[ NNSquad ] Re: pcap files of the Comcast forgeries?

[Brett Glass <nnsquad@brettglass.com>]

At 06:53 AM 12/18/2007, Russell Smiley wrote:
 
>Why would internal P2P be less likely to cause congestion?
>
>I can see that it would be least expensive since hopefully Comcast
>doesn't bill itself for traffic on it's internal network. However, if
>P2P is inclined to consume all available bandwidth then surely
>congestion is likely to occur whether the P2P traffic is constrained to
>internal nodes, or between internal and external nodes?

Exactly. When P2P software finds a fast connection or a big pipe, it milks 
it for all it can get. In fact, some P2P software, such as Kazaa, attempts
to turn turn well connected nodes into "supernodes," which consume many
times the bandwidth of an ordinary node. "Supernodes" are, in essence,
"zombies" or "bots," because they are doing the bidding of the creators
of the software at the expense of the user and his or her Internet
provider, both of which lose computing power and bandwidth.

Intellectual property issues aside, this is one of the biggest problems
with P2P software: it is designed to take resources from third parties
to accomplish the goals of the authors of the software. This is not
something that anyone concerned with privacy or security should support.

--Brett Glass