>> nothing is just designed by trial and error.
>
> The handset of classic telephones like this
> http://i891.photobucket.com/albums/ac112/stur5380/Misc%20pics/IMG_0982.jpg
> spent 14 months in design research, with hundreds of test subjects.
>
> The arrangement of buttons on a touch-tone phone was one of something
> like 15 or 18 possible patterns, with people trying out how fast they
> could dial numbers and how long they could remember what number they
> dialed.
>
> Even trivial things like "count to ten" thus take months of research.

Sure when complex human interaction is involved it's often difficult to 
avoid testing, but even so there is still lots of work going on to try 
and understand and predict human behaviour exactly to avoid having to do 
such tests.

My point was it's cheaper to avoid real physical testing, especially 
when expensive or lengthy tests are needed. If there is a feasible way 
to avoid it (like by spending more time in design or simulation) it's 
usually better.

Post a reply to this message