At the risk of both of us being perceived as sad old gits for having
nothing better to do on Saturday night than swap emails, let me just say
that I have great respect for DataDesk and Anthony's work. Apart from the
odd insult, Anthony seems to generally agree with me that no single method
is best for all circumstances. GUIs are clearly faster to use and less
iritating for informal data examination, but he agrees that you then need
to be able to justify the conclusions.
But my comment about "users who dislike error messages" was less a
criticism of the interfaces than of the users. There *is* a mentality that
objects to error messages but will accept (not question) gibberish
answers.
R. Allan Reese Email: [log in to unmask]
|