[linux-elitists] Re: Yet another mozilla atrocity (UIs, etc.)

Nathaniel Smith njs@pobox.com
Sun Oct 5 11:21:25 PDT 2003


On Sun, Oct 05, 2003 at 12:46:29PM -0400, Bill Bogstad wrote:
> I recently took a course on human factors issues and computer interfaces
> and my understanding is that the HCI people have scientifically shown
> that gui's are inherently better than command lines in most ways and the
> less you ask a user to configure the better.  I had a very negative
> reaction to this to the extent that I wish I had the time to read the
> original studies and examine their methodology.  Still, the 'best'
> research seems to say that I (and it seems some of the people on this
> list) are mutants and the interface style that we like is not right for
> most people.

You may be willing to give the researchers the benefit of the doubt,
but I'm going to simply not believe this without reading the papers in
question.  I think I've never seen a paper that actually measured
"inherent goodness" of _anything_... and many, many papers that
claimed to measure inherent goodness, but actually measured something
entirely different.  Like "ability of people with no familiarity with
the interface to accomplish a single task within a fixed amount of
time".

Now, they may have actually gotten past that; like I said, I haven't
read the papers, so I don't want to give the impression that they
_are_ wrong-headed.  But when 90% of all the research I hear described
that vaguely turns out to be confused...

-- Nathaniel

-- 
"The problem...is that sets have a very limited range of
activities -- they can't carry pianos, for example, nor drink
beer."



More information about the linux-elitists mailing list