Almost Forty Years [1961-1998]
We are where we are today—in October 1998—because of a number of innovations, and adaptive designs, among the most important of which are these:
| 1 |
A ‘modern’ computer: digital, electrical, programmable, accepting input, generating output |
|
| 2 |
Packet-switching: conceiving messages or files as packets of digitally encoded data to be routed across a network to the address they contain, rather than as proprietary messages on dedicated lines. |
|
| 3 |
TCP/IP: protocols setting the form of packets, so that they would be recognizable at intermediate nodes and at the destination. |
|
| 4 |
ARPANET: the first model implementing such a network. |
|
| 5 |
A system of "domain names" and corresponding "Internet Protocol numbers" which made it possible to assign a unique identifier to any device on the network: thus all computers, printers, &c. became addressable. |
|
| 6 |
FTP: the file transfer protocol, which enabled individual users to call up a file from a remote location. |
|
| 7 |
Time-sharing, implemented as a single computer providing computer services to one or several users, situated at terminals connected to the computer, and running programs interactively to which the operating system accorded bursts of computing resources in turn. |
|
| 8 |
The personal computer [which could then be attached, permanently or on demand, to a network]. |
|
| 9 |
The Web, based upon the pre-existing ARPANET and ftp, and brilliantly joining several features: [a] HTML [hyper-text meta-language], enabling simple formatting in ascii (plain text); [b] designation of target "pages" with a URL [universal resource locator], such as http://www.learnworld.com/; [c] a "browser" capable of matching the location of a mouse click to portions of the displayed HTML source; and [d] hyper-links, translating a click on a given graphic or text into a command to identify the URL associated with that location and send a query to the server on which it was resident asking that the source for that page be transmitted back to the querying computer. |
Some of these are associated with specific individuals, and groups, credited with inventing or designing the specific feature. [Source: Hobbes' Internet Timeline: http://info.isoc.org/guest/zakon/Internet/History/HIT.html] ARPANET was commissioned by the DoD in 1969: and the first four nodes were at UCLA, SRI (Stanford Research), UCSB, and the University of Utah.
The classic 'first' papers on packet-switching were Leonard Klinerock's [MIT] July 1961 paper "Information Flowin Large Communication Nets," the first paper on packet-switching theory, and Paul Baran's [RAND] paper, in 1962, "On Distributed Communciations Networks."
TCP followed with a paper by Vint Cerf and Bob Kahn in 1974 titled "A Protocol for Packet Network Intercommunication." [TCP: Transmission Control Program (or Protocol).] In 1982 DCA and ARPA established the Transmission Control Protocol (TCP) and Internet Protocol (IP), together TCP/IP, as the protocol suite for ARPANET.
BITNET and CSNET, the first widely used in education, the second among computer scientists, were begun in 1981. In that year, too, France Telecom deployed Minitel in France.
The Domain Name System was introduced in 1984.
The World Wide Web (WWW), developed by Tim Berners-Lee at CERN, the European nuclear research center, was released in 1991.
•
The roles of Paul Baran, J. C. R. Licklider, Donald W. Davies and others are described by Hauben and Hauben [pp. 116 ff.] They point out that Baran's 1962 report discussed how the US military could ensure communications against the effects of attack. "Baran outlined the principle of 'redundancy of connectivity' . . . The report proposed a communication system where there would be no obvious central command and control point, so that all surviving points would be able to reestablish contact in the event of an attack on any one point. . . " Baran "proposed that a packet switching, store and forward technolgy be developed for a data network."
Gary Chapman has observed how different the main features of Baran's proposal were from telephoning: the network Baran envisaged was distributed, used message disaggregation, and was marked by redundancy. And he reports that Baran has told people that he got a cold shoulder from those in the military concerned with command and control of nuclear weapons.
•
Hauben and Hauben locate the source of basic protocols in the NetworkWorking Group for the ARPANET, a group of represenatives from the initial four sites. ". . . the NWG had to implement a working communications protocol. The first set of pairwise host protocols included remote login for interactive use (telnet), and a way to copy files between remote hosts (FTP)." Stephen Crocker, a UCLA graduate student, was chair of the NWG soon after its first meeting, and he instituted the practice of issuing notes of results of the meetings as Requests for Comments (RFCs): these were intended to elicit participation and contributions. Hauben and Hauben reproduce RFC-3 which itself sets out the documentation conventions of RFCs. [Hauben & Hauben, Chapter 8, Appendix 3.]
•
Consider the terms in which Tim Berners-Lee described http (Hypertext Transfer Protocol) in 1991 [from http://www.w3.org/Protocols/http/AsImplemented.html]:
|
This document defines the Hypertext Transfer protocol (HTTP) as originally implemented by the World Wide Web initaitive software in the prototype released. This is a subset of the full HTTP protocol, and is known as HTTP 0.9.
No client profile information is transferred with the query. Future HTTP protocols will be back-compatible with this protocol.
This restricted protocol is very simple and may always be used when you do not need the capabilities of the full protocol which is backwards compatible.
The definition of this protocol is in the public domain (see policy ).
The protocol uses the normal internet-style telnet protocol style on a TCP-IP link. The following describes how a client acquires a (hypertext) document from an HTTP server, given an HTTP document address .
Connection
The client makes a TCP-IP connection to the host using the domain name or IP number , and the port number given in the address.
If the port number is not specified, 80 is always assumed for HTTP.
The server accepts the connection.
Note: HTTP currently runs over TCP, but could run over any connection-oriented service. The interpretation of the protocol below in the case of a sequenced packet service (such as DECnet(TM) or ISO TP4) is that that the request should be one TPDU, but the response may be many.
|
|
Policy
This outlines the policy of the W3 project at CERN. Whilst not legally binding, this attempts to explain my understanding of the CERN rules and the desires of the team at CERN.
Aim
The basic aim of the project is to promote communication and information availability for the High Energy Physics (HEP) community. The project is based at CERN, whose budget is provided by contributions of taxpayer's money from the European member states. It is in the interests of HEP, CERN, and the project itself that it should interwork with systems and information in many other fields, and so active collaboration with other groups is essential. To produce an information system isolating HEP from the rest of the world would be counter-productive, so the aim can be seen as furthering a global web of information.
The WWW team are all enthusiastic that information of all types should be available as widely as possible.
Collaboration
We encourage collaboration by academic or commercial parties. There are always many things to be done, ports to be made to different environments, new browsers to be written, and additional data to be incorporated into the "web". There have already been many contributions in these terms, and also with hardware support from manufacturers.
If you may be interested in extending the web or the software, please mail or phone us.
Code distribution
Code written at CERN is covered by the CERN copyright except where explicitly put into the Public Domain. The basic WWW code is in the Public Domain, the rest is covered by the conditions of distribution . In practice the interpretation of this in the case of the W3 project is that the programs are freely available to academic bodies of CERN member states and to the world-wide High-Energy Physics community. To commercial organizations who are not reselling it, but are using it to participate in global information exchange, the charge is generally waived in order to cut administrative costs. Code is of course shared freely with all collaborators. Commercial organizations wishing to sell software based on W3 code should contact CERN.
We are in the process of getting agreement to release certain parts of the WWW project code into the Public Domain.
Where CERN code is included in otherwise public domain code, that CERN code becomes also public domain.
Code not originating at CERN is of course covered by terms set by the copyright holder involved.
Protocols and Data Formats
The definition of protocols such as HTTP and data formats such as HTML are in the public domain and may be freely used by anyone. Tim BL |
•
Reviewing Conclusions
These are some of the important points to bear in mind:
| 1 |
A number of technologies had to be brought together, brought into an imaginative design, to achieve the Net and Web we know. |
|
| 2 | It was already clear to some researchers, such as Paul Baran, in the early to mid-1960s that an important task was to solve the problem of networking. What's that problem? To enable computers running on different operating systems to talk to each other. | |
| 3 | Key initial steps were accomplished by University researchers solving practical problems,with little apparent guidance from officials or institutions. | |
| 4 | RFCs (Requests for Comment) were used to elicit responses from interested parties, with great openness. | |
| 5 | Both Telnet and FTP were built in from the earliest days of the ARPANET. | |
| 6 | Tim Berners-Lee launched the Web only seven years ago. | |
| 7 | His initial statements stressed the aim of "communication and information availability" in High-Energy Physics, but he also said the project should work with "systems and information in many other fields." The aim, he wrote, "can be seen as furthering a global web of information." | |
| 8 | The initial costs of the ARPANET were borne by the US Government,through its military R&D budget. The initial costs of the Web were borne by European governments through their support of the Centre Européen pour Recherche Nucléaire (CERN) [or as it now appears to be called in French, Laboratoire Européen pour la Physique des Particules]. So these are tax-supported initiatives. | |
| 9 | HTML and the Web itself are extraordinarily simple. The Web we know results from a quest for simplicity, not complexity. The internal simplicity makes it possible to weave a resultant fabric of extraordinary complexity. | |
| 10 | This distinction is being pressed again in the quest for simple ways to query data accessible on the Web. |