for the same reason you're talking to me here on HN right now. Because a personal community of a 1000 people you vibe with is better than an entropy machine of 100 million. Can you even call something a social network if it makes the people on it anti-social?
Summary: In most online communities, 90% of users are lurkers who never contribute, 9% of users contribute a little, and 1% of users account for almost all the action.
> > The number of accounts that have posted to HN this year, divided by the number of IP addresses that have accessed HN, is 0.008. How close that is to the '1% rule' ratio depends on which is the bigger factor: users with more than one IP or IPs with more than one user. We don't know. If the former is bigger, then 0.008 is a lower bound.
> > Here's another way. The number of accounts that have posted this year, divided by the number of accounts that have viewed HN while logged in, is 0.36. That doesn't tell us much, but we can estimate the ratio of logged-in users to total users this way: logged-in page views divided by total page views. That ratio is 0.23. We can multiply those two to estimate the ratio of posters to total:
> > So the two ways of estimating produce 0.8% and 8% respectively. Both ways are bogus in that they assume things we don't know and mix units that aren't the same, but they're the two I came up with and I don't remember how I did it before. It's interesting that they're almost exactly an order of magnitude apart. That makes it tempting to say the number is probably in between, but that's another cognitive bias talking.
