Billions of devices exchange information every second without a single wire connecting them all directly. Here's the actual machinery — routers, packets, addresses and cables — behind that everyday miracle.
The internet is not one machine or one place — it's an agreement. Millions of independently owned networks around the world have agreed to speak the same technical "language," so that a message from a phone in Mumbai can reach a server in Oregon in a fraction of a second. There is no single computer running it, and no single company owns it. This guide opens up the actual mechanics behind that agreement.
People use "the internet" and "the web" interchangeably, but they're different layers. The internet is the physical and technical infrastructure — cables, satellites, routers, and the rules that let devices exchange data. The World Wide Web is one thing that runs on top of that infrastructure: a system of linked pages viewed through a browser. Email, video calls, online gaming and file transfers also run on the internet, but none of them are "the web" — they're separate services using the same underlying network.
Almost everything on the internet works through a client-server relationship. Your phone or laptop is the client — it asks for something. A server is a computer, usually running non-stop in a data centre, whose job is to store information and answer requests. When you open a shopping website, your device (the client) sends a request; the shopping site's server finds the right page and sends it back. Every server is just a computer built and configured to answer requests reliably, all day, every day.
| Role | Client | Server |
|---|---|---|
| What it does | Requests information | Stores and sends back information |
| Example device | Your phone, laptop, tablet | A data centre computer running non-stop |
| Typical action | Opening an app or website | Delivering the page, video or file requested |
When you send anything over the internet — a message, a photo, a webpage — it isn't sent as one solid block. It is broken into small pieces called packets, each carrying a fragment of the data plus labels showing where it came from, where it's going, and which piece of the sequence it is. Packets from the same message can even travel different physical routes across the network and still arrive correctly, because the receiving device uses those labels to reassemble them in the right order.
Splitting data into packets means a network doesn't need one giant dedicated wire between every pair of devices. Instead, packets from millions of unrelated messages share the same cables and routers, each hopping along whichever path is fastest at that moment — similar to how many different cars share the same road network instead of each needing a private road.
For packets to reach the right destination, every device connected to the internet is given an IP address — a unique number that acts like a postal address. Since remembering long strings of numbers isn't practical for people, the internet relies on the Domain Name System (DNS), a distributed directory that automatically translates a website name into its correct IP address the moment a link is clicked.
DNS works in a hierarchy: your device asks a nearby DNS server, which — if it doesn't already know the answer — asks a more central server further up the chain, until the correct IP address is found and sent back. This entire lookup usually happens in a fraction of a second, invisibly, before a page even starts loading.
Packets don't teleport — they physically move through hardware. A packet leaving your device typically passes through your home router, into the network of your Internet Service Provider (ISP), and then across large-scale connections between ISPs and countries — including, for most international traffic, undersea fibre-optic cables running along ocean floors. Each device along this chain that forwards a packet toward its destination is called a hop.
HTTP (Hypertext Transfer Protocol) is the agreed set of rules browsers and servers use to request and deliver web pages. HTTPS is the same system with one crucial addition: encryption. It scrambles data in transit so that anyone intercepting the packets between your device and the server sees only unreadable noise, not the actual content — which is why banking, shopping and login pages should always show HTTPS rather than plain HTTP.
Typing a web address and pressing enter sets off a rapid chain: your device asks DNS for the site's IP address, opens a secure HTTPS connection to that server, and sends a request packet through your router and ISP. The server assembles the requested page and sends it back as packets, which may travel through a dozen or more routers and even cross an ocean via undersea cable — all typically completing in well under a second.
The internet feels seamless because its complexity is deliberately hidden from the people using it. Underneath every tapped link is the same chain: a client asking, packets travelling, DNS translating names into addresses, routers and cables carrying data across the physical world, and a server answering back. None of it requires magic — just an agreed set of rules, followed by an enormous number of machines, at a scale that would have been unimaginable to the engineers who built the first four-computer network in 1969.
10 questions. Select an answer for each, then submit to see your score instantly.