>There is a link to some work by the Nielsen-Norman people Actually, it's not directly the NN people I'm glad to find out. Peter