History of the internet

from Wikipedia, the free encyclopedia
This article or section needs to be revised. Details should be given on the talk page. Please help to improve it , and then remove this flag.

The history of the internet can be divided into three phases.

In the early phase from the mid-1960s, the foundations were laid, the technology demonstrated and developed for applicability.

At the end of the 1970s, at the same time as the switch from military to academic research funding, the Internet began to grow and expand internationally . During this time, what is commonly associated with the wild phase of the original Internet happened : an economy of exchange for software and information, a grass - root -based self-organization, developing communities and the hacking spirit that knows how to bypass any restriction on access and the free flow of information .

In 1990 the commercial phase of the Internet began with the shutdown of the Arpanet . It is estimated that in 1993 the Internet accounted for only 1% of the information flows of the global telecommunications networks , while in 2000 it already dominated the majority of technical information exchange (51%) and in 2007 it was already clearly dominant (97% of bytes in the world exchanged).

Although a chronological presentation predominates in the article, it is primarily structured thematically. A chronological listing of the events can be found in the article Chronology of the Internet .

General

This article or the following section is not adequately provided with supporting documents ( e.g. individual evidence ). Information without sufficient evidence could be removed soon. Please help Wikipedia by researching the information and including good evidence.

In terms of media history, the Internet is an anomaly . Usual models of media and technology genesis in general run from the laboratory through development to application maturity and social implementation either as state military or administrative communication, as an economic control and management instrument or as a mass product of individual communication or the mass media . It is different in the case of academic data networks . In the first few years there was no separation between inventors, developers and users .

The computer science has on the net not only their subject of research, but also their communication - and publishing medium . It is both an infrastructure and a development environment that is expanded from within. Innovations are thrown around by the developer-users in the beta version , i.e. without guarantee and at their own risk, tested by colleagues and further developed. In addition, it provides the other, increasingly computerized, sciences with the same infrastructure. Access to computing resources , the exchange within a global community of peers, the To-Talk sites of preprints , publication of conference presentations and databases on the Internet - all part of the daily practices in since the 1980s, Physics and Astronomy , the Computer science itself and increasingly also in the softer sciences. After all, passing on the basic tools to the students is part of the academic teaching. Since the network, unlike most laboratory equipment, does not have a narrowly defined area of ​​application, but is a medium, student, private and leisure cultures also encounter an explosive mixture of high tech and hobbyism , science and science fiction , hackers and hippies .

Thought leader

The earliest vision of a possible worldwide computer network is a short story by the science fiction author Murray Leinster , who in 1946 in his story A Logic Named Joe was one of the first to describe a personal computer and an early vision of the Internet; There he wrote, the computer ... handles the distribution of ninety-four percent of all television programs, provides all information about weather, air traffic, special offers ... and documents every business conversation, every contract ... The computers have changed the world. Computers are civilization. When we turn off the computers, we fall back into a kind of civilization that we've forgotten how to do.

Early phase

In the late 1950s, JCR Licklider headed a research group at US armaments supplier Bolt Beranek and Newman (BBN), whose work was based on the PDP-1 minicomputer from Digital Equipment Corporation (DEC), one of the first time-sharing systems. However, computer manufacturers and most of the IT establishment believed that time-sharing was an inefficient use of computer resources and should not be pursued further. Licklider's argument was reversed, that computers are too fast and too expensive for real-time interaction for “cooperative thinking with a person”, which is why they have to divide their time between many users. Licklider was also the architect of the MAC ( Multiple-Access Computer , also known as Machine-Aided Cognition or Man And Computer ) project at the Massachusetts Institute of Technology (MIT). In 1962 he moved from BBN to the Advanced Research Projects Agency (ARPA) of the US Department of Defense , where he became head of Command and Control Research (CCR), which he immediately renamed the Information Processing Techniques Office (IPTO).

His experience with time-sharing systems enabled him to redefine from the computer as a calculating machine to the computer as a communication device. As head of the ARPA research department, he was now able to bring this paradigm shift into effect in network planning:

“The ARPA motto is that the possibilities offered by the computer as a communication medium between people dwarf the historical beginnings of the computer as a calculating machine. […] Lick was one of the first to notice the community spirit that arose among the users of the first time-sharing system. By pointing out the community phenomenon that arose in part from the sharing of resources in a time-sharing system, Lick made it easy to imagine a connection between communities, the linking of interactive online communities of people [...] "

At the same time, there was a paradigm shift in telecommunications technology from line-oriented to packet-switched concepts. It went back to parallel work by Paul Baran at the Rand Corporation (the first think tank , founded in 1946 by the US Air Force ) and by Donald Watts Davies at the National Physical Laboratory in Middlesex, England. Paul Baran published a paper in 1960 in which he took an impending nuclear war as an opportunity to develop ideas for a network that could survive such a war. The breakdown of communications into small data packets, which, provided with the destination and sender address, find their way through the network almost autonomously, was a prerequisite for the distributed, decentralized architecture of the Internet. It was also the point at which the spirits of the computer and telecommunications world parted.

The telephone operators of the time were quite interested in data communication and packet switching after it had been proven that this technology was not only feasible at all, but that it used the available bandwidth much more economically than line switching , but the primary design criteria of the national monopolies were nationwide Network security, quality of service and billing. They saw this only guaranteed by a centrally controlled network with dedicated line usage for each individual communication.

The telecommunications companies, especially in England , Italy , Germany and Japan, therefore applied a virtual channel structure to the unpredictable packet flows . In this system, too, packets from different connections are transported on the same physical line, but only up to an upper limit, up to which the capacity for each individual connection can be guaranteed. In addition, this network is not distributed, but switched via central exchanges. The specifications of this service were negotiated within the framework of the International Telecommunication Union (ITU) and standardized in 1976 under the designation X.25 . The Bundespost offered it under the name Datex-P . This creates the contrast between a rhizomatic network that grows decentrally from individual nodes on the basis of a military calculation, and a hierarchical, tree-shaped structure that is centrally planned and managed.

The ARPA research department under Licklider wrote out the various components of the new network. The Stanford Research Institute (SRI) was commissioned to write the specifications for the new network. In December 1968 the SRI presented the report A Study of Computer Network Design Parameters . At the same time, Douglas C. Engelbart and his group at SRI were already working on computational techniques for promoting human interaction. It was therefore decided that the SRI was the appropriate place to set up a Network Information Center (NIC) for the ARPAnet. The DARPA tender for a Network Measurement Center went to the University of California, Los Angeles (UCLA), where Leonard Kleinrock worked, who had written his doctoral thesis on queuing theory. Vinton G. Cerf , Jonathan Postel and Steve Crocker also worked on the UCLA team at the time .

Vinton Cerf (2010)
Robert E. Kahn

BBN was awarded the contract for the development of the packet switching technology, more precisely an Interface Message Processor (IMP). Worked there, among others , Robert E. Kahn , who came from MIT and goes back to the much of the architecture of the Internet (TCP / IP). The IMPs, forerunners of today's routers , had the task of establishing the lowest connection layer between the computers ( hosts ) networked via telephone lines . The first IMPs were shipped in May 1969.

The Internet started in the fall of 1969 when the first four mainframes at UCLA, SRI, the University of California at Santa Barbara ( UCSB ) and the University of Utah were connected. On October 29, 1969, “Io” was the first successful Internet message to be sent tentatively from UCLA to the Stanford Research Institute.

The first of thousands of Request for Comments documents ( RFCs ), which specify the technical standards of the Internet, had already appeared six months earlier . These standards are not enacted in the style of a law, but as a friendly request for comment. As the author of the first RFC, Steve Crocker justified this form with the fact that the participants were only doctoral students without any authority. They therefore had to find a way to document their work without it appearing as if they wanted to impose anything on anyone, i.e. in a form that was open to comments. RFCs can be created by anyone. They are intended as discussion papers with the stated aim of breaking the authority of what is written. In addition to the mostly technical texts, the philosophy (for example RFC 1718 ), the history ( RFC 2235 ) and the culture of the network are recorded and sometimes even composed ( RFC 1121 ). The free availability of the specifications and the associated reference implementations were a key factor in the development of the Internet. A year later, the first RFC resulted in the Network Control Protocol (NCP), a set of programs for the host-host connection, the first Arpanet protocol.

In 1971 the network consisted of 14 nodes and grew by one per month. After completion of the NCP and implementation for the various architectures, the higher-level services Telnet ( RFC 318 ) and FTP ( File Transfer Protocol , RFC 454 ) were created. Ray Tomlinson of BBN modified an email server program for the Arpanet and invented the user @ host convention. Larry Roberts wrote a mail client for this .

The network was impressive. It was time for a first public demonstration, held in 1972 at the International Conference on Computer Communications in Washington. In the basement of the conference hotel, a packet switching computer and a Terminal Interface Processor (TIP) were installed, which, unlike an IMP, could process the input from several hosts or terminals. 40 machines were connected across the United States. The demonstrations included interactive chess games and the simulation of an air traffic control system. The conversation between ELIZA , Joseph Weizenbaum's artificial-intelligent psychiatrist at MIT, and PARRY, a paranoid program by Kenneth Colby at Stanford University, became famous. Participants from England , France , Italy and Sweden were there. Representatives from AT & T , however, attended the conference, they left in deep confusion. In the same year, projects for radio and satellite-based packet networking started, the latter with institutes in Norway and England. In his doctoral thesis at Harvard University, Bob Metcalfe outlined the concept for a Local Area Network ( LAN ) with multiple access channels, which he called Ethernet . He developed the concept further at Xerox PARC before later founding 3Com .

Arpanet , SATNET and the radio network had different interfaces, packet sizes, tags and transmission rates, which made it difficult to connect them together. Bob Kahn , who went from BBN to DARPA, and Vint Cerf , who was now teaching at Stanford University, began developing a protocol to connect different networks together. They were based on the developments of the CYCLADES project. In autumn 1973 they presented the first draft of the Transmission Control Protocol (TCP) at a meeting of the International Network Working Group in England . In the following year, TCP was implemented simultaneously at Stanford University , BBN and University College London (Peter Kirstein).

"The efforts to develop the Internet protocols were therefore international from the start."

- Cerf 1993

Four iterations of the TCP protocol set followed. The last one appeared in 1978.

In 1974 BBN launched Telenet, the first public packet-switched data communication service as a commercial version of the Arpanet. Because of the DARPA funding, BBN had no exclusive rights to the source code for the IMPs and TIPs. Other new network companies asked BBN to release this. BBN was initially reluctant because the code was constantly changing, but released it in 1975.

growth

Internet growth interpreted based on the number of available hosts . The ordinate of the diagram has been scaled logarithmically because of the strong increase .

With the research funding for the implementation of TCP, DARPA had fulfilled its initial mission. In 1975 responsibility for the Arpanet was transferred to the Defense Communications Agency (later renamed Defense Information Systems Agency ). BBN remained the contractor to operate the network, but military security interests now became more important. In addition to DARPA, the National Science Foundation (NSF) also funded research in computer science and networks at around 120 US universities. Other institutions such as the Ministry of Energy and NASA started their own networks.

At the beginning of 1975 the Arpanet had 61 knots. The first mailing list was set up. Together with the RFCs , mailing lists are becoming the most important means of open cooperation between the technical community. However, in the most popular list of the time, science fiction was discussed. The Jargon File , a dictionary of hacker culture compiled by Raphael Finkel, was published for the first time, of course on the Internet.

UUCP ( Unix to Unix Copy ) was developed in 1976 at AT&T Bell Labs and distributed as part of Unix Version 7 . Facilities that could not afford a dedicated line made it possible for UUCP to exchange data with computers on the Arpanet via dial-up connections on normal telephone lines.

The new network-connecting TCP was demonstrated for the first time in July 1977 in a complex test setup. The transmission line began with a mobile packet sender in a moving car in the San Francisco Bayshore Freeway, ran to a gateway at BBN over the Arpanet, via a point-to-point satellite link to Norway , from there via cable to London on back the Atlantic Packet Satellite Network (SATNET) to the Arpanet and finally to the Computer Science Institute of the University of Southern California:

“So what we were simulating was someone in a mobile theater of war moving over a continental network, then over an intercontinental satellite network, and then back on a wire network and to the main data center at national headquarters. Since the Department of Defense paid for it, we looked for examples that could be interesting for a military scenario. "

- Cerf 1992

Experiments on packet-switched voice transmission have been carried out since the mid-1970s. TCP is designed for reliable transmission. Packages that are lost will be sent again. However, in the case of voice transmission, the loss of some packets is less detrimental than a delay. Because of these considerations, TCP and IP were separated in 1978. IP specifies the User Datagram Protocol (UDP), which is still used today for voice transmission (see ibid.). This officially ended the Arpanet experiment in 1978. The final report states:

“This ARPA program produced nothing less than a revolution in computer technology and was one of the most successful projects ARPA has ever undertaken. The full extent of the technical change that will result from this project may not be gauged for many years. "

One of the pioneers remembers the crucial factors:

“It was very exciting for me to participate in the development of the Arpanet and the Internet Protocols. One of the main reasons it worked, I think, is that there were a lot of very smart people, all working more or less in the same direction, led by some very wise people in the funding agency. The result was the creation of a community of network researchers who firmly believed that collaboration among researchers is more powerful than competition . I don't think any other model would have brought us to where we are today. "

Institutionalization

In order to continue the vision of a free and open network, Vint Cerf set up the Internet Configuration Control Board (ICCB) from DARPA in 1978 , chaired by Dave Clark at MIT . In 1983 the Internet Activities Board ( IAB ) ( renamed the Internet Architecture Board after the Internet Society was founded ) replaced the ICCB.

For the actual development work, the Internet Engineering Task Force ( IETF ) and the Internet Research Task Force ( IRTF ) were formed in 1986 under the IAB . In contrast to state standardization bodies or industrial consortia, the IETF is an open forum “according to law and custom”. Anyone can become a member by subscribing to one of the approximately 100 task-oriented mailing lists and participating in the discussions.

"In theory, a student raising a technically sound concern about a protocol gets the same amount of careful attention, or more, than someone at a multi-billion dollar company worried about the impact on their shipped systems." (Alvestrand, 1996, p. 61).

All work, with the exception of the secretariat, is unpaid and voluntary.

The development work within the IETF obeys a limited standard. The results must solve a pending problem as directly as possible and, in accordance with a hacker aesthetic of elegance, as simply and compactly as possible. They must work with the existing structures and provide connections for possible expansion. Since there is no well-defined membership, decisions are not made by voting. The IETF's credo is:

“We don't want kings, presidents and elections. We believe in a rough consensus and in executable code. "

If an interesting problem and enough volunteers are found, a discussion takes place, an executable code is written for alternative approaches and tested until a consensus is formed. If this does not happen, if the process encounters unsolvable problems, or if the parties involved lose interest, a standard can get stuck before it is adopted. In every phase of development, standards or code are published in the tried and tested RFC format and are accessible to anyone interested. This means that they are tested at an early stage by a large number of users under the most varied of conditions and that this broad experience is incorporated into the development process before a standard is officially released. The standards are open and freely available. In contrast to the ISO process, those involved in standardization cannot acquire patents and, unlike ISO, the IETF does not finance itself from the sale of the documentation of standards. Nothing stands in the way of the continuous further development of this knowledge.

Although it had no planned damage function, in 1988 the runaway computer worm owned by 23-year-old Robert Tappan Morris crippled 6,000 of the 60,000 Internet hosts that are now 60,000. The DARPA then formed the Computer Emergency Response Team (CERT), which since then has been dealing primarily with computer security problems as a preventive measure.

The Electronic Frontier Foundation (EFF), founded by Mitch Kapor in 1990, is not an internet institution in the strict sense, but it has played an important role in the USA as a public and lobbying association for the protection of civil rights on the internet.

Vint Cerf and Bob Kahn founded the Internet Society ( ISOC ) in 1992 as an umbrella organization for all those interested in the Internet and for the existing bodies such as IAB and IETF .

The following year, the National Science Foundation (NSF) established the InterNIC ( Network Information Center ), which outsourced certain services in its area of ​​responsibility to third parties, namely directory and database services to AT&T, registration services to Network Solutions, Inc. and information services to General Atomics / CERFnet.

Network research

On the initiative of Larry Landweber, representatives from various universities (including Peter Denning and Dave Farber) developed the idea of ​​a computer science research network ( CSNET ). An application for funding to the NSF was initially rejected as too costly. On a revised application, the NSF then approved $ 5 million over a five-year period in 1980. The protocol that connects the various subnets of CSNET is TCP / IP . In 1982 it was decided that all systems on the Arpanet should switch from NCP to TCP / IP - although only a few hundred computers and a dozen networks were affected, no simple operation (compare RFC 801 ).

CSNET and Arpanet were connected in 1983, but US scientists complained that the country's supercomputers were inaccessible. Astrophysicists had to travel to Germany to use a supercomputer made in the USA. In July 1983, an NSF working group issued a plan for a National Computing Environment for Academic Research . The supercomputer crisis led to the approval of a budget of $ 200 million for the establishment of supercomputer centers at various universities and the NSFnet. The NSFnet started in 1986 with a nationwide 56 Kbps backbone, which was soon expanded to 1.5 Mbps and in 1989 to 44.7 Mbps lines. At the time, Bob Kahn and Vint Cerf were already planning a test network with 6 gigabits per second. A number of NSF-funded regional networks developed around the NSFnet backbone. From the beginning of 1986 to the end of 1987, the total number of networks on the Internet rose from 2,000 to almost 30,000.

New architectures, protocols and services

In addition to TCP / IP, proprietary protocols continued to be used, but new open networks also emerged. The most important of these is BITNET ( Because It's Time NETwork ), which started in 1981 as a cooperative network at the City University of New York and established the first connection with Yale University. One of the peculiarities of BITNET is, for example, that it implements file transfer via email. In 1987 the number of BITNET hosts worldwide exceeded 1,000.

TCP / IP became the de facto standard, but it was not recognized as an official standard. A wrong path in network development began when the International Organization for Standardization (ISO) developed a reference model ( OSI model ) for its own connection-oriented internetwork standard called Open Systems Interconnection (OSI) from 1982 onwards. In contrast to the horizontal process of the Internet community, the ISO standardization process is based on a vertical, multi-layered process of proposals, elaborations and coordination that goes back and forth between the national standardization organizations, the working groups and finally the ISO plenum. All interests should be taken into account. The standard is intended to be complete in a theoretical sense. At the same time, it should be backwards compatible and abstract enough not to obstruct future developments. The limited circulation in the institutions involved in the process means that standards are only tested to a limited extent before they are adopted. When a standard is finally adopted, it is often enough already overtaken by technology. OSI has never developed very far from paper concepts into practical computer use and is now considered to have failed. Until the 1990s, however, the research and technology authorities of many countries, including Germany and Japan , decreed that OSI was the official and therefore the only network protocol into which research funds were to flow. Even the US government stipulated in 1988 that all computers purchased for use in government agencies must support OSI and declared TCP / IP to be an interim solution.

In 1983 the US Department of Defense decided to split the network into a public Arpanet and the confidential MILNET . Only 45 of the 113 host computers remained in the Arpanet. The number of computers connected to these hosts was of course much larger, especially due to the transition from time-sharing mainframes to workstations in an Ethernet LAN. Jon Postel first assigned numbers to the individual interconnected networks, then he developed the Domain Name System ( DNS ) with Paul Mockapetris and Craig Partridge with a first name server at the University of Wisconsin that translates names into numbers. At the same time, he recommended the user@host.domain addressing scheme that is common today. The new address system was institutionalized in 1988 with the Internet Assigned Numbers Authority (IANA), of which Postel became director.

In 1981 Bill Joy began at the University of California at Berkeley with a research assignment from DARPA to integrate the TCP / IP protocols into the free version of the Unix operating system maintained there. They were published in August 1983 in the BSD (Berkeley Systems Distribution) -Unix version 4.2. The computer and network operating systems were merged. Not least because of this, many computer companies, such as Sun Microsystems , which Joy co-founded , began to make BSD the basis of their workstations. The free software 4.2BSD spread rapidly. Thousands of developers around the world adopted it, laying the foundation for today's global Internet. In 1977, with the Tandy TRS-80 and the Commodore PET, the first computers for private use came onto the market, Steve Wozniak and Steve Jobs announced the Apple II . The IBM PC followed in 1981 and shortly afterwards the first IBM PC clones. With the cheap microcomputers and their ability to communicate via modem, a new generation of user cultures entered the IT and network world.

The integration of TCP / IP and local Ethernet drove the expansion of the Internet. Ethernet cards also became available for personal computers. In the early 1980s, students of Professor Dave Clark at MIT developed the first TCP / IP stack for MS-DOS . The source code for PC / IP and some simple network applications spread quickly and inspired many others to open up the PC to the Internet. Since the DOS operating system is not capable of multitasking, PC / IP could only support a single connection (one socket). For some applications (such as Telnet) the limitation is not a problem, but FTP requires two connections at the same time, a control and a data channel. Phil Karn , then employed by Bell Labs , began to write a new TCP / IP stack in 1985, in which he implemented multitasking within the application - a daring trick that worked. Developed for CP / M , Karn soon ported the code to DOS and, being a passionate radio amateur , also revised it for use on packet radio . He released the code for non-commercial use under the name of his amateur radio callsign KA9Q (compare Steven Baker, Desktop TCP / IP At Middle Age, Unix Review February 1998).

In 1979 Usenet was created , which was to become an internet-wide notice board. Steve Bellovin wrote some command line scripts for this, which allow a computer to call up messages on another computer via UUCP. Technically, Usenet is an early example of client-server architectures. Socially, it forms a public space where everyone can read and write on topics that encompass pretty much everything under the sun.

Another form of cooperative social space that also enables synchronous communication is multi-user dungeons (MUDs). Based on Tolkien's Dungeons and Dragons motifs, these worlds allow several players to move through purely text-based rooms together, kill dragons, solve puzzles and chat with each other. When game environments emerged, MUDs were later also used for educational and discussion purposes. The first of these, the MUD1, was also written by Richard Bartle and Roy Trubshaw in 1979 at the University of Essex.

In 1988, Jarkko Oikarinen's Internet Relay Chat (IRC) was another synchronous communication format.

In parallel to the Internet, local discussion forums, bulletin board systems (BBS) emerged, initially as stand-alone PCs with one or more dial-up connections. With the help of telephone lines and X.25, these small computers are also networked, for example for FidoNet , developed in 1983 by Tom Jennings.

In 1985 Stewart Brand founded the legendary BBS Whole Earth 'Lectronic Link (WELL) in San Francisco . Commercial Internet service providers such as CompuServe and AOL followed. These separate networks also set up gateways to the Internet at the end of the 1980s, through which they have been able to exchange e-mail and news since then (Fidonet, for example, 1988, MCIMail and CompuServe 1989). A number of so-called freenets were created to enable people outside of the universities to access the Internet . The first, the Cleveland Freenet , was launched in 1986 by the Society for Public Access Computing (SoPAC).

First web server at CERN

The mass of information available on the Internet became increasingly unmanageable. The need for navigation and search tools led to new developments at various research institutions. At CERN in 1989 , Tim Berners-Lee started thinking about a distributed hypertext network that would become the World Wide Web (WWW).

The Archie (by Peter Deutsch, Alan Emtage, and Bill Heelan at McGill University ) and Hytelnet (by Peter Scott at the University of Saskatchewan) services offer similar links .

In 1991, Wide Area Information Servers ( WAIS , by Brewster Kahle of Thinking Machines Corporation) and Gopher (by Paul Lindner and Mark P. McCahill of the University of Minnesota) were added. The first version of Berners-Lee's WWW has been released. The following year, the Mosaic web browser was created at the National Center for Supercomputing Applications (NCSA) . Also in 1992 the University of Nevada released Veronica , a search tool for the gopher room.

In the same year, the librarian Rick Gates started the Internet Hunt, a search game for information in which even those who did not participate could learn from the published solution strategies. In 1990 the first remote-controlled machine was hooked up to the network, the Internet Toaster from John Romkey. Vending machines , coffee machines and an abundance of web cameras soon followed .

However, the open architecture of the Internet makes it possible to intercept every communication at all nodes between sender and recipient. The answer to this problem is cryptography , but it was considered military-state secret knowledge. The first cryptography tool accessible to mere mortals was PGP ( Pretty Good Privacy ), released in 1991 by Phil Zimmermann .

In addition to texts, images and audio files were also found on the Internet as early as the 1980s, but their integration with the WWW had only just begun. The first regular radio broadcasts on the net were the audio files of interviews with net pioneers, which Carl Malamud put on the net from 1993 under the name Internet Talk Radio.

The multimedia backbone ( MBONE ) continued, via which the first audio and video multicasts were broadcast in 1992. Initially only a few laboratories with a very wide range were able to take part, but the tools developed here were soon further developed for domestic use.

The program CUSeeMe (a play on words on I see you seeing me ) offered video conferencing for the PC.

The RealAudio streaming format (1995) made it possible to access sound information in real time on the network. With MIME ( Multimedia Internet Mail Extensions , RFC 1437 ), multimedia content can also be sent in e-mails since 1993.

Based on the simplified markup language Wikitext , Wikipedia started in 2001 and, as a collaboratively edited online lexicon, has become the most extensive encyclopedia in the world.

internationalization

Europe

In Europe there were already the first dial-up connections and UUCP-based networks in the early 1980s, such as the EUnet ( European Unix Network ) established in 1982 with nodes in the Netherlands , Denmark , Sweden and England .

In Germany , the Internet was only known from the cinema (for example WarGames - war games ), as one of the German Internet pioneers, Claus Kalle from the computer center at the University of Cologne, recalls.

Mainframes communicated via the expensive Datex-P . The first computer network that had an e-mail link to the USA and a gateway to the Internet there was the European Academic Research Network (EARN), launched in 1984 . Of course, experimentation with TCP / IP was soon started - the RFCs that could be obtained by email from EARN aroused curiosity - but the climate was not favorable for TCP / IP. When in 1985 the Association for the Promotion of a German Research Network e. V. (DFN-Verein) was founded, he only represented the official OSI line.

"In Germany and Europe at that time they were completely convinced and also politically and financially supported that the protocols of the OSI world would soon be widely available and stably implemented and thus a basis for manufacturer-independent networking would exist."

The first connections from computers outside the USA were via UUCP. In 1984, for example, the JUNET ( Japan Unix Network ) was established and a first message from Kremvax caused a stir, since the USSR had apparently been connected to the Usenet since then , but this turned out to be an April Fool's joke. Similar to “The Well”, mailbox networks have been created since the mid-1980s , in addition to FidoNet the Z-Netz and MausNet, and the CL-Netz, founded in 1987 by political activists . From 1992 these networks offered inexpensive mail and news connections to the Internet via gateways.

The initiative for an IP network in Germany came from the University of Dortmund in 1988 . As part of the Europe-wide InterEUnet network, it was first connected via Datex-P , then via a dedicated line to Amsterdam and from there to the US Internet. The IT-Computer Operating Group (IRB) of the University of Dortmund operated an anonymously accessible FTP server .

“It was especially beneficial to work with the fairly fresh copies of the GNU and other public domain packages ( Emacs , GCC , ISODE, etc.). In this way, access to Netnews and Internet Mail was possible for the first time, so that one could keep up to date. "

There was a similar initiative at the computer science chair of Professor Werner Zorn at the University of Karlsruhe , which led to the establishment of the XLINK ( eXtended Lokales Informatik Netz Karlsruhe ), which also connects to the USA with New York's NYSERnet ( New York State Education and Research Network ).

The OSI regime of the DFN gradually relaxed. The X.25 -based science network (WiN) should also support TCP / IP hosts right from the start. In 1989, on the initiative of Rob Blokzijl at the National Institute for Nuclear Physics and High-Energy Physics in Amsterdam, the European network providers merged to form RIPE ( Reseaux IP Européens ) in order to ensure administrative and technical coordination for a pan-European IP network. In order to consolidate the already existing European IP networks, some network operators began to plan and build a European IP backbone structure called EBONE in 1991.

In 1992 initiatives such as Individual Network e. V. (IN) with the development of alternative procedures and structures for the provision of IP services. The IN also took an active part in shaping the German IP landscape. Last but not least, the Netnews distribution would have made slow progress without the IN cooperation.

The growth in international IP connectivity can be seen in the registration of country domains. In 1988 Canada , Denmark , Finland , France , Iceland , Norway and Sweden were added. In November 1989 a total of 160,000 hosts were on the Internet. Australia , Germany , Israel , Italy , Japan , Mexico , the Netherlands , New Zealand and Great Britain join them. In 1990 Argentina , Austria , Belgium , Brazil , Chile , Greece , India , Ireland , South Korea , Spain and Switzerland were added. In 1991 it was Croatia , the Czech Republic , Hong Kong , Hungary , Poland , Portugal , Singapore , South Africa , the Republic of China (Taiwan) and Tunisia . In 1992, the number of hosts passed the million mark. Ever smaller countries and territories such as Cyprus , Antarctica , Kuwait and Luxembourg are registering country domains. In 1997, a number of island nations and protectorates were added, so that today the entire world map is mapped onto the address space of the Internet.

Meanwhile, the spread of the Internet in Europe is very advanced, in Iceland 97.8% of the population use the Internet. In other European countries this number is comparatively low, for example 20.7% in Kosovo.

Asia

China has been conducting research in the field of computer technology since the 1950s. The research and use of Chinese computer networks first began by the Ministry of Railways in 1980. It set up wide area networks (then still: Long Haul network). The nodes in Beijing, Shanghai and Jinan consisted of PDP-11 systems, the network architecture was DNA .

The first international connection from China to the Internet took place via a gateway in Germany, University of Karlsruhe, September 1987 and came from academic circles. It was not until 1994 that the Chinese authorities gave permission for a full-fledged TCP / IP Internet connection. Subnetworks in China with internet connection are currently CHINAnet, CERnet (China Education and Research Network), CSTnet, CHINAGBN, UNInet and CNC.

In 1998, the Ministry of Information Industry (Xinxi chanyebu) was set up in China to monitor and protect network operators with international Internet access and other matters relating to the Internet. It is also busy with things related to censorship on the internet . In 1998 there were attempts in China to set up a Chinese intranet , but these were abandoned after a short time. B. China C-Net from the Sichuan Internet Development Corporation, or the CWW, also called China Public Multimedia Network , which was developed by China Telecom .

Commercialization

Stimulating and promoting the emergence of a commercial market for Internet providers was one of the goals of the NSFnet initiative. Early beneficiaries included companies such as Performance Systems International (PSI), Advanced Network and Systems (ANS, founded by IBM , MERIT and MCI ), Sprintlink and CERFNet from General Atomics, which also operated the San Diego Supercomputer Center .

Commercial ISPs were to take over the maintenance and expansion of the Internet from universities and research authorities in the late 1980s. This also created a significant market for internet-based products. Len Bozack, a Stanford student, founded Cisco Systems . Others such as 3Com, Proteon, Banyan Company, Wellfleet and Bridge also entered the router market. The first internet industry fair, Interop in San Jose in 1988, attracted 50 exhibitors and 5000 visitors.

In 1991, the NSF lifted the previous ban on advertising (the acceptable use policy ) in the public network infrastructure. This paved the way for General Atomics (CERFnet), PSINet and UUNET ( AlterNet ) in California to join forces to form the first Commercial Internet eXchange (CIX) to organize unrestricted traffic between the commercial networks.

In Germany, too, the privatization of university infrastructure began in the early 1990s . The third-party funded project Eunet of the IT-Computer Operating Group of the University of Dortmund became EUnet Deutschland GmbH at the end of 1992. In the following year, the XLINK project at the University of Karlsruhe also became a division of NTG Netzwerk und Telematic GmbH, which is in turn a subsidiary of Bull.

Turning from 1990

A turning point can be identified at the transition from the 1980s to the 1990s. The ARPANet is officially switched off in 1990. The NSF is shifting network funding from directly funding academic backbone infrastructure to providing funds that universities use to buy connectivity from commercial providers. With the dwindling role of the NSF on the Internet, the Acceptable Use Policy also ended. At first cautiously, then in an incessant stream, the advertising messages set in on the network. The network providers interconnected in the CIX marketed the Internet as a business platform. User cultures came through the gateways of the commercial BBSe, who were used to paying for information and who in turn used the channels unrestrainedly for commercial purposes. The law firm Canter & Siegel from Arizona sparked a notorious conflict when it sent mass postings ( spam ) to the Usenet in 1994 to advertise its green card lottery services . The network residents reacted violently and prevented this abuse by canceling the spam again and covering the company with massive amounts of e-mails.

WWW neighbors of Wikipedia

From 1990 onwards, targeted efforts were made to bring commercial and non-commercial information service providers into the network. Among the first were Dow Jones , Telebase, Dialog , CARL (the Colorado Alliance of Research Libraries ), and the National Library of Medicine .

The first commercial Internet provider World was launched in 1990. In 1991 the WWW was able to start its triumphal march. More than 100 countries were connected to the Internet, with over 600,000 hosts and almost 5,000 individual networks. In January 1993 there were already over 1.3 million computers and over 10,000 networks.

The then US President Bill Clinton and Vice President Al Gore made a statement about their technology policy in February 1993 immediately after taking office at a town meeting in Silicon Valley , in which the Internet already played a central role. This triggered a kind of foreshock in a geopolitical situation in which the US was in an economic crisis, Europe was on the upswing and Japan was at the top of the world. The real shock wave went across the world when Al Gore announced the National Information Infrastructure Agenda for Action (see NII) on September 15 of that year , in which he not only turned networks into a multi-billion dollar industry himself, but also a basic infrastructure for business, education, science and culture. The awareness of lagging behind the USA in a key technology sector triggered hectic activity everywhere. This was when the commercial development and mass settlement of the Internet began.

For the new generations of users there is only one piece of information that should circulate freely and as widely as possible: advertising. All other information is commodity for them. In order to make information (e.g. stock market data, teaching material, pieces of music) available to those and only those who paid for it in this promiscuous milieu, additional, complex protective mechanisms, zones with access controls and cryptographically secured rights control systems had to be incorporated into the Internet. The rights industry ( Bertelsmann , Sony , Time Warner , etc.) has been working hard since 1994 to make their goods sellable over the Internet and to secure them technically. Nothing demonstrated the new quality of the Internet better than the first cyber bank First Virtual, which started operations in 1994.

Microsoft apparently missed the advancing developments of the Internet: Bill Gates made no mention of the Internet in the first edition of his 1995 book The Road Ahead . Shortly afterwards, he turned his company on to the Internet. The first version of the Microsoft Internet Explorer web browser appeared in the same year .

After the link between hardware and software was broken, the web broke the link between specific software and information. Microsoft Network ( MSN ), on the other hand, was an attempt to establish such a coupling again: a closed format in which companies could offer information and services for a fee - provided they paid an initial fee of $ 50,000 and a share of all revenues to MS. It was a belated imitation of closed BBSs like CompuServe or AOL , which had already been overtaken by the WWW, which allowed everyone to offer information free of charge.

Domain names had previously been nothing but a mnemonic that made the underlying numerical IP addresses more manageable. With the advent of large corporations with their protected trademarks, they became an aggressively disputed territory.

The first prominent dispute over whether domain names are intellectual property was that between MTV Networks and Adam Curry . Around May 1993, Curry, an MTV video jockey, started an information service at mtv.com on his own and at his own expense . In discussions with leading employees of MTVN and its parent company Viacom New Media, it was said that MTV was not interested in the Internet, but did not prevent Curry from his activities. So Curry expanded its range of information, including a bulletin board on which musicians and representatives of the music industry could talk to each other. E-mail addresses such as popquiz@mtv.com were displayed in the television programs he moderated . In January 1994, MTVN formally requested Curry to stop using mtv.com . However, MTV broadcasts continued to refer to this address and a senior executive asked Curry to include certain information on his page in February. In the meantime, MTVN and AOL had signed a contract to offer a paid service that should include a bulletin board for music professionals that was strikingly similar to that of Curry. MTVN sued Curry among other things because of the violation of trademark claims to release the domain mtv.com . Curry's attempts to settle the dispute amicably failed. He resigned. Ultimately, an out-of-court settlement was reached in which Curry gave up the domain to MTV.

The situation was typical for the period around 1993/1994: Large companies, including those from the media industry, ignored or underestimated the importance of the Internet, while innovative individuals built up popular and free information offers through their personal commitment, just to have to watch them The fruits of their labor were discussed with the help of the legal system. After numerous judgments had ruled that domain names were subject to the trademark regime, a brisk trade began. For example, CNET bought the tv.com domain in 1996 for $ 15,000. business.com was sold for $ 150,000 in 1997 and resold for $ 7.5 million two years later.

By 1995, the commercial backbone infrastructure in the USA was set up and interconnected to such an extent that the NSFNET backbone service could be discontinued. In the same year, a number of Internet companies went public, the most spectacular being Netscape, which is based on NCSA browser technology and has the third largest NASDAQ - IPO value of all time.

In the wake of the economy, lawyers also found their way onto the Internet. As part of the process of legalization, the legislature also took steps towards regulation. In 1996, the controversial Communications Decency Act (CDA) was passed in the USA , which bans the use of indecent words on the Internet. A few months later, a court issued an injunction against the application of this law. In 1997 the highest US court declared the CDA unconstitutional. Nevertheless, during this time the supposedly unlawful area of ​​the Internet was included in the legal regulations of areas such as cryptography, copyright law and general terms and conditions.

In many countries courts and state authorities intervene in cyberspace . The People's Republic of China for example requires that ISPs and users to register with the police. A German court ruled that CompuServe had to prevent access to newsgroups that deal with sexuality in the broadest sense. Since CompuServe keeps its global information offer in stock at its headquarters in Ohio and it was not technically able to differentiate it according to individual countries, it switched off the newsgroups for all users, which triggered a mainly American wave of protests and boycotts against Germany. Saudi Arabia restricts Internet access to universities and hospitals. Singapore obliges providers of political and religious content to register with the state. New Zealand classifies computer disks as publications that can be censored and confiscated. American telecommunications companies took offense at Internet telephony services and called on parliament to ban the technology.

The self-organization of the technical development of the Internet fundamentals also changed their character. While there were no more than one hundred people at the annual IETF working group meetings in the mid-1980s, it is now often two to three thousand. Accordingly, they are no longer collective brainstorming, but rather tightly packed sequences of presentations. The actual work is increasingly taking place in small, closed groups, the design teams. While the technology of the Internet, which is more than twenty years old, is scaling surprisingly stable, the community structures are reaching their limits. The composition of the working groups also changed:

"Since the late 1980s, the proportion of academic members in the IETF has steadily decreased - and not only because the number of companies kept increasing, but also because more and more founding members moved into business."

The collective pursuit of the best solution for the Internet as a whole, according to Jeanette Hofmann, threatens to be undermined by the interests of competing companies that want to enforce their respective products. Ultimately, the sheer size, the advancing generation of engineers, and the weight of the structure that has grown resulted in the standard development tending to become more conservative and mediocre. Hofmann's conclusion: The IETF is well on the way to becoming a standardization organization like any other:

“The Internet and its community have returned to normal. At some point the fathers will have lost each other among the growing number of equal members - and with them some of the ideas and principles that surrounded the development of the Internet. "

Dotcom boom and the 21st century

Dotcom bubble on NASDAQ

In the mid-1990s, the Internet began to grow faster and faster - and by then at the latest it was also known to larger parts of the (non-academic) population. In Germany, Deutsche Telekom and various competitors (for example AOL and CompuServe ) offered nationwide Internet access at increasingly favorable conditions and massively advertised these offers.

The speed of the modems continued to increase and in Europe, the ISDN connection, a digital telephone connection designed directly for fast data transmission, was offered. The speed of the backbones continued to increase as more and more cables were laid for a lot of money.

As a result, the Internet became increasingly popular. This also made it more and more interesting economically and many larger companies began to present and advertise their products on their websites. Some private individuals went further and founded companies that only acted on the Internet and offered goods and services there. With little start-up capital, they were able to implement ideas that were well received by customers. In order to expand their business further, they obtained additional capital through an IPO. As the company name often the domain equivalent (which ends on commercial providers usually with ".com"), was this boom as a dot-com boom (. English dot to German: Point) refers.

There was also a dot-com boom in Germany - mainly due to the IPO of the former state-owned company Telekom . This IPO was massively advertised to the entire population in order to make the Telekom share known as a popular share, especially among the population, which was previously rather inexperienced. As a result, more and more people became interested in the stock market and bought shares in various other Internet start-ups.

Numerous stock exchange experts considered the share prices of Internet companies on the Neuer Markt to be overvalued - but such voices were ignored in the general euphoria.

In 2000, however, there actually was a stock market crash that ushered in a general downward trend on the stock market. Since then, in retrospect, the dotcom boom has also been referred to as the dotcom bubble . Many of the Internet startups that were founded had to close again, especially business models that were supposed to be financed through advertising alone or even wanted to make surfers pay for receiving advertising ( paid4 scene ), could not hold up. The now established Internet companies such as Amazon or eBay were not so badly affected that their very existence would be jeopardized. And so the Internet boom continued, albeit at a slower pace.

At the end of the 20th century, ecological aspects were also taken into account for the first time when setting up the Internet infrastructure. In 1999, the first web hosting provider was founded in California to operate its servers entirely with green electricity , and Greenpeace Energy has also offered this in Germany since 2003 . In 2005, the electricity consumption for Internet infrastructure and usage in Germany had already exceeded the electricity consumption for lighting.

Copyright Infringement

Due to ever greater transmission capacity on the one hand and ever better multimedia capabilities of the PCs on the other hand, the Internet is increasingly a not always legal distribution channel for practically any type of data.

As early as the mid-1990s, it was no problem for conventional PCs to save and play compressed audio files, especially in the MP3 format, which was already popular at the time. In addition, these files can be transferred in just two to three times the playing time at a simple ISDN speed; with DSL lines even faster than they are played. This soon led to a lively bartering of such files without observing copyright law. Attempts by the music industry to take action against this have only met with moderate success; In addition, the exchange systems continue to improve and operate according to the peer-to-peer principle without central, controllable bodies.

In addition, the music industry reacts very slowly to the competition and for a long time has failed to develop its own legal offers for downloading music files directly to the PC. Instead, only the file sharing sites were fought, with various reports of collateral damage such as wrongly accused persons or the obstruction of the legal use of corresponding services additionally damaging the image of the industry and giving the black copiers an image as "modern Robin Hood". It was not until 2003 that an online music store started in the USA with the Apple iTunes Music Store , which was accepted by numerous users due to the prices (99 cents / song). Since June 2004 the iTunes Store has also been available in Germany, Great Britain and France. Here, however, the files are initially blocked against further copying, so-called digital rights management (DRM). Only recently has there been a trend away from DRM restrictions, as these are poorly accepted by customers and often prove to be largely ineffective.

In the course of increasing bandwidths and hard drive capacities, this barter trade expanded to include programs and later video content, where the providers, however - probably also startled by the experience of the music industry - react faster and make corresponding download offers available.

literature

Web links

Commons : History of the Internet  - collection of pictures, videos and audio files

Individual evidence

  1. ^ Martin Hilbert, Priscila López: The World's Technological Capacity to Store, Communicate, and Compute Information. In: Science , 2011, 332 (6025), pp. 60-65; free access via: martinhilbert.net/WorldInfoCapacity.html
  2. ↑ Georg Ruppelt (Ed.): "The great humming god". Stories of thinking machines, computers and artificial intelligence. CW Niemeyer, Hameln 2003, ISBN 3-8271-8807-5 ; again: after Martin Luther became Pope and the Allies had lost the Second World War. Wehrhahn, Hannover 2007, p. 174
  3. ^ Paul Baran and the Origins of the Internet - Biography . RAND Corporation website. Retrieved April 27, 2014.
  4. ↑ October 29th is Internet Day . In: Die Welt , October 29, 2007
  5. ↑ kremvax. In: The Jargon File (version 4.4.7). Eric S. Raymond's Home Page, accessed December 2, 2017 .
  6. ↑ internetworldstats.com
  7. ↑ Jeanette Hofmann: The success of open standards and its side effects. (No longer available online.) In: Telepolis. Heise Zeitschriften Verlag GmbH & Co KG, July 23, 1999, archived from the original on September 10, 2014 ; accessed on March 26, 2014 .