, to earn forums of their own.
Perhaps the most common recurring polarized debate on IBMPC at this writing can be found on OS2-AIX FORUM, in which proponents of the OS/2 operating system and proponents of the AIX operating system debate the relative advantages and disadvantages of each system. OS2-AIX FORUM was started as a way of funneling the argument concerning the two operating systems away from OS2ARENA FORUM, which is supposed to provide a discussion home for the uses of OS/2, including its uses when compared with other operating systems. The debate has recurred several times on OS2ARENA FORUM since, has also broken out, among other places, on MARKETNG FORUM, and is the successor to a similar recurring debate pitting UNIX against PC-DOS.
The form of debate in OS2-AIX FORUM is a familiar one to IBMPC participants and probably to most people, albeit with different issues. Useful languages and operating systems seem, inevitably, to attract "hard core" or "cult" users; people who believe that their choice of language or operating system is the best one possible; that no other language or operating system is necessary for their, or by frequent extension, anybody else's, needs. Few popular languages or operating systems have escaped attracting at least some of these "cult" users, and neither neither UNIX nor OS/2 are exceptions.
Polar debate, as encountered on IBMPC, is almost always rooted in an uncompromising belief in the correctness of one's choices in computer operating system, language, or other selection, and generally begins with an uncompromising expression of that conviction. There is no requirement, in polar debate, that both sides enjoy hardened positions. Only one side need have an uncompromising perspective from which to debate. Indeed, polar debate appears to proceed best when one side is not hardened into a position; when that side believes that compromise is possible.
OS/2, at the time OS2-AIX FORUM opened, was a reasonably new operating system. It had not yet developed a set of "hard core" users. If there was a core set of users surrounding the operating system, it cames from users of its ancestor operating system, PC-DOS. UNIX, by contrast, had been in use for nearly two decades, and had a loyal following in the programming community. The opening of the debate, in outbreaks of the OS/2 versus UNIX debate on IBMPC, has often started with a discussion of some feature or prospective capability of OS/2. An appender might, for instance, wonder if the operating system could be used in real time control applications. The discussion, in such cases, is basically exploratory at this point -- a fairly routine interaction for an electronic seminar.
The debate starts in a refutation of this exploration. The refutation might assert that OS/2's kernel is poor for real time control applications and that it's too bad OS/2 wasn't UNIX, whose kernel is ideal for such tasks. If nobody challenges this assertion, the debate never develops. It is, of course, difficult to know how often this kind of claim is not challenged. It is clear in the experience of IBMPC users, however, that the challenge is picked up fairly frequently.
The challenge may well be fairly placid. The original appender might say something along the lines of "that doesn't seem all that obvious to me. It looks like I can easily program OS/2 to handle a large range of real time control tasks." It doesn't seem to matter how the challenge is phrased. Once the initial refutation is challenged, a polar debate is almost inevitable.
When such debates break out, there are a variety of recurring elements:
A third, and less laudatory, use of the "quotation rebuttal" technique is to recast the words of the person being responded to. Quotations will sometimes be presented without relevant context information, and will sometimes include words that have been changed so as to take on a different meaning than was originally intended (usually by substitution of loaded words for neutral ones).
Quotation rebuttal is probably one of the key instigating elements in polar debate. Appenders to IBMPC's SENSITIV FORUM, in discussing the technique, assert that the "quote/counterquote" style, as quotation rebuttal is sometimes called on IBMPC, creates the impression of attack, even when there is none intended.
An appender who offered a compromise solution in one of the OS2-AIX debates, for instance, perhaps asserting that both operating systems have a distinct role to play in the market, is likely to be labeled as either a UNIX basher or an OS/2 bigot by uncompromising proponents of UNIX, an OS/2 basher or UNIX bigot by uncompromising proponents of OS/2, and as an agent of compromise by nearly everyone else. Such appends will very often become the target of point by point rebuttals of quotations.
This second element sets up a characteristic pattern of this kind of debate. It is almost always the case that the most extreme proponents of one position (usually the advocates of the universal solution) take a strong position. The system they advocate has few or no faults. The alternative system has few or no redeeming features except, perhaps, popularity. Defenders of the alternative, who most likely were in the middle of another discussion when the debate breaks out, defend the system under attack, occasionally respond in kind by pointing out the faults in the universal solution, and usually wind up trying to forge a compromise position that will end the debate.
Such compromise proposals are almost never accepted, with results that are easy to predict and painful to watch or experience.
This last result is probably the most damaging. Polar debates are, in many ways, similar to a cancer. A few rogue cells rapidly proliferate appends, killing off normal discussion and eventually, all regular interest in the forum. It is exactly this kind of outcome that has led forum owners, out of self defense, to open forums like C-DEBATE and OS2-AIX, hopeful that a forum dedicated to a specific debate will successful draw the "cancerous" discussion away from their own forum.
Taken as a whole, these debates represent a curious sort of extension to what Keisler, Seigel, and McGuire (1984) have called a flame. A flame, according to this paper, is a long, and frequently emotion laden, append in which the appender expresses in great detail his or her feelings about one issue or another. The word has become an important one on IBMPC, and appenders frequently declare appends as flames by beginning them with a line saying "&colon.FLAME ON." and ending them with a line saying "&colon.FLAME OFF."
Polar debates extend the concept of flame in two ways. First, such debates are most often composed of a large number of flames, many of them expressed in the form of "quotation rebuttals". Second, the sequence of appends can be regarded as a sort of community flame, in which at least two perspectives are flamed at length. Hence it is reasonable to think of a polar debate as a sort of "flame train". The behavior associated with polar debates is generally regarded as a problem on IBMPC. There have been numerous formal and informal attempts to restrain such behavior, including:
The development of these kinds of strategies on the IBMPC computer conferencing facility will be explored in greater detail in the next two chapters.