Index Home About Blog
Newsgroups: comp.risks
X-issue: 5.57
From: decvax!utzoo!henry@ucbvax.Berkeley.EDU
Date: Tue, 10 Nov 87 00:26:42 est
Subject: Re:   A simple application of Murphy's Law (Tape Labels)

> ...When
> you attempt to overwrite a labeled tape on our system an operator message
> appears asking if you really want to write to the tape. The operator must
> have answered yes to this question...
> This is of course an example of the seeing, hearing, reading what you expect
> to see, hear or read rather than what is actually there. There appears to be
> nothing that you can do to prevent this kind of error...

Actually, no, there are things that can be done to prevent this kind of
error.  I don't think you have diagnosed it quite correctly.  I strongly
suspect that the operator saw the question and understood it, but that
he/she sees that question a dozen times a day, and the normal answer has
become a reflex.  *That* behavior is fully predictable and a conscientious
interface designer will avoid such situations.  "Do you really want to do
this?" is a question that should never be asked unless there is truly a
good chance that the answer will be "no".

Note also that the question was directed to the wrong person:  the operator,
who probably doesn't know enough about the work to judge whether the request
is a reasonable one.  Since the system insists on asking him/her questions
that would require considerable investigation to answer intelligently, the
questions will quite predictably be answered unintelligently.  It is not
unreasonable to request manual intervention when major data destruction
is requested, but it *is* unreasonable to place the decision in the hands
of someone who gets paid for throughput, not thought.

				Henry Spencer @ U of Toronto Zoology

Index Home About Blog