Last changed 11 Mar 1998 ............... Length about 900 words (6000 bytes).
This is a WWW document maintained by Steve Draper, installed at

(Back up to workshop main page) (Back up to other position statements)

Position statement

by Steve Draper
for the HCI teaching workshop 23rd & 24th March, 1998 University of Glasgow, Scotland.

When I reflect on how to teach HCI, a central point is that I myself didn't learn HCI from courses or textbooks, yet here I am teaching them. That is one of the reasons I can't just do my teaching by reproducing how I was taught. Instead, I think about what my important learning experiences were.

So my course revolves around the central exercise in which I hope to contrive for them the experience of seeing their interface designs (which actually they thought were pretty neat) fail; but then to be able to fix many of the observed (but unexpected) problems fairly easily. That is the experience that did it for me, and my main idea in teaching is to try to reproduce it for them.

A related feature of HCI teaching is (I myself believe -- but do others?) that what's important in HCI is an attitude (say, a "user-centered" attitude) rather than any particular facts or techniques. How can we teach this? How can I test this? (A real test might be how far a student would jeapordise their career by arguing with their clients or management about the need to pay for early and frequent user testing.) Such an attitudinal learning objective is rare in HE.

There are of course other angles, such as informal and formal pressures to teach traditionally (imitating what colleagues are doing in other subjects, producing exams to conventional format even though writing exam answers is not a relevant skill to an HCI practitioner). But recently I have come to think that a more important trap is trying to teach theoretical ideas that I am interested in myself. The counter-challenge to this is to observe what concepts and techniques students actually find useful when given a substantial design exercise, or if you ask them after their first year's employment what out of the material taught have they found useful. Many ideas feel satisfying while not adding much leverage in practice.

A final issue is part of a current research project ("MANTCHI"), but represents something that may increasingly affect us all as university funding reduces, the possibilities of the internet increase, and students require more flexible learning conditions. Could we form a UK collective for teaching HCI, and share expertise usefully? Anyone who uses a textbook at all has done this in one way: remote authoring, local adaptation and delivery. In MANTCHI we are doing some true reciprocal teaching collaboration, where each site both authors some items and receives others. This may be unique: normally there is a rigidly assymetric division into expert/authors and novice/buyers in both the textbook market and new CAL initiatives. Not only does the latter reduce the quality of authoring (because it means they do not have the experience of trying to use others' material and so cannot understand why people don't want theirs), but it doesn't fit the structure of HE, where experts are thinly spread across the country but all have to teach "whole" courses. Can we organise ourselves to use others' expertise for topics we are less than brilliant at in return for remote delivery of our best material? MANTCHI is also distinguished by focussing on "tutorial" material such as exercises. As I mentioned, I think these are of primary importance in HCI teaching, yet are if anything more sensitive to local delivery conditions.

In approaching this workshop, then, I focus on the following beliefs: