I have made this contrast before, and I think it is another key framing distinction in the net debate.
Network engineers draw the Internet as a cloud, because it doesn't matter to endpoints how the packets get there. The packets get routed, but the protocols are designed to cope with messiness, with buffers overflowing, and computers crashing, and wires being unplugged or ripped up by backhoes. This is the mental model the end-to-end principle encourages - the net is just where packets come from and go to, and has a big 'Somebody Else's Problem' field around it. This is one place where engineers and normal people converge - they don't think about how stuff gets there, they just enter a website or email or IM address and there they are.
By contrast, telco's and networking providers, naturally, do see the wires and the complexity, because that's what they do. They can't use the SEP method, so they fall back on thinking hierarchically, which is another way of coping with complexity. This contributes to the difficulty of getting the open network argument across to governments - the hierarchic frame is a good fit for their default approach to organisation and information flow, so regulatory capture is a likely outcome.
There are countervailing ideas in political thought across the political spectrum, from commons theory to anti-trust and deregulation. The immense success of WiFi's tiny slice of free spectrum is promising, but as Doc, Eric and Mitch and Jon point out, the attack on net neutrality is building. We need to keep trying out metaphors of openness and freedom, invisible hands and co-operation, until we find one that fits.