Any time someone advocates barriers/moats/etc in a market, I immediate ask myself, does the redistribution of profits this causes outweigh the efficiencies gained?

The small town vendor story is instructive:

Walmart destroys downtowns, but it also saves the (oftentimes poor) locals hundreds or thousands of dollars a year. In some communities, this has been a huge boon to standards of living, even though it closed the local hardware shop. In other towns, it has turned them into a wasteland of low wage jobs in strip-malls and broken a big chunk of the local economy.

The same question can be put to the web markets: Going with one or two big providers will almost certainly add to efficiency gains, and thus lower prices. Is trading away the competition worth it?

Edit: Sometimes equilibria is two or three providers, or one dominant and a few smaller (think search). In my mind the best case market is probably US coffee, where you have one dominating chain (Starbucks) that opens up the market for lots of smaller one-offs.