[thelist] User perception

[Jason Handby]

> >   http://www.cogs.susx.ac.uk/users/blayw/flyingl.html
> > 
> > "Lesson 1 - Blaming the user is always easy and always worthless."
> 
> The difference here being that "pilots are very strongly 
> motivated not to make mistakes", but Users are not. They are 
> often lazy and/or don't pay attention and/or do really dumb 
> things - one would hope pilots would not have such lax 
> attitudes towards flying. Unfortunately, we can't program our 
> web applications to account for every ridiculous scenario 
> that might present itself, and if we tried to do this, we'd 
> most likely end up making our applications *less* usable, not more.

I'm a programmer. Users frequently do things that strike me as dumb,
inattentive, lazy... And maybe I'm right; but the point is that if I put a
problem down to "a user being dumb" I'm not then able to make changes in
order to avoid the problem recurring. The problem with the blame model is
that it makes it harder to fix things. Users will always be dumb and lazy
and inattentive. I can't change users. I can change my website though.

One of the examples that Blay gives is of the old 3-pointer altimeter
design. Technically it conveys the altitude information accurately. But
pilots found it hard to read it accurately at a glance. Yes, perhaps they
should have read it more slowly and carefully, but the fact was that they
didn't. And if the investigators had just blamed "pilot error" every time
there was a crash then the situation would not have improved. Improving the
altimeter design meant that the pilots were able to read it accurately by
glancing at it, and there were fewer accidents.

I can't change how clever my users are, or the length of their attention
spans. But I can design my interfaces to take account of those parameters,
and in doing so I can solve problems that on the face of it are caused by
"user error".




Jason