Interest in usage survey and data mining?
Hi,
recently we discussed anonymous contributions to the DDTP. Potential and
actual
problems were clearly addressed - and there are diverging answers. (Why
don't they
uncover themselves? Is it better not to accept those contributions?)
Grisu provided some
initial statistics - with a very simple analisys. During this
discussion there was also
some contest on the topic "my work is more user-visible than yours"
(debconf/ddtp).
Debian provides a lot of localized software and documentation
originating from different sources.
There is pure Debian work (debconf, DDTP, Debian specific stuff such as
debhelper)
and surely more from other sources (GNOME, KDE, FSF, the translation
project). In spite
of the possibility that Jon Doe might see Debian responsible for every
third party work
compiled and distributed by Debian: I would be interested in the
reception of our work!
Would anybody dare to ask on her debian-xx-user list? The answers might
be disappointing.
However, the group interviewed and those who actually answer might not
be representative.
I'm also interested in your personal impressions.
Would anybody be interested in deeper insight in the contributor's
structure and
time series? Just out of curiosity? Perhaps there are differences among
the teams?
First results/hypothesis on the team lists which are actually (access)
statistics:
- There are people that work on the three descriptions they are
interested in - with and
without an account.
- There may be less contributors than entries in the team list.
Why? I use DHCP and get different IP addresses. When I'm not logged
in, I get varying
addresses. As I'm logged in most of the time (and do most of the
interactive work with
the same browser), but not always, I got multiple entries in the
statistics (at least two).
To be more precise, we should present results in the form "There
were $n contributions
from ISP $bar in $year". (There are at least three DDs in the area
covered by my ISP -
and me.)
- Some people just have a glance at the translation page and surf on.
- For some reason people don't even manage to create an account. (From
the German
"team" [in the sense of "everybody that hit the web site since
logging started"], e.g.
giammi and Giammi. Both entries lack e-mail address and real name.)
Did they surf away or contribute some anonymous work?
- Some long-standing contributors are "semi-anonymous'".
They do not log in, but
they do permanent work for several years, work that is very much
aprechiated
by the regular "non-anyms" (mainly reviews and corrections). And they
can be
identified, even if they don't sign their comments ... Some of them
even have
an account.
- Some former contributors have chosen to work on different subjects.
BTW, there
are people that said they don't like to work with a system that is as
unstable
as the DDTSS was a year ago. There has been a significant decrease in
donwtimes ;-)
Cheers,
Martin
Reply to: