Getting back to the core

"Kozmo started out as a business dedicated to getting consumers what they needed every day," he said. "It morphed form that into being superstore for everything you might possibly want, from books to CDs, and having deliveries as a tack-on. We're going back to the core." ~ http://money.cnn.com/2005/09/14/technology/kozmo_redux/index.htm http://money.cnn.com/2005/09/14/technology/kozmo_redux/index... Some of the failed dot-com crash-outs from the late 90's had some interesting ideas that might have worked had they stuck with the core of the original idea? Can you think of any others? My case in point is a re-born kozmo as maxdelivery. The difference? (aside from taking $USD230M from VC's) ... * charges for delivery reflect reality ($10 minimum order, and there's a $4.95 delivery charge for orders under $50) * simplifying inventory (focus stays on DVDs, groceries and other daily needs) * no VC's (company isn't beholden to investors who are more focused on growth than profitability) * sticking to favourable geography with high living density of Manhattan NYC (instead of having physical presence in 15 states) Chris Siragusa has obviously learned his lessons the hard way. Do you know of other failed sites that could have been successful getting back to the core?