2.1 The Basics
Big Dummy's Guide to the Internet

Go to Table of Contents. Visit Gifcom.

     Electronic mail, or e-mail, is your personal connection to the
world of the Net.
     All of the millions of people around the world who use the
Net have their own e-mail addresses.  A growing number of "gateways" tie
more and more people to the Net every day.  When you logged onto the host
system you are now using, it automatically generated an address for you,
as well.
    The basic concepts behind e-mail parallel those of regular mail. 
You send mail to people at their particular addresses.  In turn, they
write to you at your e-mail address.  You can subscribe to the
electronic equivalent of magazines and newspapers. You might even get
electronic junk mail.
    E-mail has two distinct advantages over regular mail.  The most
obvious is speed. Instead of several days, your message can reach the
other side of the world in hours, minutes or even seconds (depending on
where you drop off your mail and the state of the connections between
there and your recipient).  The other advantage is that once you master
the basics, you'll be able to use e-mail to access databases and file
libraries.  You'll see how to do this later, along with learning how to
transfer program and data files through e-mail.
    E-mail also has advantages over the telephone.  You send your
message when it's convenient for you.  Your recipients respond at their
convenience.  No more telephone tag.  And while a phone call across
the country or around the world can quickly result in huge phone
bills, e-mail lets you exchange vast amounts of mail for only a few
pennies -- even if the other person is in New Zealand.
    E-mail is your connection to help -- your Net lifeline.  The
Net can sometimes seem a frustrating place!  No matter how hard you
try, no matter where you look, you just might not be able to find the
answer to whatever is causing you problems. But when you know how to
use e-mail, help is often just a few keystrokes away: you can ask your
system administrator or a friend for help in an e-mail message.
    The quickest way to start learning e-mail is to send yourself a
message.  Most public-access sites actually have several different types
of mail systems, all of which let you both send and receive mail. We'll
start with the simplest one, known, appropriately enough, as "mail," and
then look at a couple of other interfaces. At your host system's command
prompt, type:
             mail username
where username is the name you gave yourself when you first logged on. 
Hit enter.  The computer might respond with
or, actually, anything at all (but you'll have to hit enter before
you get to the end of the screen). Hit enter.
     The cursor will drop down a line. You can now begin writing the
actual message. Type a sentence, again, anything at all.  And here's
where you hit your first Unix frustration, one that will bug you
repeatedly: you have to hit enter before you get to the very end of the
line.  Just like typewriters, many Unix programs have no word-wrapping
(although there are ways to get some Unix text processors, such as emacs,
to word-wrap).
     When done with your message, hit return. Now hit control-D (the
control and the D keys at the same time).  This is a Unix command that
tells the computer you're done writing and that it should close your
"envelope" and mail it off (you could also hit enter once and then, on
a blank line, type a period at the beginning of the line and hit enter
     You've just sent your first e-mail message.  And because you're
sending mail to yourself, rather than to someone somewhere else on the
Net, your message has already arrived, as we'll see in a moment.
     If you had wanted, you could have even written your message on
your own computer and then uploaded it into this electronic
"envelope."  There are a couple of good reasons to do this with long
or involved messages.  One is that once you hit enter at the end of a
line in "mail" you can't readily fix any mistakes on that line (unless
you use some special commands to call up a Unix text processor).  Also,
if you are paying for access by the hour, uploading a prepared
message can save you money.  Remember to save the document in ASCII or
text format.  Uploading a document you've created in a word processor
that uses special formatting commands (which these days means many
programs) will cause strange effects.
     When you get that blank line after the subject line, upload the
message using the ASCII protocol.  Or you can copy and paste the text,
if your software allows that. When done, hit control-D as above.
     Now you have mail waiting for you.  Normally, when you log on,
your public-access site will tell you whether you have new mail
waiting.  To open your mailbox and see your waiting mail, type
and hit enter.
     When the host system sees "mail" without a name after it, it
knows you want to look in your mailbox rather than send a message.
Your screen, on a plain-vanilla Unix system will display:
         Mail version SMI 4.0 Mon Apr 24 18:34:15 PDT 1989  Type ? for help.
         "/usr/spool/mail/adamg": 1 message 1 new 1 unread
         >N 1 adamg              Sat Jan 15 20:04   12/290   test
     Ignore the first line; it's just computerese of value only to the
people who run your system. You can type a question mark and hit
return, but unless you're familiar with Unix, most of what you'll see
won't make much sense at this point. 
     The second line tells you the directory on the host system where
your mail messages are put, which again, is not something you'll likely
need to know.  The second line also tells you how many messages are in your
mailbox, how many have come in since the last time you looked and how
many messages you haven't read yet.
     It's the third line that is of real interest -- it tells you who
the message is from, when it arrived, how many lines and characters
it takes up, and what the subject is.  The "N" means it is a new
message -- it arrived after the last time you looked in your mailbox. 
Hit enter. And there's your message -- only now it's a lot
longer than what you wrote!
        Message 1:
        From adamg Jan 15 20:04:55 1994
        Received: by eff.org id AA28949
        (5.65c/IDA-1.4.4/pen-ident for adamg); Sat, 15 Jan 1994 20:04:55 -0400 
        (ident-sender: adamg@eff.org)
        Date: Sat, 15 Jan 1994 21:34:55 -0400
        From: Adam Gaffin <adamg>
        Message-Id: <199204270134.AA28949@eff.org>
        To: adamg
        Subject: test
        Status: R
        This is only a test!
     Whoa! What is all that stuff? It's your message with a postmark
gone mad.  Just as the postal service puts its marks on every piece of
mail it handles, so do Net postal systems.  Only it's called a
"header" instead of a postmark. Each system that handles or routes
your mail puts its stamp on it.  Since many messages go through a
number of systems on their way to you, you will often get messages
with headers that seem to go on forever.  Among other things, a header
will tell you exactly when a message was sent and received (even the
difference between your local time and Greenwich Mean Time -- as at the end
of line 4 above).
     If this had been a long message, it would just keep scrolling
across and down your screen -- unless the people who run your public-
access site have set it up to pause every 24 lines.  One way to deal
with a message that doesn't stop is to use your telecommunication
software's logging or text-buffer function.  Start it before you hit
the number of the message you want to see.  Your computer will ask you
what you want to call the file you're about to create. After you name
the file and hit enter, type the number of the message you want to see
and hit enter.  When the message finishes scrolling, turn off the
text-buffer function. The message is now saved in your computer. 
This way, you can read the message while not connected to the Net
(which can save you money if you're paying by the hour) and write a
reply offline.
     But in the meantime, now what?  You can respond to the message,
delete it or save it.  To respond, type a lowercase r and hit
enter.  You'll get something like this:
          To: adamg
          Subject: Re:  test
     Note that this time, you don't have to enter a user name.  The
computer takes it from the message you're replying to and
automatically addresses your message to its sender. The computer also
automatically inserts a subject line, by adding "Re:" to the original
subject.  From here, it's just like writing a new message. But say you
change your mind and decide not to reply after all. How do you get out
of the message? Hit control-C once. You'll get this:
          (Interrupt -- one more to kill letter)
If you hit control-C once more, the message will disappear and you'll
get back to your mail's command line.
     Now, if you type a lowercase d and then hit enter, you'll
delete the original message.  Type a lowercase q to exit your
     If you type a q without first hitting d, your message is
transferred to a file called mbox.  This file is where all read, but
un-deleted messages go.  If you want to leave it in your mailbox for
now, type a lowercase x and hit enter.  This gets you out of mail
without making any changes.
     The mbox file works a lot like your mailbox.  To access it,
           mail -f mbox
at your host system's command line and hit enter. 
     You'll get a menu identical to the one in your mailbox from which
you can read these old messages, delete them or respond to them.  It's
probably a good idea to clear out your mailbox and mbox file from
time to time, if only to keep them uncluttered.
    Are there any drawbacks to e-mail?  There are a few.  One is that
people seem more willing to fly off the handle electronically than in
person, or over the phone.  Maybe it's because it's so easy to hit r
and reply to a message without pausing and reflecting a moment. 
That's why we have smileys (see section 2.4)!  There's no online
equivalent yet of a return receipt: chances are your message got to where
it's going, but there's no absolute way for you to know for sure unless
you get a reply from the other person.               
     So now you're ready to send e-mail to other people on the Net. 
Of course, you need somebody's address to send them mail.  How do you
get it? 
     Alas, the simplest answer is not what you'd call the most
elegant: you call them up on the phone or write them a letter on paper
and ask them.  Residents of the electronic frontier are only beginning
to develop the equivalent of phone books, and the ones that exist
today are far from complete (still, later on, in Chapter 6, we'll show
you how to use some of these directories).
     Eventually, you'll start corresponding with people, which means
you'll want to know how to address mail to them.  It's vital to know
how to do this, because the smallest mistake -- using a comma when you
should have used a period, for instance, can bounce the message back
to you, undelivered.  In this sense, Net addresses are like phone
numbers: one wrong digit and you get the wrong person.  Fortunately,
most net addresses now adhere to a relatively easy-to-understand
     Earlier, you sent yourself a mail message using just your user-
name.  This was sort of like making a local phone call -- you didn't
have to dial a 1 or an area code.  This also works for mail to anybody
else who has an account on the same system as you.
     Sending mail outside of your system, though, will require the use
of the Net equivalent of area codes, called "domains." A basic Net
address will look something like this:
     Tomg is somebody's user ID, and he is at (hence the @ sign) a site
(or in Internetese, a "domain") known as std.com.  Large organizations
often have more than one computer linked to the Internet; in this case,
the name of the particular machine is world (you will quickly notice
that, like boat owners, Internet computer owners always name their
     Domains tell you the name of the organization that runs a given
e-mail site and what kind of site it is or, if it's not in the U.S.,
what country it's located in.  Large organizations may have more than
one computer or gateway tied to the Internet, so you'll often see a
two-part domain name; and sometimes even three- or four-part domain
     In general, American addresses end in an organizational suffix,
such as ".edu," which means the site is at a college or university.
Other American suffixes include:
          .com for businesses
          .org for non-profit organizations
          .gov and .mil for government and military agencies
          .net for companies or organizations that run large networks. 
     Sites in the rest of the world tend to use a two-letter code that
represents their country.  Most make sense, such as .ca for Canadian
sites, but there are a couple of seemingly odd ones.  Swiss sites end
in .ch, while South African ones end in .za.  Some U.S. sites have
followed this international convention (such as well.sf.ca.us).
    You'll notice that the above addresses are all in lower-case. 
Unlike almost everything else having anything at all to do with Unix,
most Net mailing systems don't care about case, so you generally don't
have to worry about capitalizing e-mail addresses.  Alas, there are a few
exceptions -- some public-access sites do allow for capital letters in
user names.  When in doubt, ask the person you want to write to, or let
her send you a message first (recall how a person's e-mail address is
usually found on the top of her message). The domain name, the part of the
address after the @ sign, never has to be capitalized.              
     It's all a fairly simple system that works very well, except,
again, it's vital to get the address exactly right -- just as you have
to dial a phone number exactly right.  Send a message to tomg@unm.edu
(which is the University of New Mexico) when you meant to send it to
tomg@umn.edu (the University of Minnesota), and your letter will either
bounce back to you undelivered, or go to the wrong person.
     If your message is bounced back to you as undeliverable, you'll
get an ominous looking-message from MAILER-DAEMON (actually a rather
benign Unix program that exists to handle mail), with an evil-looking
header followed by the text of your message. Sometimes, you can tell
what went wrong by looking at the first few lines of the bounced
message.  Besides an incorrect address, it's possible your host system
does not have the other site in the "map" it maintains of other host
systems. Or you could be trying to send mail to another network, such
as Bitnet or CompuServe, that has special addressing requirements.
    Sometimes, figuring all this out can prove highly frustrating. 
But remember the prime Net commandment: Ask.  Send a message to your
system administrator.  He or she might be able to help decipher the
    There is one kind of address that may give your host system
particular problems.  There are two main ways that Unix systems
exchange mail.  One is known as UUCP and started out with a different
addressing system than the rest of the Net.  Most UUCP systems have
since switched over to the standard Net addressing system, but a few
traditional sites still cling to their original type, which tends to
have lots of exclamation points in it, like this:
     The problem for many host sites is that exclamation points (also
known as "bangs") now mean something special in the more common systems
or "shells" used to operate many Unix computers. This means that
addressing mail to such a site (or even responding to a message you
received from one) could confuse the poor computer to no end and your
message never gets sent out. If that happens, try putting backslashes in
front of each exclamation point, so that you get an address that looks
like this:
Note that this means you may not be able to respond to such a message
by typing a lowercase r  -- you may get an error message and you'll
have to create a brand-new message.
    If you want to get a taste of what's possible through e-mail,
start an e-mail message to
Leave the "subject:" line blank.  As a message, write this:
              send quote
Or, if you're feeling a little down, write this instead:
              send moral-support
     In either case, you will get back a message within a few seconds to
a few hours (depending on the state of your host system's Internet
connection).  If you simply asked for a quote, you'll get back a
fortune-cookie-like saying.  If you asked for moral support, you'll also
get back a fortune-cookie-like saying, only supposedly more uplifting.
      This particular "mail server" is run by Oregon State University.
Its main purpose is actually to provide a way to distribute agricultural
information via e-mail.  If you'd like to find out how to use the
server's full range of services, send a message to its address with this
line in it:
              send help
You'll quickly get back a lengthy document detailing just what's
available and how to get it.
     Feeling opinionated?  Want to give the President of the United
States a piece of your mind?  Send a message to president@whitehouse.gov
Or if the vice president will do, write vice-president@whitehouse.gov.
     The "mail" program is actually a very powerful one and a Netwide
standard, at least on Unix computers.  But it can be hard to figure
out -- you can type a question mark to get a list of commands, but
these may be of limited use unless you're already familiar with Unix.
Fortunately, there are a couple of other mail programs that are easier
to use.