Re: Estimating number of existing Debian users
Hi,
> > This is a wild extrapolation, so the number is pretty unreliable,
> > especially the usage of popcon is too small to be significant;
>
> A worse problem shows up if you look at the popcon graph for
> apt-listbugs and compare it to the general popcon graph. While the number
> of popcon users has increased by about 1/3 since etch released, there is
> negligible change in the number of apt-listbugs installations reported in
> popcon.
This sounds a bit wrong, yes. And especially considering merkel.d.o
was down for a while after etch was released (users would have
uninstalled apt-listbugs to work around the problem).
> The 30000 apt-listbugs users you're seeing in your logs could all be
> users who installed it before etch was released. Probably not entirely
> the case, but I see little in the numbers to say otherwise, unless you
> also have some numbers for how many were using it before etch release?
apt-listbugs in sarge used osdn.debian.or.jp. Very old values from
2004[1] is available, where it counted 4000 unique
IPs. (Hmm... apt-listbugs installation increased by 800% in the last 3
years, which sounds pretty remarkable)
[1] Masato Taruishi presented about apt-listbugs in Debian Miniconf in
China, However, I don't seem to be able to find a reference.
Japanese-language version is avaulable from
http://www.netfort.gr.jp/~dancer/column/2005-debianmeeting.html
http://www.netfort.gr.jp/~dancer/column/data/debianmeetingresume200510.pdf
regards,
junichi
--
dancer@{debian.org,netfort.gr.jp} Debian Project
Reply to: