[jacquesm]

I really think the network effects on FaceBook are what keeps it where it is. I know people that would love leave FB, except for one little detail: all their friends, family and co-workers are on it and they use it for everything from planning birthday parties to holidays and staying in touch. That's a very difficult thing to attack head-on.

Craigslist is another example of such a service, you need the market and the goods, if you have only one (for instance by crawling CL) then you still won't have the market.

That's a really nice example of 'good enough' execution combined with a chicken-and-the-egg situation for any followers.

Of some things there can only be one, something similar happened in NL with the Marktplaats website, when Ebay wanted to gain marketshare here they found they just could not beat it so they ended up buying it.

[vinceguidry]

I think you're taking idea effects and clothing them in the trappings of execution. In order for a business to get really big, it has to provide 10x the value that the current incumbent provides. Microsoft beat out IBM because software had many times the potential that hardware of the time had. Sure, IBM failed horribly on software, but that's because IBM didn't think software was important, not because it couldn't build decent software. The thinking behind the idea gives you the zeal to execute on it.

If the idea is right, meaning it is 10x better than the current best offering, and it's executed properly, then network effects can go screw themselves. Nobody took search seriously before Google did, that gave them a leg up over the competition. (who were they again?) It's just not the right time to start a Facebook or Craigslist challenger. Maybe in a few or ten years the market will change such that most people interact with the Internet differently enough that it's worth building a social network to serve their needs. Right now it would be foolish to do so, because their needs are already being more-or-less perfectly served by Facebook, ditto for CL.