[prev in list] [next in list] [prev in thread] [next in thread] 

List:       kde-look
Subject:    RE: Standard questions and answers
From:       "Glen Parker" <glenebob () nwlink ! com>
Date:       1999-09-15 19:28:07
[Download RAW message or body]

> > Your examples do not reflect the real world dialogs.
>
> Real world dialogs should change. That's my point. Real world dialogs
> should ask simple, unambiguous questions, and give simple answers for
> the user to choose from. The user should know exactly what the question
> is and the exact meaning of the answer they are giving. You can't get
> simpler than Yes or No. OK is not too bad, Cancel can be ambiguous.
>
> > Ok/Cancel is excellent for confirmation.
>
> OK is excellent for confirmation, when there's only one button:
>
>   -----------------------
>   Something has happened!
>
>             OK
>   -----------------------

Why does 'OK' make sense here?  Are you asking a question or an opinion?
The user has no recourse at this point, so the button should say 'Whatever',
or 'Go Away', or maybe 'Close' would work ;-)

> Cancel means nothing in a confirmation dialog, and doesn't mean much at
> other times. When was the last time someone asked you a question (in
> English :-)) and you replied "cancel". Just because it's a computer
> doesn't mean we have to use computer-speak. Quite the opposite in fact.

The last time somebody asked me if I was sure I wanted to do something, I
sayed 'no', and then did it anyway ;-)  The question is ambiguous, and so is
the answer.  The point is, you don't care if the user is sure or not, you
care whether they are going to do it anyway.

> I'm not saying OK/Cancel don't have a place. Sometimes they are the best
> option, but Yes/No is much easier to understand and therefore normally
> much better.
>
> > Yes/No is only needed if there is a true choice between two identical
> > alternatives it is like.
> >
> > if ok do
> > else nothing
> >
> > and
> >
> > if y do
> > elseif do
> >
> > There is a semantic difference.

My point exactly.  Much more consicely than I could do though ;-)

> We're talking about human computer interaction. This is about usability.
> This is about people who may or may not be happy working with computers,
> getting things done. The above shows that you are still thinking like a
> programmer.

Bah!  Computers have no real-world counterpart, they are a *new* real-world
concept, and they must be learned.  Sure, we should make them as easy as
possible to use, but we shouldn't force-fit them into some role they aren't
equipped to fill.  What I mean is, the concepts should be consistent and
simple, and users should then be educated about the new concepts.

My favorite example is of how (was it MS, or Apple?) directories were
renamed to folders, because people already know what a folder is.  I call
bullshit!  I have never gone to a filing cabinet and found an infinite
hierarchy of files within folders within folders.  The concept of
directories is *new*, and not that hard to understand.  By calling them
folders, the proper learning process is short-circuited.  Now people have to
learn new meanings for old words, which is more confusing than just learning
a new word with the new concept (damnit ;-).

Glen

[prev in list] [next in list] [prev in thread] [next in thread] 

Configure | About | News | Add a list | Sponsored by KoreLogic