Re: [Usability] user levels, etc.



On Mon, 2001-11-12 at 17:53, Maciej Stachowiak wrote:
> On 12Nov2001 05:29PM (-0500), Luis Villa wrote:
> > On Mon, 2001-11-12 at 17:04, Havoc Pennington wrote:
> > > > And if almost nobody will ever use a preference, then why have it at
> > > > all? 
> > > 
> > > Basically because the "almost nobody" is really loud and complains a
> > > lot. ;-)
> > 
> > <rant>
> > The other reason is that because developers who decide (to quote maciej)
> > that 'people really shouldn't need' some feature are displaying the
> > worst kind of inflexibility and (to be blunt) arrogance.
> 
> Having too many preferences in the UI has a usability cost, and this
> is borne out by user testing. It becomes more difficult for most users
> to find the settings they really care about. 

1) That's absolutely true.
2) That's not an excuse to remove the settings. That's an excuse to find
a good way to hide them. 

> Ignoring even the UI issues, extra preferences, even hidden ones, have
> a cost in code complexity and make testing more difficult, since the
> number of possible configurations grows exponentially.

That's /sometimes/ a very good argument, and a necessary one [I'm told,
without knowing lisp, that sawfish is a great example of this]. However,
often times it's just another poor excuse. I refuse to believe that
using $DESKTOP_DIR instead of a hardcoded ~/Desktop (for example) adds
additional overhead and complexity.
 
> I think GNOME should focus primarily on making the 90%-of-users feature
> set work as well as possible, even if that means removing some of the
> .1%-of-users features entirely.

I think that the next generation of computer users will fit more and
more into that .1 and less and less into the 90%. My mother is terrified
to change anything on her desktop, but my little sister has thought
themes were the coolest thing on earth since she first found out about
them at the age of 7. In other words, old people are scared of choice
and young people who have grown up with computers embrace choice. We
should be planning for that next generation of users, not for the
current one.
 
> Your rant above is an example of the classical Unix/free-software
> attitude that tends to end up with that result (the idea that lots of
> preferences == easy to use).

I do not claim, anywhere, that more preferences == easy to use. However,
they can bear no additional usability cost if well structured and
well-hidden from the beginning user who is unprepared to deal with the
complexity.

Furthermore, I /do/ claim that more preferences (in the long term) ==
better to use, and further, that those that claim that they know best
about which set of simple preferences are best are often at best wrong
and at worst extremely arrogant in their choices.

So... the idea of an advanced-pref-editor-thingy is a perfectly good
compromise. One person or any small team dictating that 'I know what is
best for all of you' is /not/ a good compromise, especially when the
rule of thumb is to err on the side of simplicity and not on the side of
choice.

Luis




[Date Prev][Date Next]   [Thread Prev][Thread Next]   [Thread Index] [Date Index] [Author Index]