That's because it's a classical chicken and egg problem - the idea is only valuable if a lot of people use it.
Here's a back-of-the-envelope calculation:
If I go to Hacker news, what are the chances there'll be someone to chat with? Well, according to http://www.internetworldstats.com/stats.htm there are currently around 2 billion Internet users, and if we assume that at any given point 10.000 people are on Hacker news then one out of 200.000 global users are available to chat on this site at this moment. What this means is basically that unless you have an installed base of millions there will be noone to talk to. And you won't get that base because there's noone to talk to...
It's simple math, and noone seems to do the numbers before they launch something like this.
Here's a back-of-the-envelope calculation:
If I go to Hacker news, what are the chances there'll be someone to chat with? Well, according to http://www.internetworldstats.com/stats.htm there are currently around 2 billion Internet users, and if we assume that at any given point 10.000 people are on Hacker news then one out of 200.000 global users are available to chat on this site at this moment. What this means is basically that unless you have an installed base of millions there will be noone to talk to. And you won't get that base because there's noone to talk to...
It's simple math, and noone seems to do the numbers before they launch something like this.