The ARPANET was designed and built between 1967 and 1972 and although it was based on the concepts outlined in the RAND report, it was itself not designed with survival of a nuclear attack in mind. Originally the system´s primary purpose was to permit researchers and scientists to communicate and transfer data. The networked computer concept was funded by an agency within the US Department of Defense called the Advanced Research Projects Agency (ARPA ). Since most of the host computers had little if anything in common, all of these individual computers were linked together using a sub-net of identical computers called "Interface Message Processors" (IMPs). Each IMP, or node, in the network would be responsible for sending, routing and receiving information. Messages would be broken down into different parts. Then packets with the information parts would be sent along separate routes to their eventual destination (packet-switching). In this system no one node would act as the central or primary sending unit. If one or more nodes in the system went down, the packets would simply be routed through other operational nodes and thus sent on to there destination. The IMPs communicated using "Network Control Protocol" (NCP). Within the ARPANET a  host, through its IMP,  would connect to another host within the network through its IMP. All hosts within the ARPANET were limited to communications within that one network. The networking of computers was an incredible achievement, however it is unlikely that every person or organisation wishing to network could afford to purchase a second computer (an IMP) in order to do so. Eventually, the ARPANET gave way to what we know today as the Internet. The Internet links networks that would normally be incompatible into what appears, on the surface, to be one huge unbroken network of systems. This was made possible through the creation of the "Transmission Control Protocol" (TCP) and "Internet Protocol" (IP). The new protocols made it possible for a computer in one network to exchange data  with a computer in a completely different network. No longer would a subnet of IMPs be required.

          During the time period that the ARPANET was being built a number of other networks were being developed. These other networks were also based on  packet-switching technology. France had the Cyclades network. The British were developing the NPL system. In other parts of the US there was the SATNET satellite network, and the ALOHA packet-radio network in Hawaii. A short time latter the US Department of Energy established MFENet for its researchers. At about the same time, with a grant from the National Science Foundation (NSF), CSNET was established for the Computer Science community. All of these systems were developed independently of each other and were in one way or another incompatible and/or incapable of communicating with each other on there own. TCP/IP provided a solution. Probably the most significant idea behind the new protocols was that they were designed to support a concept called "Open Architecture Networking". This meant that the network providers were free to choose the structure or architecture of the networks themselves. They could  then be made to interwork with other networks through a meta-level "Internetworking Architecture" using computers as gateways between the different networks and making hosts responsible for end-to-end transmission of packets together with error correction and retransmission if necessary. So long as one had a gateway, hosts utilising the protocols could communicate with any other host in any other network utilising the same protocols.

            The idea of open-architecture networking was first introduced by Robert Kahn shortly after he arrived at DARPA in 1972. It was Kahn who decided to develop a new version of the NCP protocol which could meet the needs of an open-architecture network environment. In 1973 Kahn asked Vint Cerf to assist him with the design of the new protocol. Cerf had been deeply involved in the development of the NCP protocol. The new protocol is principally two separate protocols, however they are both so necessary that they are customarily referred to as one, TCP/IP. TCP breaks information into different packets at its source and then puts them together at its final destination. IP does the addressing of the information ensuring that the packets can be sent across multiple nodes and networks using different standards. The new protocols usage became wide spread as other networks began using TCP/IP to communicate with ARPANET. Another key idea behind TCP/IP software was to make it public-domain, which means that the use of, and access to, the code is made available to anyone wishing it. 

         The following is a list of events which I believe were especially significant to the expansion of the Internet. This list is by no means meant to represent all significant events in the development of the Internet.

• January 1, 1983 ARPANET makes the transition from NCP to TCP/IP
• 1984 The British JANET announces its decision to use TCP/IP
• 1985 The U.S. National Science Foundation announces its decision to use TCP/IP for its NSFNET program
• 1985 US Federal Agencies share the costs of developing a common infrastructure. The Federal Networking Council is formed to coordinate this sharing 
• 1985 Through the "Coordinating Committee on Intercontinental Research Networking" international cooperation is established with European research communities such as RARE, TCP/IP is the protocol of choice
• 1988 The NSFNET instructs its regional networks to seek commercial, non-academic customers to help subsidise there expenses, thereby lowering rates for everyone and expanding there services
• At the same time the NSFNET through the enforcement of the "Acceptable Use Policy" stimulated the development of private long-haul networks by denying usage of the NSFNET on a national scale unless that usage was in support of Research and Education