[GnomeMeeting-devel-list] Knowing most users (was Re: [Fwd: Re: Quick question])

[Christian Rose <menthos menthos com>]

lör 2002-12-21 klockan 18.07 skrev Matthias Marks:
> > > I know many ppl that like the current design (including me), so why
> > > change things most of the users like ?
> >
> > You actually know the majority of GnomeMeeting users? ;-)
> 
> I've been around on #gnomemeeting for about one year, i think i know what i'm 
> talking about...

My remark was an ironic one. The ironic point was about claiming to know
the majority of GnomeMeeting users, or if you want,
$any_free_software_app's users.

Consider this: Out of all pieces of software on your machine, how many
irc channels or other forums belonging to those pieces of software do
you regularily visit? For those that you don't, is it because:
1) you just never use that software, or
2) because you don't care enough about that piece of software to spend
time using its forums, or haven't bothered to check out what forums it
has?

I don't know about you, but I know that I don't subscribe to the mailing
lists or visit the IRC channels of the majority of software I use on my
machine. Still, I consider myself to be fairly actively engaged in the
free software arena, spending large parts of my time (besides
translating and handling translations) reading many mailing lists and
using irc, so I would assume that many people that aren't that active
are using mailing lists and irc even less.

In order to claim that one knows most users, one has to know the total
number of users, or pick a representative sample. To know the total
number of users is very difficult, and in the case of free software
extremely difficult. One can get an idea from the number of downloads
from the web site, but even that doesn't take into account the number of
users that get it from other sources, like for example from their
distribution. Since many people don't bother to upgrade many pieces of
software from the versions that are provided with their distribution,
one could expect that the total number of users is significantly larger
than the number of downloads from the GnomeMeeting web site.

The last statistics that I could find that Damien sent to the list is
from one year ago, so one could assume that the figures have grown from
then, but here they are:

3228 downloads in 8 days (which translates to 12105 downloads a month).
For simplicity's sake, we can assume that GnomeMeeting releases at least
one new version every month (I don't have any statistics on that, but I
assume it's not very far from that, the last couple of months seem to
match that).
To know what number of new users this represents, one has to assume a
ratio of users that are new "downloaders". Let's assume that 40% of all
downloaders have already downloaded GnomeMeeting at one point in the
past. That would give 7263 new users just from the web site each month.
If we consider that these represent only a part of the total number of
GnomeMeeting users, we easily have tens of thousands of GnomeMeeting
users.

Anyway, the exact figures aren't that important. My point is only that
it's silly to claim to know tens of thousands of people, and it's also
silly to claim that the extremely tiny fraction of that that visit
#gnomemeeting (and thus liked the application that much or care about it
that much that they bothered to do so) would be a representative group.

Don't you agree?


Christian