What is an IP Address?

Start here. An IP address is a number that helps networks deliver data toward the right device or network. It answers a practical question: where should this traffic go next? It does not, by itself, prove who a person is, who owns a company, or who has the authority to speak for everyone using a network.

An address tells networks where to send data

When you open a website, send a message or join a video call, your device is part of a network conversation. Data is divided into packets, and each packet needs a destination that other networks can understand. The IP address gives routers a usable destination and helps them move the packet closer to it.

That is why an IP address is closer to a location in a delivery system than to a name on an identity card. A network may use it to reach a device, a gateway or a whole block of infrastructure. The address can change. Several devices can appear behind one public address. A device can also use a private address inside a local network. None of these details turns the number into a complete description of the person using the connection.

One idea, several kinds of address

IPv4 uses 32-bit addresses, which is why it is commonly written as four numbers separated by dots. IPv6 uses a much larger address space and is written in hexadecimal groups. Both serve the same broad purpose: helping networks distinguish destinations and exchange traffic.

A public address can be recognised across the wider Internet when routing is configured for it. A private address is used inside a home, office or data centre and is not normally announced as a public destination. Dynamic addresses may be assigned for a while and later reused; static addresses are intended to remain stable for longer. These are operational choices, not statements about human identity or political authority.

What happens when you open a website

A domain name is useful for people, while an IP address is useful for the networks carrying the request. A name service can return an address. Your device then sends packets to a local gateway, routers compare the destination with their routing information, and independent networks pass the packets onward. The reply follows a route back to a reachable address.

This is a chain of coordination. Names, numbers, routing rules and local operators each do different work. Keeping those layers separate makes the system easier to understand: the service that records a number is not necessarily the network that carries the traffic, and neither is automatically the owner of every activity that depends on the number.

The number is not the person

An IP address can be useful evidence about a network path or an operational record. It is not reliable proof of a person’s identity, motives or legal responsibility on its own. A shared gateway, a mobile network, a virtual private network, a carrier-grade NAT or a compromised device can all complicate the relationship between a visible address and a human being.

The same discipline applies to institutional records. A database can say which party is recorded as responsible for a number resource. That record matters for routing, abuse handling and continuity. It does not automatically prove ownership of every downstream business, nor does it give the record keeper a mandate to decide for every operator or user affected by it.

Note 2 begins with this distinction: an address book can help a system coordinate without becoming the sovereign of everything written in it. Note 49 follows the next step, asking how a technical reference becomes accepted authority.

When the address book becomes authority

Coordination is necessary when two networks must agree that a number is unique and usable. The danger appears when a necessary record becomes the only accepted route to recognition. If operators cannot change the institution that keeps the record, the administrator’s practical power may become larger than the technical task that justified it.

Participation does not solve that problem by itself. An administrator may invite people to meetings, publish a process or describe itself as speaking for a community. Those actions can be useful, but attendance is not the same as authorization. A technical role does not automatically become political representation across countries or continents.

Note 73 gives readers a direct way to examine that claim: who was actually authorized to decide, who bears the consequences, and what can affected participants do when the decision is wrong?

The better question is whether you can leave

A resilient coordination system should preserve the useful facts without making one administrator irreplaceable. It should be possible to verify who controls a resource, keep an accurate history, transfer that control and continue operating when the coordinating service changes.

In practice, that means designing for:

• unique identifiers and records that can be checked;
• clear evidence of operational control;
• portable records that do not disappear when a provider changes;
• auditable updates and an understandable dispute path;
• replacement coordinators that cannot create conflicting claims; and
• continuity for networks that are already carrying traffic.

This is the practical meaning of decentralization in this context. It does not mean that nobody coordinates anything. It means coordination remains a service with boundaries, evidence and an exit. Note 72 develops the proposed rights of uniqueness, portability and continuity in full.

Why the exit has to exist before a crisis

Dependence is easiest to ignore while every record is accepted and every provider is cooperative. Once routing, contracts, security policies and customer operations all rely on one institution’s continued recognition, leaving can become expensive even when the underlying network is still working.

That is why portability and replacement should be tested during normal operation. The point is not to predict that every administrator will fail. The point is to avoid discovering during a dispute that the system never designed a way to keep the network running without the same gatekeeper.

Three things to carry forward

1. An IP address is a destination used by networks, not a complete identity of a person.
2. A registry record can coordinate uniqueness without granting its administrator unlimited political or commercial authority.
3. A durable Internet needs accurate records, verifiable control and a real way to change coordinators while the network continues to work.

If this is your first step into Lu Heng’s argument, begin with Note 2, then follow the questions of authority, representation and replaceable coordination through Notes 49, 72 and 73.