Diagramming Robust AI
Mary Murphy and Leo Wong
hello at albany.net
Tue Mar 31 21:40:41 EST 1998
Frank Sergeant wrote:
> In article <352045E5.AAA71C20 at albany.net>,
> Leo Wong <hello at albany.net> wrote:
> > You may or may not sympathise with this:
> > "Behind every user-friendly interface is a terrific human contempt."
> > - Ellen Ullman
> Hopefully Ellen gives enough reasons and/or examples so her
> reader could decide whether to sympathise with it or not. As it
> stands, I do not think we have sufficient data to decide. Would
> anyone who has read Ellen care to present the arguments?
This might help:
Pretty graphical interfaces are commonly called
"user-friendly." But they are not really your friends.
Underlying every user-friendly interface is a terrific
The basic idea of a graphical interface is that it does
not allow anything alarming to happen. You can
pound on the mouse button all you want, and the
system will prevent you from doing anything stupid....
To build such a crash-proof system, the designer
must be able to imagine - and disallow - the dumbest
action. He or she cannot simply rely on the user's
intelligence: who knows who will be on the other side
of the program? Besides, the user's intelligence is not...
programmable; it cannot protect the system. No, the
real task is to forget about the intelligent person on the
other side and think of every single stupid thing anyone
might possibly do.
In the designer's mind, gradually, over months and
years, there is created a vision of the user as imbecile.
The imbecile vision is mandatory. No good, crash-
proof system can be built except it be done for an
The prettier the user interface, and the fewer odd
replies the system allows you to make, the dumber you
once appeared in the mind of the designer.
-- Ellen Ullman, "Out of Time: Reflections on the
Programming Life," in Brook and Boal, _Resisting the
Virtual Life_. San Francisco, City Lights, 1995.
hello at albany.net
More information about the Neur-sci