Saturday, February 09, 2008

 

My Interview in WorldChanging

I did the interview below with my friend Jon Lebkowsky, and it first appeared under Jon's byline in WorldChanging. (I first heard the phrase, "Freedom to Connect" from Jon's lips!) Below, Jon has elicited the most concise, coherent explanation of Network Neutrality and Structural Separation I've ever done. (In my very humble and completely dis-interested opinion, of course . . . What do you think?) I also explain some of my motivations for doing the F2C: Freedom to Connect conference. I'm reproducing it verbatim but you can read it at WorldChanging too.

David Isenberg is founder and coordinator of the annual Freedom to Connect (F2C) conference, held in Washington, DC (actually Silver Spring, Maryland). This year's conference is March 31-April 1 at the AFI Silver Theatre in Silver Spring. The conference, focused on net neutrality and net regulation, is a must-go if you're concerned about the future of the open Internet.

JL: "Net neutrality" has always been more complex than public conversations would suggest. As coordinator of the F2C conference, and as one of the more knowledgeable participants in the conversations about the Internet's future, how do you see the issue today? What's the best case future of the Internet, and what are some of the risks we should be thinking about?

DI: Here is my multi-layered answer to Network Neutrality is Complex, Discuss.

The first answer is that this meme is largely promulgated by
people who want Network Neutrality to be complex. The telcos
and their henchmen are fond of saying, "Nobody knows what it
is. You can't even define it." The subtext is, "So how could
we possibly have a law about it?"

Then along came the AT&T-Bell South merger agreement, with a
perfectly adequate definition of Network Neutrality agreed to by
both sides. It defined Network Neutrality as Internet service
that does not privilege, degrade or prioritize any packet
transmitted over an Internet-connected network based on its
source, ownership or destination. Problem solved . . . except . . .

. . . as long as there are networks in the real world, not just in
some Platonic ideal space, there will be congestion. And network

operators will need to deal with the congestion. So they will
have to decide which traffic gets throttled or filtered and which
traffic gets through. Therefore, every Network Neutrality rule
needs a Network Management Exception.

This means that we must decide what network management is reasonable
and what isn't. This is where things do get complex.

The telcos, because they have a voice service, have an overwhelming motive to
discriminate against other voice services. The cable companies,
because they have a video entertainment service, have an
overwhelming motive to discriminate against video services . . .
all in the name of effective network management, of course.

In fact, the two big examples of Network Neutrality violations in
the U.S. have been (a) a telephone company (Madison River)
discriminating against Internet telephony and (b) a cable company
(Comcast) discriminating against Internet video entertainment.
In both cases, the company said they were simply managing their
network.

So the debate about Network Neutrality quickly becomes a debate
about what constitutes reasonable network management.
In findings by Odlyzko and Coffman, Odlyzko and the entire
Internet 2 research team have shown that a network is only as
congested as its most congested link, that (a) in virtually every
case, network congestion can be alleviated by replacing a very few
nodes or facilities with more capacious ones, and (b) in
every case studied, it is easier to add capacity than it is to
impose more complex management schemes. But these findings
are derided by the operators as "academic," even when derived
from real-world experience.

So, once again, those with a financial interest in keeping
things complex rule the day. As long as network management is
about so-called quality of service, i.e., rationing, rather
than simply providing more than enough capacity so rationing is
not necessary, then it remains complex and subject to operator
bias . . . and operator financial interest in what is carried.

At this point, I must agree with the carriers. It would be
a disaster if the FCC, or Congress, or any governmental agency,
told the carriers how to manage congestion on their networks.
The rule book would be overly long, overly complex, overly
subject to interpretation, dispute, and litigation.

To simplify, I have proposed – and continue to propose – that
the carriers should simply be prohibited from having any
financial interest in what they're carrying. Then we could
let the carriers manage their networks without fear that they'd
discriminate against traffic that competed with their interests
because they would have no competing interests.

This solution, which is called Structural Separation, would
guarantee and assure that any network management was for
the sole purpose of congestion control, because that would
be the carrier's only interest. Any bias in favor of one kind
of packet or against another would be, by definition, entirely
innocent. As such it could be easily remedied.

Structural Separation would make Network Neutrality simple.
Enforcement of the proposition that carriers could have no
financial interest in what is carried would be via audit.
There would be no big fat complex rule book. There would
be no government intervention in processes the government is
not competent to govern. The carriers would behave, and
the network would be de jure neutral. Simple.

But the political process to get from "here" to "there" would
be as complex as the carriers could make it.

JL: Structural Separation doesn't sound like a concept you could build a whole conference around, so I assume that this year's Freedom to Connect will be about much more. What subjects are you covering and how are they related?

DI: You're right. Structural Separation is just one scenario. The huge
barriers to getting there, spelled t.e.l.c.o (and c.a.b.l.e.c.o.) are
mostly uninteresting. Funny that the telcos and cablecos resist,
even though under Structural Separation the shareholders would
probably make more money, the employees would still be employed and
the customers would be served, perhaps better than they've ever been
served. The IDEA of Structural Separation sounds like death to the
telco because it is so antithetical to their culture and their
business model.

Not only that, Structural Separation is waaaaaay too simple to spend
a whole conference on.

F2C: Freedom to Connect – and I remember that the originator of that
phrase was a wise man, initials JL, from Austin TX – is a conference
designed to look at any Internet topic we want to look at.

Conferences are not so much about exchanging information, you can do
that on the Web. Conferences are more about meeting your friends,
making new ones and building relationships. The most important parts
are the breaks, the hallway conversations, the meals and the
parties. I try build in lots of that. I try to make the food good
so people arrive early for breakfast, stick around for lunch and
enjoy the breaks. I try to make a killer evening reception – I'm
working on something super this year.

I also try to make the conference sessions themselves as interactive
as possible so it is as much of a conversation between the speakers
and the audience, and among members of the audience, as possible.
Again, the goal is relationship building, not information
transmission. In fact, much of the research I trust indicates that
real learning occurs best in the context of trusted relationships.

So to that end, I am inviting speakers who I admire and trust, who
have done good work in lots of different fields, and invite them to
tell stories (rather than advocate positions or convey information)
and challenge them to give the best talk of their lives. For
example, I am inviting John St. Julien, who keeps
Lafayette Pro-fiber
Blog.
As a result, I expect we're going to learn a lot about the
challenges Lafayette, Louisiana faced when it decided to build its
own municipal fiber network. We can look up the demographics of
Lafayette, the cost-per-home-passed, the megabits-per-whatever, the
price-per-month and even the overall economic benefit . . . what I
really want John to do is tell the story of his experience.

You can see from who I've invited that my predominant interest is in
infrastructure, and how the economics of infrastructure is different
from the economics of end-user goods. So I've invited Dirk van der
Woude, who is in charge of
Amsterdam's world-leading fiber network
and
Tim Wu, a fount of new ideas in regulations to spur wireless
innovation. I've also invited several people who are working to keep
our infrastructure free, such as security expert
Bruce Schneier, Open
Rights Group founder
Suw Charman, Electronic Frontier Foundation
Chairman
Brad Templeton, and OneWebDay founder Susan Crawford.

But in addition, I've also invited some people I admire who see the
Internet and its applications as an emerging critical infrastructure
for society, such as author
Clay Shirky, non-profit sector activist
Katrin Verclas and political activist Andrew Rasiej. Since this is
an election year, we're going to try to get the technology policy
advisors to whoever the major presidential candidates are in late
March, too.

Also, because the Internet's problems pale next to the big one –
Climate Change – and because there's a strong hypothesis afoot that
we could actually use the Internet to ameliorate Climate Change, I'm
putting together at least one session, and maybe two, on how the
Internet can save the planet through reduced travel, traffic
congestion management, active monitoring and management of energy
uses, real-time energy auctions, and computation-shifting to places
with renewable energy. This topic will be addressed by Kevin Moss,
BT's head of corporate social responsibility, Bas Boorsma, of Cisco's
Connected Urban Development program, Robin Chase, founder of ZipCar,
and
Bill St. Arnaud, optical wireless researcher and keeper of http://
green-broadband.blogspot.com/
.

All-in-all, the F2C: Freedom to Connect program will present a
NetHead perspective that is under-represented inside the beltway.
Thanks for asking :-)

JL: Do you get many DC policy insiders at the gathering? Do you think you have an impact on them?

DI: I actively try to recruit federal government staff to
come to F2C. This year there are already four FCC staffers
on the roster and I certainly expect many more. In the past
we've had not only FCC folks but people working telecom
policy issues in Congress, in the Congressional Research
Service, at the Federal Trade Commission, and from several
other agencies. Having these folks in the room is a key
goal of F2C for me. They don't get exposed to NetHeads from
outside the beltway very often! Does this have an impact on
them? I hope so!

Also, we've had several FCC commissioners come speak at F2C,
and we've had former Chairmen Hundt and Powell.
Two years ago Representative Rick Boucher gave an excellent
detailed insider's view of the negotiations on the
comprehensive telecom bill – the one that stalled out
because it did not have strong enough Network Neutrality
language.

To me, one of the highlights of all five years of F2C came
during a panel of congressional staff members from the Senate and
House discussing pending telecom legislation. In the Q and A,
somebody asked why they had not talked about more unlicensed,
Wi-Fi-like spectrum. The panelists' answers collectively
indicated that it wasn't on their radar screen, and the
F2C audience started booing and heckling. The staffers
looked shocked . . . Now, I have no idea whether this
incident influenced any of them, but I sure hope so!

Technorati Tags: , , , , , , , ,


Comments:
I think this could have been much simpler. Suprisingly, the entire EU legislation and preamble describing net neutrality and structural separation is about half this length. It does not need to be even this difficult. Non-discrimination, interconnection and access: that all there is to it. You don't even need structural separation if have net neutrality. None of the major players: EU, Japan, South Korea, China, India have structural separation. Its helpful, cleaner, and easier for regulators, but not required.

Uncle Mike
 
Post a Comment

This page is powered by Blogger. Isn't yours?