Building on Robert's point ... if the environment is new enough, you may 
not know what the key behaviors are that your personas should exhibit. So 
in that situation, your design and your personas will co-evolve, each 
influencing the other.

What you know about the new environment may be limited to a conceptual 
model, some scenarios built on it, and perhaps an architectural model that 
explains how the solution will be structured.

The initial testing that you do on your scenarios and conceptual models 
provides inputs for persona development. You can postulate personas for 
this artificial environment as a way of envisioning how various target 
audiences will be served and how they will respond. These personas will be 
spectators at the research events that you hold to validate the scenarios 
and the models. That is, as you evaluate the data, and struggle to explain 
the variation in the responses, you'll start to see persona qualities 
emerging. There's no way to validate these vibes that you're getting about 
the personas until later, because until the environment is defined well 
enough to build some kind of design prototype, users can't respond to it.

This happens in training also. The reason that training is designed around 
objectives and tasks is that if you ask a student what they need to know, 
they usually can't tell you. Only an experienced person who knows the 
environment that needs to be learned can say what needs to be in the 
training. The objectives and tasks serve to outline the expertise of the 
experienced person. Given those, you can evaluate the student's preparation 
against the objectives and tasks, and figure out where their skill gaps are.

In both cases (radically new environments and other learning situations), 
it's necessary to do some modeling before trying to draw conclusions about 
how users would respond, or about how representative users would respond. 
What makes this an obstacle to persona development is that you may not be 
able to determine at first what is significant about the users in order to 
understand how they vary.

User behavior is emergent, and emergent phenomena have to be observed. When 
the situation is new enough, you have to build it and observe it before 
you'll know what the typical behaviors are. You might have a very nice 
repertoire of personas, and yet you might not be able to put the 
differences that matter into those personas until (as late as when) your 
users get a bit of experience with the environment that you're creating.

This is why experience design around behavior is such a great idea. If you 
observe the behavior first, and then design an environment to support it, 
your personas are already live and active, and your typical users can just 
fit right in.

Best wishes,

Bruce Esrig

At 05:33 PM 11/13/2007, Robert Reimann wrote:
>Hi Oliver,
>
>Personas are a powerful and widely applicable design tool, but they do
>have their caveats. ....
>
>Personas are descriptive of behavior patterns as they currently exist.
>They are not, on their own, necessarily predictors of behavior change
>in the wake of disruptive technologies (though they can help designers
>understand this with the right supporting data). Finally, as Adrian
>points out, personas are of limited use without accompanying scenarios
>that explicate context-based activities, tasks, and goals. ....

>Robert.
>
>On Nov 13, 2007 12:19 PM, oliver green <[EMAIL PROTECTED]> wrote:
> > Hi everyone,
> >
> > I am trying to understand the finer nuances of using personas.

________________________________________________________________
*Come to IxDA Interaction08 | Savannah*
February 8-10, 2008 in Savannah, GA, USA
Register today: http://interaction08.ixda.org/

________________________________________________________________
Welcome to the Interaction Design Association (IxDA)!
To post to this list ....... [EMAIL PROTECTED]
Unsubscribe ................ http://www.ixda.org/unsubscribe
List Guidelines ............ http://www.ixda.org/guidelines
List Help .................. http://www.ixda.org/help

Reply via email to