Tuesday, 30 July 2013

A Short History of Internet

The Internet and the WWW are not the same thing! The Internet is a communication structure which began in the 1960's and the WWW is a protocol, a particular way of communicating and displaying content which is only some 10 years old.

How does the Internet work?
The Internet is a packet-orientated network. That means that the data you transfer is divided in packets. This principle is not new, it was already used in the 1960s. So what happens when you transfer data across the Internet's various networks?
The networks are linked by special computers, the so-called Routers. A Router checks where your packet (your data) goes and decides in which direction to send it. Of course not every Router is linked with every other Router, they just decide on the direction your data takes.
So if the Routers know where the data is going, there must be some kind a address. Of course, there is an address, namely the IP - protocol. As I mentioned above, the data transferred with IP is divided in packets. This is handled by another protocol, the TCP.
It was soon discovered that the IP - addresses (that are, in fact, just numbers) are of easy to handle for computers, but not for us humans. So the Domain Name System was introduced in 1984.

A little history

In 1964 Paul Baran of the RAND corporation proposed the principles of a new network which was to be built for robustness and flexibility. This new network would have no central authority. It would continue to operate if some of the network was damaged or destroyed. The principles of this network were that all the nodes would be equal in status, each could send and receive messages.All the messages would be sent in packets, each with its own address. These packets would be sent at one node and would arrive at another one. This may seem rather obvious, but what was new was that the way the packets went through the net was not important. That means that if one node was destroyed, the rest of the nodes would still be able to communicate. This is of course inefficient and rather slow, but extremely reliable. The Internet still uses this method nowadays, and there has been only one collective crash so far.

The first test network built on these principles was installed in National Research Laboratory in Great Britain in 1968. Shortly afterwards, the Pentagon's Advanced Research Projects Agency (ARPA) wanted to installed a more advanced network based on the same principles in the USA. The network consisted of four high speed computers. In 1969, the first node was installed in UCLA.

By 1971 there were 23 nodes on ARPANET: The first node (1969) was in UCLA, other nodes were in the Stanford Research Institute, the University of Utah and the UCSB.
ARPANET was constructed because computer time was precious and expensive at that time and the ARPANET offered the scientists possibilities to share their computers using long distance computing. This is nearly unbelievable nowadays, for instance a normal PC has 256 Megabytes of RAM today. This is very sharp contrast to the University of Utah's computer. This Honeywell 516 mini computer had 12 Kilobytes of RAM!

1972 was a key year. Ray Tomlinson of BBN invented the first e-mail program. Scientists used it for communicating with each other, of course for sharing results of their experiments and as humans do, gossiping! (Each user had his/her own e-mail address.)

The first international nodes were set up in 1973. These were located in England and Norway. The growth of ARPANET was possible because you could use any platform to connect to it. (This is still the case with today's Internet.)
In 1974 Vint Cerf and Bob Kahn published "A protocol for Packet Network Internetworking" which specified the design of a TCP.

UUCP (Unix-to-Unix Copy) was released in 1976.

USENET was established using UUCPin 1979.


The TCP/IP protocol was established for ARPANET in 1982. This protocol became standard (instead of NCP) on 1st January 1983. The name "Internet" was first used in 1983.
ARPANET split into ARPANET and the military segment, MILNET. MILNET became integrated with the Defense Data Network created the previous year. The new protocol standard and even more the split-up were important cut-overs for ARPANET, keeping in mind that it was originally created for military purposes.
Thanks to TCP/IP and its decentralised structure, ARPANET grew and grew during the early eighties. The Name Server system was developed at the University of Wisconsin.
The number of hosts broke 1.000 in 1984 and the Domain Name System (DNS) was introduced.
In 1986 the National Science Foundation (NSF) wanted to make supercomputers useable for research projects, so they decided to link five super-computing centres. First they wanted to use ARPANET for connecting the computers, but ARPANET's bureaucracy and shortage of staff kept NSF from using this solution.
So the NSF built their own network using the IP-protocol of ARPANET. NSF linked the five centres. (56 kps). But apparently they could not link the universities with this network, simply because they didn't have enough money for building cables to every university.
The solution: The schools and universities of one region were linked together and this network was linked to one of the supercomputers.
The "traffic" in this network increased steadily and so the computers and the lines were soon to slow to handle the massive amount of data.

NSF signed a contract with Merit Networks to increase the performance of the network in 1987. The computing centres and lines have been upgraded ever since.
We now had 10.000 hosts. Two years later this number increased tenfold to 100.000 hosts.

In 1990 ARPANET ceased to exist, but its users scarcely noticed that because ARPANET's functions were continued.
WAIS and Gopher were released in 1991.
The WWW was invented at CERN (the European institute for particle physics) situated in Switzerland. Originally, WWW was developed only for scientific communication.

The widely acknowledged father of the WWW is Tim Berners - Lee. Tim BL was the driving force behind the development of the WWW. He wrote the first WWW client and the first WWW server and defined standards such as URL, HTML and HTTP while working at CERN. Prior to that, he was a founding director of Image Computer Systems and a principal engineer with Plessey Telecommunications in Poole, England. He is now a scholar at MIT.

The WWW was released by CERN in 1992 and the number of hosts broke 1.000.000.
One year later, the first browser, Mosaic, was released. The growth rate of Internet was an incredible 341% and it still grows exponentially although there is a platoing effect.

In August of 1991 files were available for download on four newsgroups (alt.hypertext, comp.sys.next, comp.text.sgml and comp.mail.multi-media).
By October there were mailing lists, namely www-interest@info.cern.ch and www-talk@info.cern.ch.

January 15th 1992 the first line mode browser was made available by anonymous FTP.

March 1993 the WWW measured 0.1% of the NSFNET backbone traffic.
By September that year the WWW measured 1% of the NSF backbone traffic! In December, the WWW won the IMA award and the New York Times wrote an article about it.

In May 1994 the first International WWW Conference (also known as "The Woodstock of the Web") was held. VRML was conceived at this event.
In the same year the IW3C2 (International WWW Conference Committee) was founded by NCSA and CERN in Boston.

On December 14th, the first W3 Consortium meeting was held in Cambridge (USA). On 16th, CERN decides not to continue WWW development due to budget conditions and transfers the WebCore project to INRIA (Institut National pour la Recherche en Informatique et Automatique, France).

Source: http://www.olinda.com/Webdesign/1_History/history.htm