Error Messages Are Evil

Richmond richmondmathewson at
Sun May 11 15:43:44 EDT 2014

On 11/05/14 21:48, Alejandro Tejada wrote:
> Recent article published by Don Norman.
> "Error messages punish people for not behaving like machines.
> It is time we let people behave like people. When a problem
> arises, we should call it machine error, not human error:
> the machine was designed wrong, demanding that we conform
> to its peculiar requirements. It is time to design and build
> machines that conform to our requirements.

Indeed: but how?

Mind you, if Donald Norman (who has been banging on about Usability 
theory and 'affordance'
for years) wants to write about machine errors, he should at least 
correct his human error and
get his English grammar sorted out:

"the machine was designed wrong"

is a simple grammatical error any person who wants to be taken 
seriously, and has any academic
pretensions, should not make.

"the machine was designed wrongly"

Obviously Donald Norman doesn't know that verbs are modified by adverbs, 
not adjectives:
that is HUMAN ERROR.


"It is time to design and build machines that conform to our requirements"

Well, oddly enough, all machines that I know of are designed by humans, 
and are very rarely,
if ever, designed to annoy the people who use them, but in conformance 
to their requirements.


Donald Norman started his career years ago by making some blindingly 
obvious remarks about
door handles being put on the wrong way round, or on the wrong sides of 
door . . . and he did
have a point; now he, as a "one trick pony" has extended that into areas 
which do not connect
with door handles.


What Norman might have done is criticise GUI, and in very many cases the 
criticism would be valid.

What Norman conveniently overlooks is that millions of people use 
computers with
"badly designed" interfaces, "badly designed" keyboards (he had a right 
royal rant about the QWERTY
keyboard) and don't seem to feel an urge to get up from their collective 
bottom and radically
redesign everything.

The same could be said for the efforts of the late Jeff Raskin.


Error messages are a necessity, not because computer systems are 
designed badly, but because
humans and computers are completely different things that work in 
completely different ways.

If babies had error messages parenting would be 1000 times easier.

All an error message is is a computer's way of telling us it doesn't 
understand; because a computer
is, frankly, a very stupid mathematical calculator, and we humans are 
not. If a computer did not
throw up error messages we would never know when we were failing to get 
a machine to do what we wanted it to do: that would make life far more 
difficult than any error message.
> Stop confronting us: Collaborate with us."

Computers never "confront" us; they are not capable of that. All a 
computer does is tell you it does
not understand what you have told it to do.

Accusing a computer of "confronting us" is a socking great 
anthropomorphism which only serves to show that Norman has very little 
understanding of what a computer is and what it can do.

The fact is that a computer can ONLY do what we tell it to; and it ONLY 
"understands" a load of electronic pulses. Clever people have made our 
lives easier by designing graphical representations
of what goes on inside a computer and nicer ways of getting a computer 
to do what you want it
to. Some people are not quite as clever as other people, and they have 
designed less effective
ways of getting a computer to do something.


"Error messages punish people"

"punish" ; utter rubbish.

Error messages are more important than Norman realises.

Before he makes any further pronouncements of this sort Donald Norman 
needs to do the
following to things:

1. Go on holiday to a country where he doesn't speak the language and 
nobody there speaks his.

2. Get time allotted to himself on a VAX machine (if there are any left) 
and learn a spot of
Assembler language, and then try and type an e-mail message to his best 
friend using only
Assembler language on the VAX.


It's amazing how purified I feel after a rant of that sort.

But, having had to read about 3 of Norman's book and attend interminable 
lectures on
Usability theory at the "University" of Abertay I feel very strongly 
indeed about what he says, and
have given it some considerable thought.


More information about the Use-livecode mailing list