There is something common with some of these ideas worth noting: Dating sites, social travel (social anything), and Craigslist killer all depend on network eff

There is something common with some of these ideas worth noting: Dating sites, social travel (social anything), and Craigslist killer all depend on network effects. A common argument for starting a new business is that "hey, someone else already has this idea, they've mapped out the market and proved it can make money." This works really well for things like restaurants where there is no network effect, no incentives for new users to go with well-established providers or disincentives for existing users to switch providers. In the social/dating/craigslist space, you don't just need to build a better X, you need to convince people to leave the old X (or pray that they've never heard of the old X before and won't consider it) and sign up for your new X. In terms of dating, social and craigslist at least you're only dealing with one type of customer. As soon as you turn to local events/businesses you needs to do this for two (the people looking for events and businesses, and the promoters and business owners). Cable networks are similar businesses, but I think the key distinction is that they charged money for their services. Somewhat paradoxically, I think this makes it much easier for people to switch than if all the services are free - there is the immediate incentive of "save $10 a month!" I think this is why AOL and Prodigy managed to make so much money. Then all the free (ad-supported) online services came along and basically wiped out the paid services, and intensified the network effect and its disincentive for switching to better services.