Sunday 19 November 2006

History and Evolution of the Internet Part II

At first, it was thought that the network could take advantage of the existing telephone network infrastructure for this kind of information routing. However, when the first remote computers located at MIT and Berkeley were linked together, it was found that the phone lines were just too slow to allow for a successful transfer of data and the running of programs between the two locations. But the experiment was successful in one regard, this was the first time two computers had been linked to each other to form a ‘Wide Area Network’ or WAN. Finally, in 1966-67, the new head of computer research at APRA, Leonard Roberts, proclaimed a blueprint for an interconnected computer network.

It was to be called the APRANET. Following this announcement it was discovered that a few other research institutions were also studying the advantages of computer networking for speedier transfer of data and communication. Independent of each other, and in fact in total ignorance of similar research, scientists at MIT, the RAND Corporation and the UK based National Physics Laboratory were working in the same field. The first functional version of the ARPANET integrated within itself the best features of all these investigations.

This was just the hardware part. Now the need was to develop a suitable software platform which could integrate these computers and enable the transfer of data between them. This was the impetus towards the development of the ‘interface message processor’ or IMPs, the work on which was finished only in 1968. Finally in 1969, IMPs were loaded on to the computers located at the remote locations of UCLA and Stanford. This was the first computer network of its kind which enabled the students at UCLA to ‘login’ in to the Stanford computer, gain access to its database and transfer data between the two locations. The success of this experiment led to the addition of four more host computers to ARPANET and the establishment of research centers at Utah and Santa Barbara. By the December of 1971, ARPANET had linked together 23 host computers. This was the first true fully functional computer network. At the First International Conference on Computers and Communication in the October of 1972, the ARPANET was presented to the public. Another resultant development of this conference was the formation of the IWG or the Internetworking Working Group. Its main task was to coordinate the research happening in this field.

Simultaneously, research was being carried out on the means to increase the functionality, flexibility and expansion of this system. 1972 marked the launch of the prototype of the program we know today as e-mail. The early 70s also mark the development of host-to-host protocols which would enable two remote computers to be merged together so that they function as one host (though only for the period that they are linked together). However, the most important innovation of this period was the development of the ‘transmission control protocol/internet protocol’ or TCP/IP.

This was a language which would enable disparate computer networks to interact with each other on a common platform. Finally we come to the one innovation which definitively determined the present nature of the internet as we know it today. The scientists working on computer networking decided that such a network would have an ‘open architecture’. In this sense, they remained true to the basic idea of Licklider’s ‘Galactic Network’. Firstly, in an ‘open architecture’ system, every network is given the freedom to develop applications which would be specific to that particular network. At the same time, ‘open architecture’ insured that these networks would not need to undergo any modification to join the larger complex of interconnected computers.

No comments: