When is it okay to discriminate? When it improves the performance of the network. That was one of the messages of Carnegie Mellon's Dave Farber, known in some circles as the Grandfather of the Internet, at a session on network neutrality hosted by The Progress & Freedom Foundation Tuesday. The National Press Club event was the public debut of PFF's book on the subject, Net Neutrality or Net Neutering: Should Broadband Internet Services be Regulated? It's edited by PFF's Tom Lenard an the Free State Foundation's Randy May (Randy being a former PFF colleague). I'll direct you to its listing on Amazon, despite that company's position on NN.
Dave had already spoken on net neutrality the day before at another event; he's been the focus of a lot of attention since the thoughtful NN white paper he co-authored with Wharton's Gerald Faulhaber, Berkeley's Michael Katz, and Vanderbilt's Christopher Yoo. At the start of that paper, the authors state "[t]he Internet needs a makeover. Unfortunately, well-intentioned Congressional initiatives, aimed at preserving the best of the old Internet, threaten to stifle the emergence of the new one." At our event, Farber went so far as to speak blasphemy, arguing that TCP/IP may have taken us about as far as we can go. (He refrained from saying that explicit blasphemy the day earlier when debating Cerf, a co-developer of TCP/IP, but did say the Internet needed the freedom to grow and evolve.) Dave made a compelling argument that legislation could have unintended consequences on the technology underlying the Internet, preventing new architectures from emerging that could better meet consumer needs.
As for discrimination, he said "sometimes you need to discirminate." He was referring to the constant actions taken across the Net to prioritize traffic, such as Akamei's service, and said the network simply can't function robustly without some prioritization. Mark Cooper, author of one of the essays in Net Neutrality or Net Neutering, found himself trying to agree with Dave while not. He said he was "okay with functionality," referring to what Dave was talking about, but said he favored functionality "only if it is nondiscriminatory." Of course, Dave had just made clear that functionality by its very definition was discriminatory. It's that sort of confusion that Dave is referring to when he talks about the dangers of legislation.
Dave said non-technologists (folks like Mark, presumably) that support NN "think we're at utopia in the Internet," but "changes are afoot... the current TCP/IP protocola is near its end." Still, the closest Dave came to taking a direct shot at Mark was to rub his ear and comment that he was in danger of losing his hearing from Mark's shouting next to him. PFF's Adam Thierer, author of another essay in the book, was a bit less cautious. "Mark is a science fiction author," Adam said, critiquing his contention that network operators desire a scarcity model by pointing out the principal rule of networking is to maximize traffic. Randy was even less restrained. "Mark is downright misleading," he said, noting the FCC Title II regulatory nirvana cited by Mark never applied to cable.
When introducing the debate, Tom said broadband was, if not in its infancy, at least in its early adolescence. Dave agreed wholeheartedly with that (and thus dismissed Mark's argument, which involved a 30-year timeline dating back to Computer II). Even Mark admitted that broadband itself was fairly new, and some of the developments on it such as P2P are even newer. If we can all agree that we are very early in the evolution of this new consumer-friendly technology, and if a professor whose students created much of what we know today as the Internet tells us that more positive changes are afoot if only we keep our regulatory hands off a little bit longer, can't we all agree to go ahead and demonstrate just a wee bit of patience?