JUNE 14, 2011  INET NEW YORK


     ******

10.30 am Panel : Pushing Technolgy Boundaries


>> DAVID SOLOMONOFF:  Introduce Lesley Daigle, chief technology officer of
the Internet Society.  She's actively involved in shaping the Internet's
evolution for more than a dozen years and following her work with the
Internet engineering task force, IETS, and being the chair of the Internet
architecture board, IAB, Daigle's role with the Internet Society is provide
strategic leadership on important technical issues as they relate to ISOC's
ongoing programs.

>> LESLIE DAIGLE:  Good morning.  Thank you, David.  Let's see if I can get
this technology to work.  There we go I think.

The panel session we have this morning, I'm going to give some brief
introductory remarks, and we will hear considerable discussion from our
esteemed panelists.  We have Hunter Newby,  CEO, Allied Fiber; Link Hoewing,
VP of internet and technology policy at Verizon; Chris Libertelli, Senior
Director, Government and Regulatory Affairs at Skype; and Nick Gall, vice-
president, Gartner Research.

I'll give opening remarks to do scene setting for the panel.  We will hear
from the panelists' perspective, give everybody an opportunity to provide
opening remarks, and then we will get to some dialogue and Q and A and
hopefully we will have a robust discussion.  I'm certainly encouraged by the
level of interactivity in the last session.  I'm looking forward to hearing
from all of you some more.

Terms of reference.  First of all, what is the Internet?  It's easy to say
that the Internet is that thing that we turn on in the morning, and is the
thing that we always get the phone call from a spouse when it isn't working
at home.  But really fundamentally, if we want to be crisp in talking about
major Internet issues, we do have to have some crisp terminology in terms of
what actually is the Internet.

From a technical perspective, the Internet actually is the system of
interconnected networks that are performing according to the Internet
engineering task force's protocol standards, and best practices, for
communicating with resources and end points reaching globally throughout the
Internet.

And all those words are measured and important in terms of global nature of
it, in terms of the Internet working, so on and so forth.

Then, we can say that an Internet service is providing the connection to the
Internet for an end point or a network, and either directly or through
proxy, to provide access to the rest of the Internet with nondiscriminatory
best practice, best efforts services.

That is kind of key as well.  You probably will get your Internet connection
from an Internet access provider, and that access provider is providing
essentially the connection over a last mile, whatever that last mile may be,
whether it's DSL, cable connection, carrier pigeons, which has been done, or
any other form of connections.

These companies are responsible for the experience over that connection.
That all seems fairly straightforward and maybe a little bit grinding
through some details, but there are some points of confusion.

The Internet is based on the Internet protocol standard, IP standards.  It
turns out that those standards are very useful for networking, either useful
for other forms of networking as well and it is not at all uncommon at this
point to be providing a number of services over IP in a network, usually
over a particular network and not an Internetwork, and some of those
services you may be familiar with, your home telephone service may now be
being offered over an IP network.

But that is as distinct from the Internet which is a service being offered
over the IP network.

There are also so-called over the top services.  These are other services
offered other than in your access, retail access network, that you may be
using over the top of the Internet.  Fundamental concepts, Internet as a
service.  So Internet access retailers can and should be clear about what
they are offering you by way of Internet service, what is the expected
performance of that link, what is, what should you actually expect out of
the service?  And then they should deliver that.

That though should be considered separately from any other IP-based services
they may be offering, may be offering you, may be offering on the same
infrastructure.  The key point there is that as long as that Internet
service that you are offered is happening according to the service levels
that have been advertised, you shouldn't have to care what other things that
your retail access provider may be doing with their infrastructure.

Then at the Internet Society, our focus is on ensuring that that Internet
service isn't just the least common denominator service offering, but that
in fact, it continues to flourish and provide the global networks that is
the basis for so many innovations that we have come to know and love,
certainly not the least of which is the worldwide web, so that is it for
setting the scene for the terminology today.

Today's panel, as you see in your programs, is actually set with some
framing questions for discussion.  I'm not going to run through these
questions.  But certainly, keep them in mind as the sorts of things that the
panelists had in mind when they were putting together their remarks.

Then at this point I would like to pass it over to the panel to make their
remarks.  First, can I invite you, Hunter, to make your remarks?

>> HUNTER NEWBY:  Certainly.  Thank you.  I'm the CEO of Allied Fiber.  I
thank the Internet Society for asking me to be here and all of you for
taking time out of your day to come and listen and participate.  I find it
an honor to be here in the presence of such great people and thinkers and
builders and contributors to our society today.

What I'd like to focus on, in terms of what our common understanding is,
everybody coming in the room and having a one-on-one level course under our
belts, definition of terms.  You did touch on a couple of them lightly.
Leslie just did this.  Internet protocol is not the Internet.

That is really important concept to understand.  VOIP does not mean voice
over the Internet.  The worldwide web is not the Internet and the Internet
is not the worldwide web.

Net neutrality is not Internet neutrality.  It's network neutrality.  There
is a big difference between the definition of those terms, as per the FCC
and you will find in a court of law.

The Internet as much as it is a collection of IP networks that are
interconnected, begins as the physical layer.  And I spend most of the past
ten-plus years at layer 1, actually 0, I suppose, defining a set of
standards and rules for interconnecting disparate networks at the physical
layer.

Without the physical layer, there is no layer 2, and there is no layer 3,
and there is no public layer 3 or any other layer in the stack.

Try and bring this into focus, some of the motto I suppose, a shared open
and transparent Internet is what the Internet Society seeks.  In my mind,
the best way to ensure that is at the physical layer.  If you have access,
to fiber, dark fiber of your own that you can light, that is meaningful.
That has physical proximity to other networks that are on different fiber
systems and paths, you can light that fiber, put multiplex and Ethernet
switches and routers on it and connect to those other IP networks without
oversight or regulation, which is what is being discussed here at the higher
layers and basis of an open and free Internet.

I completely agree with that.  I think that needs to manifest at the civic
layer so there needs to be an appreciation for what occurs at once.  The
interconnections for public IP networks that create the basis of the public
Internet take place in buildings.  Sometimes they are referred to as carrier
hotels.  Sometimes they are referred to as Internet peering points or
exchange points.  They are physical locations on planet earth, with a
longitude and latitude, and actual specific geography.  Not a cloud.  Not in
reference to air.

These buildings are quite important.  And as it relation to the Master
Switch, which Tim mentioned, and Egypt which comes to mind, there was a
particular building in Egypt that was commandeered by the government, and
that is basically where the internet was shut off.  Think about it and
project it in the U.S., where are those buildings; who owns them, who
controls them and what happens within them, who makes the rules, who is
allowed to connect to whom.  How do you connect between those buildings?
Because that is the physical consistency of the public Internet and all
other networks that exist within the U.S. and any other countries.

I would like to the conversation at that level, and then we can discuss the
shared open and transparent Internet above it from that vantage point.
Thank you.

>> LESLIE DAIGLE:  Thank you, Hunter.  Link, would you like to make some
remarks?

>> LINK HOEWING:  I should start with a disclaimer.  I don't think we ever
used carrier pigeon for any of our services to date, but interesting idea;
hadn't thought of it before.

I spent a lot of time thinking about Internet policy, and the truth is that
aside from some of the key policy questions like privacy, the hardest thing
for me to fathom is in fact defining what the Internet is.  There are wide
array of definitions, and Leslie talked about some, and most of them go back
to the technologies of the protocols involved in making the Internet work.

All of these are important factors to me.  But what makes the Internet the
Internet is the cooperation and multistakeholder arrangement that undergirds
the operations.  It's unique in that sense, decentralized, network of
networks that do actually interconnect and generally works quite well.

One good definition that helps explain it better for me is this one.
Voluntary agreement among network operators to exchange traffic for their
mutual benefit.  This definition too has its flaws.  After all it's focused
on the network providers and networks themselves, not on the users.  It also
doesn't talk about what other things make the Internet valuable.  But it's
the focus on the voluntary nature of the agreement and the collaborative
nature of how these agreements are reached that is an important concept.

There are incentives for these companies to interconnect and exchange
traffic.  This also underscores to me how well these interrelationships
generally work, what brought this home to me was a recent study by the
packet clearinghouse that was done a few weeks ago, and in that study they
surveyed peering arrangements on the Internet and reached the following
conclusion.

The Internet or network of networks consists of 5039 providers which are
interconnected with one another in a sparse mesh, each of the
interconnecting links takes one of two forms, transit or peering.  Of the
total analyzed agreements that they looked at, 99 percent were actually
handshakes, not even written agreements.  In their view this actually
demonstrates how well these interconnection processes work and how well the
companies work together to take the traffic back and forth.

This is one indication to me that we are having success in making sure the
Internet continues to operate.  Interconnection process seems to work pretty
well.

But none of this is easy work.  Interconnecting is one thing.  Making
traffic work and makings it get to the consumer as they want to get to it
them is another thing.  On our network every day 100 million people connect
in one form or another.

About 2-1/2 billion text messages are sent back and forth, hundreds of E-
mails are exchanged, gigabytes of video are exchanged.  Yes, we still
connect a billion and a half phone calls every day as well.

We have to work with other networks to make this work, work with background
providers and companies globally to make this happen.  New research from
analysis confirms the Internet is becoming more global and better reflects
the reach of the Internet particularly into less developed corners of the
world.  Not too long ago traffic transited through the U.S. no matter where
it was headed or where it originated.  Many ISPs worked together in putting
facilities in less developed parts of the world, and connection points
sometimes called ISCs have sprung up around the world to allow
interconnection among various ISPs globally.

This means a lot of traffic can generate and transit back and forth
regionally rather than transit through major countries, which is another
major development that is positive.

All of this suggests that the cooperative and competitive dynamics of the
Internet play out over the world every day generally without incident.  Yes,
there are flare-ups, disputes, conflicts.

How could there not be with thousands of companies trying to interconnect
and send data back and forth.  But based on the PCH report most of these
networks interrelationships seem to work out positively in the long run.

This success is reflected in part in the investments companies are making to
upgrade and advance their networks.  We have been investing 16 to $17
billion a year for many years now, in part to build out fiber to home
technology and build the new 4G LTE based mobile broadband network.

We have built a global fiber network that connects to 2700 cities and 150
countries, which has 450,000 root miles of fiber which is enough to circle
the globe 18 times.

All these networks are expanding capacity.  In the U.S. it is clearly
leading in 4G deployment, deploying more quickly and seeing devices deployed
more quickly.  Contrary to popular belief, our exact phones are advancing,
we are beginning to implement 100 gigabit technologies in our back phones
and fiber to home networks can offer a lot of speed.  In fact, we have
tested one gigabit connection to a home recently and successfully.

We are pursuing new approaches to reach consumers as part of this investment
process.  One of our leaders at the CES show this year, consumer electronics
show, said it best:  No one company can do it alone.  This requires
collaboration, and competition.  All these factors are part of the Internet.
We have operationalized lab technologies to begin to try to test LG and 4G
technology more aggressively.  We did have a lab in Waltham which is
operational.  It lets entrepreneurs take devices and try them in operating
4G environments.  There are no nondiscrimination or nondisclosure
agreements.  They come in, test the device.  We help them engineer it so it
works well on the network.

If they want to work with us on developing the product, further, fine.  If
not, they can walk out and do it any way they want to.  We have venture
capital companies to help them get financing if they want to move forward.

It is another model which you wouldn't have seen ten years ago, an open lab
trying to get people to develop LG technology so they move forward quicker.

Innovation, collaboration and competition, these are fundamental.  I think
they are being reflected every day in what is going on in the investment,
advancement of these networks.  Consumers do have more choices and there is
a lot more going on than there used to be, in many ways, especially new
business models and new approaches to the consumer.  A Harvard University
paper put it well.  They said older technologies like electricity were
disruptive.  But then led to a period of stability as companies learn how to
adopt new technology.

We have entered a new phase in that pattern, disruptions followed by
stabilization has itself been disrupted.  Businesses and social institutions
constantly find themselves racing to catch up and learn the steadily
improving and changing technologies in communications and IT space.  It is
not the old pattern anymore.  We get new innovations and they die down or
get adopted widely and we go on for something new.  It continuously happens.
New innovation and new technologies.

I'm generally an optimist.  There are challenges.  There are four issues I
want to mention that are challenges we have going forward.  First, even
though we are deploying 4G technology now and LTE technology, I believe we
are going to need more spectrum quickly.  The administration is focused on
that.  It is an important thing.  It is going to take time to make that
happen.  But it is important we start moving forward with that quickly.  We
got our lead in 4G deployment in this country in part because we did DTV
transition, so we have options in the last part of the last decade, that
helped us deploy more quickly.

Secondly, there are important consumer issues that need to be addressed.
One of them is privacy.  New legislation by senators Kerry and McCain offers
a lot of promise as a framework for privacy in our view.  We have an
endorsement, it's a good mechanism within that bill.  The white paper was
well done as well.  Good models for how to do a good privacy framework.

Third, innovation is moving rapidly.  It is imperative regulation be looked
at with great caution.  There may be situations that arise that cause debate
or concerns that make people believe we need to step in with a government
role.  I think the multistakeholder approach we have used works well, and we
should try to do that as much as we can.

I believe, even though we believe in the multistakeholder approach, it is
fraying at the edges.  There is constant pressure for intervention of all
kinds.  We need to finds ways to keep pushing back on that.  That is an
important aspect from Verizon's standpoint to keep that process in place,
which is one of the reasons we are a major member of the ISOC and involved
in what they are doing.  It is important work.  Thank you.

   (Applause.)

>> LESLIE DAIGLE:  To Chris, if you would go ahead.  Thank you. 

>> CHRIS LIBERTELLI:  Thanks.  I'm going to make two or three points,
hopefully very briefly.  When you have people like Vint Cerf (inaudible)
(faint audio).

So we can get into questions.  I'm more interested in your questions than me
downloading to you in a one-way conversation.  I spent four years working at
the FCC (inaudible) public policy work for Skype.  I feel like I need to
test that bias (inaudible) come at this from the perspective of some of the
work in government now at Skype, as we transition Skype into a new ownership
structure hopefully in the next few months.

The two or three points I want to make are about the public policy
environment.  (Inaudible) less public (inaudible) important decisions going
to be made, decisions about your money (inaudible) worthy of our attention.

The first point I want to make as I was coming up here on the train from
D.C., I reread parts of Jonathan's book, Future of the Internet and How to
Stop It, read it in combination with Master Switch.  They are two books that
are the best discussion of jurisdiction and how public policy is expressed
on the Internet.  Those are the two best discussions.

What I took away from the two pieces is, have a fundamental problem with
jurisdiction in the space.  There is a public policy consensus we should
have this kind of interconnection regime or that kind of universal service
regime is built on assumption of national sovereignty.  And for Skype there
are limits on territorial jurisdiction and the reach of government to go
after, regulate something which are largely offshore.

I don't know if you appreciate that when you download Skype and agree to our
terms of service, you are virtually traveling to Luxembourg and agreeing to
a term of service with the company based on Luxembourg law.

That says something about ways Internet companies touch national boundaries,
and limits what regulators in places like Brussels can do to establish an
interconnection regime that are pro consumer.  I think that leads you to a
world where best practices and voluntary arrangements are likely to be more
important than government policies, at least in the short term.

The debate reminds me about the debate about repatriation and moving money
offshore back to the United States.  There is a limit (inaudible) into the
United States, but there is likewise a limit to the telecommunications
environment.  That means that industry is going to drive some of the
solutions, not government.  But that isn't to say that government doesn't
have an important role when it comes to the access part of the network.

Things like filling out wireless licenses or dealing with traditional
(inaudible) one of the reasons it has been difficult for (inaudible) ATT
mobile merger, we are in a space where we haven't re-articulated what
success looks like in terms of public policy.

What I mean by that, around 1996 when congress last passed the Telecom act,
there was a bipartisan consensus we were going to pass this bill and the end
state would be something called intermodal competition.  What does that
mean?

That means cable companies would compete with Link's company.  Link's
company would have a wireless network that would compete with AT&T, and you
could (inaudible) against that goal.  I'm not sure we have a clear
articulation of that goal today.

For Skype it's about multi-modal competition, which is to say access layer
interconnection is important and should have competition, innovation at that
layer.  Absolutely.  But the real gain from our perspective is happening at
the higher layers of the protocol stacks.  We should have competition at the
access layer, but also articulation of what success looks like at the
application layer.

What that looks like, policy of maximum openness, maximum competition,
innovation between companies like Skype and Google and others, building
applications higher up in the protocol stacks.

The third point I want to make is about what is happening in D.C. that
affects the way the networks are interconnecting or how they are funded.
That has to do with universal (inaudible) set of industry negotiations, that
it might surprise you to know are going to move to $8 billion of our
financial resources, our money from one provider to the next with the goal
of supporting a ubiquitous supportable network.  And the Obama
administration rightly in our view changed that dynamic to make it about
supporting broadband.  That is what the right articulation of what success
looks like.

But I would say like net neutrality debate, where we have a great deal of
public involvement, this one is unlike net neutrality debate in that we have
much less public involvement in how that process is playing out.

   (Faint audio.)

One public policy issue (inaudible) how it's defined, universal service
reform effort, how subsidies are parsed out to particularly smaller rural
companies, the questions that will be answered in that proceeding are
critically important if we are going to have ubiquitous interconnected
broadband enabled Internet.  And it's one that I think we all should do more
to engage the FCC on, particularly as the industry comes forward with a
proposal and suggests ways for policymakers to (inaudible)

I'm going to leave it there.  I'm much more interested in your questions and
happy to answer any questions you have about Skype and what our future looks
like.  Thanks.

   (Applause.)

>> LESLIE DAIGLE:  Thank you, Chris.  And Nick, would you like to round out
the remarks?

>> NICK GALL:  Sure.  Wow.  Sounds really complicated.  Sounds like a really
complicated problem, just hearing the former three panelists address this.

So, I'm certainly not going to try to solve that problem here.  In fact, I'm
simply going to try to frame the problem.  I think that is three-quarters of
the battle.

There is no one way to frame it because this is a wicked problem.  It is now
a wicked problem because I'm from Boston.  We like to say wicked in front of
things.  Two design makers in the '70s coined the problem to frame a class
of problems that divide easy solutions.  They have ten criteria for
determining whether something was a problem and one applies here especially.
Their criteria number 9, which is the choice of explanation of the problem,
determines how it will be addressed, i.e., the framing issue.

How we frame the problem is going to have a huge impact on how we think we
have solved it.  Let me frame it from the point of view of the end point,
from the user.  That is where I come from.  I've been using the Internet
since 1980, thank you, Vint, thank you, Tim, for at least the 90 part of it.
I've made career decisions as well as job choices based on my access to the
Internet.  For me this is crucially important as well as publicly important
as well.

In that time one of the key things that struck me about the Internet is, and
the topic of this panel, is about innovation.  The Internet is about
innovation.  The topic of our panel discussion today is the future of
innovation in the Internet.

One of the problems here is where do we do the innovation?  Which layer or
layers do we do the innovation?  That is the crux of the matter for me, the
frame I want to put on it.  If you put in it that frame, it is just the good
old end-to-end argument all over again.

Do we do the innovation at the end points, where companies like Skype and
Google and Microsoft and whoever, Tim Berners-Lee, in his lab doing it on
his own, can they innovate without restraint at the end points, and the
network has to accept the protocols they put on top, whether they are
hostile or not.

Or do we want to innovate down at layer 0, for one or two or three or four;
that would be great if you could innovate everywhere, but we can't.  To get
the controversy going, I disagree the traditional disruption stabilization
cycle that Link referred to has been disruptive.  I don't think so.

As protocols become more widely established, more general purpose in nature
and more independent with upper levels, their innovation slows down, right?
Still on IPV4.  You can innovate underneath and on top of IPV4, but you
can't change IPV4.  At least it has been a struggle, 20-year struggle.
There is an issue about the stability of some layers being the enabler of
innovation and experimentation of other layers.

One of the best works I've ever read on this issue from an economic business
perspective is Design Rules, Power of Modularity by Clark.  It has applied
to other industries including the Internet.  The main thrust is the option
value created by a set of interfaces, technology, can create unmanageable
designs, something that no one can really control; they can't control the
innovation, can't control the outcomes, can't control the profit chain.
That leads to the problem of who makes the profit when that happens.
Innovation of profit in at least a capitalist society go hand in hand.
Where we decide that the most profits should be, will determine where we
have the most innovations.  I look forward to the debate and the questions.

   (Applause.)

>> LESLIE DAIGLE:  Great.  Thank you very much, Nick, although I would like
to push back on one of your points that we can't change IPV4, an area near
and dear to the (overlapping speakers) in fact had an event last week and I
can tell you that this morning, in Chicago, at a cable meeting, Best Buy is
talking about what they should tell their customers on the floor in two
minutes or less about IPV6.  We are making progress.

All right. 

>> NICK GALL:  I should say it slows down.  Doesn't stop.  It slows down. 

>> LESLIE DAIGLE:  Indeed, doesn't take away from your overall point.  Thank
you all panelists.  Before we open it up to the floor and before I start
asking probing questions, I would like to know if any of you thought of
interesting questions to ask each other to kick the dialogue off and warm
up.

>>  Are the Bruins going to win game 7? 

>>  Absolutely. 

>> LESLIE DAIGLE:  I will start with one of our discussion questions.  The
Internet started as an inherently collaborative network, indeed
Internetwork, so what is a general Internet in this day and age, and what
are the qualities of it that need to be preserved, or nurtured, in order to
ensure it continues to provide the environment for innovation?  And are
there any that are in fact not relevant?  Any takers?

>> NICK GALL:  I'll take a crack at that.  I'm anxious to get to the
questions.  I left out one of the parts I'm going to use for my framing of
the Internet.  It's the definition used in the FCC's open Internet report,
back in December, which I thought was pretty good overall.  Let me read it
to you verbatim.  I'll summarize it.

The Internet is a mass market retail service by wire over radio that
provides the capability to transmit data and receive data from all or
substantially all Internet end points.

I'll stop there.  There is a little more.  But that is the essence of it.
The Internet is, with all due respect to Verizon and others, it is not layer
0.  It is not the net work itself.  It is the end point.  That is the
Internet.  It is the end point.  Everything else is a means to an end of
that set of end points.  I can have the world's most sophisticated, most
inexpensive communications pathways.  But if I can't reach out to the world
and reach virtually all end points, I don't have the Internet.  For me, the
key issue here is when it comes to neutrality, is can I have access to
virtually all the end points.  I looked for a definition of end point in the
300-page whatever report.  I didn't find a good one.

I would characterize end point as not just the Website, or just supports,
though that is part of it, but it is any protocol or API, if you will, that
that end point wants off a route, so I'm happy to have people have access to
any end point, end point can use any protocol it wants to on top of the
Internet.  We have the public Internet.

>> CHRIS LIBERTELLI:  A few words about that definition.  I think we spent
way too much time last time we were trying to figure out the right words to
use (inaudible) commission really (inaudible) part of the FCC order to get
it just right.  I understand why they did that.  They were trying to come up
with a tight definition that encompassed the problem they were trying to
solve.

   (Faint audio.)

But for Skype, the question was about consumers experience the Internet.
What I mean is there is a set of expectations built up over the last ten, 15
years, around what Internet access is.  The carriers know what Internet
access is.  Consumers when they walk into a Verizon or AT&T or Cricket
store, by Internet access, they have come to understand what that is, and it
means the ability to go anywhere you want to on the web, to connect up a
device and run Skype, etcetera, etcetera

That is the more important perspective to look at the definition of the
Internet from.  And the most important, one of the most important words in
that definition is radio.  There should be no distinction in terms of public
policy between (inaudible) and wireless.  Yes, wireless networks need to be
more aggressively managed because of (inaudible) I'm not challenging that
notion.  But we are going to go to a world soon where once that conversation
which may begin on this device that may be connected to a physical network
(inaudible) end of the conversation (inaudible) in that environment it
should all be seamless (inaudible) wireless piece of that definition.

   (Faint audio.)

>>  One of the reasons, my comments, optimistic about where we are going
with the Internet, I do believe the expectation that consumers have about
what they should be able to do on the Internet are widespread and
acknowledged.  Everybody knows that including us as providers.  That
provides a lot of reassurance, and in fact they are going to be able to get
what they want in terms of connectivity to the various sites they want to go
to.

Another piece of that that helps, and that is why the investment piece is an
important discussion, make sure we try to encourage continuing advancement
of the networks so they can do more with it too.  It is not just getting to
the site.  It is also being able to do things they want to do on it.  Some
of those things require a lot of capacity.  Some require better latency for
example.

On the LTE network we are deploying, we are (inaudible) low latency network.
Chris and I were talking before the event here.  I had the comment
repeatedly from people, it's as good as my (inaudible) in terms of
connectivity (inaudible) good latency.  All those take time and engineering
to do.  It's a lot of work, but it's paying off.  The more capacity and more
advances we have in these networks, the more people can do with them and the
more (inaudible) will happen.

Only criticism I have of what you said, we are an integral part of this too.
The way you are discussing it, it sounds like we are not part of the
technology that is the Internet and the truth is we are.  We are innovating
in different ways but we are providing innovation.  We are experimenting
with different business models.  If you went into a Verizon store ten years
ago, we had a phone you would have spent 18 months testing on, tightly
integrated app you could or could not buy depending what you wanted and
fairly limited number of features.

Today you go in there and you might as well have Apple and Google and others
in that store with you, because people are driven by those decisions first
and foremost and not just by the network.  The environment has changed
because it's a collaborative environment.  You have to work with other
partners to make the value of this get to the consumer.  It is not as easy
as it used to be.  There is competition and more choice than there used to
be.

>>  I would add to the comments made by the panelists that the way I see
this is the ying and yang.  If you look at the symbol, it represents to be
balanced.  It is not 50/50, straight line through a circle.  It's a wave
which has connotation of what we do here.

It's not going to always be equal.  There is always sometimes going to be a
little more, a little less, but in the end it's all balanced.  I believe the
applications drive the network.  I believe end points are important.  But I
also believe without the core there is no way to connect the end points.

Necessity being the mother of invention, you can't create an app, people
can't access the app.  If there is only one network, then there will be
controls.  And I go back to Ben Franklin in the free press.  Free press
didn't mean the paper was free, like it didn't cost money to buy it.  It
meant that the information in the paper wasn't under the oversight of the
crown, of the king.

In the net neutrality parallel, if you have a single network that provides
you all of your content, it can decide for you what you will see and not
see.  The root of openness is based upon multiple physical paths, multiple
printing presses in different places owned by independent people that print
these things with the content that they choose to put in it which is
relevant to those that read it and not censored in any way.  That is the
parallel that I draw.

If there is anything that I think needs to be preserved about the Internet,
to keep it open and neutral it's carrier neutral Internet connection
facilities in the U.S. and around the world, that are the basis of the
Internet.  These are peering points of the carrier cells that I referred to
earlier for without layer 1 there is no 2 or 3, etcetera.

There is no core.  Then there are no end points.  I appreciate all the other
perspectives.  I completely agree, Verizon is central to the development of
the core network that applications like Skype write over.  And I would argue
that innovation at layer 1 in physical fiber itself with single mode, SMF
28, SMF28E, SMF 28E plus LL which is a particular product type of a
particular vendor.  Low loss fiber that exists today that was created within
the last five years did not exist ten or 20 years ago.  It's more efficient
for carrying 100 gigabit lambdas, and which then supports 100 gigabit
Ethernet, which supports LTE, which then supports -- it's all connected.

There is nothing isolated about any piece of the stack.  It's all connected.
We must all appreciate each layer and its role, for if we take any of those
layers for granted, they will be taken from us.

>>  Can I respond? 

>>  Briefly. 

>>  I agree it's all connected.  But one of my favorite metaphors for the
Internet, which came from David Clark, the first chairperson of the IGS, is
the hourglass diagram, hourglass metaphor.  For me that is what
distinguishes the kind of innovation which is amazing innovation, down to
layers 1 and 2, at the bottom of the hourglass, that can go on as much as we
want, and we need it because we need more band width, radio, land lines and
so on.

The top of the hourglass is obviously all the cool applications, APIs like
the web and so on, that get built on top of IP and TCP.  But that narrow
waves, the bottleneck in a way, for better word is the IP protocol.  That is
the part that in a way has to be stable, neutral, noncompetitive, open, all
the key words, we don't have to do that below.  We don't have to do it
above.  Believe me.  Everybody competes.  Skype competes with other VOIP
providers and so on.  That is cool, where we all cooperate.  Where we have
to be neutral is just (inaudible) although everything is connected, the
power of the way the Internet was modularized or layered was there is that
one key information hiding layer that lets us go about innovating both ends
of it, top and bottom of it, without interfering with one another.  We have
to preserve that.

>> LESLIE DAIGLE:  If I may make a comment on your comment, certainly we
need interoperability, narrow waist.  The extent we have interoperability at
other layers, we do derive other benefits for being able to use building
blocks, is one of the reasons why we argue for instance for keeping DNS
infrastructure open and not impeded by implementation of policy responses.
I think we heard this morning, interoperability is certainly key for some
applications as well.

I do actually have another question, but I'm going to hold it for later.  I
would like to come back to the question of whether we currently have
Internet service over wireless, by the earlier definition, and the extent to
which we don't, how will that impact future innovation or will it change.
Data plans today (inaudible) point to point connectivity between (inaudible)
like to leave that thought with you and turn to the floor, and start taking
some questions from you all.  I see the microphone is ready in the aisles.
Please do identify yourself.

>>  My name is Patrice Lyons, corporate counsel, corporation for national
research initiative.  I've been involved with the Internet for many years.
I'm a lawyer by profession.  I get picky about words and what they mean.

A lot of words flew by me this morning like public versus private Internet.
We have the Internet, or open Internet versus open architecture, which is
much more serious because you are talking about the open architecture
Internet here.

But the question I wanted to ask, and my bar association tackled this
several years back, and we did it in the context of 802.11 in an Internet
environment, and basically, I have a great interest in copyright law, so we
were looking how to take the various elements that are involved in a managed
information environment and how you would structure it, and what then might
be necessary to change legally, if anything.

One of the first things that we have really struggled with back then, and
really met some tentative conclusions, is that the layered approach wasn't
helpful, over, under, you are talking about bits after bits, you are
managing the environment.  You are identifying you have to have different
ways to track things, different ways to protect things, security was most
important thing that we were concerned with.  So what we looked at then was
a more integrated approach.

And at the time, my corporation and the president Bob Khan, with something
to do with the TC/PIP, we put a paper in the working group on governance
where instead of the layered approach as the definition in 1995 of the
federal networking council of what it meant to be integrated Internet,
instead of just layered approach, we suggested that there also be an
integrated approach that you wouldn't close the door to one versus the
other.

I seem to have heard today just mostly all about layers.  But have you found
it more acceptable in some of your applications, Skype perhaps, the
integration of the different elements in the environment.

>>  I would say we talk about open architectures, I'm translating that to
mean things like our open API program, which is a set of (inaudible) (faint
audio).

Link into Skype and use Skype at the application layer.  My reaction to your
comment, I don't have experience specifically with your project.  We need to
be aggressive about why are we going through these definitional battles?
What problem are we trying to solve by defining different OSI layers or
coming up with alternatives.  In public policy terms, it's always about
whether consumers are harmed or whether the access providers or another
entity is restricting output, doing things to harm consumers.  That is the
touch stone.

I do feel like particularly last summer at the FCC we spent an inordinate
amount of time coming up with definitions that are likely to be outdated by
technology anyway.  We should have some humility around this definitional
exercise, because it often doesn't have any kind of enduring relevancy.

>> LESLIE DAIGLE:  Thank you.  I do see that there are a number of questions
in the back channel.  I will take some of them, although I will observe some
of them appear to me to be particular questions to the panelists about their
company affiliate's business practices as opposed to the general topic of
this panel.

>>  One company in particular.  I don't know why   (chuckles.)

>> LESLIE DAIGLE:  I will use my moderator prerogative to not take those
questions.  If there is a panel general point that you want to make, I
invite you to reframe.

But before we go to those, any other questions from the floor at the moment?
If we can get a mic to Vint.

>>  From Google, this is a question that goes to some of the things that
Patrice mentioned.  If we are overly strict about layering, there are
potential hazards.  We understand that if you don't know what the network is
doing, you can't adapt to it.  So the idea of layers, you know something
about what is going on in the layer below.

   (Faint audio.)

Not a layer bigot (inaudible) architecturally what made the Internet
remarkably flexible and resilient is having standard interfaces that are
stable.  The point Nick made about the narrow waist is the example of that.
The stability has permitted applications to be developed that are on top of
those interfaces on the presumption that they aren't going to change
dramatically.  That makes them hard to change.  That is why IPV4 and IPV6 is
such a struggle.  The point I'd like to make, it is not as much the layering
as it is having stable interfaces, various parts of the architecture, allow
people to rely on them.

It is true, however, that there are coincidences.  So for example, there are
physical things like the fibers that Hunter talks about or radio frequencies
that we talk about in the wireless world, and they correspond to places
where there is standardized interface, procedures by which you modulate the
signal are examples of those stable interfaces.

If I can take advantage of the microphone to ask Link a specific question,
that Leslie alluded to but didn't get discussed, with regard to IP protocol
itself and wireless world and (inaudible)   (faint audio.)

Can we anticipate that IPV6 will be acceptable from the handset, so that
programs that are running in the handset actually have IP addresses, and
similarly, will we see that (inaudible) not sure about plans for (inaudible)
access to the IP layer in those two media

>>  We have IPV6 in the background (inaudible) I'm going to check into that.
I don't know what the time frame (inaudible) it is not right now capable of
doing that.    (Faint audio).

>>  One comment (inaudible) completely agree with stability at the
interface, and it all comes down to one word that you can underline, bold
and put an exclamation point after it:  Control.  Everything is about
control.  It all starts with dirt.  I'm not kidding.  Dirt right-of-way.
Access to land.

And then the land, you can build the roads and you have the rights of ways
to buildings, and physical layers strictly as it relates to stability,
access to a building, being able to go through the door and go into the room
where all the other carriers are and physically connect to the other
networks.  The ability to have that and to presume that you can have that
creates stability and predictability in all the other functions of whatever
your business is in a network world.

That controls the underlying assets and rights to create that environment
for the good of all people potentially, assume taken for granted.  That
level of stability is what I believe we should seek to protect, so that all
the other layers can be equally protected and stable.

>> LESLIE DAIGLE:  Thanks.  I'm going to take a back channel question at
this point.  We have from Aaron Sumrich from Rutgers, relying on consumer
expectation alone to define the scope and definition of the Internet is very
limiting.  Most consumers don't have the expertise to describe needs and
wants that are not currently within their scope of service.  Thoughts from
the panel.

>>  One, I disagree with that.  My dad is a 78-year-old Internet user, and
he knows as much about what he wants; I'm not saying he knows the
technology, but he knows what he wants.  He talks to a lot of other people
and he also understands what they have.  I think in general terms they do
know what they want.

Can they translate it into how you change your network to do that?  No, of
course not.  Another thing about the Internet that I've always valued is
that there are so many other people on the Internet who are experts at this
stuff, watching themselves and offering their own expertise to make things
better, that over time these things get fixed because of them as well, good
and bad.  We have seen examples of that repeatedly.

That is one of the reasons that this model is so valuable in my view.  You
get a lot of input constantly, and not just from traditional consumer
angles, my service isn't working or my warranty needs to be updated or
whatever, you get it from people who contribute either by making changes
themselves or suggesting changes or participating in ISOC.  But in terms of
consumers, a lot of them recognize and appreciate what is out there today,
and they usually demand it from us and we have to try to provide it.

>>  Can I make two points about that?  First, every one of the billion
dollar web companies today, including some in the room, started by web
consumers, Internet consumers, not by the network providers.  They are
consumers.

Secondly, it's the consumers that create the pull for innovation.  It is the
fact that somebody offers a cool new service that is kind of flaky back in
the original days, and people are, wow, this is so cool, I can call anywhere
in the world, and everybody begins downloading software and using it, and
that is what creates the innovation.

The consumers in my opinion are the absolute driver, the engine of the
innovation in every possible way.

>> LESLIE DAIGLE:  It's fair to say it is not always clear when a user is,
quote-unquote, is just a user or an inventor of something huge.  That is an
important aspect of the Internet.  I saw hands go up.  In the back.

>>  Good morning, Ron Andrew of dot support LLC.  On the discussion of
pushing technical boundaries, specifically what you said in the last
question, we are talking about development at end points and innovation at
the edge.  The filtering probably certainly is detrimental to true net
neutrality.  Most consumers have no idea that it's happening to them.  It
has come to light in this last period.  How can filtering by ISPs, Google,
Facebook and the like be combatted?  Who is going to take this up, at which
level or layer or which organization will do the deciding on the behalf of
the consumers?

>>  I can tell you who is not going to take it up.  And that is the federal
telecommunications regulator, FCC, because the way the statute works, the
things that you are describing, and it's funny, I learned about this too for
the first time this morning, when we talk about net neutrality, inside
beltway definition of that is asymmetric which means it is provided to
companies that provide (inaudible) (faint audio).

Nondiscrimination guarantee (inaudible) happening lower down (inaudible)
because the statute limits the agency authority, which is our view
(inaudible) higher up, without jurisdiction (inaudible) you move over to
other parts of the government like the Federal Trade Commission.  It is not
the FCC.  It may be something more like a consumer protection rationale for
intervention around the kind of content you are describing.

>> LESLIE DAIGLE:  My additional comment to your question is, I hope that
you are an ISOC member, and I hope you are an ISOC New York chapter member,
because you too can represent the fight for the good side.

Over there, please. 

>>  Thank you.  My name is Terry (inaudible) as to who I am.  But I was the
one (inaudible) practicing a lucrative practice and conned into doing policy
and safety societal work, I don't know how many years ago.  When we talk
about consumers, we need to recognize that the Internet is no longer one
way.  That is how it used to be.

When we first looked at the societal steering group and societal, Internet
Society, societal committee, we really need to now look at the need for
bringing the societal nonheat into the room.  I'm a lawyer.  If I were smart
enough, I would have been an engineer.

But I'm not.  But I'm on Dr. Phil and Good Morning America and Today Show,
and I know what the consumers want and I call in for other people in the
room and say, what are they talking about?  I think it's time that as all of
us who are in technology start engaging the consumers, I have two of the
teens from my teenager program here today, and Verizon uses them on
absolutely everything and Google does and everybody else does when they look
for experts.  Am I wired (inaudible) Skype, one of the top five technologies
that they love the most in this country.  I called Microsoft.  I didn't know
who to call (inaudible) last Wednesday with IP 6, how can I compete with
them.

What we need to do is all start recognizing that networks, no networks, who
is in the room, who is not in the room, regulators or no regulators, in the
end this is all run by the consumers.  When somebody decides they want this,
they are going to get this.

They are driving things differently.  Unless you engage them, unless you
listen to them, unless the Internet Society brings in young people and
regular consumers like 78-year-olds who really get this, we are lost.

My challenge I guess, less of a question, is building on the other question
about consumers, is we can define it all we want, but we need to make sure
everybody is at the table.  That means any consumers there, and I am happy
to help again, whatever you need me to do, but it's time we start doing
that.  We need to make the Internet Society relevant across all aspects of
technology, how people's lives are being impacted.

>> LESLIE DAIGLE:  Great.  Thank you.  Does the panel have any comments or
thoughts you would like to add to that?

>>  Wholehearted agreement that consumers have to be there.  The end point,
frame it any way you want, but it's consumers that drive innovation.

One challenge, if by some means, providers were able to dampen the
enthusiasm for new services or what have you in the consumer base, we might
have a problem.  If it weren't possible to see that your friend had a cool
app that you don't have, and now you want to get it, you need to get it off
the web or the Internet, if that were somehow blocked, maybe there might be
a problem.  But as long as that access to new innovation, as an end point is
there, consumers will keep driving the things.

>>  As you look at the competition part of what things need to do, if one
network designed you can't have it, you will go someplace else.

>>  Let's put that on the table.  That is the 800-pound gorilla, elephant,
what have you, in the room.  That is, it would be relatively simple if the
lower layer providers of the world, Verizons and so on, just wanted to
compete at that level.

But unfortunately, this open capitalistic society we have, everybody wants
to compete at every level.  The fear is that the providers of the pipes will
give preference to the apps they put on the pipe against the other apps that
are on the pipes.  That is what in essence is, it's not all about, it's a
lot about.  Maybe we should somehow address that.

>>  If I can leap in there, that was the essence of the point of the
definitions that we ran through at the beginning of the session.  If you are
offering those other apps over the same pipes, that is fine.  But be clear
about what you are offering as the Internet, and be sure you keep offering
it as advertised in order to enable competition.

>>  Can I give two analogies to get us going?  One is, electrical network.
Originally, the providers of the electricity were the appliance
manufacturers as well.  They sold appliances, the apps for the electrical
grid, as well as the grid.  We outgrew that by and large.  The other one is
the road network we all use.  Imagine if the people who built the roads also
competed as automobile manufacturers.  Keep those in mind.

>>  I'm still unclear, when you say give preference, what do you mean?
Apple store is the Apple store.  We have no control (inaudible) same thing
with Google store.  We don't manage those stores.  How are we going to give
preference --

>>  (Inaudible) Skype. 

>>  There you go. 

>>  I'd like to add something onto that.  (Overlapping speakers). 

>>  This is going to happen on a wide scale.  Today, for example
(overlapping speakers) apps that are free today on our smart phone, nobody
has blocked those.  People are using them today.  (Overlapping speakers).

>>  You hit the nail on the head.  Your comment is assuming there is an
Internet that is open.  Then the consumers drive everything.  That is off
the table, because of what Nick just said, for two reasons.  The two reasons
are if there aren't multiple networks to compete, you can't just go from
this one to that one.  They don't let me have it, so I'm going to go to this
one.  What if there is only one?

Nick's comment about the higher layer operators deciding to get into even
higher layers, and start to compete, and great point about the road, who
built the Grand Central Parkway?  Vanderbilt.  Why?  So he could drive his
car to the city.

Imagine if that was the only car that was allowed to the Grand Central
Parkway.  There wouldn't be any traffic problems, but you wouldn't be able
to use the GCP.

The other problem is not about the Internet, but about network neutrality at
the FCC.  What Chris said about access to the Internet, so let's say you are
a consumer.  You want to consume WikiLeaks.  The net neutrality says the
government will protect the rights of the citizens to access all legal
content.  What is the definition of legal?  Who defines what legal is?  What
if the government said WikiLeaks is illegal?  And they had a piece of paper
in their hand that they could tell every ISP in the United States to block
that domain.

Now you are a consumer.  You want to consume that content.  But you can't,
because legally it's not allowed.  We have multiple issues going on here.
Multiple legal issues, consumer issues, and you said it, and so did Nick.
It comes down to the physical paths.  If this guy won't sell it to me, I'll
go to this network.  There has to be another network for that to happen.
That occurs, unfortunately for everybody that wants to talk about higher
layer applications, which I love, that occurs down in the lower boring
layers, where everything is kind of like nothing happens, like the bottom of
the ocean we think don't change much.

Nick is spot on with that.  Absolutely accurate statement.  You have to
watch out for when the guide itself, the dark fiber decides he wants to
light the network and no longer wants to sell dark fiber.  The guy that lit
the network sells IP.  He doesn't want to sell clear channel transport.  The
guy that sells IP starts offering voice over IP services and doesn't want to
allow competing services over his network.  And so on.

>> LESLIE DAIGLE:  I'd like to take us in a slightly different direction.
We have a comment, question from the back channel from   Ethan Wechsler.
Adding network capacity won't decrease (inaudible) which is choking the
Internet.  How do we decrease spam with a pointer to sources, short term or
longer term.  I would suggest the panel consider it more broadly than just
spam, but all forms of unwanted traffic, and the extent to which that is a
defensive challenge consuming resources and getting in the way of investing
in innovation.

>>  Part of it is, back to the collaborative nature of the Internet that
companies have to work together on this.  They haven't sent us to do that
because a lot of spam is using up capacity that we want consumers to use and
not be wasted on spam.

For example, we not long ago did combine with other companies to sue a major
spammer.  We actually did win.  The FCC has enforced this as well.  There
are organizations that do focus with ISPs working together on trying to
combat spam.  It's a never-ending battle.  I'm not saying we have solved it.
Frankly, when you are talking about sending out millions of E-mails and
hoping you get one hit out of a million, it is not economically too hard to
see why people do this.

But I think there is progress being made.  There is also increasing
improvements in the way that spam filters work.  We certainly have had
complaints from time to time about increase in spam, but over time I'm
seeing less of that.  I think we are getting better at it.  Continuing work
is needed.  It's a collaborative process.  It's a (inaudible).

>>  Other comments? 

>>  We live in an open free society.  I get junk mail.  I'd like to cut down
on the junk mail.  But I can't stop the post office.  We will shut down the
Internet on Saturday and nobody will get spam.  Does that make everybody
happy?  Spam-free Saturdays!  Other innovation.

>> LESLIE DAIGLE:  Chris, you wanted to leap in?  No takers?  All right
then.

I'd like to come back to the wireless question.  And briefly, following up
how Vint was framing it, as it stands today, my understanding is, it's
possible to get data plans for any number of devices, some of them built and
delivered by companies that have no wireless access network interest
themselves, Apple, but it is not possible to do point-to-point connections
over IP, between handsets for instance, because of the way wireless networks
are architected.

My point is, the wireless is clearly key to the future.  And this is, my
point is from a discussion of wireless providers on a different continent,
so I'm not looking at you, Link, but the point stands and to do better
framing of it, at what point do we expect to actually see devices connected
to a wireless Internet be Internet (inaudible) IP as full on Internet
service nodes.  Or is it not a problem?

>>  I don't understand the gist of the question, to be honest with you.
What are you driving at?

>> LESLIE DAIGLE:  What I'm driving at, I can put my handset on a data plan
today, and I can access a number of things over the Internet based on the
applications that Apple has elected to provide me to put on the Internet.
But according to Vint, I cannot then get my handset to connect directly to
another device sitting next to me on the same wireless network except via a
proxy in the core of the Verizon service.

Not trying to pick on the Verizon service.  This is just the way wireless
networks work today, in part because of concerns about what happens if you
actually provided direct Internet access, open Internet access to a device
on accessing over wireless.

>>  If you are talking about what happens when mesh technology for Internet
becomes mainstream, that has a lot of cool implications.

Great New York Times article about Internet in a suitcase over the weekend,
which is exactly that.  Get back into the Egypt or any other country that
shut it down, the Department of Defense or one of the intelligence agencies,
I forget which, has now a project under way, millions of dollars (inaudible)
to create these capabilities.  Besides the Department of State doing this,
certainly the DOD is extremely interested for war fighting itself to create
mesh networks from the battlefield on up.

It's more a matter of when, not if, and again it changes the end-to-end
argument.  So who knows what will happen because of it?  Chris, you had
something you wanted to say?

>>  I wouldn't say, I don't have the topology we are discussing here, but
this notion of connecting sideways, smart phone to smart phone without the
need for a centralized infrastructure that can be shut down by an
authoritarian regime (inaudible) (very faint audio).

Version of what Skype did at the application layer (inaudible) powerful
stuff particularly as radios become more smart, software, middle ware,
creates ad Hoc networks where they are most needed without reliance on a
centralized infrastructure.  (Inaudible)   (very faint audio).

>>  Skype do it at the access layer too? 

>>  Long story there, but we need to partner with a company (inaudible)
router based sharing (inaudible).

>>  I'm going to try to clarify Leslie's question earlier, repeat it a
different way, and hope to get discussion going, which is I think that she's
asking, when would a company like Verizon or other wireless providers give a
static IP address to a wireless handset?  If I wanted to, I could install
the Apache web server and have a Website live on the Internet, full
Internet, have my hand-held device be a full Internet solution.  That is the
question.

>> LESLIE DAIGLE:  Thanks, clearer reframing. 

>>  I understand the question.  A lot of this is driven by changes in
technology and the market.  I don't know where this is going to head,
either.  A lot of this is future.  I don't know.  We made a lot of progress
to date.

Some of these kinds of questions are things consumers ask.  There will be
reactions in the market.  But I can't tell you what kinds of things will
happen.  That is what you are asking.  Will that happen?  I don't know.  Of
course it is possible.  A lot of things are possible.  It doesn't mean it is
always going to happen right away.  I don't know the answer to the question.

>>  At a higher layer, talk about it from a registry standpoint, when we
move away from (inaudible) architecture from (inaudible) alert which is the
root of (inaudible) dialing perspective (inaudible) move to real DNS IT
architecture, sort of hierarchy, from the stack higher down, and then think
of it from the resolution base to SRV (inaudible) domain based identifier,
then the necessarily, you have an IP address and that will enable IP
applications and we will be there, I think, my personal opinion.  LTE has a
lot to do with that.  Because most of the apps that are running today on the
wired public Internet, Internet, they are not that great on wireless
networks because of the fact that it's dynamic, that it's routing
dynamically, and that the throughput is not always there.

I look at my phone, and I'm in amazement how it goes from edge to GPRS to
GMU to 3G toggling while I'm standing still in my house.  I couldn't run the
same app consistently and get quality of service.  It has to do with issues
back (inaudible) but LTE deployment on a broad scale in any country provides
the pathway for full-on IP networking, which then will, like Link said, the
consumers are already using, it will just meld.  It is a time situation.
You have to upgrade everything.  I believe it will happen.  For sure we are
on that path.  When?  I'm not sure.

>>  When you look at how fast we progress from 2G to 3G to now 4G, we are at
100 million, in Verizon's case we deployed to 4G already, by end of 2013,
all the territory we covered today, which is 290 million people can get
access to LTE, so just in three or four short years, we essentially deploy
to an entire country.  And we deploy feeds that are much faster than any DSL
available or even cable modem in some cases, and latency is great.

These networks are going to offer new things.  We will see how it plays out,
what consumers want.  That is part of what we are doing with the networks,
try new business models, see what they want, see if we can meet the needs
better.  They are good questions, good technology issues.  But do consumers
want them, do we demand them.  These are all things we see in the
marketplace.

>> LESLIE DAIGLE:  I'd like to thank the panelists for dealing with that
question so well.  We are not done yet.  We will let them have, like to give
the panelists the opportunity for 30 seconds each final words.  We can go in
inverse order.  Nick, you want to start?

>> NICK GALL:  Sure, I wish innovation could happen at every layer of the
stack simultaneously with equal speed.  We can't.  If I'm going to err on
the side of innovation, I would prefer to have it happen at the end points.

>> LESLIE DAIGLE:  Chris. 

>> CHRIS LIBERTELLI:  I want to pick up on the consumer point that was
mentioned earlier, two points.  One is that as citizens, we need to push our
politicians to define what (inaudible)   (faint audio).

Innovation environment we want to have.  Secondly, keep up the engagement.
What we saw last summer was good, lot of transparency with fair amount of
public discourse around net neutrality rules.  There are more important
battles being fought in particular around universal service.  If ISOC can be
part of that discussion, we will come up with better rules.

>>  Chris's point about engagement is important.  Multi stakeholders, we do
believe that whole model for how the Internet activities to be worked on is
really important and it drives a lot of change.  The kind of questions that
you are asking today are good ones.  They are kinds of things that help
drive people to start saying we should look at that.  Maybe there is
something new we can offer.  That is how this process works.  That is
positive in my view.

The other quick thing, we didn't spend a lot of time, we did talk about
consumers, we did not spend enough time talking about fundamental issues
they care about.  We talk about networks and technologies and things we like
to see but those don't always affect consumers.  Privacy affects them.  On-
line safety affects them.  A lot of industry players are spending time on
this.  This is important.

We need to get this right.  Those affect technologies you want to see
deployed, because if we can't get it right, it is going to be harder for
them to want to use them.  We need to focus more attention on them.  That is
why I mention privacy in my initial remarks.  Multistakeholder approaches
are better than having government demand how those ought to be done.

>>  I add too what Link said, one thing that wasn't discussed which is a
whole other discussion and topic and day probably, is cost.  How does it all
get paid for and every layer.

But going back to one of the original things, I believe Nick mentioned was
framing the discussion, so my final comment would be, framing.  I'm a big
believer in defining terms.  I can't stand people use words interchangeably
in the conversation and walk away and think they understood what the other
said but they don't.  They have a different definition of the underlying
word.

It is challenging and problematic, and I spend time personally and
professionally, writing about, speaking about, trying to educate everyone,
starting with a basis.  So I would encourage everyone to get to the root of
the word that each of you have in your mind that you make up, and there is
consistency of your knowledge of what it is that is going on.  And when you
start to have conversations, as tedious as it may be, make sure you are on
the same page with the other person.  That way you can be constructive and
productive.

Thank you. 

>> LESLIE DAIGLE:  Great.  Great words to end on.  Thank you for coming and
participating.  I thank the panelists for their contribution.

   (Applause.)