Marshall, the means used to characterize bandwidth utilization and Internet traffic statistics can be as slippery and loopholely as tax rules. I frankly don't know what the total impact of Napster was or could be in the future, except to say that in some pockets (like university dorms, for example) the impact was severe enough to warrant putting in fatter pipes both to the outside world and internal to the campus backbone. Also, I'm sure there's a way, but I personally don't know how to go about collecting data on this distributed kind of application (as opposed to a more centralized one) in such an orchestrated way so as to make sense, allowing one to make such claims. If someone here can explain how this is accomplished, please do.
What appears to be fairly certain is that future peer-to-peer applications will have a greater impact, overall, on network traffic than they have had in the past (as new types of applications are created), the degree to which still remains to be seen. But forecasts and arguments that I've read suggest that p2p traffic will be appreciably greater, from a proportional standpoint, in the future than they have been to date. One of the riddles here is how to monetize such activity, and how to compensate the ISPs and NSPs whose backbones must support the traffic.
To a great extent, this has been a missing component in the overall calculus, and for this reason I'm a bit suspect about where the additional bandwidth will come from to support it. Then what? I don't know, but SPs have not been shy about blocking routes and filtering certain types of applications in the past when those routes and apps were adversely affecting their business. Here, too, I'd like to hear some additional comments and corrections. Someone kindly add to this. |
