USENET costs--who's paying for what?
Brian Reid
I've been taking readership data for USENET lately. This data consists of
summaries from 100 sites about the number of readers of various newsgroups.
I then collect it all back at glacier, add the traffic data, and produce
this result. I'm not proposing to do anything with this information, but other
people might.
TOP 40 NEWSGROUPS IN ORDER BY PER-READER COST
+-- estimated total number of people who read this group, worldwide
| +-- approximate $US per reader per month spent sending
| | this group, worldwide.
| | +-- (rating) percentage of the total user population
| | | who read this group
| | | +-- (share) percentage of newsreaders who read
| | | | this group.
V V V V
1403 $33.84 0.66 2.44 net.religion.christian
1348 24.15 0.63 2.34 net.philosophy
2505 21.65 1.17 4.35 net.politics
1789 18.51 0.84 3.11 net.religion
1211 16.98 0.57 2.10 net.abortion
3716 13.10 1.74 6.46 net.space
4982 11.91 2.33 8.66 net.singles
2807 11.51 1.32 4.88 mod.techreports
5326 10.94 2.50 9.25 net.women
4308 10.30 2.02 7.48 net.music
4087 9.55 1.91 7.10 net.micro.amiga
4569 9.51 2.14 7.94 net.sources.mac
12649 9.31 5.92 21.98 net.sources
6799 8.83 3.18 11.81 net.sources.games
1624 8.62 0.76 2.82 net.comics
1665 8.36 0.78 2.89 net.sport.hoops
2918 7.41 1.37 5.07 net.cse
10171 7.33 4.76 17.67 mod.sources
2505 7.15 1.17 4.35 net.startrek
3317 6.72 1.55 5.76 net.sf-lovers
4005 6.46 1.88 6.96 net.auto
867 6.36 0.41 1.51 net.sport.hockey
2518 5.33 1.18 4.38 net.kids
2463 5.28 1.15 4.28 mod.computers.vax
1486 5.28 0.70 2.58 net.ham-radio
2188 5.24 1.03 3.80 net.motss
1500 5.11 0.70 2.61 net.tv.drwho
1541 4.93 0.72 2.68 net.games.pbm
3124 4.87 1.46 5.43 net.micro.apple
5767 4.86 2.70 10.02 net.micro.pc
4858 4.74 2.28 8.44 net.movies
1803 4.71 0.84 3.13 net.nlang.india
1403 4.63 0.66 2.44 net.cycle
6083 4.62 2.85 10.57 net.micro.mac
2766 4.49 1.30 4.81 net.music.classical
6331 4.19 2.97 11.00 net.arch
2408 4.08 1.13 4.18 net.nlang
2064 4.05 0.97 3.59 net.aviation
4143 4.03 1.94 7.20 net.news.group
5574 3.95 2.61 9.68 net.audio
--
Brian Reid decwrl!glacier!reid
Stanford re...@SU-Glacier.ARPA
Adrian Kent
>
>I've been taking readership data for USENET lately. This data consists of
>summaries from 100 sites about the number of readers of various newsgroups.
>I then collect it all back at glacier, add the traffic data, and produce
>this result. I'm not proposing to do anything with this information, but other
>people might.
>TOP 40 NEWSGROUPS IN ORDER BY PER-READER COST
> Brian Reid decwrl!glacier!reid
> Stanford re...@SU-Glacier.ARPA
point out two things:
(1) the figures quoted in this list are spuriously accurate (estimated world
audience to the nearest person, cost per person to the nearest cent.) In fact,
there's no indication how either the readership or the cost were estimated,
and so no reason to place any faith at all in the numbers.
(2) newsgroups have, in the past, been abolished for completely spurious
reasons: history has been rewritten, 'votes' have been taken on the basis of
one pro-abolitionist's claim about the contents of his private mail - end of
group.
adrian
Brian Reid
> (various blah blah woof woof about my data)
The discussion of the methodology, its accuracy, and its meaning is taking
place in net.news.group (where it belongs).
--