Home

New networks and network standards appear

One of the limitations that the Department of Defense put on ARPAnet was that only universities conducting defense-related research could use their network. This limitation was the driving factor in the creation of Usenet and other inter-university networks. Another such non-ARPAnet university network was formed in 1981 by CUNY and Yale Universities. The students called it BITnet ("Because Itıs There" network) because the network connected the computers using the protocol that was packaged ("there") with IBMıs mainframe computers. Again, BITnet used existing telephone lines to connect the computers together, and given its low cost many other institutions joined the network. By 1990, over 3000 institution-sites were connected to the network enabling them all to exchange e-mail (again note that this communication was most likely not totally for research purposes) (Moschovitis). The National Science Foundation also recognized the limited access of educational institutions. Their first answer to this problem was a grant given to the University of Wisconsin, Madisonıs chair of the Computer Science Department, Larry Landweber. The network that he created in 1981, CSnet, was chiefly used by academic and industrial researchers in the field of computer science. The network was in fact a precursor to the forthcoming NSFnet.

ARPAnet in Oct., 1980

The mid-eighties brought many technical advances that created many of the Internet standards that are still used today. In 1983 ARPAnet forced all of its servers to switch to the TCP/IP protocol. Many see this as another defining moment because it standardized the protocol that the Internet uses to this day; since ARPAnet led the way, many other networks followed. Also in that year, name servers were introduced to the network, again at the University of Wisconsin under the supervision of Larry Landweber. <T> The job of the name server was to translate the numbers of a physical network address (in the form of 123.456.789.000, standardized by TCP/IP) into actual names that were understandable to the users. Each machine would have its own name that ARPAnet users could remember, however each server also had to keep a list of the hosts that a user could connect to. The implication of this was that any site that a serverıs administrator did not think was appropriate (or perhaps contained commercial content) could be excluded from the list, thus making it very difficult to route to a "blacklisted" server. This process was made much simpler in 1984 when Domain Name Servers (DNS) replaced the name servers. The advantage of DNS was that if oneıs own server did not know the address of a particular name, it would forward the request to another server or to a centralized list. This eliminated the need for each server to hold large files and allowed the number of Internet servers to grow more rapidly. The use of DNS was also aided by Cisco Corporationıs introduction of Internet routers in the same year. <T>

E-mail the
Author / Webmaster

©2000 John Thomson, Jr


Next...