Everyone talking at once: Notes on chat-centered teaching

Online antiquity

In the future, everyone will be a hacker for 15 minutes

Space and time

Varieties of power

Will it float?

Teaching English at a university with a heavy, mostly core-curriculum workload, I had the good fortune in the mid-aughts to be handed online teaching opportunities with no structure or guidance. This translated to enormous freedom. These were the early days, with the nightmare of televised "distance learning" was still a fresh memory. Primitive courseware platforms strode the earth in search of sacrificial victims to the gods of bad software design and worse pedagogical assumptions. I was already transitioning from literature scholar to full-time computer geek, so I welcomed the challenge of designing my own courses and programming my own tools. These included a quizzing/testing Web application, a threaded discussion forum, an alarmingly successful plagiarism detection system, and a text-chat environment.

[Note: I write this in June 2021, after the Covid-19 pandemic forced all teachers to improvise online solutions. Doubtless a period of academic reflection on the results of this dire experiment, which I have had to observe as a complete outsider. My experience allows me to hope that true innovation resulted from these "uncertain times." I just hope it does not involve Zoom.]

The once-a-week chats were the only meetings of the entire class in real-time. The chat pedagogy that I read about was distinctly unimpressive -- chats were for low-intensity open discussions, or private one-on-one meetings. Class discussions took place exclusively on asynchronous discussion forums (which I would also be using). Chat was not viewed as a usable medium for groups of more than a few students, and the tools were accordingly meager, with modest maximum logins and crude administrative controls. As I launched into this work dissatisfied with the existing expectations and tools, Amid these unsatisfying expectations and tools, I had little suspicion that chat would reveal inherent strengths.

What I wanted was something as informationally rich as a conventional class meeting, with meaningful dialectical play between instructor and students, and among students. After some false starts (AOL Instant Messenger, weird Java applets, fast-refresh HTML pages, and other things I've pushed from memory), I settled on IRC as my chat software. I was familiar with IRC from the Linux hacker world, and I knew it was free and open-source software that could handle thousands of logins and conversations without crashing. It was old, well-tested, and somewhat disreputable software that appealed to online criminals as much as to Linux system administrators. This made some of my more technologically (or criminally) informed students raise their eyebrows when I assigned it for classroom use. Another advantage was that I could run my own IRC server and have complete control as the oper (operator, i.e., the boss).

Another advantage of IRC was its associated bot program Eggdrop, as well as Eggdrop's endless special-function extension scripts. IRC requires a bot to guard ownership of a channel and to perform other housekeeping functions, such as managing user identities and ejecting abusive users. These conventions evolved to meet the communication needs of hacker communities, which meant that they had to be robust enough to thwart talented pranksters. This was a pretty good fit for a random selection of college students.

I added more sophisticated capabilities as well. Eggdrop kept track of the word-count of each student (helping my grading), prepared chat logs that were posted to the Web automatically (helping students study), and supported a general-purpose chatbot that I named Socrates (helping me amuse myself). Socrates could provide answers to questions posed by either the instructor or students, ranging from simple word definitions to statements about the themes covered in the class or the class rules and requirements. Alas, my preserved chat logs didn't capture the output from Socrates, which has been lost.

With the technology in place, the remaining questions were about how to structure the ~60 minutes of class time. I had been teaching the world literature surveys for years, but with only one hour per week of real-time contact, I had to become a much more disciplined teacher.

There were two general goals to the class. I wanted to acknowledge and discuss the strangeness experienced by these first- and second-year students when they confronted ancient texts. I wanted them to be comfortable with the uncertainty of that strangeness. This meant welcoming the most random imaginable responses from the class. Yet I also intended to model more sophisticated models of reading, and for each class I naturally had in mind a number of specific themes to put across. I prepared a list of text quotations for each reading selection that I could drop into the chat, guiding discussion by eliciting student responses to specific examples.

So far this is not very different from a class that meets physically. Chat, however, opened opportunities missing from conventional classes. Academic skeptics of online pedagogy doubted that the rich give-and-take of in-person class discussions could be duplicated over a remote, text-only medium. They thought this at the time, anyway -- the early aughts. Such critics must have overlooked the behavior, in their own households, of adolescents busily abandoning both telephony and email in favor of text chat for their most urgent personal communications.

Among other novelties, chat exploded the temporal confinement of the conventional classroom, where everyone speaking at once is mayhem. In the IRC window, everyone speaking at once is a surveyable field of knowledge. Much of it was repetitive, but you quickly learned to skim through repetitions. Adding your own comments might interrupt reading, but there was plenty of opportunity to catch up on you had missed.

This was a conversation unlike a real-world dialog, where the past slips away, becomes contestable, and then forgotten. It differed, too, from a static transcript (all those books in libraries!) that is encountered long after the participants are gone. The technology facilitated a viscerally immediate discourse that was also preserved and reviewable. As machines will do, this machinery ginned up what had been a naturally slow and discontinuous process, changing the experience and the outcome, even when not breaking radically with past practice.

Also different is chat's native power structure. Granted, the teacher, now an oper, is just as controlling, if not more so, than usual and I certainly spent a lot of effort keeping it that way. What does change is the relationship between one student and another student, and between one student and the classroom collective. In a large, conventional, core-curriculum classroom, an instructor is grateful if 10-15% of the students spontaneously talk during class time. The others don't talk because of various reasons: the anonymity of crowds, lack of self-confidence, lack of interest, lack of preparation, peer social pressure, intimidation caused by the "good" students, gender, race, cultural norms that stress modesty, neuro-atypicality, introversion, physical challenges, learning disabilities, and so on. Also: Limited available time. To speak out in a physical classroom requires commitment to a competitive social game at which all are not equally skilled, nor, in an ideal world of discourse, should they need to be.

I required (as teacher and oper) everyone to talk in the chats, period. Part of the course grade depended on a word count of chat contributions. This was both possible and desirable. It's possible because electrons are cheap and no-one is overwhelmed by the cacophony. Chat participation was non-competitive -- if 30 people made the same correct comment, good, they all get credit. Originality and uniqueness are highly overrated. For a moment, the classroom dialectic became an online poll, and what's wrong with that? Everybody went on the record. Nobody got humiliated for being the 27th correct answer (or incorrect answer, for that matter -- correctness is assessed elsewhere). It's desirable because we say we want everyone to participate, but we usually don't build pedagogical frameworks where they can, or where they are rewarded for doing so.

It's possible that the way I ran these classes was too idiosyncratic to be generalized. I was able to surf the comments streaming across my screen and pick out the tactically optimal ones to help push the conversation forward. The students, for their part, had no trouble keeping up. The chat transcripts speak for themselves.

I found the experience stimulating, and to show it off, I invited numerous people to join the chats as observers. Some colleagues refused, because they already knew what to think of such things. The hostility to online teaching that I encountered at my university led in large part to my departure from the academy after 28 years. I would have welcomed engaged and informed debate, but in the absence of that, I defected to the less (or at least differently) insular tech sector.