How Does the Internet E-Mail System Work?

A Brief E-Mail Primer

How Does the Internet E-Mail System Work?

How Do I Get E-Mail If I'm Not on a Network?

One of the nice things about the Internet e-mail system is that to use it, you don't have to know how its plumbing works. The missives you fire out, and the dispatches that come your way, are all routed automatically, without any help from the likes of you and me. However, one of the characteristics of the Internet is that its innards are never that far away. The programs we use to surf the Net are only a thin, shiny veneer covering the Net's insides.

So it pays, then, to know at least a little about the underlying mechanisms that make the Internet go. This information will come in handy, for example, when you need to set up your Internet e-mail software. It will also give you just enough background to let you pose intelligent questions to your system administrator, just in case something goes awry.

Internet E-Mail Addressing

I like to think of the Net as a giant city where the houses are computers. A neighborhood where the houses are connected with side streets is like an individual network connected via cables. In turn, each neighborhood is connected to other neighborhoods via larger roads and avenues or, for longer trips, by highways and expressways. The point is that in any city you can get from your house to any other house by traveling along a particular set of streets, roads, and highways. The Internet works the same way: You can "travel" to other computers on the Net by "following" the various communications lines that make up the Net's infrastructure.

This metaphor serves quite nicely to explain how electronic mail works. The first thing you need to know is that each user on the Internet has a unique e-mail address. Just like the full address of a house, your e-mail address tells everyone the exact location they can use to send messages to you. To help you understand the structure of these addresses, consider this generic e-mail address:

user@provider.com

This address has four parts:

user: This is the recipient's user name. Most user names are a single word representing the person's first or last name, or a combination of the two names. Some companies insist that the user name be both the first name and last name separated by a period (for example, paul.mcfedries). Other e-mail systems will use different conventions. CompuServe, for example, uses two numbers separated by a period.

@: This symbol (it's pronounced "at") separates the "who" part of the address (the part to the left of the @ sign) and the "where" part (the part to the right of the @ sign).

provider: This is the Internet name (that is, the domain name) of the user's access provider. It tells you where the user's mailbox is located.

com: The last part tells you what type of organization you're dealing with. See Table E.1 for a list of the main organization types. Notice, too, how provider and com are separated by a period. You pronounce this as "dot," so the entire address reads as "user at provider dot com."

Table E.1. Internet Organization Types

Type What it represents

com Commercial business
edu Educational institution
gov Government
int International organization
mil Military
net Networking organization
org Non-profit organization

This type of e-mail address is a relatively straightforward affair. Here's one that's a bit more complex:

biff@math.utoronto.ca

Again, the address begins with a user name (biff, in this case) and the @ sign. The utoronto part means this person is located at the University of Toronto. Why doesn't it have an edu at the end? Well, many newer e-mail addresses are forgoing the old "type" designations in favor of geographical designations (or geographical domains, as the pocket protector crowd calls them). In this case, the ca tells you that the University of Toronto is located in Canada. Table E.2 lists a few other common geographical domains. (Note that, in a bit of geographical chauvinism, most U.S. e-mail addresses don't bother with the country code.)

Table E.2. Some Internet Geographical Domains

Code The country it represents

at Austria
au Australia
ca Canada
ch Switzerland
cn China
de Germany
dk Denmark
es Spain
fi Finland
fr France
jp Japan
nz New Zealand
uk United Kingdom
us United States

The math part of the address is called a subdomain, and it's used to narrow things down a bit. Universities and other large organizations usually have various networks. The University of Toronto, for example, probably has separate networks for the Mathematics department, the Physics department, the Chemistry department, and so on. The math subdomain tells you that this person is part of the Mathematics department's network.

When you know a person's (or a company's) e-mail address, you're in business. You just fire up your e-mail software, compose a message, tell the software the address to send the message to, and then fire it off. In the same way that the post office can use the address on an envelope or a package to track down the recipient, so too does the Internet e-mail system know how to locate the specified e-mail address. You just sit back and let the Net do all the dirty work.

How Does My Mail Get There from Here?

When you dress up an e-mail message and send it out into the cold, cruel world of the Internet, it actually goes on quite a journey. The route it takes, the places it visits, and the sights it sees vary from message to message, but the basic itinerary is easy to spell out:

A special e-mail computer on your network (or on your access provider's network) wraps up the message in a electronic "envelope" (that has, like a regular envelope, the recipient's address and your return address, among other things) and sends it out. The Internet mechanism involved here is called SMTP (Simple Mail Transport Protocol), which is a member of the TCP/IP family. The provider's computer that handles outgoing mail is usually called an SMTP server.
Because there will only rarely be a direct connection between your network and the recipient's network, the mail will make a number of stops at intermediate networks along the way. At each stop, another e-mail computer temporarily stores the message while it figures out the best way to relay the message toward its ultimate destination. When that's done, the message is sent on its merry way, and the whole thing repeats at the next digital roadside rest stop.
Eventually, the message finds its way to the recipient's network, where an e-mail computer routes it to the person's mailbox. (If the recipient uses a networked e-mail program that doesn't understand Netspeak—such as Microsoft Mail or cc:Mail—the message will have to go through another computer called a gateway that converts the message into something the program can make sense of.)

This might sound like an awful lot of work just to send an e-mail note, but it's quite efficient for two reasons:

The message is traveling at nearly the speed of light, so even the longest hop between networks happens in less than the blink of an eye.
Almost all the Net's e-mail computers use SMTP to transport the messages.

How Do I Get E-Mail If I'm Not on a Network?

If you use an access provider to do the Internet thing, there's no direct way to get an e-mail message onto your computer (because you're not jacked into a network full-time). You could use the Telnet service to log in to your access provider's network and read the mail directly from your mailbox, but most people prefer to store their private messages on their own hard disk.

To solve this problem, your access provider has probably set up a special account for you on one of its computers. This is called a Post Office Protocol (POP) account, and the computer it's set up on is called a POP server. POP essentially turns the server into the electronic equivalent of a post office. When people send messages to your e-mail address, the messages are stored in a special mailbox all your own. You can then use an e-mail program such as Windows 95's Windows Messaging, Windows 98's Outlook Express, or the Mail component of Netscape Communicator to log in to this account, grab the messages, store them on your computer, and then clean out the mailbox.

NOTE: MASSIVE MISSIVE MATH
Today, it's estimated that between 30 and 60 million people use e-mail and, on the Internet alone, send well over a billion messages a month. These messages range in length from a few dozen words to a few thousand, but let's take 500 words as the average. That means Net e-mail types send more than 500 billion words a month scurrying around the world. Too big a number to comprehend? OK, let's knock it down to size. 500 billion words a month is the equivalent of sending the manuscript of a 400-page book 2,000 times every hour of every day.

Table of Contents

Internet E-Mail Addressing

How Does My Mail Get There from Here?

How Do I Get E-Mail If I'm Not on a Network?