@sozobe,
One thing worth mentioning is that Abuzz failed right in the middle of the dot com bust. They did a lot wrong and very little right, so there's a number of reasons to point to, but the main reason was:
overvaluation.
I could have made abuzz profitable in one day, but nobody could make abuzz profitable to the tune they had been banking on. The NY Times was spending big bucks on abuzz hoping for a big payoff. They could have survived as a marginally profitable site very easily but that wasn't what they had been budgeting for. When the dot com bubble burst, they had to stop bleeding money on pipe dreams and went for cutting losses instead of turning it into an ad-supported profitable niche site.
They had the wrong business model (selling the software as a service to companies and product placement within the community), executed it poorly (other people do the "expert as an advertisement" better than they ever did), and their software was very
very poorly made.
You could use anyone's username, you didn't need to confirm your account in any way (e.g. no need for the email to be real), they didn't filter sockpuppetry (e.g. you could vote as many times as you could make a fake account) and did all that while trying to rely on the user input that they never fully sanitized.
It was really horrendously bloated code, with no security. Hell I could make people say things in their own account just by sending them a specially crafted email. I did it to bobsal once when he insisted it was not possible. I sent him an email with an iframe of a page that launched a bunch of popups to specially crafted abuzz urls. The urls were there by design, so that you could reply from an html email with a reply form.
So when he got the email, it displayed in his preview screen in outlook and launched a bunch of popups, each posting something to a different thread.
It was a joke in terms of technology, but they could have made it a profitable site by putting ads on it. They tried selling the ads themselves instead of accepting a lower CPM with networks and never managed to sell their inventory.
Making abuzz profitable could have been done overnight. They'd needed to have dropped 90% of their paid staff (keeping ones on to make the folksonomy actually work and code stability into the program) and signed up for third party ad networks to sell their inventory.
But that wouldn't have qualified as a success for them, so they cut their losses.