View source for Web history

__NOTOC__
The World Wide Web was developed by English physicist [http://en.wikipedia.org/wiki/Tim_Berners-Lee Sir Tim Berners-Lee] and Belgian computer scientist Robert Cailliau, while they were working at CERN in Geneva, Switzerland. They proposed an information management system using [[hypertext]], a concept dating back to the 1960s, which would be accessible via the [[Internet]].

==Precursors to the Web==
The concept of [[hypertext]], which underlies the Web, originated in the 1960s, with projects such as Ted Nelson's [http://en.wikipedia.org/wiki/Project_Xanadu Project Xanadu], Douglas Engelbart's [http://en.wikipedia.org/wiki/NLS_%28computer_system%29 oN-Line System], and Andries van Dam's [http://en.wikipedia.org/wiki/Hypertext_Editing_System Hypertext Editing System]. Both Nelson and Engelbart were inspired by Vannevar Bush's 
[http://en.wikipedia.org/wiki/Memex Memex], a micro-film system which Bush imagined in the 1945 essay "As We May Think".

Before developing the World Wide Web, Tim Berners-Lee built an earlier [[hypertext]] system at CERN in Switzerland in 1980, which he called [http://en.wikipedia.org/wiki/ENQUIRE ENQUIRE]. This stand-alone database used bi-directional [[hyperlink]]s and provided an editor for easy creation of the links.

Berners-Lee's original proposal for the web was modeled after the [http://en.wikipedia.org/wiki/Dynatext Dynatext] system, for which CERN had a license. Dynatext was a powerful tool for reading documents annotated with [[Standard Generalized Markup Language]] ([http://en.wikipedia.org/wiki/SGML SGML]), but it was considered too expensive for widespread use, since the license imposed a fee each time any document was changed.

==Tim Berners-Lee launches the Web==
[http://en.wikipedia.org/wiki/Tim_Berners-Lee Tim Berners-Lee's] innovative contribution was to develop a practical way to combine [[hypertext]] with the [[Internet]].

In March 1989 Berners-Lee wrote a short document called [http://www.w3.org/History/1989/proposal.html Information Management: A Proposal], describing an information management system using [[hypertext]]. This first paper envisaged browsers on client computers that would have read-only access to hypertext documents and databases stored on several different servers. The documents would be transmitted over the network in a standardized hypertext format.

In November 1990, Berners-Lee and Robert Cailliau published a more formal proposal, titled [http://www.w3.org/Proposal.html WorldWideWeb: Proposal for a HyperText Project]. This proposal described the web as a network of links that would allow you to navigate from one node to another, where each node was a hypertext document. It was noted that a world-wide web would require a standard access protocol for requesting documents from remote servers, and a standard information format for the transmitted documents. The proposal also mentioned the possibility of providing a keyword search option (this would later be provided by [[search engine]]s).

By Christmas 1990, Berners-Lee had built the basic tools necessary for a working Web:
*the first [[web browser]], called WorldWideWeb (later renamed Nexus), which was also a Web editor
*the first [[web server]] (running on a NeXT computer using the NeXTSTEP platform)
*the set of first [[web page]]s, which described the project itself (archived [http://www.w3.org/History/19921103-hypertext/hypertext/WWW/TheProject.html here])

Berners-Lee presented the WorldWideWeb browser to the CERN ECP/PT group on February 26, 1991 (a link to the presentation can be found under that date in [http://www.w3.org/History.html this timeline].

On August 6, 1991, he posted a short summary of the World Wide Web project on the alt.hypertext newsgroup, which marks the beginning of the Web as a publicly available service on the Internet.

==Development of URI, HTTP and HTML==
Along with the first [[web browser]] and [[web server]], Berners-Lee also developed the basic Web protocols, including:
*a system for naming individual documents on the Web: the Universal Document Identifier (UDI) - now called [[Uniform Resource Identifier]] (URI)
*a network access protocol: the [[HyperText Transfer Protocol]] (HTTP)
*a publishing language for Web documents: [[HyperText Markup Language]] (HTML)

===Universal Document Identifier (UDI) - now called Universal Resource Identifier (URI)===
The 1990 [http://www.w3.org/Proposal.html proposal] described a Web link as an ASCII string from which the browser would deduce how to contact an appropriate server and request a particular document. In [http://www.w3.org/History/19921103-hypertext/hypertext/WWW/Addressing documents from 1992] this link was being called a "hypertext name" or "file name". The general format of a hypertext name [http://www.w3.org/History/19921103-hypertext/hypertext/WWW/Addressing/Addressing.html was specified] as follows: 

:scheme://host.domain:port/path/path#anchor

In the above chain, "host.domain" identifies the server (with an optional specification of the port) and "path/path" is the pathname leading to the file on the server (with an optional specification of an anchor within the document).

The "scheme" (see [http://en.wikipedia.org/wiki/URI_scheme URI Scheme] is related to the protocol used to access different types of documents. The 1992 documents mentioned for example the following schemes:
*file: a scheme for any protocol used by the browser for access to files on remote machines, such as [[File Transfer Protocol]] (FTP)
*news: a scheme for any protocol used by the browser for access to news articles and newsgroups, normally using the NNTP protocol
*http: a specific protocol developed to search and retrieve documents on the Web (the [[HyperText Transfer Protocol]])

The following name under the "file" protocol referred to a file on one of the cern servers (and to an imaginary anchor 123 within it):

:file://cernvax.cern.ch/usr/lib/WWW/defaut.html#123

The "file" scheme has since been replaced by the "ftp" scheme, since [[File Transfer Protocol]] has become the standard internet protocol for file transfer.

The general form of a name under the "http" protocol [http://www.w3.org/History/19921103-hypertext/hypertext/WWW/Addressing/HTTPAddressing.html was specified in 1992] as follows (the parts in brackets are optional):

:http://hostname[: port]/path[?searchwords]

For example, the name of an HTML document on one of the cern servers was:

:http://info.cern.ch/hypertext/WWW/TheProject.html

Berners-Lee originally called this sort of hypertext document name a [[Universal Document Identifier]] (UDI). However (see the footnote [http://www.w3.org/DesignIssues/Model.html here]), this was changed to [[Universal Resource Identifier]] in discussions within the URI working Group of the [[Internet Engineering Task Force]] (IETF).

Meanwhile, the term [[Uniform Resource Locator]] (URL) was introduced to denote a string that provided an address for a resource, while the term Uniform Resource Name (URN) came to represent a string that merely names a resource. URL and URN are types of URIs (see [http://en.wikipedia.org/wiki/Uniform_Resource_Identifier URI].

An [http://www.w3.org/Addressing/URL/url-spec.txt early document] specifying the syntax and semantics of the URL was published in March 1994.

In June 1994, the IETF published Berners-Lee's [http://tools.ietf.org/html/rfc1630 RFC 1630], which defined a formal syntax for all URIs.

===HTTP===
(To be written)

===HTML===
(To be written)

==Difference from other Proposed Systems==
(To be written)

==Mosaic Web Browser==
(To be written)

==World Wide Web Consortium (W3C)==
(To be written)

[[Category:Web protocols]]
[[Category:Draft]]