Re: [Usability]GNOME feedback, time for a new usability test?

[Christopher Warner <zanee kernelcode com>]

I think a new usability test needs to take place but I don't think there 
is anything special that needs to be done. You take the majority of 
tasks an average user (secretary, student, developer, etc etc) need to 
do daily and then give them gnome. Each and every group and/or person 
will do it differently. What i'm doing right now is taking ordinary 
people and telling them to use Gnome; they have there own tasks to 
complete (reading email, writing a paper, checking out their favorite 
site etc). In event what I'm doing is taking Gnome and asking them if 
they can't get what they need to do done. If they can great, if they 
can't they tell me why not. I'm putting together a report and will 
release it eventually if anyone wants to check it out.  I'm using a pool 
of 10 users,  each of different age and background with computers. There 
is no time frame for them to complete any task, however if they set 
forth to complete a task; they must complete it. The level at which they 
become fed up or are pleased is determined by gnome and the usability. 
Depending on them being fed up or pleased I move on to the next user.

-Christopher Warner

Roberto Rosselli Del Turco wrote:

> In the recent debate about GNOME 2 interface (button order, undo 
> button, top panel, more config options for advanced users etc.) 
> "users" were mentioned very often, but, AFAIK, there's been no 
> usability test comparable to the one Sun performed in March 2001: is 
> something like this scheduled for the near future? As soon as the 
> major distros will ship GNOME 2(.0.1), and more users will start using 
> it, time will be right to think of a new usability test to gather 
> information about the new GNOME experience.
>
> The SUN report was really helpful (IMHO), hope the SUN people will go 
> for a bis. In case this won't happen (soon), though, we could setup an 
> usability test to be performed by GNOME contributors/users. That would 
> consist of:
>
> * taking SUN study as a model, write down a list of questions ("What 
> do you think this icon is for?") and tasks ("Change fonts", "Add xxx 
> to the panel") to be performed by the user;
>
> * add a "standard" GNOME 2 configuration and have the tester create a 
> new user from scratch using that configuration;
>
> * the tester asks questions and has the user perform the tasks;
>
> * the tester adds to the report relevant information about the user 
> (new user, Windows user, etc.) and comments about his behaviour (icon 
> xxx not really explicative);
>
> * the tester send the report to a central location, where they'll be 
> catalogued and organized.
>
> In this way the usability people could gather lots of information 
> about how GNOME 2 is receveid by old and new/prospective users. WDYT?
>
> Ciao