IEEE Consumer Electronics Magazine - September 2018 - 47

building my own networks using the
same principles as the Internet. I was
fortunate to learn about the technologies as they developed and work with
some of the designers firsthand. I knew
that a local network wasn't even a network, it was just a shared wire or radios.
The networking was done entirely in the
connected computers. I didn't want to
just dial up and connect one computer to
an online service. I wanted my entire
home network to be interconnected with
the rest of the Internet and Microsoft's
campus network. At that time, you had
to get a separate account for each computer just like you did for each phone
line. After all, that's the way dial-up
modems worked. I took a different approach because I was interconnected to
a network, and all the computers would
share a single connection.
At that time, the term broadband
was used to describe a fat pipe that the
provider would use to sell services. This
is why AT&T paid a high price for my
local cable company, MediaOne; they
expected to make money just as Minitel
had by selling phone calls and cable TV,
reading meters, and gaining new revenue streams from e-commerce, meter
reading, and whatever else they could
offer. By using the intelligence in my
computers, all I needed was one shared
connection, and all AT&T saw was a
jumble of packets that all looked the
same. My job at Microsoft allowed me
to gain this capability built into Windows. Users no longer needed a network professional to set up a home
network; they could just buy what they
needed at any computer store! I can't
claim all the credit, but I assume that
this contributed to AT&T being bought
by Southern Bell Company (SBC).
Today's AT&T is really SBC.
We have today's regulatory system
because the business model of telecommunications and the country's needs
for connectivity were not a good match.
In the days of telegraphy and, then,

The problem, with or
without neutrality, is that
we're increasingly adept
at programming around
the network.
telephony, the high capital costs and little differentiation required a regulatory
agency to ensure an orderly marketplace. We start by recognizing that
moving intelligence outside of networks
inverts the model.
The Internet is not something we
acquire through a broadband pipe.
Instead, we turn the pipe around and
originate the services from within our
own homes or offices. We use that
broadband pipe, and any other facilities,
as commoditized resources. This means
we need locally owned infrastructure
that is more akin to sidewalks and roads
than train tracks. I'm careful to use the
word infrastructure rather than utility to
avoid the idea that we're consuming
anything anymore than we consume
sidewalks when we take a stroll. Water
and electricity are metered by usage. It
doesn't make sense to talk about using
up a supply of ones and zeros. There is
no scarcity of Internet.
We pay for sidewalks as a community. We join together to pay for the
paths in an apartment complex or as a
city for paving the paths. Sidewalks are
not strictly necessary. We have them
because they facilitate walking and
make the city a better place, and, like
sidewalks, ambient connectivity is free
to use. Today, each innovative application, such as medical monitoring, requires
a separate negotiation with carriers who
do not gain much revenue from devices
that generate little traffic. We avoid
depending on relationships with myriad
carriers just to ensure connectivity. Ambient connectivity allows users to unleash
major innovation. Imagine communi-

cating without a monthly fee merely to
connect. That monthly fee will soon
seem as strange as paying just to cross
the street.
As we transition to ambient connectivity, we can continue to use the existing
telecommunications infrastructure as just
another wire. That's the crux of the
problem for the providers: they are,
indeed, just another wire with all the
value being in applications. During this
transition, we need network neutrality
more than ever to ensure that the carriers
don't fight the future by abusing their
stewardship of our vital means of communicating. This is harsh for them but,
it's just business. Companies like Comcast and Time-Warner have moved on and
are now in the content business knowing
that the networks are no longer the focus
of their business. Verizon and AT&T are
following along. They may or may not
succeed in this strategy. Time will tell.
The battle over network neutrality is
framed in the existing regulatory framework that treats the Internet as just
another telecom service rather than something new. We must look ahead, not
backward. We must seize opportunity to
add trillions to the economy. Just think
about what would happen if we simply
reduced everyone's Internet and cellular
phone bills by perhaps US$100 each
month and returned nearly US$1,000
a year to every family in America
while providing a level playing field
for new businesses.

ABoUt the AUthor
Bob Frankston (IEEECEMagazine@
bob.ma) is best known for writing VisiCalc-the first electronic spreadsheet.
He is a Fellow of the IEEE.

reference
[1] J. H. Saltzer, D. P. Reed, and D. D. Clark.
(1984). End-to-end arguments in system
design. [Online]. Available: http://web.mit.edu/
Saltzer/www/publications/endtoend/endtoend
.pdf

september 2018

IEEE Consumer Electronics Magazine

http://web.mit.edu/Saltzer/www/publications/endtoend/endtoend.pdf http://web.mit.edu/Saltzer/www/publications/endtoend/endtoend.pdf http://web.mit.edu/Saltzer/www/publications/endtoend/endtoend.pdf

Table of Contents for the Digital Edition of IEEE Consumer Electronics Magazine - September 2018