My impression of the article is that the author was saying it's difficult to get into business selling only what or who is nearby, since both of those are close

My impression of the article is that the author was saying it's difficult to get into business selling only what or who is nearby, since both of those are close to becoming (if they're not already) commodities. The trick with location-based services is to provide novel interaction or functionality or customization on top of these commodities. Quoting the OP: "As a LBS start-up, you need to think about adding distinctive value for users; differentiating on location is an oxymoron." There are plenty of companies who will be hugely successful in the LBS space. Maybe Google, maybe Loopt, maybe Brightkite or Yahoo or CloudMade... If you think you can be the next company to track where everyone and everything is (massive infrastructure requirements!) -- this article is suggesting you are probably on the wrong track. (side note: Yahoo's FireEagle is also something to watch in this space, as are the recently launched developer products from CloudMade.)