Transcript EFF / TCPA Roundtable - Oxford Internet Institute

Societal Implications of
Networking: A Personal
Perspective
David J. Farber
Distinguished Professor of Computer Science and Public Policy
Carnegie Mellon University
School of Computer Science
Dedicated to Jon Postel
Paul Baran
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
First some history
The UC Irvine days
A group of kids at UCLA
An IMP at UC Irvine
Kids at UC Irvine
The lay of the network land
The early network
Rational
Computer sharing
File Sharing
A smallish group of friends
Primitive facilities
The path to EMAIL
QuickTime™ and a
TIFF (Uncompressed) decompressor
are needed to see this picture.
What was the net used for
Silver futures
Research in making the net work
The crashes
Hack and hack
The migration to TCP/IP
Slash Cut
Toward a Computer
Science Network
Why?
Getting it funded
Must pay for it after 3 years
The path to a network
PhoneNet
Internationalization
Commercialization
An Explosion of networks
Not all TCP based
NSFNet
NREN
The regional networks
The commercialization of network
services
The evolution to where we are today
From the “good old boys”
to the mob
The evolution of the purposes of the net
Who cares about sharing computers after
the pc (except for super-computers)
People wanted to communicate by
Email
Pictures
Documents
Voice
Video
Everyway
The merging of computing and
communications will drive the
future.
 we went from computers without networks ->
computers with networks -> networks with
computers ->
 the cloud inverts
 the total network is the real action
 the computers are no longer the motivating
item as in the early days
 no longer can we develop computers without
paying attention to the networks
 no longer can we develop networks without
wondering if our computer architectures are OK
The network is changing our
economy
 Buggy whip industries are in trouble
 Even the old “monopolies”
 Some realities re com based businesses
 Telephone vs cable
 Wireless
 Unlicensed (WiFi)
Politics 101
USA
Will the internet escape regulation
The reality of inside the beltway
Security and privacy may be the wedge
The legacy businesses -- will they “give
up”
The “new” era
 The new Chairman of
the Commission
 Mike Powell
 What are the critical
issues they face
 Spectrum , spectrum,
spectrum
 Unbundling
 Economic
relationships
 The internet!!
Security and Availability Will Be
the Drivers
 The current internet is vulnerable
 Points of potential vulnerability
 Service protocols
 Commercial arrangements
 Old software
 Hard to design security in after the fact
Broadband Access
 Will it be open?
 What does that really mean
 Why worry
Technology is easy -society is rough
Show stoppers or at least slowers
Nation States
taxes
culture
rights
Individuals
privacy
1984 fears
big government
big brother
big business
Intellectual property
problems
copyright bill (S)
off to jail with the president of Upenn
show down and maybe stop the value
of the net
the de-evolution of the net to TV
What should we do -society
defend our liberty
understand the world is not all the
same
educate
We Are at the Beginning of
a Great Change
The next 10 years will be as wild as
the last 25
Technology will change dramatically
All optical nets
Wireless of many varieties
Communicating computers everywhere
The Real Next Generation -Optical Networks
 Implications
 Protocols IP?
 Processor architecture
 Software systems
 Will it be a fully distributed system -- back to the
past
NLR Footprint Topology
SEA
POR
SAC
NYC
CHI
OGD
DEN
SVL
CLE
FRE
PIT
KAN
RAL
NAS
STR
LAX
PHO
SDG
BOS
WAL
OLG
DAL
ATL
WDC
What differentiates optical networks
from where we are now
bandwidth-latency is a major difference
 speed affects the ability to compute in a
timely fashion
implies difficulties in
routing
maintenance
actual hardware of switches
may and most likely will imply changes in
protocols
interface to computers
structure of software systems