Understanding IPv4: The Foundation of the Internet

Open a website on your laptop and a reply finds its way back across several networks. IPv4 supplies part of the addressing system that makes that journey possible. It tells routers where a packet is going; it does not own the network or decide who should govern it.

That distinction makes IPv4 easier to understand. Start with delivery, then look at how addresses are shared and managed.

What do the four numbers mean?

An IPv4 address looks like 192.168.1.10. Each of its four parts represents eight binary digits, so each part runs from 0 to 255. Together they form a 32-bit address. There are 4,294,967,296 possible combinations, including ranges set aside for private networks and special purposes. They are not all public addresses available for a website.

The IPv4 specification describes source and destination addresses in packet headers. Think of the destination as the address on a parcel. It helps with delivery; it is not a description of the person receiving it.

How does a packet get there?

Your device sends traffic toward a router. The router consults its forwarding information and passes the packet to the next appropriate link. Other routers repeat the process until the destination network is reached. The complete route is not written into an ordinary IPv4 address.

A network prefix groups addresses that can be handled together. The notation /24, for example, says that the first 24 bits identify the prefix. If that sounds unfamiliar, the guide to address structure and prefixes works through one small example.

Does every device need its own public address?

No. Your phone and laptop can have different private addresses while sharing a public address through your home router. Another household may use the same private numbers inside its own separate network. The private-address specification makes this reuse possible; those internal addresses are not advertised as destinations on the public Internet.

A common form of network address translation, or NAT, keeps track of connections so replies reach the correct internal device. It saves public addresses, but adds state that the network must maintain. For the mechanism and its trade-offs, read how NAT shares an address.

Finite numbers and unavailable addresses are different claims

The size of IPv4 is fixed. That does not tell a business whether it can obtain suitable addresses, at what price, or on what terms. An allocation pool, an address already in use and an address available through a transfer or lease are different things.

In Note 45, Lu Heng challenges the leap from an exhausted registry pool to the claim that IPv4 cannot support further economic activity. His argument asks readers to examine actual availability and cost, then question the administrative power justified in the name of scarcity. A useful decision starts with the operator's real options, not just the number of bits.

What does IPv6 change?

IPv6 uses 128-bit addresses and a different packet format. It offers much more address space. Adding it does not automatically remove the need to reach customers using IPv4. Running both protocols is one way to serve both groups, with operational work on each path.

IPv4 and IPv6 also both need deliberate security configuration. An address format alone neither encrypts a conversation nor decides which connections should be allowed.

Why Lu Heng brings operators into the argument

For someone running a network, addresses connect engineering to investment and control. Lu Heng argues that the people building networks and bearing their costs should be able to choose how those networks develop. In Note 29, he connects IPv4 transferability and markets to his case for decentralizing the registration layer.

The practical question is therefore larger than “Which protocol is newer?” It is: what serves the users, what does it cost to operate, and who gets to decide? The IPv6 introduction takes that question further through the example of an online shop.