Skip to main content
ScienceVerse
TechnologyDifficulty 2-4

Internet

The Internet is a global network of networks that moves data between computers as independent packets using the Internet Protocol (IP), with addresses of 32 bits (IPv4) or 128 bits (IPv6). Protocols layered on top add reliability (TCP) and human-friendly names (DNS). The World Wide Web is not the Internet: it is one service that runs on it — linked resources identified by URIs and fetched with HTTP, invented by Tim Berners-Lee in 1989.

Sign in to save this concept.

Protocol responsibilities

  • IP (RFC 791; IPv6 RFC 8200): addressing and forwarding of independent datagrams; each carries a Time to Live that is reduced along the route, and the datagram is destroyed if it reaches zero.
  • TCP (RFC 9293): reliable, in-order byte stream using sequence numbers, checksums and retransmission.
  • DNS (RFC 1034): a consistent, hierarchical name space mapping names to resource records such as addresses.
  • HTTP (RFC 9110): stateless, request/response application protocol for the Web's URI-identified resources.

The Web (W3C: an information space of URI-identified resources) sits entirely at the application layer. Its transport has changed — HTTP/2 multiplexed over TLS and TCP, HTTP/3 over QUIC and UDP — without changing HTTP's semantics.

Common misconception: TCP's reliability is end-to-end, not hop-by-hop: routers in between do not acknowledge or retransmit TCP data.
Full explanation — the complete reference version every reading depth is based on

A network of networks

The Internet is made of many separate packet networks — home, school, business and mobile networks — joined by routers (called 'gateways' in the original specifications). Every connected host speaks the same family of Internet protocols, so a phone on mobile data can exchange data with a server on a fibre network on another continent.

How data travels

  1. Your device splits data into IP datagrams, each carrying a source and destination IP address.
  2. Routers forward each datagram towards its destination independently; datagrams may take different routes.
  3. Each datagram carries a 'Time to Live' that is reduced along the route, so a datagram that cannot arrive is eventually destroyed instead of circling for ever.
  4. IP itself makes no promises: datagrams can be lost, duplicated or arrive out of order.
  5. TCP, running on the two end hosts, numbers the data, detects losses and errors, retransmits, and hands the application an in-order stream.
  6. DNS turns names such as example.org into the IP addresses that routing actually uses.

Addresses

232=4 294 967 2962^{32} = 4\,294\,967\,296

The number of distinct 32-bit IPv4 addresses (our calculation); IPv6 uses 128-bit addresses, giving 2¹²⁸.

Worked example (our calculation): a 32-bit address allows 2³² ≈ 4.3 billion distinct values. IPv6 enlarges the address from 32 to 128 bits — four times as many bits, but 2⁹⁶ times as many possible addresses.

The Internet is not the Web

The World Wide Web is an information space of resources (pages, images, data) identified by URIs and linked together. Tim Berners-Lee invented it in 1989; the Web originated at CERN, and he wrote the first server and browser in October 1990. Browsers fetch web resources with HTTP, which runs over the Internet's transport protocols. Email, video calls, online games and app updates also use the Internet without being part of the Web.

How it began

The ARPANET's first node was installed at UCLA in September 1969, and four host computers were connected by the end of that year. On 1 January 1983 the ARPANET switched its host protocol from NCP to TCP/IP, the protocol suite the Internet still uses.

How we know

Every protocol described here is defined in a public standard (an IETF RFC) that anyone can read: IP in RFC 791, IPv6 in RFC 8200, TCP in RFC 9293, DNS in RFC 1034 and HTTP in RFC 9110. The history is from 'A Brief History of the Internet' by its builders and from the W3C, the body Berners-Lee founded to develop web standards.

Assumptions and limits

The clean layered picture is an ideal. IPv4 and IPv6 run side by side, devices such as firewalls and address translators keep state that the original design avoided, and a reliable TCP stream can still be slow when packets are lost, because missing data must be resent before later data is handed over.

Common misconception: Misconception: 'the Internet and the Web are the same thing'. The Internet is the global infrastructure that carries packets; the Web is one application built on top of it, alongside email, messaging, streaming and many others.
Info: Connections: the Internet is a special case of a computer Network, and every host on it is a Computer. Its physical links include the optical fibres and wireless LANs described under Networks.

Ask ScienceVerse

Still curious about Internet? Ask a question, get hints, take a short lesson or try a challenge. The tutor answers only from this concept's approved sources, and says so when it has none.

Ask the tutor about this concept on the full tutor page.

Connections

Prerequisites

Understand these first:

Guided learning path

See everything to learn before this, in order, with your progress:

Related concepts

Check your understanding

Take a quick check of two to five questions, with an explanation for every answer:

See the neighbourhood of Internet in the Knowledge Galaxy

Sources and methodology

Claims marked “awaiting scientific review” cite the sources listed but have not yet been signed off by a scientific reviewer.

Content status: published 1 October 2026.

  • Scientific review: this version has not yet been signed off by a scientific reviewer.
  • The Advanced explanation has not yet been reviewed for age suitability.