I guess if academics think we know the meaning of research competition,
your situation vis-a-vis Cynthia shows us what _real_ competition can mean
in terms of not being able to openly share ideas and data! still, something
can be salvaged as long as you two keep your discussions general enough to
avoid proprietary concerns.
Your account of the design and usability problems you've seen does seem to
me to exemplify my hypothesis. Perhaps I was too simplistic to say that the
designers don't want the input. I didn't mean to say that they don't
_think_ they want the input, or even feel the need for it. What I meant is
that the design process, the customary practices they associate with good
design, tends to seek to close itself off from _some kinds_ of user input
(see earlier message).
What you seem to be reporting is that although they want the input, they
don't _trust_ it, and in trying to make it 'trustworthy' what they are
actually doing is limiting its usefulness. This is very close to what I
imagined, though I did not quite know how the designers would put things.
The mistrust is based, I think, on their sense of how important the design
process is, and so how risky it would be to be guided or misled by
untrustworthy data. So this leads them to minimize the risk, which to them
is the risk of bad design, but to me is the risk of taking the design
process substantially out of their control. They minimize, as you say, by
insisting that trustworthiness equals quantitative measures, but this
exactly limits the data to the kind they can control: i.e. they control
what measures are to be assembled, and so what potential kinds of
implications the data can have. Educators do just the same thing with
quantitative testing measures or quantitative research data. What is
dangerous is open-ended, uncontrolled, ethnographic studies that might lead
to design approaches that did not originate with the designers, or even
perhaps agree with their philosophy.
Imagine, for instance, that designers of a certain social category
background inevitably see things from their own perspectives (gender,
social class, culture, etc.). If ethnographic data from users who were very
Other to them produced directions for design that were alien to their
thinking, they would be quite uncomfortable with the results, I think. It
calls into question the assumption on which their design competence rests:
that a design expert can design for _anyone_. This is a universalist view
that is antipathetic to real cultural differences, and it is a founding
principle of the power of one social caste to legislate for all social
categories. It is exactly parallel to the notion that research on while
male college students reveals universal psychology, or that studies of the
expertise of white upper-middle male class technical specialists prescribes
the ideal cognitive strategies for all humans, or that [same category]
curriculum designers can say what all students need to know. If this
assumption crumbles, so does most of their power, in each case. What would
happen if you had to hire women to design for people who were not out to
'act masculine' in every tiny detail of their lives? if you had to hire
African-Americans from poor, urban backgrounds to design for people with a
distaste for authority and impersonality in interaction? (these are just
hypothetical guesses, but I hope the general point is clear)
I am not saying that you cannot design for people unlike yourself, but I am
saying that you cannot expect to be equally good at designing for everyone,
and that as cultures and habitus diverge, your competence declines.
Ethnography threatens the principle that all uses are alike, or more deeply
the idea that one system of classification, constructed from any one
position within the social structure, can take account of all relevant user
differences. That is a principle of modernism. It is even a keystone of
classical Marxism, and one of its key weaknesses in my view. Of all the
designers who need the humility of openness to ethnographic Other-ing, it
is social engineers who need it most (and who are least likely to sacrifice
their power by embracing it).
To end on a practical research note, it would be interesting to find out
how the designers react to open-ended ethnographies of use/users,
especially when the data conflicts with their own ideas of what _should_ work.
JAY.
---------------------------
JAY L. LEMKE
CITY UNIVERSITY OF NEW YORK
JLLBC who-is-at CUNYVM.CUNY.EDU
---------------------------