[Sigia-l] UML - still not fit for the UCD?

[Chris Chandler]

Stewart, et al

I'm working in a large organization with several different engineering teams all using some form of use cases to specify
and document functionality.

This summer we struggled mightily with the question of the relationship between our wireframe/interaction documentation
and their use casing. The wireframing/paper-prototyping began first, but then one engineering team insisted on doing a
set of "business use cases" which then were used to write "technical use cases." I documented, cajoled, argued, insisted
on the functionality I felt was crucial to the redesign of our eccomerce engine, and then I sat through endless hours of
review meetings (often by conference phone, as I am in Los Angeles and that team was in Orlando) to make sure they
captured most of the same information in their use cases.

The Los Angeles group, including IA and technical teams have decided never to do this again, and we are planning two
different approaches for the next set of projects. One project will be building an interactive prototype as the main IA
deliverable/specification. I'm excited by that prospect, but alas, I'm in the other group, which has to continue working
with the team in Orlando.

Sorry for the long story, I guess I wanted to say that your question has been on my mind for a couple of weeks now.

Luckily, the problem of use cases not fitting the needs of UI designers has been documented and addressed very well in
at least one place. If you, like I, are going to be forced to work with Use Cases as a methodology, I strongly encourage
you to look at the work of Larry Constantine and Lucy Lockwood, who together wrote a book called Software for Use. You
can find many interesting articles on their method, which they call "essential use case" or "task case" - modeling on
their website www.foruse.com

In particular, you might look at this paper:

Usage-Centered Engineering for Web Applications
http://www.foruse.com/articles/webapplications.htm

(as an aside, they are quite careful in distinguishing their work from "user" centered design, of which they have many
criticisms)

You can read reviews of their quite excellent book here:
http://www.amazon.com/exec/obidos/tg/detail/-/0201924781/

Good luck.

-cc






----- Original Message ----- 
From: "Stewart Dean" <stewart at webslave.dircon.co.uk>
To: <sigia-l at asis.org>
Sent: Wednesday, November 12, 2003 6:53 PM
Subject: [Sigia-l] UML - still not fit for the UCD?


>
>
> I'm after peoples experiences of UML as I'm wondering if I've given it too
> hard a time in the past. I personally have had classes on UML and have used
> it to communicate ideas to technical teams but is it true that some folks
> use use cases as a main deliverable within user centered design?
>
> I have found that use case often take the focus off what the user wants to
> do and instead focuses upon how technically things are going to be done
> with servers being given equal status to some degree to users, i.e.
> actors.  Defining technical solutions before defining the user needs I
> consider working against user centered design. Can UML really be use to
> define user interaction that isn't fraught with technical domination?
>
> My caution was echoed by this article in Usability news
>
> "The Usability and UML seminar in Scotland this month concluded that basic
> UML (User Modelling Language) is seriously restricted, and restricting, in
> modelling complex, collaborative human activities involving computer-based
> systems...."
>
> full story.
> http://www.usabilitynews.com/news/article272.asp
>
> I'm interested in peoples war stories related to UML.  How have people
> adapted UML or over come technical encroachment of people attempting to
> empose UML?
>
> Regards
>
> Stewart Dean
>
>
>
>
> ------------
> When replying, please *trim your post* as much as possible.
> *Plain text, please; NO Attachments
>
> Searchable list archive:   http://www.info-arch.org/lists/sigia-l/
> ________________________________________
> Sigia-l mailing list -- post to: Sigia-l at asis.org
> Changes to subscription: http://mail.asis.org/mailman/listinfo/sigia-l