Re: Creating Personas to Enhance GNOME's UX

[Allan Day <allanpday gmail com>]

Hey Allan,

> Defining personas is very important for User-Centered Design. In fact,
> it should be one of the first activities needed to build a system with
> good UX, based on JJG's Elements of User Experience. We need to know
> who we're designing for before we design, right?
> 
> 
> There's an ongoing effort for this, but it's not gaining much
> traction. Fortunately, GNOME.Asia is interested in helping out by
> doing research in Asia. I hope you can help us do research in other
> areas. We also found this GNOME Marketing User Survey effort. We can
> build on this existing work if it's fit. 
> 
> 
> Let's start by surveying GNOME users about their demographics,
> computing habits, what they use GNOME for, and how much they use
> GNOME. Would that do? If so, let's draft the questionnaire :)

Thanks for taking this up! I'm sure that a set of personas would be
beneficial. Also, I like the idea of using research to inform the
creation of personas. Purely fictional accounts lack the complexity and
richness of those which incorporate research results. They are also far
less convincing.

I see that this message has already had some responses. I'm afraid that
I don't have time to go through all that right now, but I do want to
give a few initial reactions.

My view is that the usability project should focus on personas as a way
to inform GNOME UX design. (They could maybe even become a part of the
HIG.) Our first step should be to decide how to make a set of personas
that are effective in that context. Then we can decide whether we can
collaborate with marketing and what kinds of research strategies will be
most appropriate.

That said, let me jump the gun a little and say that I'm unsure that a
survey is the best research approach here. In my understanding, the
purpose of this research would be to add richness and validity to our
personas. On the validity front, our aim should be to be able to claim
that 'people like this really exist and we want GNOME to be well-suited
to their needs'. This needn't require that our personas are statistical
representative of the GNOME target audience (something which is beyond
our practical abilities anyway).

If we want to add richness to our personas, qualitative interviews or
even ethnography might be more appropriate. They are better suited to
producing statements like 'Mary is a busy mother. She really hates
computers and wishes she could spend her time eating cheese instead of
having to use one.' Interviews are more likely give you nice quotes to
use. Observations give you a better sense of the contexts in which
computers are used.

But anyway, let's decide what work we want these personas to do and go
from there. I've moved the existing persona material to the GNOME wiki
[1]. It would be great to develop that page.

Thanks again,

Allan

[1] http://live.gnome.org/UsabilityProject/Personas/