With help of the experience from the mini-network, which has been set up at NPL by Donald Watts Davies one year earlier, the plans for an ARPA network are getting more precise.In 1968, the research department of ARPA, governed by Licklider, calls for proposals of the different components needed for the network.
The first IMP
Stanford Research Institute (SRI) receives the assignment of enunciating the specifications for the new network. BBN (Bolt, Beranek and Newman) gets the tender for developing the technology for packet switching, to be exact, for the development of Interface Message Processors (IMPs) in December. Robert E. Kahn was responsible for their crucial system design of the ARPANET. Only a short time before, Kahn asked for time off from MIT (Massachusetts Institute of Technology) to work for “Project ARPANET” all out.
The purpose of the IMPs, which can be seen as the precursor of a router, was to establish a connection on the first layer between two computers (hosts) connected via a telephone line.