UML/UP integration with usage-centered design
- This posting to another eGroup (<mailto:email@example.com>) may be of
interest to participants in this one.
While I applaud the efforts of the organizers [of TUPIS2000 at UML 2000] and
support enthusiastically the exploration of the topic of integration of
user-interface and task
modeling with UML and UP, I do have several concerns about the forthcoming
colloquy at UML 2000 and the tone and direction that may be implicitly
First, I am concerned about this group and its discussions being somehow
seen as representative and its deliberations taking on some--even if
informal--decision-making role in the proposal, recommendation, and
promulgation of notational and methodological practices in these areas.
While I was among the initial list of invitees, work commitments to clients
in North America simply preclude my participation. The current demands for
training and design consultation in usage-centered design are such that
little time remains even for reading submissions and none for writing a
workshop paper. Unfortunately, this leaves the forthcoming discussion
without any direct representation for usage-centered design in general and
essential use cases in particular, arguably among the most widely practiced
approaches to model-driven, task-based user interface design.
I am also concerned about how readily such workshop forums can lead to a
certain kind of democratization of ideas that, while laudable in principle,
all too readily can lose track of history and the weight of experience.
There is a vast difference between, on the one hand, recently minted
modeling schemes or academic proposals proved on a project or two and, on
the other, notations and processes like usage-centered design, or OVID for
that matter, which have proved their efficacy on literally hundreds of
projects of varying scales at large numbers of sites around the world and
have been refined through real-world practice and feedback over a period of
many years. Even in this realm, caution and skepticism must be maintained.
One widely used and much-touted method, for example, produced two of the 10
featured systems in the Interface Hall of Shame
Finally, and it is a small matter, I want to point out that usage-centered
design is not a "brand name product." Believing in calling a thing what it
is, we refer to the process we have shepherded into shape over nearly a
decade as usage-centered design because it is a design process centered on
usage, as distinguished from one centered on users or anything else, for
that matter. Perhaps we should have paid more attention to "branding" and
issues of "brand recognition" and devised a clever acronym, but we have been
busy designing usable systems and figuring out how to teach others to design
usable systems. In any event, UCD is not a suitable acronym for
usage-centered design, as it has been used in the past for user-centered (or
worse, "user-centric") design, which is not a process or method so much as a
philosophy and is not centered on use but on the people using a system. In
our experience, this small shift in emphasis--from the people to what they
are doing, from profiling users to understanding and modeling tasks--leads
to marked gains in the usability of the end product.
All that said, I hope everyone involved in TUPIS2000/UML2000 has an
interesting and successful conference. I am sorry I cannot be there to join
in and contribute, but I eagerly look forward to the results.
Director of Research & Development | Constantine & Lockwood, Ltd.
58 Kathleen Circle | Rowley, MA 01969
t: +1 978 948 5012 | f: +1 978 948 5036 | www.foruse.com