摘要：This article is part of a pair, the first of which looked at the state of play in IP multicast routing [0]. In this article, we look at the broader problems and future activities with multicast. We divide the areas into routing, addressing, transport, security, operations, and research.There has been quite a bit of debate about the nature of compelling applications for multicast recently.[44] It is certainly the case that we do not completely understand the "market" for multicast—this is at least in part because multicast does not yet provide a complete set of functions for all the applications and services we might imagine. This is a typical "chicken and egg" situation, though: To put an extreme version of the argument, the application writers do not see any multicast deployed; the Internet Service Providers (ISPs) do not see any multicast applications; and the router vendors do not see any multicast service demand from ISPs. (The same problem afflicts IPv6, Integrated and possibly Differentiated Services, and mobile IP, of course.)
