I have not read Tom Friedman's The World is Flat but I thoroughly enjoyed Ed Leamer's review essay "The Flat World, a Level Playing Field, a Small World After All, or None of the Above?" in the March 2007 Journal of Economic Literature (gated).
Leamer quickly makes the point even if there are forces that seemingly point to a more level playing field, there will always be clustering. Texts in urban economics usually begin with the thought that in a featureless-plain world, there would still be cities as long as there are agglomeration economies. The range and the ambit of these economies may change but they still determine (for example, as per Leamer) how many Hollywood's there are and how much of the world they serve.
And relationships must still be formed because they facilitate transactions. " ... Friedman misses the distinction between markets and relationships, and thus misses the potential for policy measures that might facilitate the formation of long-term relationships between workers and employers. We economists do a linguistic disservice when we call a relationship-free, frictionless outcome "perfectly competitive." The Luddites had it right when they complained thet there is nothing oerfect about the outcome at all. Frictions are our friends. Frictions give us the peace of mind thst we will still have our jobs when wake up tomorrow. Frictions reduce the chances that one party will try to 'hold up' the other, absconding with the lion's share of mutual benefits. Frictions give us the confidence to make relationship-specific investments from which great returns can flow. It's the friction we call 'falling in love' that allows the human species to flourish." (p. 123).
The essay elaborates many related points and will surely (I hope) show up on many syllabi -- even if it will not be the best-seller that the book is.