The Internet

The Internet is a computer network consisting of interconnected networks that use standardized, open (nonproprietary) communication protocols. It's designed to scale to support an increasing number of users.

In short: a network of networks, used to send and receive information.

No single entity owns or controls the global Internet.

The pieces of a network

Building a network means making trade-offs

There is no single "ideal" network. Every design decision buys you something and costs you something else:

1. More direct connections = more speed
2. More connections = higher cost, more redundancy, more fault tolerance

That second one cuts both ways: the extra connections that make a network resilient are the same ones that make it expensive to build.

An affluent, densely populated area is likely to have a strong overall network because the ISPs believe it will be worth the investment.

Internet Service Providers make trade-off decisions like these every time they build out an area.

The Internet and the World Wide Web are not the same thing

The World Wide Web is a system of linked pages, programs, and files. It runs on the Internet.

The Internet is the underlying network that carries the data. The Web is one of the things built on top of it — email, video calls, and online games are others. Confusing the two is a common mistake and a reasonable exam target.

Scalability

Scalability is the capacity for a system to change in size and scale to meet new demands.

The Internet was designed to scale, and it has: it keeps working as more and more devices come online. How it manages that comes down to the protocols it runs on, which is the next section.

Redundancy is the inclusion of extra components so that a system can continue to work even if individual components fail — for example, by having more than one path between any two connected devices in a network.

Fault tolerant means a system can continue to function even when individual components fail.

This matters because elements of complex systems like a computer network fail at unexpected times, often in groups.

Routers choose the path

A router is a type of computer that forwards data across a network.

Information sent between two devices doesn't always take the same path. The routers in the network decide the path each time.

That's only possible if more than one path exists. In an optimal network there are multiple paths available, so that if one is down, traffic can still get through.

The more redundant the network, the more fault-tolerant the network is.

Redundancy costs money

Extra paths mean extra connections, and every connection costs something to build and maintain. This is the same trade-off that shows up when designing any network — redundancy is bought, not free.

The DNS is redundant too

Redundancy isn't only about cables and routers. The DNS is a group of servers working together to deliver IP addresses, and each server acts as a backup for the others, so the system keeps operating if an individual server fails.

Protocols

A protocol is an agreed-upon set of rules that specify the behavior of some system.

In class, when every message in the Internet Simulator reached everyone in the group, you solved it by inventing a protocol — an agreed format that made clear who a message was from and who it was for:

Sender – Message – Receiver

That is exactly what the real Internet does. The Internet Protocol (IP) sends data across the Internet by assigning unique numbers — IP addresses — to each connected device. Every message includes the sender's IP address, the information being sent, and the recipient's IP address, so it can be routed properly.

An IP address is the unique number assigned to each device on the Internet.

Why open protocols matter

Internet protocols are open, shared, and standardized. Anyone who wants to use Internet technologies — build a website, create a smartwatch — just needs to use the appropriate protocols for that purpose.

The openness of these protocols is a key reason the Internet has grown the way it has.

Packets

A packet is a chunk of data sent over a network. Larger messages are divided into packets, which may arrive at the destination in order, out of order, or not at all.

Packet metadata is data added to packets to help route them through the network and reassemble the original message.

A datastream is information passed through the Internet in packets.

The Internet is unreliable by default

When you can only send a limited amount of bits per message, a longer message has to be broken into pieces — and once you do that, you discover the problem: the pieces don't necessarily show up in the right order, and some don't show up at all.

By default, the Internet is inherently unreliable. Information sent over it may be dropped or may arrive out of order.

This isn't a flaw to be fixed at the wire level; it's a consequence of routers independently choosing paths. The fix is a protocol layered on top.

TCP

The Transmission Control Protocol (TCP) is an error-checking protocol that ensures transmitted packets are received by the recipient and placed in the correct order.

TCP solves both problems at once: it confirms every packet arrives, re-sending anything that was dropped, and it puts the packets back in order using their metadata.

TCP will be on the test.

UDP

The User Datagram Protocol (UDP) sends packets quickly with minimal error-checking and no resending of dropped packets.

It trades reliability for speed — useful when a late packet is worse than a missing one, like a live video call.

UDP will not be on the test, but knowing it exists helps explain why TCP's guarantees are a deliberate choice rather than the only option.

HTTP and DNS

IP addresses can change for a given device for all sorts of reasons, and it makes no sense for every device to track the address of every other device on the Internet.

The Domain Name System (DNS) solves this: it's the system responsible for translating domain names like example.com into IP addresses. It's a hierarchy of servers dedicated to keeping track of IP addresses worldwide.

HTTP (Hypertext Transfer Protocol) is a protocol for computers to request and share the pages that make up the world wide web.

How the DNS helps the Internet scale

As more devices come online, individual devices don't need to track any of the new addresses — the DNS handles it. And the DNS can always add more servers to its hierarchy to support the growth.

The digital divide is differing access to computing devices and the Internet, based on socioeconomic, geographic, or demographic characteristics.

Who is affected

Internet and computing access differ depending on socioeconomic, geographic, and demographic factors. Groups more likely to have less access include:

• Low-income households
• People in rural areas
• Older adults
• Some minority groups

Why it matters

The divide isn't only about convenience. As more essential services move online — schoolwork, job applications, healthcare, government forms — the gap between people with reliable access and people without it grows into a gap in opportunity.

These factors need to be considered when governments, schools, businesses, and individuals implement policies that depend on these technologies.

A policy that assumes everyone has a fast connection and a personal device will work well for some people and fail others entirely. Recognizing that in advance is the point.

It connects back to network design

The digital divide isn't an accident — it follows from the trade-offs made when networks get built. An area where people can pay more is more likely to get more direct connections and higher speeds. The same economics that shape a network's design also shape who ends up on the wrong side of the divide.