Information Technology Reference
In-Depth Information
CHI and MIS: Two Disciplines Divided by a Common Language?
My first presentation as a CHI researcher with an MIS audience, in 1990 at UCLA, ended badly.
It had gone very well until my host asked a question. It seemed meaningless, so I replied cau-
tiously. He rephrased the question. I rephrased my response. He started again, then shrugged and
stopped. The next day I gave a job talk on a different topic at UC Irvine. When I next saw my
UCLA host, he seemed astonished to hear that Irvine had hired me.
Later I identified the origin of my confusion and loss of face. My host and I attached different
meanings to the word “users.” To me, and to everyone in CHI, it meant hands-on computer
users. My host understood it in an MIS sense that included people who might never use a key-
board, who decide what software is needed and read printed output or reports. His question
homed in on users who were not hands-on users, but to me all use was hands-on. I couldn't under-
stand him.
This misunderstanding was not minor, personally or conceptually. There was a fundamental
distinction between the CHI and MIS views of human-computer interaction. CHI formed soon
after the arrival of the PC. It was primarily about a single user sitting at a keyboard and display.
Most early CHI researchers were cognitive psychologists interested in the mind of a person using
a computer. MIS, with its broader perspective, employed two terms: “users” and “end users.” In
CHI, these were rolled into one. Someone who did not face a keyboard was not a user and was not
of any interest. Few of us ever said “end user”; we found it strange when others did.
This linguistic confusion was not unique. Working in industry and discovering practices that
were incomprehensibly ill suited to developing usable software, I became interested in organiza-
tional behavior. Guided by a colleague in a marketing research group, I began to explore MIS. My
colleague submitted an article to an ACM transactions journal titled
Successful Implementation of
Office Communication Systems
, using “implementation” in the MIS sense, to mean organizational
deployment. But to me—and the editor—“implementation” was a synonym for coding or program-
ming. The paper was accepted, but the editor forced a change of title. It became “Strategies for
encouraging successful adoption of office communication systems” (Ehrlich, 1987).
Just prior to participating in a panel on task analysis at a European conference, I discovered to
my dismay that this MIS-oriented group defined “task analysis” differently than I did. To them, it
meant an organizational task analysis, considering tasks as components in a broad workplace set-
ting. In CHI, it meant a cognitive task analysis, breaking a simple task into components; for exam-
ple, is “move text” better thought of as select-delete-paste or select-move-place?
Two Communities, Two Terminologies
(Grudin, 1993, p. 113) identifies these and other terms
used differently by MIS and CHI researchers. Within CHI, but not MIS, an application was a
single-user, personal productivity application; anything more was a system. The term “evalua-
tion” differed in connotation: Because a first principle of CHI is iterative design, “evaluation” is
a desirable element of the design process, whereas in MIS it often referred to an assessment car-
ried out long after the design and thus “not where the action is” (Herb Simon, quoted in Bødker,
1990).
Given that MIS and CHI share theoretical and methodological ground, why dwell on differ-
ences? Finding and exploiting the common ground requires understanding and working around
the differences. Conflicting word use, structural differences in the organization of scholarship,
and related problems impede effective communication.
This essay presents a view of the history of human-computer interaction as a framework for
approaching these issues. It is one perspective, one distillation. It builds on earlier histories and
will be followed by others.
Search WWH ::
Custom Search