Repeated Games and Reputations is primarily about infinitely repeated interactions, where players play a fixed stage game in every period and discount future payoffs. All three ingredients of this canonical setting are very well, and very beautifully, motivated. As argued in Osborne and Rubinstein, players may entirely ignore, in their strategic thinking, the existence of the horizon, although this horizon may be finite in some physical sense.1 Given the objective of avoiding backward induction from the ultimate period, an infinitely repeated fixed-stage game is the simplest framework for studying long-run interactions. It is fairly tractable, thanks to dynamic programming methods, yet it is rich enough to capture most important strategic components of intertemporal incentives. Finally, as the book itself argues, discounting is more useful for studying behavior (or strategies), while no-discounting (limits of the means) may deliver a simpler tool for studying payoffs. Not surprisingly, this canonical setting has attracted the attention of many researchers, and the literature studying it is growing so rapidly that it is becoming ever more difficult to keep track of the entire literature, even for those who have contributed to it. In this context, a book such as this, written by authors who have both followed the literature closely and contributed significantly to it, has a very important role to play.