There seem to be several things that stand to inhibit whether or not IMS will see enough wide spread acceptance to be relevant. One is that consortiums like this seem always to be fundamentally unware of the basic nature of the Internet. Another is that this kind of deep packet inspection will consume at least one, may two orders of magnatude more network computing resources of virtually every kind, from bandwidth to appliance memory. By the time IMS becomes complete enough to actually be relevant, its ability to actually yeild usefull functionality could easly be eclisped by new Internet protocol suite developments. To state that solar-power wireless mesh networks may be a big enough presence to allow anyone who wants to completely bypass local service providers and IMS might not be that big a stretch, given the pace they've kept up so far (I've always thought that 3G meant the number of generations to delivery, in human terms). Its also interesting that the article mentions that IMS at one point depended on IPv6, and now will work with IPv4 - further evidence of how far the consortium is from grasping the nature of the beast they are dealing with.