Re: GNOME user survey 2011 (v4)

[Alan Cox <alan lxorguk ukuu org uk>]

> Gathering feedback does not necessarily require an online user survey.
> As stated, for a project which currently targets, among others, users
> who do not care what parts of their operating system can be labelled
> "GNOME" a survey is not a very reliable way of gathering feedback.
> 
> Have you ever tried to explain, to a person who doesn't have an
> interest in software, what GNOME actually is?

Yes - my MBA research was into Linux desktops some years ago and did
involve looking at end users attitudes. The quick summary from then would
be:

Most users used the desktop they got by default (whether because they
didn't know to to switch or were never annoyed enough to bother I didn't
have time to find out)

The managers wanted a system that was a free exact clone of windows
look/feel because change was expensive (training, lost time etc)

The technies in the organisation often inflicted their personal
desktop preference on the entire company.

If I wanted to look at the "Gnome 3 is crap" assertion I think I would
tackle it a bit differently as so much online updating is going on
nowdays.

Collect statistics from a few Fedora and other mirror sites, correlate
downloads together by IP/time and other evidence, and look at how many of
them download which desktops or combination of desktops. Repeat this over
time and plot graphs. Distro popularity shifts may also provide some
evidence for this.

The trouble is while that will tell you about movement and popularity it
will not tell you why. So it's a way to evaluate the claim "Gnome 3 is
crap loads of people are changing or holding back on updating
their desktop" but it's not going to answer useful things. There is a bit
of value in knowing if lots of people hate or love Gnome 3, but the real
value is knowing how it could be better for users, and counting downloads
won't do that.

And if real non-technical end users are like the ones I dealt with then
asking them probably won't help either. Particularly in the business
world to many of them at the time Gnone was 'click on this splodge in the
morning to write letters' 'click on that thing in the corner to turn it
off'. They are not decision makers either - impress their boss 8)

The more interested and technically motivated people on the other hand
can tell you stuff, "power users" particularly. They tell you stuff that
reflects a particular use and understanding case though. Similarly you
can learn an enormous amount by seeing what people are struggling with
and what they do to the desktop - eg the various 'how to fix Gnome 3'
pages tell you a lot about what people wanted and which is non-obvious
for configuration. They are also from people who liked it enough to
persevere so made an effort.

> I urge you to consider the fact that if the majority of people
> subscribed to desktop-devel-list don't have faith in idea of an online
> user survey, an online user survey is probably not going to much have
> effect on the views of the people who contribute to the discussions on
> desktop-devel-list, and since most of the GNOME community read
> desktop-devel-list you can probably extend this to all of the other
> GNOME mailing lists and IRC channels as well.

Some days I think Miguel got the Ximian monkey dead right, except
that there should have been three of them.

Alan