As Douglas Adams said about the universe, the Internet is big, really big; you won't believe just how vastly, hugely, mind-bogglingly big it is. On a typical Sunday, as I'm writing this, internetlivestats.com estimates that the world's 4.5 billion Internet users have fired off around five hundred million tweets, performed four billion Google searches, sent nearly two hundred billion emails, and moved an astounding five quintillion bytes of data (that's five billion billions!) across the countless cables that connect our computers, phones, televisions, and, well, pretty much everything.
It's not strictly true that the Internet was intended as a communication network that could withstand a nuclear strike—in its first incarnation, it connected only four computers, all at research universities on the West Coast. (The first message sent over the new network, dubbed ARPANET, consisted of two letters: “LO.” That sounds appropriately heraldic, though really researchers were trying to send the word “login” until the computer crashed.) But still, it was the 1960s, the Cold War was always a hot topic, and the early Internet engineers were military-funded and thinking existentially.
That's why the Internet doesn't have a central clearinghouse, like an old-fashioned telephone switchboard, that all traffic must pass through. Instead, the network is deliberately decentralized by using millions of specialized networking devices called routers, each with its own map (a route table) of places to forward traffic.
Internet traffic itself breaks down into tiny chunks of data called packets. Each packet can take a different route to its destination, sometimes traversing several different “hops” between routers along the way. If one router were to disappear, say because Khrushchev got frisky with some ICBMs, theoretically other parts of the network could continue communicating as the routers ping each other and figure out new traffic paths in real time. This is what we call a packet-switching network, and it has changed the world forever.
If you've been a citizen of the Internet for any length of time, you've probably encountered the notion of an IP address, which is the string of numbers and punctuation that identify your computer online.
IP stands for Internet Protocol, and it's more than just numbers. IP defines the structure of a data packet. IP addresses identify each packet and tell them where to go. All Internet routers and switches must understand IP to ensure that packets reach the correct destination.
Internet Protocol is that rare beast in the tech world: a universal standard. Everything speaks IP. Voice chat? That's voice-over-IP (VoIP). Remote desktops? PC-over-IP (PCoIP). Learn IP, and you have grasped the core of the Internet. The rest of the networking technologies fit around it like melee diamonds around the central jewel in a ring.
Most introductions to computer networking start by trying to explain something called the OSI model, which is a list of seven “layers” of networking, some of which are basically fantasy. Nobody worries about all seven of those layers in the real world. These are the important ones for you to know:
It would be wrong to think of the Internet as a single organism, like a brain, where every node is connected with equal weight. Rather, the Internet is balkanized, made up of many smaller networks that must figure out how to communicate with each other. Your Internet service provider (ISP)—like Comcast or AT&T—runs a mini-internet that forwards data to other ISPs to keep you online.
This process is far from straightforward and quite a bit beyond the scope of this book—it involves fiscal and political calculations, as well as constant vigilance over who to trust. (Imagine if I could convince your routers to send traffic into my evil network, pretending I was a valid destination!) Just know that there is some pretty interesting math called the Border Gateway Protocol involved, and it works way better than you would have any right to expect.
Finally, a quick word about the Domain Name System (DNS). DNS itself is a web service (some might even say a “cloud service”) that translates the friendly names of websites, like wiley.com, into IP addresses so that your computer can communicate with the remote server to download the web page. If your computer hasn't talked to wiley.com before, it will have to contact the nearest DNS server. That server might need to talk to another server, and so on, up to a small number of authoritative root name servers around the world. The root name servers have the final word on what domain names map to which IP addresses.
If that sounds like a possible weak spot for the ol' nuclear apocalypse, well, you're not entirely wrong. Hackers frequently target root name servers with distributed denial-of-service attacks (see “The Great Cloud Heist (and How to Foil It)” in hopes of seriously damaging the Internet. Thankfully, though, the name servers are pretty resilient by now, and most computers cache the DNS names in case of a temporary outage, so it's probably okay.
From massive undersea cables to the little modem in your closet, the Internet connects the modern world. It's brought along memes, trolls, Instagram, and a complete revolution of our lives and work.
We barely mentioned the word “cloud” in this chapter, and yet we didn't really have to. Without the Internet, the concept of cloud would be meaningless. You'd be back in the mainframe days, plugging your terminal into the central computer. The Internet puts more than just a world of information at your fingertips: it can turn any laptop into a supercomputer.
18.227.0.192