Last changed 7 Feb 1999 ............... Length about 900 words (11,000 bytes).
This is a WWW document maintained by Steve Draper, installed at

Web site logical path: [] [~steve] [this page]

Lurking and Learning

Contents (click to jump to a section)

Based on a message to ITFORUM by Steve Draper on 13 Dec 1997.

This is an attempt to summarize, at least for my own benefit (see 1 below), the issues around benefiting from dialogue, as illuminated for me by the ITForum discussion on LBO and lurking. (I don't have a full paper on LBO (Learning By Onlooking cf. vicarious learning cf. lurking) yet, but here are some chunks: 1, 2, 3, 4.)

I will call those who generate utterances (spoken or written) "speakers", and those who receive them "listeners".

Dialogue, whether spoken and face to face or written and mediated by e-mail, is both a social activity that needs managing and coordinating, and a personal activity with private benefits.

Learning benefits

What is magic, but not altogether obvious, about human dialogue about conceptual matters (as opposed to making joint practical decisions) is that although the dialogue will look coordinated (turn taking, apparent agreement), in fact detailed studies show that what the participants take away from it as personal gains is often quite uncoordinated. For instance Miyake's study of "constructive interaction" showed that when two people discussed her set problem (how a sewing machine works) both made a lot of progress through stages of understanding, but these were not in step: just because A articulated a new level of explanation did not mean B absorbed it at that point. Later work by Christine Howe and others with children similarly showed that when two children of unequal conceptual level with respect to a problem (e.g., the physics of what determines whether an object floats or sinks) discussed the issues both progressed: it was not that the less advanced received knowledge from the more advanced, who gained nothing. Apparent agreement in the dialogue did not reflect the private beliefs of the participants, and the dialogue was more valuable to them individually and privately than a transcript predicted.

There are three distinct sources of learning value for participants.

(1) The value to a contributor of generating a contribution. If I don't after all e-mail this, I will still gain this value. That is why writing lecture notes is valuable, even if I were never to read them again. That is why writing a private diary can be a learning experience. This is Laurillard activity number two, and it is learning-generative by itself.

(2) The value of reading, skimming, or overhearing others' inputs — the value of non-interactive reception.

(3) The value of receiving others' responses to one's own inputs. This is powerful in improving one's current ideas.

The value of (3) is what sometimes makes contributors complain about lurkers: if lurkers replied, contributors would benefit more. It is also true that lurkers gain (2) but miss out on (1) and (3), so such complaints are not totally selfish: lurkers could gain more by contributing.

Dialogue management: the group problem

However contributions by all in a big group would reduce the interaction to chaos. Managing/coordinating the joint activity has various aspects, including managing the volume and agreeing a topic; but it is also a matter of urgency or time constraints: where this doesn't fit an individual's schedule, they will not be able to contribute actively but can still benefit personally at no cost to others. All of this applies equally to face to face discussions, e.g., seminar groups.

The listener's (time) management problem

The distinction may be made, for any approach to (time) management, between urgency and importance: between what must be done soon because there is a time limit (time priority) and what deserves selection because it has a high value (value priority). (Many of us make everyday mistakes from failing to distinguish these two, e.g., breaking off a useful conversation to answer the phone which is urgent but often unimportant.) Any dialogue has time dependent (urgent) aspects, but these are often unrelated to the personal value of taking part.

A listener on an email discussion forum will be taking both urgency and importance into account in deciding whether to contribute (reply). Because dialogue requires a timely (urgent) response, this is only possible from listeners who not only value the content but can make it urgent as well, so that they have time to read and reply within the typical time response rate of that forum. (What that time is depends upon the community, and NOT the technology. Some local email communities have a standard response time of a few minutes, others of hours, others of days. What matters is the common behaviour, which sets the common expectation, and hence the interpretation of what length of delay means a deliberate failure to reply, or the end of a topic.)

This was vividly described by Jane Moch in a message in which she distinguished three areas of interest, each of similar personal value, and all addressed to some extent through internet communications; however only the one that is also urgent for her because it is needed from hour to hour in her job gets responses because they can be immediate. In the others, she values the reading but seldom can respond in a timely way. Lurking, then, may generally be the response of listeners for whom the material is valuable but not urgent.

The speaker's management problem (getting feedback)

In addition there is a value for contributors less related to learning and more to do with their management of their own production. Silence (in response to an utterance or posting) is ambiguous between:

Contributors in general may need at least some positive reassurance about the value of what they have delivered in their presentations, but, especially over new media, this isn't automatically bound into what gets discussed. Besides the other cases (the silence of disinterest or contempt), there have been some on ITForum of the first: messages that were very highly valued by many recipients but too much so to be responded to. They seemed to deserve more knowledge and effort than anyone had available for putting into a response. If this applies to every single listener, this is unpleasant for the speaker. I now think that you have to invent new "feedback" habits there that don't come naturally, such as interrupting the conversation to ask the other end to change their camera shot: it feels unnatural to have to speak to do something you would normally do tacitly by moving your own head, but how else can they know what I need?

It's not only the lurking discussion here that has had me thinking about feedback practices in new media. I've recently been doing some tutorials over video conference links (see for some elementary notes on this).


The ITForum format of both unsolicited discussion and scheduled discussion papers is especially successful because it accommodates both the issues above (importance and urgency, value of the content and timeliness of responses). Without deliberately introducing topics with sponsors, some e-mail lists languish because they do not have a mechanism for dealing with the management issue of time control while addressing the value issue of worthwhile topics. However there is a further issue that ITForum does not have an explicit code for, although it has seen some examples of good practice. This is the issue of how to design a contribution or manage an exchange to maximize interaction — to reduce monologue and increase dialogue. It is to do with contributions that are more like stating a problem than giving an answer; more to do with connecting with many people's experience and less to do with reporting the last word on something.


Miyake, N. (1986). Constructive interaction and the iterative process of understanding. Cognitive Science, 10(2), 151-177

Howe, C.J., Tolmie, A. & Rodgers, C. (1992) "The acquisition of conceptual knowledge in science by primary school children" Brit. j. Dev. Psy. vol.10 pp.113-130

Howe, C. J. (1991). Explanatory concepts in physics: towards a principled evaluation of teaching materials. Computers and Education, 17(1), 73-80

Howe,C. (various) A list of some of the many more publications by Howe

Web site logical path: [] [~steve] [this page]
[Top of this page]