Paul Baran’s contribution to the development of the
Internet, the part he played and why his contribution was most significant
Introduction
This report describes the significant contribution that Paul Baran
(pictured left [1]) played in the development of the Internet. The report
describes the part he played by explaining his ideas on building a distributed
network and transmitting information over the network in packet-sized pieces
rather than as a whole. The report shows how his ideas developed and the
problems Baran had in persuading people that his ideas would work. The report
concludes with an explanation as to why I feel Paul Baran’s contribution
was the most significant in the development of the Internet.Distributed Network
Baran, born in 1926, started working for RAND in 1959. From here he was to ‘embark on the project that would change the world’ [2]. RANDs prime sponsor was the Air force and 1959 was a critical time of the ‘Cold War’ when, Baran states ‘the US and USSR were building hair-trigger nuclear ballistic missile systems’ [2]. These missile-control systems were not robust and would not survive attack. Baran’s challenge was to build a more survivable network so the US and USSR would not blow each other up because of control failure.
Baran worked on building a distributed network of nodes (see figure left [2]) with no central point of failure, plenty of redundancy and more bandwidth than the military would know what to do with. The US military had already rejected an idea to use the US network of AM radio stations, because they wanted more bandwidth. Baran worked on avoiding a centralised system because it could easily be disabled by one strategic blow. His distributed network of nodes would each connect to its neighbour creating a significant amount of redundancy, for example, should one node fail there would always be another route to the destination.
Baran concluded that the signals travelling through this distributed network would have to be digital. Digital signals, which are just 1s and 0s can be rejuvenated, even corrected if they arrive in error and can travel through many nodes to reach their destination. Analogue signals would deteriorate at each switched node, they cannot be rejuvenated and would be unrecognisable at the destination [3].
Message blocks
Baran’s next idea was to break each message into regular sized message-blocks. Each block would then take its own path to the destination before being reconstructed into the original message. He proposed an algorithm, or computer procedure, for the node to decide which path each message-block would take. Each node would perform a kind of store-and-forward routine, rapidly moving message-blocks on to the next node until delivery [2]. Each node would keep an updated table of ‘hops’ (how many nodes passed through) it would take to reach a destination and so distinguish the best possible route, a lot like routers do today and is known as packet-switching.
Open Literature
Baran, born in 1926, started working for RAND in 1959. From here he was to ‘embark on the project that would change the world’ [2]. RANDs prime sponsor was the Air force and 1959 was a critical time of the ‘Cold War’ when, Baran states ‘the US and USSR were building hair-trigger nuclear ballistic missile systems’ [2]. These missile-control systems were not robust and would not survive attack. Baran’s challenge was to build a more survivable network so the US and USSR would not blow each other up because of control failure.
Baran worked on building a distributed network of nodes (see figure left [2]) with no central point of failure, plenty of redundancy and more bandwidth than the military would know what to do with. The US military had already rejected an idea to use the US network of AM radio stations, because they wanted more bandwidth. Baran worked on avoiding a centralised system because it could easily be disabled by one strategic blow. His distributed network of nodes would each connect to its neighbour creating a significant amount of redundancy, for example, should one node fail there would always be another route to the destination.
Baran concluded that the signals travelling through this distributed network would have to be digital. Digital signals, which are just 1s and 0s can be rejuvenated, even corrected if they arrive in error and can travel through many nodes to reach their destination. Analogue signals would deteriorate at each switched node, they cannot be rejuvenated and would be unrecognisable at the destination [3].
Message blocks
Baran’s next idea was to break each message into regular sized message-blocks. Each block would then take its own path to the destination before being reconstructed into the original message. He proposed an algorithm, or computer procedure, for the node to decide which path each message-block would take. Each node would perform a kind of store-and-forward routine, rapidly moving message-blocks on to the next node until delivery [2]. Each node would keep an updated table of ‘hops’ (how many nodes passed through) it would take to reach a destination and so distinguish the best possible route, a lot like routers do today and is known as packet-switching.
Baran published his networking ideas in the open literature in 1964. The research was not classified or patented. Baran felt his ideas belonged in the public domain and it made sense for the US to share the information enabling safer control over its (and the USSRs) nuclear missile control systems.
Having published his work, Baran then spent years trying to persuade people that his ideas would work. In 1967 other research teams were meeting to discuss a revised network idea. A basic distributed topology had been devised but no idea of how to actually make it work. Fortunately a British researcher called Roger Scantlebury, who had also been working on the message-switching idea, told them of Paul Baran’s work and that it had been gathering dust on Pentagon desks for years [2].
Significant
The significance of these ideas in the development of the Internet meant a great change from circuit switching to packet
switching. Circuit switching enables a conversation to take place by connecting
the phone lines of those in the conversation. This means that these lines
are solely for the use of this conversation. It is estimated that ‘up to
50% of a typical voice conversation is actually silence’ [4] which seems
a great waste of resources. Packet switching is far more efficient and cost
effective. The packet-sized information messages that Baran proposed only
uses the phone line for the fraction of the time it takes the message to
be sent, allowing the line to be utilised far more efficiently. Baran (pictured left more recently [5]) had difficulty in getting AT&T to accept his ideas because it was such a great difference from what was already in place and working. Instigating Baran’s ideas meant shifting their whole telephone system from analogue, circuit switched technology to digital, packet switched technology. AT&Ts telephone system was massive, it worked, not very efficiently but the technical gurus working there could not see why anything had to change.
Conclusion
I feel that Paul Baran’s contribution was the most significant in the development of the Internet because digital circuit switched technology is so important to us today. Our telephone exchanges all use digital circuit switched technology. This technology allows the telephone exchange to work more efficiently connecting our phones, computers, fax machines and any manner of digital device. Without this technology our computers would never have been able to be connected to the phone lines. Without this connectivity via the phone lines the Internet would never have grown as it has because no one would have been able to connect to it.
Moving from analogue to digital communications was a crucial step in the evolution of the Internet [3] and it was Baran’s work on digital communications that lead to this move and the beginning of the evolution. Other researchers were developing similar ideas but it was Baran who created mountains of technical papers which were later referenced by other researchers. I feel it is right that Baran was credited with the invention and, I also feel that we would not be as far advanced as we are today without Baran’s work.
References
[1] Griffin, S . Internet pioneers [online] Available from: http://www.ibiblio.org/pioneers/baran.html [Accessed 05 Jun 2003]
[2] Naughton, J. (1999), A Brief History of the Future, London,Phoenix
[3] T171 Module 2 Section 2.5, A Survivable Network, [online] Open University. Available from: http://t171.open.ac.uk/t171/module2/section2/network.cfm?DisplayMode=live&CFID=1174458&CFTOKEN=11003647 [Accessed 20 Jun 2003]
[4] T171 Module 2 Section 2.7, Circuit-switching versus packet-switching, [online] Open University. Available from: http://t171.open.ac.uk/t171/module2/section2/switching.cfm?DisplayMode=live&CFID=1174458&CFTOKEN=11003647 [Accessed 20 Jun 2003]
[5] Stewart, B . The living Internet[online] Available from: http://livinginternet.com/?i/ii_rand.htm [Accessed 05 Jun 2003]