Alen Peacock wrote: > An update: > http://heartbeat.skype.com/2007/08/the_microsoft_connection_explained.html
Sounds a lot like my wild speculation of yesterday, that the size of the network plus some poorly-chosen (in hindsight) arbitrary, reboot-affected supernode-selection thresholds led to a cascade failure. They didn't blame a "25 sigma" or "once in 10,000 year" event, but did mention a "perfect storm"! - Gordon _______________________________________________ p2p-hackers mailing list [email protected] http://lists.zooko.com/mailman/listinfo/p2p-hackers
