The problem with starting another Ebay is Metcalfes law that states that the value of a network is proportional to the square of the number of users of the system. They have a lot of users, and thus their network is valuable - even if you try to sell Nigerian stamps from 1800 there will probably be a bidder. A startup will have big trouble getting the first million users, simply because there aren't already a million users. It's a catch 22 situation.
The same is btw. true for facebook, etc. Their value is not in their technology but in their size.
True, attracting enough buyers and sellers will be a challenge, even if a scam-free model is invented.
This actually reminds me of the story of the Arizona Stock Exchange, which was technically a better model for selling stocks, but couldn't convince enough traders from NYSE and Nasdaq to give it a try (http://www.businessweek.com/1997/09/b3516129.htm).
The same is btw. true for facebook, etc. Their value is not in their technology but in their size.