TABLE 1.2
| Number of Internet worms, viruses, and hacking attempts reported, 1988–2003 | ||||||||||
| Total incidents reported (1988–2003): 319,992 | ||||||||||
| An incident may involve one site or hundreds (or even thousands) of sites. Also, some incidents may involve ongoing activity for long periods of time. | ||||||||||
| SOURCE: "Incidents Reported," in Cert/CC Statistics 1988–2004, CERT Coordination Center, Carnegie Mellon Software Engineering Institute, February 13, 2004, http://www.cert.org/stats/cert_stats.html#incidents (accessed October 25, 2004). Reproduced by special permission of the Carnegie Mellon Software Engineering Institute. | ||||||||||
| 1988–1989 | ||||||||||
| Year | 1988 | 1989 | ||||||||
| Incidents | 6 | 132 | ||||||||
| 1990–1999 | ||||||||||
| Year | 1990 | 1991 | 1992 | 1993 | 1994 | 1995 | 1996 | 1997 | 1998 | 1999 |
| Incidents | 252 | 406 | 773 | 1,334 | 2,340 | 2,412 | 2,573 | 2,134 | 3,734 | 9,859 |
| 2000–2003 | ||||||||||
| Year | 2000 | 2001 | 2002 | 2003 | ||||||
| Incidents | 21,756 | 52,658 | 82,094 | 137,529 | ||||||
Internet had its origins on October 4, 1957, when the Russians launched the Sputnik satellite into orbit with four military rockets. The news of Sputnik, a beeping steel sphere the size of an ottoman, sent the United States into a frenzy. At the time, the United States and the Union of Soviet Socialist Republics (USSR) were engaged in what has become known as the cold war, a period of sustained military buildup and ideological conflict, and Americans were fearful that Soviet satellite technology could be used to spy on the United States or to launch missile attacks on American targets. The one advantage the United States thought it had, technological superiority, now seemed tenuous.
In response, the U.S. government formed the Advanced Research Projects Agency (ARPA) within the Department of Defense a year after the Sputnik launch. The central mission of this new agency was to develop state-of-the-art technology to stay well ahead of the Soviet Union. One of the first things on ARPA's agenda was to create a system by which the ARPA operational bases could communicate with one another and their contractors via computer. They wanted the system to be resilient enough to survive a nuclear attack
John Licklider, a scientist at the Massachusetts Institute of Technology (MIT), was appointed to oversee the computer research program at ARPA in 1962. He conferred with some of the leading minds in networking technology at the time, including the MIT graduate student Leonard Kleinrock and Paul Baran of RAND. Their solution, first published in 1967, was a nationwide network of ARPA computers known as ARPANET. In this network, a user on any computer terminal in the network would be able to send a message to multiple users at other computer terminals. If any one computer was knocked out in a nuclear attack, the remaining stations could still communicate with one another.
For this network to function properly, the researchers established that the computers would have to transmit information by first breaking the information down into discrete packets. These packets were then to be sent along high-speed phone lines and reassembled upon reaching their destination at another computer. At the time, telephone conversations traveled across dedicated telephone wires in one long stream of data from one user to another like a single train traveling along a track. While this was adequate for chatting with distant relatives, it did not work well when one computer attempted to send data to multiple other computers on the network. By packetizing information, the information became much more flexible. Much like cars on a highway, the packets could be routed easily to multiple computers. If one packet of information went bad in transmission, it did not disrupt the stream of data transmitting from one computer to another and could easily be resent. Packets could also carry information about themselves and where they were going; they could be compressed for speed; and they could be encrypted for security purposes.
After two years of engineering the parts needed for ARPANET, the researchers at ARPA set up the first four computer centers in the network. They were located at the University of California, Los Angeles (UCLA), Stanford Research Institute, the University of California, Santa Barbara (UCSB), and at the University of Utah. Between these nodes, AT&T had laid down telephone lines capable of transmitting data at 50 kilobytes per second. The first test of the system commenced on October 29, 1969, when Charley Kline at UCLA tried logging into the Stanford system. Upon encountering the letter "G" in the word "LOGIN," the system crashed.
A Loose Affiliation of Networks
Needless to say, the researchers at UCLA worked out the problems, and a little over two years later, the ARPANET was fully functional and had forty nodes linked to it. Figure 1.2 shows the ARPANET in September 1971. Throughout the early and mid-1970s, the development of networking technologies progressed slowly. Ray Tomlinson invented the first e-mail program in order to send typed messages across the network in 1971, and a year later, the first computer-to-computer chat took place at UCLA. Bob Metcalfe at Xerox in 1973 developed the first Ethernet to connect computers and printers in a large organization together. Mike Lesk at AT&T Bell Labs in 1976 put together a program entitled Unix-to-Unix-copy protocol (UUCP) that allowed Unix computers, typically used by academics, to communicate with one another over the phone lines.
Such technological developments as these allowed people and organizations that were not connected through ARPANET to set up networks of their own by the late 1980s
FIGURE 1.2
and the early 1990s. One of the largest of these was the Computer Science and Research Network (CSNET), established by a number of universities with help from the National Science Foundation (NSF). These universities, which included the University of Wisconsin and Purdue, recognized the advantages in resource sharing and communication the ARPANET provided the Ivy League and West Coast schools and wanted to develop similar capabilities. Usenet was established initially to connect researchers at Duke and the University of North Carolina and eventually spread all over the country. BITNET was formed to connect computers in the City University of New York system. Most of these smaller networks used standard telephone lines to operate. They were primarily set up to transfer scientific data, share computing resources, post items on bulletin boards, and provide e-mail.
One big problem was that these different networks could not readily communicate with one another. Each network used different methods to identify the computers within the network. A computer in one network could not even recognize the computers in different networks, and information packets sent out from one network could not navigate the other networks. The situation would be analogous to a state in the United States having its own postal services and unique postal address system indecipherable to people living outside of that state. A letter sent to a relative in that state would never reach its destination.
During the 1970s, two clever engineers, Vint Cerf and Bob Kahn, devised the Transmission Control Program and the Internet Protocol (TCP/IP). This suite of programs created a universal address system that could be installed on any existing network. Once installed, the machines on the network could recognize and send information to a machine on any other network, provided they also had TCP/IP. In 1983 ARPANET started using TCP/IP, which was already being used by CSNET. The merger created the first collection of interconnected computer networks, and the "Internet" was born. To this day, each machine on the Internet has a unique IP address that identifies that machine on a network. Servers typically have permanent IP numbers assigned to them, while most personal computers are given a different number by an Internet service provider (ISP) each time the user begins a new session.
In the year the Internet was born, home computing was still in its infancy. Commodore 64 had just made its debut, sporting a one megahertz microprocessor and 64 kilobytes of random access memory (RAM). Relatively few people owned home computers. Most of them used their machines for basic business applications, such as word processors and spread sheets, and for playing games. Home users did not have direct access to the Internet. Relatively low speed modems were widely available in the mid-to late 1980s, and people could dial directly into servers owned by CompuServe, Quantum Computer Services (later to be renamed America Online [AOL]), and Prodigy. These services allowed people to post messages, go into chat rooms, play games, or send and receive e-mail. None of these services connected to the Internet, and e-mails could only be sent among people on the same service.
The only people who could surf the Internet freely were those who had access to powerful mainframe computers through a university, the government, or a large corporation. The Internet, however, was a very uninviting place in the 1980s. Users connecting to the Internet had to know exactly what they were looking for to get it. To reach another computer or server on the Internet, users had to key in the IP address for that computer, which consisted of a string of up to twelve numbers, such as 216.183.103.150. To maneuver around on a server, commands had to be typed in computer code on a prompt line, and cryptic directories had to be sifted through. There were no Web browsers, colorful Internet pages, or search engines.
By 1984 the dedicated name server (DNS), developed by the University of Wisconsin, was introduced, making the Internet somewhat more user friendly. A DNS is a computer server on the Internet with a database that pairs domain names with IP addresses, giving people the ability to type in a name instead of a twelve digit number to reach a destination on the Internet. An Internet browser in 2004 contacts one of many DNSs each time an address, such as www.yahoo.com, is entered into the address bar. Most Internet service providers have a name server filled with the names and IP address numbers of widely used sites. Once the browser makes the request of the name server, the name server sends back the IP address number, which for Yahoo is 216.109.118.70. The Internet browser then uses this IP address to access the site the user wants to visit (Yahoo in this case).
Along with these name servers, a dedicated name system was also put into place so that no two names would be the same. Domain names with a minimum of two levels were decided on. The top-level designated the country or economic sector a computer is in (.com or .gov), and a unique second-level domain name designated the organization itself (NASA or Google). Host names, such as "www," specified the actual machine in the domain. Information Sciences Institute was put in charge of managing the root dedicated name server in 1985 for all domains to make sure that no two were alike and to track who was registered for what name. Some of the first domain names to be registered were symboics.com, mit.edu, think.com, and berkeley.edu.
A Major Expansion in the Mid-1980s
In 1986 Internet use expanded exponentially when the NSF installed new super computers and a new backbone for the U.S. Internet service, giving rise to NSFNet. Today, Internet service providers (ISPs) and cable companies have their own backbones, which are all tied into one another. When a home user connects to the Internet via phone, digital subscriber line (DSL), or cable, they connect to a bank of modems called a point of presence (POP) that is owned by the service provider (See Figure 1.3.) Each POP from each service provider, be it AOL or Comcast, feeds into a citywide network access point (NAP). These NAPs are connected to one another via backbones that consist of bundles, or trunks, of fiber optic cables that carry cross-country transmissions. The first NSF-funded backbone consisted of 56 kilobytes per second wire to connect the access points. The wire was laid down by AT&T. The NSF also provided five supercomputers to route traffic between the NAPs and bundles. In 1988 the NSF upgraded NSFNet when they installed supercomputers that could handle 1.544 gigabytes of traffic per second and fiber-optic line that could transfer information at 1.5 megabytes per second. Computers are only able to recognize and deal with information in digital binary form, wherein all data is made up entirely of ones and zeros. Computers process the long lines of complex, binary computer code in bite-sized quantities known as bytes. Each byte consists of a string of eight ones and zeros that can be used to represent binary numbers from zero to 255. In binary, one is 00000001, two is 00000010, nineteen is 00010011, and 255 is 11111111. A thousand bytes equals a kilobyte, a million bytes equals a megabyte, and a billion bytes equals a gigabyte.
The creation of the NSFNet ended the transmission bottlenecks that existed in the early Internet. The network also provided access for most major research institutions and universities. Academic departments and government agencies across the country jumped at the chance to set up servers and share information with their colleagues. From 1986 to 1987 the number of hosts (machines with a distinct IP address) on the Internet jumped from 5,000 to nearly 28,000.
The NSF strictly prohibited the use of its site for commercial purposes. Though such a rule may seem harsh, it had the intended consequence of fostering the development of private Internet providers. In 1987 UUNET became the first commercial Internet provider, offering service to Unix computers. Nineteen eighty-nine saw the introduction of The World, which was the first commercial provider of dial-up access, and computer scientists at McGill University in Montreal invented Archie, the first Internet search engine for finding computer files.
An Internet for Everyone
In 1991 Tim Berners-Lee at CERN, the European Organization for Nuclear Research, located in Switzerland, introduced the three technologies that would give rise to the World Wide Web. The first of Berners-Lee's technologies was the Web browser, a program that allowed a user to jump from one server computer on the Internet to another. The second was the hypertext markup language (HTML), which was a programming language for creating Web pages with graphics and links to other Web pages. The third was the hypertext transfer protocol (HTTP), a command used by the browser to retrieve the HTML information contained on the server's Web site. In concert, these three innovations led to the Web as it became known in the early twenty-first century. On his server, Berners-Lee created the first Web site at CERN. As the technology spread, many more Web servers and sites quickly came into being.
With these technologies, people were no longer required to use complex computer codes and sift through cryptic directories to retrieve information from other computers on the Internet. A user with the browser on his or her computer was able to simply type in the name of a server on the Internet along with the HTTP command (i.e., http://www.yahoo.com), and the browser began the process of gathering information automatically. The browser first contacted a DNS server to get the IP address of the server. Once the browser connected with the Web server, the browser sent out the HTTP command. The HTTP told the server to send the browser the HTML code for the specified Web site. Upon receiving the HTML code, the browser read the code and simply formed the page the owners of the server wanted the computer user to see (Yahoo's home page, for instance).
At the same time that these great strides were being made in establishing the Internet, the U.S. government began to take a more active role in its development. In 1991 Senator Al Gore (D-Tennessee) introduced the U.S. High Performance Computing Act into Congress. The act set aside over $2 billion for further research into computing and to improve the infrastructure of the Internet. Though most of it was earmarked for such large agencies as the NSF and NASA, some of the funds were placed into the hands of independent software developers.
Marc Andreesen used a federal grant received through this act to develop the Mosaic X Web browser, which was released in 1993. The browser was one of the first commercial browsers to employ the HTML program language and the HTTP, and it became the first browser to be embraced by the general public. It was easy to set up, easy to use, and backed by a full customer support staff. It displayed images in an attractive way and contained many of the standard features used on Microsoft's Internet Explorer today, such as the address prompt and "back" and "forward" buttons. Tens of thousands of copies of Mosaic X were sold.
Once Mosaic X became popular, more and more Web sites employing HTML and HTTP were posted. According to Internet historian Robert H. Zakon (http://www.zakon.org/robert/internet/timeline/), in June 1994 there were approximately 2,700 Web sites; by June 1995 there were an estimated 23,500. Growth in hosts during the same period reflected an increase from 3.2 million in July 1994 to 6.6 million in July 1995. The Web was growing at such a rapid rate that the NSF created the Internet Network Information Center (InterNIC) as an agency to handle domain names. The InterNIC contracted with Network Solutions to handle domain registration. By 1995 the companies that ran the older dial-up services for home users, such as CompuServe, American Online, and Prodigy, brought their clients to the Internet and offered Internet service for all. Internet network providers, such as MCI and Qwest, began laying fiber optic cables and communications networks at a breakneck pace. Advertising appeared on the Web for the first time (the first banner being for ZIMA), e-shopping appeared on the Internet, and many companies such as Netscape went public.
In the following years the Internet has gained a firm foothold in American life. As of July 2004, Zakon estimated the number of Web servers at more than 52 million, and the number of hosts, computers with a registered IP address, at more than 285 million.
User Comments Add a comment…