The last issue of Communication of the ACM is about "entertainment networking". One of the paper is interestingly dealing with the design of networked games to support users located around the world. As described by Cormac J. Sreenan and Kobus van der Merwe in their introduction:
Jeremy Brun et al. consider how to design networked games to support users located around the world. In this context they explore distributed servers and the relationship between network latency and fairness in networked games. Differences in latency can lead to inconsistency in game state, giving certain users an unfair advantage over others. This problem is exacerbated in situations in which decisions concerning the game state are made on distributed servers, rather than through the more conventional centralized server architecture. The authors identify two techniques that are useful for mitigating these effects: trading inconsistencies and judiciously selecting the location of the distributed servers. The article should be of particular interest to companies interested in scaling-up game servers for use by a truly global subscriber base.
Why do I blog this? I like the authors stance "A game can be considered playable if its users find its performance acceptable in terms of THE PERCEPTUAL EFFECT OF ITS INEVITABLE INCONSISTENCIES" and I wonder to what extent it would be possible to do seamful design (i.e. using the latency as an element of gameplay).