>>> Anyway, just pushing back on the religion that P2P has developed. >>> It's >>> fun. It's sexy. It's really hard. But it's often pointless. >> >> The advantage of p2p is that it reduces cost of entry in a certain >> market >> space. No one denies that anything p2p cannot be done central. However, >> sometimes economics justify using p2p as a solution. And in Skype's >> case >> experience gained from Kazaa was no doubt helpful. > > What economic argument could possibly support the avoidance of central > rendezvous? Servers are cheaper than engineers. By orders of > magnitude. Building and maintaining complex p2p systems requires > hundreds of engineer hours, billed at high rates. A server costs > $50/month. Even if you only value your time at $5/hour, if it saves you > 10 hours a month, it saves you money. > > Economics are unquestionably on the side of appropriate use of central > services, bootstrapping, authentication, and rendezvous being the big > three. What numbers are you using that oppose this?
None. Maybe, we missed each other but I had mentioned in earlier posts that these three require centralized service for Internet scale deployment. What does not necessarily require a centralized service is a central file repository for filesharing, or a central media relay to support call establishment for nodes behind NATs/firewalls. It is in these contexts that I prefer a p2p solution to a centralized one. -s _______________________________________________ p2p-hackers mailing list [email protected] http://lists.zooko.com/mailman/listinfo/p2p-hackers
