I subscribe to a few Usenet groups and it does seem to be slowing down lately. Although the activity in the groups I subscribe to is decent, most of the groups offered are effectively dead.

I was looking at the numbers on Google groups, which lists Usenet groups amongst others with statistics, and a few interesting this emerged. Note through that since I'm using Google groups for stats, some of the groups listed aren't Usenet groups and would affect the results, but I expect the results should be similar. It's sort of like using the statistics for overall fruit sales to show the sales of apples. The problem is that I couldn't find stats for just the Usenet alone.

1. The most popular group is news.admin.net-abuse.sightings with 5584 subscribers and 24,150 messages in the past month.

2. Google lists over 4 million groups, or (according to my calculations) precisely 4,196,088.

3. Over 4 million, or 4,002,393 groups received no messages in the past month, effectively making them "dead" groups. 209,038 groups have absolutely no members, while 3,595,378 have less than 10 members. Note that a "member" is likely anyone who ever posted to that group, so someone who only posted a single message could be classified as a member.

4. Many groups are mistakes or variations on the same title as other groups, such as bainaries instead of binaries. This accounts for many of the dead groups.

5. From the numbers it appears that 193,695 out of 4,196,088 groups, or about 4.6% of the groups have any activity at all. over 95% of the groups are dead. 15,282 (about 8% of the "active" groups or less than .4% overall) get at least 100 messages per month, or roughly 3 per day, and could be considered active. Those active groups getting less than that are marginal and could go either way. 

6. Of the 13 highest activity groups, many appear to be auto-posters and restricted groups. 3 have only 1 member while 1 has only 4 members. One of them is for the Dow Jones.

7. A surprising number of the most active groups are foreign language groups. However, alt.tv.american-idol is amongst one of the 14 most active with 1,231 members and 14,282 messages per month.

8. About 60,000 (59,930) have been posted to in the last day. About 158,000 have been posted to in the last 10 days.

9. Comparing Google groups for rec.arts.tv: Google groups stated 3,122 subscribers although I counted less than 100 posters for the past day. This discrepancy is partially due to posts not getting through and that fact that I'm comparing one day's activity to one month's, although the same people tend to post so over 3,000 is too big a discrepancy. This is most likely the result of Spammers who are filtered out. As a result, most of these numbers are inaccurate.

Although these are stats for Google groups and not the Usenet specifically, one thing seems clear. The problem is not that the Usenet is dying, as much as it's just not getting rid of it's garbage. The Usenet is alive, but in pieces. It needs to get rid of those groups which are no longer active. 95% of the Usenet is dead, but the rest is alive and active.

Many news-servers do try to clean off dead groups since they only serve to slow the system down. As such, when one tries to get a listing of newsgroups they'll often see only the actual active groups. So a newsserver might list only 10,000 groups. One I use has about 30,000 groups. But not all are active.

Referring to the Usenet is like referring to the Internet. It's a collection of things, not one specific thing. Individual groups may die but the Usenet continues. However, overall Usenet activity is something that can be measured. If the News Admin group is the most popular Usenet group, then we know there's at least 5,500 people using the Usenet. The problem is that there's no clear number as to how many use the Usenet overall. It could be 10,000 or it could be 100,000, if not higher.

But when we look at it from an ISPs point-of-view, even 100,000 may not be all that much. There are millions of people using the Internet, so it's possible that an ISP may be running a server for only a small fraction of it's users. A newsgroups that gets 100 posts a day may be getting them from 50 to 100 different servers.

Based upon the net-admin group, let's say there's 5,000 servers and 100,000 users. That's 20 users per server. Even a popular system like AOL may have only 1,000 Usenet users and 1,000,000 Internet subscribers. So why support a system for only .1% of your users?

For an ISP that may have only 100,000 subscribers, that .1% would work out to only 100 customers. Chances are they aren't going to drop your service, and even if they do it may not make much difference. Out of the 100 they might loose only 2 or 3. And it's more likely that the customer will stay with the service and find another Usenet server.

Supporting a Usenet server might cost the ISP $100/month but if they're only making $20/month per customer they could afford to loose a few. And if you're talking net profits, the ISP might be making only $2 net, so even loosing nearly half those customers would leave the ISP better off. And for and ISP with only 10,000 customers, they're better not having a server than having one for only a few people. Even loosing 5 customers at $20/month would mean that they break even.

The Usenet will not die from lack of popularity alone. While individual groups may rise and fall, the Usenet as a whole continues to adapt. The problem is that those running servers are often shutting them down because it's no longer financially reasonable.

And since the advent of the Usenet, forums have appeared. Those are the greatest competition to the Usenet. New users aren't bothering with the Usenet anymore, they're using forums instead.

I expect there will always be those that offer Usenet service as long as it's financially reasonable, and there will always be those that wish to use the Usenet. There will also be groups such as rec.arts.tv which are likely to never die since they deal with constantly popular topics. The problem is whether these people will have the Usenet access. Those who don't know where to look for free usenet access after loosing their servers are unlikely to subscribe, thus furthering the diminishing activity. And even free servers don't last forever; one often finds themselves switching servers each year or two.

So in summary, the Usenet is not dead, although it's under constant attack. There are those who will continue to use it as long as they can, but it doesn't seem to be drawing in many new users. Groups do die, but the end of a newsgroup doesn't mean the end of the Usenet. Still, with service being shut down and forums drawing away potential new Usenet users, it's hard to say if the Usenet will still be around in ten years.

If the Usenet dies, it will be with a whimper unheard by most.