Re: [Nautilus-list] Nautilus user testing at MIT

["Blad, John Erling" <john erling blad aftenposten no>]

>> Here's the scenario.  We ask the testers to find a document, open it,
>> make
>> a change, and save it.
>> 
>> They find the document.  They open it, but they open it in the Nautilus
>> window and then make the change and then try to save, but since they
>> aren't
>> in an edit window, of course, they can't.
>
> Part of the problem here was that the old text viewer let you change the
> text but had no way to save changes. This was an obvious flaw in the text
> viewer. The new text viewer does not have this flaw, so this problem
should
> be greatly reduced.
>
> Of course, some people will still expect to be able to make changes in the
> text viewers, and be surprised/frustrated when they can't. But at least
the
> UI won't mislead them into thinking that they can.

So they can't change the text, but why do they try to do that? Is there
anything
that can be done to stop that *before* they try? Learning by doing something
wrong and then to have to redo the operation is not very clever.. and which
one
do the user blame? The application? Him/her -self? The programmer?

>>Eventually the dither around
>> and
>> find an editor.  Today, they found it in the left panel of the Nautilus
>> window (hadn't knoticed that before), but it isn't obvious.  The
>> Windows/Mac intuition is that if you double click on a text document, it
>> opens in an editor.
>
> I can't help but gripe about this misuse of the word "intuition". It's
true
> that Nautilus does not behave exactly like Windows or Mac in that by
default
> you view documents in Nautilus and have to take a different step to open
> them in external applications. But this has nothing to do with
"intuition",
> unless you cheapen "intuition" to mean simply "expectation based on using
> somewhat similar programs in the past". If you follow that route, then you
> start calling everything new "non-intuitive" because it's not exactly like
> its predecessors.

At least this is two different problems, one is how you make something
intuitive
for someone that is used to an other environment, the next is how you make
something intuitive in this environment. More generally I gess most users
make
some kind of schemata for "computers" and more spesific ones for Macs,
Windoze,
KDE and Gnome. As so many people migrates from Windoze you can't simply
dismiss
the users knowledge from this particular environment.

>> The other thing, and this isn't really nautilus, I think,but Gnome is
>> that
>> when you try to launch something there is no indication that it is
>> launching so we see people launching two, three, 16 instances of an
>> application.  They are looking for the hour glass or something.
>
> I agree that this is one of GTK/Gnome/Linux/Unix's big problems with poor
> feedback. I hope we can find some way to address it in the future.

Problem no A1 in user interfaces on computers is the ability to interact
with
the system without getting any other response than the respons from the user
interface. For example if you presses a button then nothing usually tells
you
that the action is done. Some GUI's (nearly all) even fake a response
without
reflecting the systems real state.

John