The World Wide Web is a global information system in which resources are identified by addresses, connected through hypertext links, and accessed using standardized software and communication protocols. It operates primarily over the Internet and supports documents, multimedia, data, and interactive applications. Developed at CERN beginning in 1989, the Web combines a shared addressing system with technologies for exchanging and presenting information. A web browser is its most familiar user interface, but automated programs also interact with Web resources. (home.cern)
The Web and the Internet
The Web and the Internet are related but distinct. The Internet is the interconnected network infrastructure that transports data between computers. The Web is an information system built on that infrastructure. Internet technologies also support services such as email and file transfer that do not require Web pages or browsers. A Web interface to email is therefore one way of accessing an email service, not the underlying email system itself. (internetsociety.org)
The distinction is also architectural: Internet protocols provide communication between networked systems, whereas Web technologies identify resources, request or modify them, and describe their representations. The Web is consequently neither a single program nor one centrally maintained collection of documents. (rfc-editor.org)
Origins and early development
In March 1989, Tim Berners-Lee, then working at CERN, proposed an Internet-based hypertext system for sharing information across different computers. The proposal addressed information-management problems in a research environment whose projects, personnel, and computing systems continually changed. In November 1990, Berners-Lee and Robert Cailliau formalized the project in a further proposal. (home.web.cern.ch)
During 1990, Berners-Lee developed the first Web server and the first browser-editor, called WorldWideWeb. He also created early versions of the addressing, communication, and document-formatting technologies that became the Web’s foundations. The first website, hosted on his NeXT computer, explained the project itself. (home.cern)
Two developments helped the Web spread beyond its initial research community:
- Open availability: On April 30, 1993, CERN placed its World Wide Web software in the public domain. A subsequent release used an open licence, supporting wider dissemination and implementation.
- Graphical browsing: NCSA Mosaic, released for Unix in 1993 and subsequently for Windows and Macintosh, helped popularize the Web through an accessible graphical interface. It was not the first Web browser. (home.cern)
In October 1994, Berners-Lee founded the World Wide Web Consortium (W3C) at the Massachusetts Institute of Technology, in collaboration with CERN. Its purpose included coordinating standards so that independently developed browsers, servers, and websites could work together. (w3.org)
Core architecture
Web architecture distinguishes three closely related functions: identification, interaction, and representation. Separating these functions allows technologies to evolve without requiring the whole system to be redesigned. (w3.org)
Identification and links
A Uniform Resource Identifier (URI) identifies a resource. A Uniform Resource Locator (URL), commonly called a Web address, additionally provides a way of locating or accessing it. A resource need not be a stored document: it may be a service, a changing collection of information, or another identifiable entity. Its identifier is distinct from the particular representation returned when it is accessed. (w3.org)
Hyperlinks connect resources by referring to their identifiers. Links can cross organizational and geographical boundaries without requiring the linked documents to reside on the same server. This shared addressing system makes independently published material part of a navigable information space. (w3.org)
Communication through HTTP
The Hypertext Transfer Protocol (HTTP) defines a request-and-response interface between clients and servers. A browser can request a representation of a resource, submit information, or perform another supported operation. The server responds with a status code, metadata, and, where appropriate, content. (rfc-editor.org)
HTTP is stateless: its basic semantics do not require a server to remember a sequence of earlier requests. Applications can nevertheless maintain sessions using mechanisms such as cookies, which allow a browser to store information supplied by a server and return it with later requests. This supports functions such as session continuity and user preferences. (rfc-editor.org)
HTTP/1.1, HTTP/2, and HTTP/3 share core semantics but differ in how messages are transported and encoded. HTTP/3 uses QUIC rather than TCP as its transport. HTTPS provides HTTP communication over a secured connection, protecting the exchanged data against network interception and alteration. (rfc-editor.org)
Representation and presentation
A response supplies a representation of a resource rather than necessarily exposing its internal implementation. The same resource may offer different representations, such as different languages or media formats. (w3.org)
The principal technologies for interactive Web documents are:
- HTML — Hypertext Markup Language: describes document structure and meaning, including headings, paragraphs, links, forms, and embedded media.
- CSS — Cascading Style Sheets: controls presentation, including layout, typography, and visual styling.
- JavaScript: a programming language used to implement interactive behavior and manipulate documents through browser interfaces. (w3.org)
HTML is a markup language, not a general-purpose programming language. A document can remain useful without scripting; JavaScript adds behavior but is not a requirement for every Web page. (html.spec.whatwg.org)
Pages, websites, and applications
A Web page is a document presented through a browser. A website groups related pages and resources, while a Web application provides interactive functionality through Web technologies. These categories overlap: a single site may contain reference documents, forms, multimedia, and application interfaces. (html.spec.whatwg.org)
Web content may be static or generated dynamically. A server can produce a response when it receives a request; browser-side scripts can subsequently change the displayed document or obtain additional information. This makes the Web both a publishing medium and a software platform. (html.spec.whatwg.org)
A browser does more than display text. It interprets document structure, applies styles, runs supported scripts, processes forms, and coordinates access to associated resources. Accessibility also depends on how browsers expose this information to assistive technologies, including screen readers. (html.spec.whatwg.org)
Discovery and search
Links provide direct navigation, but search engines offer another route to Web content. Search systems such as Google discover pages using automated crawlers, analyze accessible content, store information in an index, and retrieve relevant results in response to queries. A search therefore usually consults an existing index rather than inspecting the entire live Web at that moment. (developers.google.com)
Search coverage is not identical to the Web itself. Pages may remain unindexed because they cannot be discovered, require authentication, block crawlers, or are excluded during indexing. Even a publicly accessible page is not guaranteed to appear in a search engine’s results. Search engines are services operating on the Web, not authorities that define which resources belong to it. (developers.google.com)
Standards and interoperability
The Web’s interoperability depends on published specifications implemented by independent organizations. W3C develops standards across the Web platform; the Internet Engineering Task Force specifies protocols such as HTTP; and the Web Hypertext Application Technology Working Group maintains the HTML Living Standard. These bodies specify technologies rather than exercise centralized ownership over Web content. (w3.org)
HTML’s development illustrates the balance between continuity and change. Its specifications must accommodate existing documents while supporting evolving applications. The Living Standard model maintains the specification continuously rather than treating a numbered release as permanently complete. (html.spec.whatwg.org)
Interoperability is not automatic. Different implementations, unsupported features, and assumptions about particular devices can affect how content behaves. Accessibility likewise requires coordination among specifications, content authors, authoring tools, browsers, and assistive technologies. (w3.org)
Security, privacy, and access
Security and privacy
Security and privacy involve more than encrypting a connection. Cookies can support useful application state but can also enable tracking. RFC 6265 explicitly discusses privacy concerns associated with cookies, including their use by third parties. (datatracker.ietf.org)
A page obtained securely may still attempt to load resources over insecure connections, creating mixed content. Such resources can undermine confidentiality or integrity. Web security specifications therefore define mechanisms for blocking or upgrading insecure resource requests within secure documents. (w3.org)
W3C’s ethical principles identify privacy, security, personal agency, and reduced centralization as goals for the platform. These are design commitments, not guarantees that every website or service satisfies them. The document also recognizes that different goals can involve trade-offs. (w3.org)
Accessibility and unequal participation
Web accessibility concerns whether people with disabilities can perceive, understand, navigate, and interact with content. It depends on the entire system: suitable document structure alone cannot compensate for every limitation in browsers, tools, or assistive technologies. W3C’s accessibility work includes guidelines for content, authoring tools, and user agents. (w3.org)
Participation also depends on equipment and network conditions. W3C’s ethical principles explicitly recognize users with low-bandwidth connections and low-specification devices. A platform that is globally addressable is therefore not necessarily equally usable by everyone. (w3.org)
Persistence and reliability
A Web address does not guarantee that its resource will remain available indefinitely. Content can move, change, or disappear, leaving references unusable. W3C’s architecture consequently treats persistent identifiers and stable resource identification as important properties of a durable Web. Publication and linking provide access to information, but do not by themselves establish its accuracy or permanence. (w3.org)
References
- The birth of the Webhome.cern
- World Wide Web at 35home.web.cern.ch
- Historyw3.org
- Architecture of the World Wide Web, Volume Onew3.org
- RFC 9110: HTTP Semanticsrfc-editor.org
- About the Internet and How it Worksinternetsociety.org
- A Brief History of the Internetinternetsociety.org
- NCSA Web browser ‘Mosaic’ was catalyst for Internet growthnews.illinois.edu
- HTML Standard, Edition for Web Developershtml.spec.whatwg.org
- Web Hypertext Application Technology Working Group (WHATWG)whatwg.org
- RFC 6265 — HTTP State Management Mechanismdatatracker.ietf.org
- In-depth guide to how Google Search worksdevelopers.google.com