Chapter 102·Beginner·10 min read
IP Addresses Explained: How Every Device on the Internet Gets Found
What is an IP address, really? A plain-English explanation of IPv4 and IPv6, public vs private addresses, why we ran out of addresses and how NAT saved us, and how your device gets an address in the first place.
July 31, 2026
The previous chapter revealed the internet's first surprise: the network doesn't route by names like example.com, it routes by numbers. This chapter is about those numbers — IP addresses — the addressing system that lets any of the billions of devices online find and reach any other.
Without addresses, nothing else works. DNS exists to produce one; packets exist to be delivered to one; routing exists to find a path to one. So this is the foundation everything sits on.
What an IP address is for
An IP address (Internet Protocol address) is a numeric label assigned to a device on a network. Its job is exactly what a postal address's job is: so that data can be delivered to the right place, and replies can find their way back.
That "replies" half is the crucial part. When your browser asks example.com for a page, the request carries your IP address as the return address. The server has no idea who you are otherwise — it sends the page back to the address on the envelope. Every conversation on the internet is two-way, so every participant needs a findable address.
IPv4: the original, and its famous shortage
The addressing scheme that built the internet is IPv4, and you've seen its format everywhere:
192.168.1.1Four numbers, each 0–255, separated by dots. Under the hood each is 8 bits, so an IPv4 address is 32 bits total — which yields about 4.3 billion possible addresses.
In 1981, 4.3 billion sounded like more than anyone could ever need. Then the internet ate the world: every phone, laptop, server, smart bulb, thermostat, and car wanted an address. The mathematics were unforgiving.
IPv6: addresses beyond comprehension
The permanent solution is IPv6, which expands the address from 32 bits to 128 bits. It looks different — hexadecimal, separated by colons:
2001:0db8:85a3:0000:0000:8a2e:0370:7334The number of addresses this provides is genuinely hard to state meaningfully: 340 undecillion — that's 340 followed by 36 zeros.
To make it concrete: there are enough IPv6 addresses to assign billions to every grain of sand on Earth and never run out. The shortage problem is solved forever — which is why adoption, though gradual, keeps climbing. IPv4 and IPv6 run side by side today; most networks speak both.
Public vs private: why your laptop's address is a secret
Here's a puzzle. If IPv4 ran out years ago, how does your home — with a laptop, two phones, a TV, and a smart speaker — have addresses for all of them?
The answer is that most devices don't have a public internet address at all. There are two kinds:
| Public IP | Private IP | |
|---|---|---|
| Visible to | The whole internet | Only your local network |
| Who has one | Your router (one per home, typically) | Every device behind the router |
| Example ranges | Globally unique | 192.168.x.x, 10.x.x.x, 172.16–31.x.x |
| Reusable? | No — must be unique worldwide | Yes — every home reuses the same private ranges |
Your devices get private addresses that mean something only inside your home network. Millions of homes all use 192.168.1.x simultaneously with no conflict, because those addresses never travel out to the internet. Only your router has a public address, and it represents the whole household to the outside world.
NAT: the translation trick that stretched IPv4
So how does a reply to your private laptop find its way back through the single public address? Through NAT — Network Address Translation — running on your router.
When your laptop sends a request, the router rewrites the return address to its own public IP and remembers that this particular conversation belongs to your laptop. When the reply comes back, the router looks up its notes and forwards it to the right internal device. Dozens of devices share one public address, each conversation tracked separately.
NAT is why IPv4 didn't collapse the moment we ran out — it lets one public address stand in for a whole network. It's a clever hack (and it complicates some peer-to-peer scenarios, which is part of why IPv6, where every device can have its own public address, is the cleaner future). But for the everyday internet, NAT is quietly doing this translation for nearly every request you make.
How your device actually gets an address
You've almost certainly never typed an IP address into your own device to get online. That's because addresses are assigned automatically the moment you join a network, by a protocol called DHCP (Dynamic Host Configuration Protocol).
The exchange is quick: your device joins the Wi-Fi and essentially broadcasts "I'm new here, does anyone have an address for me?" The network's DHCP server (usually your router) replies "here, use 192.168.1.24 for the next while," along with a couple of other essentials — including the address of the DNS server it should ask for name lookups (which is exactly the next chapter). It's the digital equivalent of being handed a visitor badge and a map on arrival.
Recap
- An IP address is a device's mailing address — data is delivered to it, and replies come back to it; every internet conversation needs one on each end.
- IPv4 uses 32-bit addresses (
192.168.1.1), giving ~4.3 billion — which ran out in the 2010s as the world came online. - IPv6 uses 128-bit addresses (~340 undecillion), permanently solving the shortage; the two run side by side today.
- Most devices use private addresses meaningful only on your local network; your router holds the single public address, and NAT translates between them — the hack that stretched IPv4 for decades.
- You never set your address by hand — DHCP assigns one automatically (plus your DNS server) when you connect.
You now know what an address is. Next: how the friendly name you actually type gets turned into one. Continue to How DNS works.