ELI5 · Networking & the web

What happens when you type a URL.

Press Enter and a quick relay race begins: find the address, open a line, lock it, ask for the page, draw it.

It feels instant: you type an address, hit Enter, and a page appears. But in that half-second a whole relay race runs, each runner handing off to the next. Miss any leg and the page never shows up.

Here is the route in plain English: turn the name into a number, open a connection to it, lock that connection so nobody can snoop, ask for the page you wanted, and finally paint it on screen.

  1. semicolony.dev — go!
    semicolony.dev
    1

    You type a name and hit Enter. That name means nothing to the network yet.

  2. What’s its number?
    DIRECTORY
    2

    First the name becomes a number: a quick directory lookup (DNS).

  3. server hello hello back got it
    3

    With the number, your browser opens a connection to that machine (the TCP handshake).

  4. Line locked.
    the site
    4

    Then both sides agree a secret so the line is private and ID-checked (TLS / HTTPS).

  5. GET the page, please.
    server GET /
    5

    Now the real ask: “send me this page.” The server replies with the page’s files.

  6. 6

    Your browser draws those files into the page you see — text, images, the lot.

The whole relay that runs every time you open a page — most of it invisible.

Why it is a chain, not one big step

Each leg depends on the one before it. You cannot open a connection until you have the number, cannot lock it until it is open, cannot ask for the page until it is locked. That is why a single slow leg — a sluggish DNS lookup, a far-away server — holds up everything after it.

It usually skips ahead

Your computer remembers recent answers, so most visits skip whole legs. The number is often cached from last time, an open connection can be kept and reused, and the browser may already hold parts of the page. The full relay mostly runs on the very first visit.

The real version How an HTTP request works →
Found this useful?