Stephen Downes

Knowledge, Learning, Community

Jan 21, 2002

I had the opportunity recently to participate in an online audio symposium with Technology Source. It was by no means my first encounter with synchronous learning online. Nor, probably, will it be my last. But let me confess up front that I am not a fan of synchronous learning on the web. At least, not the way synchronous learning tends to come out in the wash.

My Technology Source experience was typical. We scheduled the event for noon, Eastern Time (you have to schedule these events for Eastern Time because the vast bulk of participants are from that time zone. It still sort of works at five or six p.m. in Europe, a reasonable nine a.m. on the west coast, and if you're in Hawaii or Australia, well, forget it). The 45 minute session featured some interplay between editor James Morrison and myself supplemented with questions typed by participants in an online chat forum.

I think we all know that online synchronous conferencing facilities aren't yet up to snuff. Even with a relatively simple system like HorizonLive there is a certain amount of preparation. My browser - Netscape 6 - was not supported. The sound came through Real Media with only a few buffering breaks (of course, I had to turn the sound down because the audio was generated using a telephone conference). And midway through the session I managed to break the entire system by loading a web page that popped itself out of the conference's frameset.

But it doesn't matter. The technology will get there. I've worked with state of the art full fiber videoconferencing systems, with mobile href="http://www.picturetel.com/">PictureTel systems, with a telephone connection from Australia, with teleconferencing in northern Alberta, with voice over IP, with text-only MUD, with text-only IRC, with text-only floating Java client, with Centra Symposium, with Microsoft's None of these systems work properly out of the box (or straight from the download, as the case may be). With any of the text-based messaging sevices, there are always firewall issues to consider (and the Java clients generally don't work through a firewall, which is why my online chat is a simple web-based chat I wrote myself). With the programs that require audio, you always have to tinker with the microphone and headset (you can't really use speakers in the office) to set the sound levels. The video systems that deliver quality video require special connections, and the systems that don't require special connections do not deliver quality video. Even my office telephone causes me pain, especially when features like voice mail are added. It's nothing an experienced technocrat can't handle, but it's always a fuss and bother.

As I said, though, the technology will get there. So let's suppose that we were working with the perfect full duplex high bandwidth multi-party text or audio or video conferencing system. Let's suppose it came pre-installed, so that all I had to do was will it into operation, and it started, connected, with all participants present (without technical problems of their own) and ready. Indeed, assume you could do everything you do in an in-person meeting: share files, use the whiteboard, demonstrate an application... everything, that is, except spill coffee on your neighbour's notes.

Perhaps it's just me, but - I still wouldn't like it. Not because I don't like interacting with people - I do. Not because I don't like the technology - I do (or I would, in a world with perfect technology). No, I wouldn't like it because the system delivers exactly what it promised: a fully interactive synchronous conference. A meeting, say. A seminar or class. A forum. Something that starts precisely now and ends precisely then. A nice neat chunk, an hour of time, an hour out of my life, an hour containing all the frustrations and foibles of in-person synchronous communications.

Let's begin with the most obvious inconvenience: the need to be at a set place at a set time. For any meeting or seminar involving more than a half dozen people, this proves to be extraordinarily difficult. People today (even students) are busy. They have lives outside the classroom or the meeting room. An in-person meeting cuts them off from the outside world (or, will cut them off so long as it's considered rude to use the cell phone or read your email while talking with a group of people). When we add to the fact that one person's noonhour is another person's midnight, scheduling becomes even more of an issue. And with large groups, scheduling must occur by fiat because the planning would take longer than the meeting. Which means that the participants must arrange their lives around the event, much in the way the student must wake at the crack of dawn to attend that 8:00 a.m. math class.

My seminar for Technology Source had about 40 people registered, yet a maximum of eight people made it into the system. Sure, some people had technical problems. But as my the email in my inbox made clear, scheduling remained the greatest issue. And I suspect scheduling was an issue for the many hundreds (ok, dozens) of people who would have logged on if only they had that hour free.

My next - and more important - concern has to do with what I would call the "rate of information transfer" problem. This has several manifestations, but let me introduce it by analogy, and then I'll apply it to the world of online conferencing.

Simply stated, the problem is this: while requiring much more bandwidth, the rate at which information can be transferred using audio or even video conferencing is much slower than the rate at which information can be transferred through other media. Or, to put it another way, we read a lot faster than we listen or watch. Odd, isn't it? But if you take the transcript of a radio program or the script of the television news, you will find you can read (and absorb the content of) the text in a fraction of the time it would take you to listen to or view the same information. That is why television news can only cover headlines and scandals, even though you spend more time watching the tube than you do reading the backgrounders in the morning paper. Now your reading time may vary (though audio and video time never does), so you may experience this phenomenon to more or less a degree. But for the vast majority, audio and video are slow.

As I said, this manifests itself in several ways.

I am writing this article in the context of responding to an email enquiry (and in my typical fashion, the poor writer is getting nothing like what she wanted - this is the danger (and reward) of asking me for advice). And she wrote, "it is far easier and quicker to develop quality 'online instruction' than a stand-alone asynchronous course." As someone who has spent a lot of time standing in front of a classrom, I wholeheartedly agree. It took a lot less time to prepare an in-person class. It takes less time (but not as little time - you have to plan ahead a bit more when you are working with technology) to teach the same material using conferencing. It takes a lot longer to write the same content, and even longer to prepare it as quality asynchronous educational material.

But the ease of creating quality instruction in an audio or video format (and here I include the in-person class as a type of audio or video format) is exchanged against the quantity of information than can be transmitted in the same time. The reason why it takes longer to prepare the same material as text or as asynchronous learning is that in the latter two modes you can do more and do it more precisely. In a typical classroom, for example, you can plan on responding to a certain number of questions and comments (which it's wise never to plan for the full hour). These questions come up on an ad hoc basis - they are asked by students, which makes them as ad hoc as you can get. By no means are all of the questions asked - there isn't time, and there is no guarantee that the question asked will be the one more people needed asking.

In a synchronous environment, the delivery of informational content proceeds at the speed of one of the slower learners. Not the slowest - these people are simply left behind (we designate their inability to keep up as "failure"). But not at the rate of the median, either, necessarily (unless the instructor's delivery style or the student's classmates have managed to intimidate the questioners into silence). For people who do not need the answer, the time spent answering the question is essentially wasted time - no transfer of information is occurring because they already had the information being transferred.

The same sort of thing happens at meetings. I am sure we have all had the experience where one of the participants didn't read the background documents and hasn't taken the time to prepare his or her reports. So the assembled participants spend a certain amount of time reviewing background information. They are then subjected to an inarticulate oral presentation of a lengthy report - the presenter, after all, "didn't have time" to put thought to paper and ink. As for useful support information - graphics, diagrams, charts - forget it. If you want a visual representation of the report, you have to generate it yourself (fortunately, you have plenty of time).

There is a trade-off in the time saved by the presenter and the time lost by the listener and we can actually approximate some reasonably sound calculations. It may take a person five minutes to prepare a one hour oral report as opposed to spending four hours writing the same content. The content, if read, would take each person an average of fifteen minutes. If each person's time is valued at the same rate (all bets are off if you are dealing with $1000 per hour consultants or the company CEO) and the quality of information is otherwise the same (all bets are off if the presenter can't write) then the break even mark for this meeting is four people (i.e., 3:00 hours x 1 person (the presenter) = 0.75 hours x 4 persons (the listeners)).

A similar logic exists in the classroom. It may take the instructor three hours to produce the lesson, but the lesson that would be taught in a one hour segment can be absorbed by the student in fifteen minutes if it is presented in written form (or written with suitable graphics). Depending on the value of the time (in some cases a teacher's time is worth more, in other cases - such as when you are presenting a brief to management (and are expected, notice, to present most of the content in written form) - the student's time is worth more. Either way, the trade-off will still be low unless the student's time is valued near zero (which explains why having the professor save three hours at the expense of a hundred students wasting, collectively, 75 hours, still looks like a good deal in universities today).

Now by this point you may be thinking that I believe all instruction and all interaction must be asynchronous. Not at all! I still use the telephone (albeit, reluctantly, and I won't answer it when I'm at home). And I still participate in face-to-face meetings - even this morning, my colleagues and I had a staff meeting (well, actually, we had some coffee and talked about the Director and griped about computer support). There are cases where you want synchronous interaction, and not only because you want to keep the content of your comments off the record.

First of all, while the calculations presented above were presented from the point of view of the rate of information transfer, there are other factors to consider. For it is not always appropriate to maximize information transfer. Absorbing information at the highest rate possible is a lot of work. That's one reason why many people prefer television to reading: it's easier. You don't need to pay full attention to what you're watching. You can chatter with your family and friends (but not at the movie theatre, because that's serious, and we're expected to concentrate), flip channels, look at the newspaper, eat, clip your toenails - any number of activities.

Video didn't kill radio, and the internet won't kill video, because listening to the radio or watching television make fine background activities for web surfing (indeed, I don't know anyone who surfs the web in silence). I have yet to go through an audio conference online or off where I was not also reading my email or sketching an outline for my next opus. I am still devising techniques for bringing my laptop into videoconferences so I can surf while waiting for the person at the other end to collect his or her thoughts.

I think that this in large part explains why audio and video conferencing receives such high marks from participants. The less you want to be at the class or the meeting, the more you want to use some form of conferencing, because it's easier and - especially if you're online - you can do other things at the same time. By contrast, you may note that if a person needs to learn something important, when they're interested and engaged, they have much less patience for the less productive moments characteristic of any synchronous event.

Another case where synchronous communication is more appropriate is where there is no obvious presenter of information. In such a case, all participants - not just the presenter or the instructor - are faced with the trade-off between preparing detailed materials and spending extra time in the meeting. This is especially the case where a high degree of interaction is required - where a piece of information must go back and forth a number of times, being processed at each interval. We see this, for example, in collaborative writing exercises, where the sentence may undergo four or five successive revisions in the space of a few moments. In such a case the extra work involved in performing the task asynchronously far outweights the savings in presentation time in a synchronous exchange.

This requirement - interaction - is often cited as a justification for the use of synchronous conferencing facilities in online learning. And where interaction actually occurs, this use is justified. But in so many cases, what is being passed off as "interaction" is in reality a "presentation." For example, it is a presentation when the bulk of the communication is handled by one person, even if some questions are allowed - a case assumed by the designers of most online classrooms (you know, the ones where everybody but the instructor must 'raise a hand' in order to be allowed a few moments of speaking time). And for example, it is a presentation when the number of people attending reaches more than a handful - if a person spends fewer than five minutes speaking, then this person is quite properly thought of as a recipient of information, not a participant in its creation.

Finally, synchronous communication is useful in cases where it does nothing else but break the monotony of bland, grey learning materials. Sure, stimulating, engaging multimedia learning materials can be designed in such a way that a learner will think a slow synchronous conference is a rude interruption, but few organizations have the resources to develop such learning and provide instead a few slabs of content in a WebCT interface. And there is nothing - trust me, nothing - exciting about taking a course using WebCT.

People need a break. They need some colour in their lives. They need to hear the occasional sound of another human voice. And if the most interaction they get in a day is an automated self-test quiz, they begin to lose hold of their foundations, to drift, to begin puttering in the garden rather than pruning prose, anything, just for a break in the routine, an off-hand witty comment even if the informational content is exactly zero. How else do you explain the student's typical use of live chat - gossip, silly remarks, discussions of the weather (I've seen it all). It's not that they want to send educationally useless comments, it's that they want some break in their day-to-day routine.

My email correspondant sent me a well researched article which pointed to the rise of what has come to be called 'blended learning'. She pointed to the increase of in the number of synchronous courses being offered online both at local institutions and as reported by online learning research firms. She pointed to the the recent emphasis on multi-modal learning as outlined, for example, in much of Tony Bates's recent work. She pointed to examples of instructors being excited by the ease of creating live e-learning and to student surveys showing that they much prefer some synchronous delivery over and above their WebCT marathons. It's hard to argue against these trends, even for some as intolerant of slowness in learning as myself.

But we need to understand what it is when we are creating synchronous learning. What we are not doing - very well - is transmitting information. For students (or classes) that require that information be served in large gulps, the synchronous classroom will be very frustrating. And what we are not doing is serving large numbers of students. As instructors time and again have noted, there is an upper limit to the number of people who profit from a synchronous session - Yahoo limited its chat rooms to 20 people (back in the days when it still had chat rooms), and that would be (roughly) my experience as well (I still remember an ambitious - and chaotic - CADE session we had online in 1995 with 40 people).

But finally: why not consider abandoning the constraints and weaknesses of synchronous learning altogether? Let me propose what to me seems to be the ideal use of the technology: asynchronous synchronous learning. Once again, I'll introduce it by analogy.

One of the most successful programs online - after email, of course - is an instant messaging program called ICQ. It has spawned, as might be suspected, imitations from both AOL and Microsoft (AOL finally broke down and bought the company a few years ago - just as well; there are about as many ICQ users as there are AOL members).

ICQ has two absorbing features:

  1. It allows for instant messaging (that's the synchronous part), but
  2. It's always on: you can send and receive messages whenever you want (that's the asynchronous part)

Now - were I setting up an environment where I wanted to provide live, in-person support for, say, a software application, I might do it with ICQ. In the old days, I would have set up a telephone help-line with a bank of help-desk operators, but the problem with that is that you can only accomodate one user at a time, which means (and we all know about this from dealing with our ISP) everybody else sits on hold and waits. With ICQ, a help-desk operator can receive a number of enquiries at the same time - people aren't put on hold, and he or she can deal with the easy ones first and the hard ones later (or interrupt that marathon two-hour help session to send out some quick answers).

A slight improvement on this at the help-desk end is to create an environment where people can wait, chat with each other, and listen in to the public parts of the help. This is common on IRC, where 'IRC-help' channels are set up for multiple users. An operator - usually a volunteer - sits in the channels and answers all questions. Unfortunately, there is a scarcity of operators, so people can't linger and chatting is discouraged. Moreover, IRC is a little too public - sandwiched between the help channels and the serious discussions of foreign policy are the sex video channels and the warez exchange boards. Also, because IRC is text-based, you have to pay attention to what's going on. You can't minimize it and wait for your question to come up.

But the logical evolution of IRC - and the logical evolution of the use of synchronous technologies in education - is the 'always on' videoconference classroom. Actually, it would be more like an 'always on' videoconference study hall than a classroom, since participants aren't logged in to listen to some instructor drone through a half-prepared lecture on the symbolism in James Joyce, they are there to interact with each other by-the-by as they do other work. An instructor is present, of course, to participate, answer questions, and generally fulfill the role of coach, moderator and encyclopedia. But the instructor doesn't hog the bandwidth - students can read online texts, watch instructional videos or work through a programmed envrionment if they want some sort of information dump.

Not that the instructor stays silent and simply responds to questions. He or she 'rides the wave' of the conversation. If the time and the mood and the participants are ready for something extended - five minutes on Dickens, perhaps, or an analysis of 'meaning' in the Tractatus - then the instructor is there. No preparation needed or desired: the strength of synchronous presentations is in the ad-libbing (by finding out what the instructor says 'off the cuff' you get a fine tuned sense of what's important).

The asynchronous synchronous learning environment is structured like ICQ, breaking as needed into separate 'rooms' (analagous to IRC channels) - though more control is exerted to weed out the porn exchanges. Silly, meaningless dialogue is allowed - even encouraged, sometimes - because of the importance of variety and personal relationships in learning. Participants both see and hear each other - and everyone else on the same channel, because they all have web cams. The channel may point into some real-life (RL) training rooms, and the RL room may have a big screen display. Full duplex always-on sound is the rule: if the noise is distracting, hit mute.

Participants flit in and out of the room as they desire, though what typically happens is that students in the same class congregate in the same room at roughly the same time (this is not organized by the institution, it is organized by the students themselves). There's always an instructor on hand, though several may converge on a room during the busy times. What structure there is is set by the instructor, who (of course) is also engaging in an ongoing email dialogue with the students. But there is no such thing as being 'late for class' and if the instructor feels something worth saving coming on, he flips the 'record' option and saves the live session in the archives for later viewing.

Now I admit: we do not yet have the bandwidth to do this yet. But the bandwidth is less than a decade away (and is already present in corporate intranets). Nor do we have the software to do this yet, the existing packages having concentrated on stifled and lockstep in-class delivery. But neither is this beyond the pale - nothing new is required, only an integration of existing elements. And the pedagogy - the pedagogy. We already know how to do this. It's how we structure - to a large degree - professional conferences, not to mention universities and offices. The fun stuff, the useful stuff, the stuff where something actually happens, occurs in an informal atmosphere.

So yeah - I guess there is a future for synchronous learning. But not the future people imagine just yet.



Stephen Downes Stephen Downes, Casselman, Canada
stephen@downes.ca

Copyright 2024
Last Updated: Mar 29, 2024 10:06 a.m.

Canadian Flag Creative Commons License.

Force:yes