- We need
to go back to the late sixtiesand early seventies.
- In that
period the first packet switching network, called the Advanced Research
Projects Agency Network or ARPANET for short was put into operation and
Dr.Kahn had a major role in the overall architectural design.
- The aim
of ARPANET was to connect computers or hosts in order to timeshare resources.
ARPANET used the innovativeconcept of "packet switching", an
alternative to the traditional "circuit switching" techniqueimplemented
by telephone networks.
- In October
1972, Dr. Kahn organized a large, very successful demonstration of the
ARPANET capabilities at the International Computer Communication Conference.
- This event
was the origin of several national and international research projects.
- For example,
in Pisa we had RPCNET, the first Italian packet switching network in a
- At the
same time, telecommunication operators started looking with great interest
at the potentialof this technology to provide data services to their customers.
- In the
seventies, ARPA developed two other packet network technologies, one for
groundbased packet radio (called PRNET) and the other for broadcast geostationary
satellites (named SATNET).
of these packet networks could communicate through ARPANET.
Dr. Kahn's idea was to get hosts to communicate across multiple packet
networks of rather arbitrary design without knowing the network technology
- This idea
paved the way to the Internet era. In the spring of 1973, Bob Kahn (then
at ARPA) and Vint Cerf (then at Stanford) started working together on the
detailed design of this idea. The outcomes of their fruitful collaboration
were many and strongly innovative.
- They realised
that a "gateway" (now known as a router) was needed between networks
to accommodate their different interfaces and route packets of data.
- In addition,
their collaboration led to the creation of a new protocol, calledthe Transmission
Control Protocol, or TCP, which as we all know is still used today, after
more than 35 years from its original design, to carry out communication
between processes running on separate hosts connected to the same or different
- In order
to support packetized voice, in 1978, Vint Cerf split the TCP into two
- The result
was TCP and the Internet Protocol and this justifies the acronym TCP/IP.
- The User
Datagram Protocol (UDP) was thus created and many services, including the
modern voice over IP service (VoIP), work today on top of UDP.
- The widespread
success of personal computers and the deployment of Ethernet Local Area
Networks during the late eighties dramatically increased the number of
networks making up the Internet.
TCP/IP was easily able to encompass both technological shifts from mainframe
to personal computers and from wide area to local area networks.
- At the
same time, TCP/IP has enabled the many rapid and accessible applications
on the Internet that we rely on today, including email, the World Wide
Web, Instant Messaging, Peer-to-Peer transfers, and a wide range of collaboration
and conferencing tools.
development shave helped make Information Technology a critical component
across the industrial world. At the time it was designed the TCP/IP protocol
was recognized as being very innovative.
I am sure that even Bob and Vint never imagined that they had created the
essential underpinnings of today's planetary Internet.
a billion users are now connected to the Internet. Only people working
in computer networking know how it really works. Very few people know how
it got here.
hardly anyone knows that a SATNET node was installed in Pisa in the middle
of the 80s.
- This was
the third European node, after the ones installed at the University College
of London and NTE in Norway.
towards the end of the 70s, Bob Kahn visited CNUCE, an Institute of the
CNR, the Italian Research Council, located in Pisa, for discussing the
extension of SATNET to Italy.
- At that
time CNUCE was about to finish a similar project at a European level named
STELLA (Satellite Transmission Experiment Linking LAboratories).
- This meant
that CNUCE had the necessary competences for collaborating with the SATNET
- In that
period fruitful discussions occurred between researchers from Pisa and
from the USA. On various occasions Bob and Vint visited CNUCE.
- And on
one of them the configuration of the Italian SATNET node was set up. Well,
it took almost one year before CNR approved the acquisition of that node.
But a few days after this happened there was a technological enhancement
in the SATNET node which moved from a small and cheap minicomputer to a
very powerful and expensive multiprocessor computer, named butterfly gateway.
- The butterfly
gateway was officially announced in one of the International Cooperation
Board meetings, which I was attending.
- You can
imagine how I felt! I remember I informed all the attendees that I had
decided to give up working on the Italian SATNET node project.
- To have
the new order approved it would have taken another whole year. And in any
case there might have been another technological improvement in the meantime!
Bob brought forward the coffee break and suddenly disappeared into a corner
- When the
meeting restarted, Bob stood up and said: "Luciano, I have some very
good news for you. We will finance the Italian butterfly gateway so that
you can get on with the activation of the Italian node!"
- I must
confess this was the first and last time in my life in which bureaucracy
has produced a positive result.