metcalfe understands the issues only too well. only the truly
blessed can blindly exploit half-truths. he is not blessed and
he is not blind.
unfortunately, mr. metcalfe is living in the world of suits.
very expensive suits. the type worn by what we might not so
affectionately call the 'marketeers'. it doesn't matter what
a small group of technically saavy internet engineers might
think, this is yellow journalism and it sells papers. metcalfe
is the figure-head of a well-laid plan of calculated risk. no
matter what the outcome of his predictions, he and his paper will
claim the high road. if there are catastrophic failures in
the internet, he will claim victory and will continue to report
on the 'clean-up.' if not, before his rantings become tiresome, his
handlers will recall him. he will again proclaim victory
(remember, he was the nagging conscience of the internet for
oh so many months who fortold and therefore, prevented disaster).
he will move on, perhaps attacking some other large and mis-understood
entity, making predictions that have reasonable probability and
limited risk until the risk pays off.
i have no doubt that his campaign will succeed. he is bright, has
built companies, used to be a technically saavy engineer. the trick
will be to bide our time, he will go away in the end. hopefully his
winnings will be modest this time.
He is also living in the world of the user. One who perhaps made the wrong
choice in choosing an ISP. One who perhaps doesn't know how to tell the
difference between local congestion at the ISP, congestion in core routers
and congestion at an overloaded WWW server.
So, IMHO, he has a point. Service is often quite poor. But he's pointing
the finger in the wrong direction. The problem is with incompetent or
greedy ISP's who grossly oversell their lines or build broken networks or
sell too many virtual WWW servers on one poor 486 server.
This is one of the reasons why the Internet now needs some sort of
centralised performance monitoring and trouble ticketing system. If you
can point the blame at an ISP's congested T1 or a company's overloaded WWW
server, and make this info public, then people will soon learn that there
is many, many more than just one reason for congestions and therefore we
also need many, many solutions.
But one thing is sure. The folks on NANOG are not good at telling their
story, at marketing their successes and trumpetting their greatness to the
world. If people would start doing this, then you would see more press
coverage of Internet operations realities and more people, even
journalists, would get an inkling of how things really are going.
Michael Dillon ISP & Internet Consulting
Memra Software Inc. Fax: +1-604-546-3049
http://www.memra.com E-mail: email@example.com