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

List:       mozilla-ui
Subject:    Re: User testing the Mozilla UI
From:       randolph () cyber-dyne ! com (Randolph Fritz)
Date:       2000-01-26 18:29:33
[Download RAW message or body]

In article <388EB382.8BCC552@weirdness.com>, Jerry Baker wrote:
>
>OK. I ordered Tog's book from Barnes and Noble, but it doesn't ship for
>3 - 5 weeks. I am anxious to get to work on this.
>

The book, alas, is out of print and not easily available; I'm
trying to find out why.  My copy is off in storage, but maybe
I can dig it out in a week or so.  Meantime, I refer you to the
section "Involving Users in the Design Process" of the Macintosh
Human Interface guidelines; you can read it on the web at
<http://developer.apple.com/techpubs/mac/HIGuidelines/HIGuidelines-56.html>.

>
>I am worried about pollution. That is, each time I test someone, they
>will have had the previous experience of Mozilla to draw upon. I will
>run out of Mozilla virgins pretty quick. At least I will be populating
>the mid-level user category as I show Mozilla to the newbies.
>
>I am not sure whether I should force them to do task a, b, and c, or
>just let them do their thing and take notes. Do I tell them, "now add a
>bookamrk", or do I just hope they try if I want data on how well they
>work the bookmarks?
>

I have a few suggestions here, but take them as suggestions only; you'll
have to make the final decisions.  The HIG has some good general ideas
here, too.  In general, it seems to me that your first task is to find
out how people react to Mozilla at two different stages of expertise:
(1) after they've just encountered  it and (2) after they have been
using it for a while and the initial familiarity has worn off.

I'd say, work out some usage scenarios (high-level sequences of tasks).
Since (I believe) the expectation is that users will be able to just
pick up Mozilla and use it, I think it would make a lot of sense to
find out how your users typically use a browser and then have them
try their typical jobs on Mozilla.  If they make it through the inital
learning period, then you can go deeper.  When running tests, I highly
recommend you adapt Apple's "Ten Steps for Conducting a User Observation,"
<http://developer.apple.com/techpubs/mac/HIGuidelines/
 HIGuidelines-73.html#MARKER-9-41>

>I also worry about some of the weirdness of the current widgets and
>other bugs freaking out the new users. Especially things like bugs
>17321, 24497, 16806, 20087, etc. These are things where the current
>behavior is bizarre enough to a new user that they may become
>disoriented IMHO. I guess testing will tell.

Bugzilla's not working at the moment, so I can't look those up.  In
any event, yes, testing will tell.  IME, the problems users will have
are not usually the ones one expects, with the notable exception of
Fitts law (target-hitting) problems.

I hope you do take this on--it is, in my view, very badly needed.

Randolph

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