It seems to me that in this field there is an eternal tension between technologists who can make things, users who take them into their own lives, and companies/management who are essential for it all to happen.
Wallflower at the Web party was in yesterday’s New York Times (re-posted here, no registration required) explores the management/techie tug of war that seems to have driven Friendster into the ground. It highlights three main pitfalls:
1. They focused on developing exotic features at the expense of basic functionality (their lag times were HUGE, and the site had numerous errors), but the added features only dragged down performance further
2. Their initial board members were Silicon Valley big names who didn’t get the concepts and weren’t representative of their user base, which lead to a disruptive series of CEOs (each with their own agendas) jerking around the engineers
3. They were offered $30 million by Google in 2004 – and held out for more. Now that they’re ready to sell, they can’t even get $20m (contrasted, of course, with Myspace‘s $580m price in when sold to News Corp. July 2005)
It’s well worth a read.
—————————————————- 
Interesting to note, by the way, that Jonathan Abrams (CEO of Friendster) started the social networking site because he wanted a facility to raid his friends’ address books for potential dates. ‘”Basically, Jonathan wanted to meet girls,” said Mark J. Pincus, a Silicon Valley entrepreneur who provided Abrams with some of the seed money to finance his project at the end of 2002. “He told me himself, he started Friendster as a way to surf through his friends’ address books for good-looking girls.”‘ (Also quoted from above article)
It’s a pity that the execution has had its ups and downs, because the theory is still a method of choice for finding dates. Just this weekend I’ll admit I went to a party, out to meet people through those I already know — friends-of-friends are far more reputable and interesting than random people in the pub! In my opinion, one of Friendster’s strengths was the way its model fits into the ways in which we interact anyway, particularly (in this case) when looking to meet people to spend time with. Much of the functionality seems to have been picked up by Myspace (which hasn’t claimed a purpose, just allows users to interact as they want), but with Friendster’s recent influx of VC cash I’ll be interested to see if they manage to grow into and exploit their chosen niche.