Re: GNOME user survey 2011 (v2)

[Federico Mena Quintero <federico ximian com>]

On Mon, 2011-08-01 at 18:35 +0300, Felipe Contreras wrote:

> What do you think?

Keep in mind that Gnome 3 just hasn't been around for very long.  Right
now Gnome 3 is most likely only being used by technical people, Linux
enthusiasts, etc. - it has not trickled down to end users yet.  We may
have found the top, obvious problems in Gnome 3, but not the long tail
of bugs.

I think the survey would be much more useful with free-form answers,
instead of fixed options.  That way you may be able to eliminate one
level of indirection ("oh, most people said 'somewhat', now let's make
another survey to find out why").  While fixed options let you do
amazing web hackery, free-form answers lead to more insightful results.

Back in 2006 I did a survey of the Gnome deployments (i.e. somewhere
around the middle of the 2.x series).  These questions were posted to
various places:
https://live.gnome.org/GnomeMarketing/QuestionsForDeployments

And this is the final report with the results:
http://people.gnome.org/~federico/docs/gnome-deployments-2006/index.html

(At the end of the report there are links to the raw results that people
sent in.  Download them and read them; they are short and quite
interesting.  Where the links say, "primates.ximian.com", please replace
that with "people.gnome.org" - I can't re-generate the HTML report with
that change just now, unfortunately.)

Let me tell you about some things I learned from that survey:

* Free-form answers work really well.  It *will* take you some time to
read them and see how to categorize or weigh them, but it's worth the
effort.

* Being able to publish the raw results really helps; this way other
people can help you extract statistics.  Do ask permission from the
respondents to post their replies so that other people can study them.

* That survey probably gave too much importance to the system
administrators themselves - not surprisingly, "better admin tools" was
the most requested thing.  However, the survey *did* get us good insight
into end-user's problems.

* Those pie charts in the report make no sense.  Apologies for
chartjunk.

It may be very valuable to ask the heads of deployments what they think
of Gnome 3 so far, even if they haven't evaluated it yet on their users
- these people have a very good sense of what will work well for people
in the "real world".

By the way, the main page for the deployments is this (I don't know how
up-to-date it is):
https://live.gnome.org/GnomeMarketing/GnomeDeployments

About two years after that survey got published, some people had the
idea of making it periodic - unfortunately I lost track of them and of
that effort.  You may want to look around for them; they'll have
interesting thoughts, I'm sure.

  Federico