This file should be called EMAIL025.TXT
Version 0.2.5 (beta release)
17 July 1993
This rough version is missing 8 out of 28 chapters and 1 out of 5
appendices.
Copyright (c) 1993 by John E. Goodwin. All Rights Reserved.
You may make and distribute verbatim copies of these course notes for
non-commercial purposes using any means, provided this copyright notice
is preserved on all copies.
For information on taking the internetworking course, contact
John Goodwin (jgoodwin@adcalc.fnal.gov)
P.O. Box 6022
St. Charles, IL 60174, U.S.A.
E-MAIL 101
If you like those little machines that give you 24 hour access to your
bank account, you'll love the Internet. I suppose there are still
people who, given a choice, will go to a drive-through teller just so
they can deal with a "live person" instead of a machine. But even those
people will admit that it is nice to have the option of doing things for
yourself, on your own schedule, anywhere. Do you remember what it was
like before automatic tellers? Banks closed at 3 p.m. on weekdays.
Each Saturday you had to guess how much money you would need for the
following week. If you were wrong you had to cash a check at a food
store (and maybe buy something you didn't want). And if you were out of
town? Well, there were always credit cards.
We don't do that anymore. I think many people go to the automatic
teller because they like the privacy of handling their own business
without having to explain it all to someone else. And we like the
illusion of having access to our "own" account anytime we want.
There are disadvantages to using an automatic teller card too--you may
have to pay a fee each month or even for each transaction--and you have
to remember to deduct those fees from your account balance or you will
bounce checks. But I'll bet you feel pretty competent using an
automatic teller and don't lose much sleep worrying over the fees.
This course is designed to give you that same sense of freedom and
competence with the Internet that you have with an automatic teller
machine or the telephone. With a home computer, a modem, and
communications software, you can connect to other computers over the
phone line to exchange electronic mail (E-mail), trade files, or search
for information. Many of those computers are connected to the worldwide
network called the Internet. Some few of them will--for a fee--let
*you* connect to the Internet. From there you can dial any of 900,000
or more computers, send E-mail to any of 25 million people, and access
hundreds of free, informative services.
In short, you are on the verge of a new method of communicating with
people and machines called "internetworking." Internetworking lets you:
o Avoid playing phone tag;
o Sign up to receive special interest electronic newsletters and
journals;
o Access hundreds of information services and document collections in
exactly the same way--no need to have hundreds of sets of different
instructions or hundreds of (expensive) special purpose software
packages;
o Find and communicate with other people who share your interests.
Internetworking is an essential skill for the '90s. Your children will
find it as common as viewing television or using the telephone. It
still has a few rough edges--but we'll explain those.
There is actually no single network owned by one company called the
Internet. Instead, many medium-sized networks have grown together to
create a "phone system" that connects together nearly a million
computers. Many hundreds of these computers allow some form of public
access. You can get the latest news or weather, download information
about Government programs or high-tech products, search on-line library
catalogues and databases, download free software, and do many other
things, with little or no monetary investment beyond the cost of your
home computer.
Using the Internet need not be expensive: you can get on the Internet
for as little as $10 a month if you own (1) a home computer, (2) a $50
modem, and (3) some communications software (under $100). There are
more expensive ways to connect to the Internet, of course. These ways
make sense for businesses or organizations that make heavy use of the
network. But in this course we will discuss methods that cost in the
$10-$40/month range. These methods are suitable for exploring the net
after hours and for casual use. We will provide some basic information
about more expensive methods of connecting (Appendix C) so that you can
make informed decisions if your networking needs should increase in the
future.
Internetworking well means mastering a whole host of skills--connecting
two computers together using the Internet is just the beginning. You
have to learn methods for transferring information from the remote
computer to your own. This is a complex task that may involve using a
file transfer protocol and compression techniques.
Because the information world is so vast, your biggest problem will most
likely not be connecting to the Internet. It will be finding what you
want. Thus, this course covers not only the mechanics of making a
connection and transferring files, but techniques for locating material
as well. And of course you will want to be savvy about the costs of
different connection methods. This means estimating whether it will
cost you more per Megabyte to transfer the information or to have it
faxed to you by a friendly librarian.
This course is intended for the general public--students,
businesspersons, librarians, teachers, writers, journalists--in fact
anyone who needs to find information and communicate with others.
Whether you are researching a paper, writing an article, or trying to
get technical information on a product, you will use these techniques
over and over.
Chapters marked with an asterisk are omitted from this edition.
COURSE CONTENTS
1 : The Past and the Future of Internetworking
2 : What Is the Internet?
3 : How Do I Connect to the Internet?
4 : Who Pays for the Internet?
5 : Internet Basics
6 : Getting on the Internet Step by Step
7 : Programs and Pictures
8 : File Compression Methods for Faster Transfer
9 : What to Do When You Only Have E-mail
10 : Employee Development: How to Get Your Employees Internetworking
Part II Special Concerns
11 : Special for Businesses
12 : Special for Students and their Parents
13 : Special for Writers, Journalists, Publishers, and Printers
14 : Special for Elementary and High School Teachers
15 : Special for Librarians
16 : Special for Scholars
17 : Special for Churches, Synagogues, and Mosques
Part III Research, Organization, and Writing
18 : Research Methods I: Basic Navigation Methods
*19 : Research Methods II: Usenet Newsgroups
20 : Research Methods III: Advanced Techniques
*21 : Organizing Information
*22 : Information Structures
*23 : Boolean Logic
*24 : Writing for an Internetworked World: Basic Problems
*25 : Writing for an Internetworked World: Getting Through to your
Audience
Part IV Resources
*26 : The Internet Address Book
27 : Bibliography
*28 : Glossary
Appendix A. Computer Hints for the *Really* Green
Appendix B. Using a Modem
*Appendix C. Technical Details of an Internet Connection
Appendix D. Just Enough UNIX
Appendix E. The Ten Best Things To Get If You Only Have E-Mail
The Past and Future of Internetworking
There is an old fashioned way to connect with other computers and share
information and there is a modern way. It is helpful to compare the two
methods briefly in order to make contact with methods you may already
know and to show off the advantages of using the Internet.
The old fashioned (ca. 1980) method of making contact with other
computers is through a bulletin board service (BBS). Bulletin board
services grew up in the late 70s as a method for sharing software,
talking, playing games, etc. with a personal computer. They range in
size from small special interest Bulletin Boards with a local following
to giant national boards like CompuServe, GEnie, and The Source. You
access a bulletin board with a modem and communication software by
dialing a telephone number. Of course, if you don't live in the local
area of the BBS you have to use a long distance carrier. This may add a
couple dollars per hour to the connect fee.
The basic services offered by a Bulletin Board and by the Internet are
similar:
(1) Access to a host computer (Internet TELNET command)
(2) File transfer capability (Internet FTP command)
(3) The ability to contact other BBS members individually (Internet
E-mail)
(4) The ability to post messages for general consumption in any of
several catagories. (Usenet Newsgroups)
The difference is that whereas each BBS has its own dialup procedure,
menu interface, file transfer methods, billing policies, and so on, THE
INTERNET USES ONE METHOD FOR ALL COMPUTERS IN THE WORLD. The savings in
terms of the "learning curve" is staggering. Once you know how to use
anonymous FTP you know how to get information from *hundreds* of
providers. It is like the difference between using the postal service
or using a special courier for each person you write a letter to. Once
you know how to address the envelope and put the stamp on, you can write
anyone.
The Internet has the added savings that any communications software you
buy for it works with all providers. You do not have to buy (or
customize) special software for each information provider. Thus you can
use one familiar graphical user interface ("windows" program) to connect
with any computer. The usual situation where you have to buy a special
"client" program to connect to each kind of "server" is replaced with a
situation in which you have a single program that any "server" out there
has to comply with. This standardization is the main advantage of using
the Internet.
Bulletin Boards are still around. In fact, one of the easiest ways to
connect to the Internet is through a national bulletin board service.
One disadvantage of this method is that--as of this writing--national
BBS's like CompuServe offer only E-mail. You can't FTP or Telnet from
them. And they often charge per message for E-mail, so using them can
be quite expensive. There are better ways.
What is the Internet?
The best way to think of the Internet is as a communications medium like
the Telephone, Television, or the Postal Service. Using the Internet
you can send a any written text by E-mail. This is rather like mailing
a letter and having it arrive in seconds--three days in the most
backwards parts of the world. Using a special protocol called File
Transfer Protocol you can transfer text files that are too long to mail
(over about 50 pages) or even transfer graphics and programs. If E-mail
is the equivalent of "talking" to a person, then Telnet, the third main
Internet service, is equivalent to telephoning a computer. As long as
you know the password for logging on to a computer, you can access and
search any of nearly a million computers. Details of E-mail, FTP, and
Telnet are contained in Chapter 5, Internet Basics.
This chapter puts the Internet into context. Rather than concentrating
on the trees that will occupy us in later chapters, it paints a big
picture of the computing world in which the Internet has evolved. When
you pick up a telephone receiver you know you can dial households,
businesses, or government offices. You can dial 800 numbers or 411 for
information. You know how to get the time or weather, get your credit
card balance, or leave a message on an answering machine. In short, you
have a good idea of what might possibly be at the other end of the line
and a great deal of experience with negotiating their various
intricacies. But you are new to the Internet. Some sense of "what's
out there" in this new world is necessary to avoid getting lost in the
thickets of acronyms, numbers, and procedures developed by different
vendors.
As we approach the middle of the 90s, the normal working situation in
offices is approaching something like this: there is a Local Area
Network (LAN) connecting together personal computers, workstations, and
mainframes of different makes. The LAN (pronounced like "land" without
the "d") may be connected to other LANs as part of a Wide Area Network
(WAN). The WAN may or may not be part of the global network called the
Internet. In colleges, universities, and research laboratories it
likely is part of the Internet; in the commercial world, except for a
few high-tech companies, it likely is not. But the difference between
academia and the commerical world is rapidly becoming blurred.
In addition to the LANs and WANs there are many, many home and office
computers that *could* be part of the global network using a modem-to-
host connection. These computers can be the portable computers of
outside salespersons connecting to the central office to file a report,
a computer in a home-operated desktop publishing company connecting
briefly to the Internet to get a graphic for a newsletter, or a parent
sending E-mail to their child at college.
Getting Over Shell-Shock
Let's face it. Not many members of the public--even the computer
literate public-are on the Internet. There are three reasons that using
the Internet for the first time can be rather intimidating, even though
it is actually rather simple to use, when you get down to the nitty-
gritty of internetworking:
o Getting on to the Internet can be a little bit complicated;
o The capability of logging on to computers you've never used before
by its very nature means facing unfamiliar--and hence uncomfortable--
situations; and
o The world is a very big place.
I like to think of the first problem--getting on to the Internet--by
remembering what it was like using an "alternative" long distance
service before the breakup of the Bell monopoly. People who used the
alternative carriers had to dial all sorts of access codes--very often a
local access number, a credit card number, a security code, *and* the
number of the party they were calling. They knew that whatever came
after that was going to be easier.
That's what getting on the Internet is like. You may have to dial a
local access number, get your modem settings right, and type the right
magic combination of words; but after all that, actually *using* the
Internet is simple. We'll talk you through the initial steps--after a
while (and some frustration) it will be as unconscious as unlocking your
front door or tuning a television set.
The second problem is a little more substantial. Using the Internet,
you can get yourself into situations that are, well, experimental.
Because the Internet gives you the freedom to "go anywhere" and "do
anything"--at least if you know the passwords--you can uncover strange
incompatibilities and unfamiliar systems. I call this experience "shell
shock". At some point you will likely find yourself face to face with a
computer program that expects you to type a command you don't happen to
know.
You can mostly avoid such situations by only trying things about which
you have good information. The situation is not much different from
using a telephone: if you stick to well-worn paths like dialing local
numbers or simple long distance calls, you will have little trouble; but
if you start dialing other countries or special numbers you may be in
for a surprise or two.
When you do log on to a new (previously unknown) computer, you can
expect to come face to face with something called a "shell prompt".
Shell prompts look like this:
% (or some other obscure symbol, like a dollar-sign)
or this:
mail> (a favourite--means you're in some sort of mail program)
or like this:
prez23:
(means 23rd command since you logged into computer "prez").
A prompt means the other computer expects you to type a command for its
"shell", or "command interpreter". The shell is the outer layer that
you, the user, interact with.
Two other types of "user interfaces" you might encounter are:
o menu systems that give you choice of numbers
o "window" systems or graphical user interfaces (GUIs, pronounced
"gooey").
Menu systems are popular on bulletin board services, and usually present
no problem to the novice. Their weakness is that they get cloying after
about five minutes. Most menu systems that are designed to be used for
that length of time or longer have a "command mode", where you get--you
guessed it--a shell prompt.
Even windowing systems (you know if you have one of these) very often
give you a window that "emulates a terminal", i.e. that gives you a
shell prompt inside.
So, you see, in each of the three common user interfaces--command line,
menu-driven, or graphical user interface--you will likely encounter, at
some time in your life, a shell prompt. Advice on things that will
likely work to get you started is given here and in Appendix A. Some
experience of other kinds of computers, especially computers that use
the UNIX operating system is useful. Some tips about using UNIX and
other operating systems you may not have encountered before is given in
Appendix D.
If you do get stuck, whether by being experimental or just by accident,
it is helpful to remember a few points:
o You can always disconnect from a remote service by using your
communications software to "hang up".
o If the computer gives you a strange symbol like a percent sign or a
dollar sign and just sits and stares at you, you can try "help" or "?"
to try to find out what the computer expects, or else try "exit",
"quit", "bye", "logoff", or something similar, to return to where you
were before.
o Many times, when you log on to a system, you will get instructions
on how to get further help or how to "escape" back to your own system.
You should remember these or write them down!
o As a last resort, exit the communications program (and all other
active programs) and shut off your computer, turn off your modem, and
disconnect it from the phone line. Be sure to do these steps in the
order prescribed. It is unwise (though tempting) to simply turn your
computer off and on, or to pull the plug on your modem with your
computer running.
More suggestions for the inexperienced are given in Appendix A.
The final hurdle to using the Internet is that the world, even the world
of the Internet, is indeed a very large place. When using the Internet
you have to decide:
o where to go;
o what information you want (and where it might be); and
o how to get to it (and get it back home intact).
For a system as vast as the Internet, these are hard decisions. Often,
the only strategy that works is to explore and try different things.
This course is designed to get you over the initial hurdles, give you a
fair grounding in methods that work, and point you in the right
direction. The exploration is up to you. As an initial orientation, we
describe the "three worlds of the known Internet" in the next section.
The Three Worlds of the Known Internet
The Internet, like ancient Gaul, is divided into three parts. These
parts are not so much territories as worlds, each with their own sets of
assumptions, favorite dialects, and favored equipment. We may
conveniently refer to them as the "PC world", the "UNIX world", and the
"Mainframe world". Here's a dossier on each of the three worlds:
World: Personal Computer (PC)
Typical Equipment: IBM PC and clones (85%) , Macintosh (15%)
Conventional Operating Systems: MS-DOS, OS/2
Windowing Systems: Windows, MacOS
Typical Size: PC ($1k to $5k typical; few $100 used)
Clientele: Mixed--Business, Home users, just about everybody
World: (mostly UNIX) workstation
Typical Equipment: Sun Workstation, VAXStations, other vendors
Conventional Operating Systems: UNIX (two major dialects), VMS
Windowing System: X Windows
Typical Size: Workstation ($5k and up)
Clientele: Engineering/Scientific users; more and more businesses
World: Mainframe or Minicomputer
Typical Equipment: IBM (various), Digital Equipment VAXes
Conventional Operating Systems: VM, VMS, UNIX
Windowing Systems: X Windows, if available
Typical Size: Minicomputer or Mainframe ($10k to millions)
Clientele: Big Business, Universities, Government
The neat picture of three worlds is distorted somewhat by a tendency for
each of the worlds to have two (or a few) major options, either in
choice of equipment, operating system, or vendor. So, for example, the
PC world is split into two camps, the MS-DOS people and the Mac people.
Similarly, the UNIX world is split into the "BSD" workstations and
"System V" (i.e. "5") workstations.
The three worlds are reflected somewhat in networking. The Internet is
dominated by minicomputers and workstations running UNIX and VMS, with
an ever increasing influx of PCs running MS-DOS and Macs. Machines
running other operating systems often put a "UNIX-like" foot forward, so
the user can almost believe he or she is dealing with UNIX machines.
There is a certain sense that the UNIX software is the standard for the
Internet. Software tends to appear on UNIX machines first and then be
"ported" to other machines.
What makes it possible for all these disparate machines to talk to each
other is the "Internet Protocol" (known more formally as TCP/IP, for
Transport Control Protocol/Internet Protocol). TCP/IP can be thought of
as a set of rules for two computers to use when they communicate with
each other, even if they are not from the same vendor.
We are used to thinking of computer systems as having "software" and
"hardware", but it is closer to the truth to say that complicated
systems like the Internet have many levels--in the case of the Internet
as many as seven--ranging from "very software" to "very hardware". Each
level has its own set of rules, called its protocol. The TCP/IP
protocol belongs to two of the middle levels. At the moment, the most
common protocol for the two most "very hardware" levels is "Ethernet"
(looks rather like the coaxial cable used for cable TV), while the "very
software" levels are completely dependent on the vendor. In fact, it is
this profusion of levels which lets the Internet work on just about any
kind of hardware and with software from many different vendors.
Anyway, the Internet grew up as several medium-sized networks, all
having diffent "very hardware" and "very software", but using the TCP/IP
protocol for their middle layers, were connected together. Two of the
first, and biggest, nets to adopt the Internet Protocol were ARPAnet--
Internet was first designed for this one--the network for what used to
be called DARPA (the Defense Advanced Research Projects Agency of the
U.S. Department of Defense), and NSFnet, a network connecting
universities and government laboratories for the U.S. National Science
Foundation. These and other large networks form the "backbone" of the
Internet. But today there are hundreds of smaller nets hooked on to the
backbones.
There *are* big networks that don't use TCP/IP. For example, in the
context of IBM mainframes at large universities and research
institutions, BITnet (The "Because It's Time" Network) emerged. This
large worldwide network does not use the Internet protocol. BITnet can
be reached from the Internet through special translators called
gateways, but it is definitely a different network. Occasionally one
encounters problems that can be traced to this fact.
The Future of the Internet
The future of any technology is difficult to forcast, and I do not
profess to know what the future holds for the Internet. Some
predictions that various forcasters have made for internetworking (and
telecommunications in general) are:
o A proposal for a data "superhighway" called the NREN (National
Research and Education Network) will pass the U.S. Congress. This is an
upgrade for the Internet.
o Commercial use of the Internet will become more common and new
schemes for charging for its use will emerge.
o The Internet will be handed by the government over to AT&T and the
other "telecoms", who will charge so much to access it that the whole
scheme will collapse.
o Optical Fiber will replace Coaxial Cable (Ethernet protocol) as the
most common standard for LANs.
o The Internet will enter the home over ordinary phone lines.
o The Internet will enter the home over existing Cable TV coaxial
cable.
o The Internet will enter the home through newly strung optical fiber
as part of a unified system for Telephony, Cable TV, and the data
communication, using [insert your favorite protocol here] as a standard.
o Personal Computers will replace telephones, answering machines,
stereos, CD players, and VCRs--maybe even TV!--as a single, universal
device for home use. Sounds like a good thing to connect to the NREN.
o Computer and telephone technology will become so intertwined that
it is hard to tell the difference. One product, already on the market,
is described as "[a handheld] alphanumeric pager, an XT-compatible
computer with a backlit screen and PCMCIA Type III slot, a fax/modem, a
cellular and land-line phone, and a voice recorder"!
You are welcome to believe all or none of these predictions.
How Do I Connect To the Internet?
Connecting to the Internet involves several steps:
(1) Getting your modem and communications software working together
(2) Connecting to a provider over the phone lines (or a LAN)
(3) Using Internet services
For the first step you will have to rely on the manuals that came with
your modem and software. Appendix B contains a discussion of some of
the obscurer terminology associated with modem settings. You might want
to read it if your manual is not well written.
Actually, you do not have to know about the second step in great detail.
Mostly it is a matter of knowing enough to intellegently choose a
provider. Each provider will have a specific set of steps--modem
settings, access numbers, passwords, etc.--that you need to follow in
order to get from you to the provider. Don't lose hope! Once you get
there you've finished the hardest part. Chapter 6 contains very
explicit instructions for connecting to one particular service, DELPHI.
Step 3 is the subject of the rest of this course--what you can do once
you're on. The basics are discussed in the next chapter.
Connection Methods
In theory, there are three ways to connect to the Internet from a
personal computer or workstation:
(1) Your PC may have a direct connection. This means that it is part
of a Local Area Network (LAN) that is in turn connected to one of the
component Wide Area Networks of the Internet. Your computer will have
its own Internet Protocol (IP) Address. This type of connection is
common in offices, especially of high-tech firms, but definitely not for
home use.
(2) You may have a connection to a "host" computer that is directly
connected to the net. If you can use a modem to connect your home PC to
the mainframe at the office and the mainframe is on the net, then you
can get an Internet connection that way. But what if your office
doesn't have a mainframe on the net? You can still subscribe to a
service that makes a host computer available to the public. This is
presently the cheapest and most common method for public access to the
net.
(3) There is a connection method in between cases (1) and (2) called a
Serial Line Internet Protocol (SLIP) connection. You dial up a special
host computer--just like case (2)--called a SLIP-server. The difference
is that the SLIP server gives you a temporary IP address and talks to
your computer using the Internet Protocol. This requires your
computer's software to speak SLIP. Your computer thinks it is using
case (1) even though your actual connection is closer to step 2. The
advantage of fooling your terminal into thinking it has a direct
connection is that it can use all the fancy, free software developed for
computers with direct connections.
The software for a SLIP connection is being built into all new
communications software. If you just bought or upgraded your software,
you probably have SLIP capability already. The catch? SLIP is too slow
on a cheap 2400 baud modem. But if you buy a fast FAX modem it works
fine. That's one reason that a $350 FAX modem is a good investment. It
is fast enough for SLIP and gives you the ability to send and receive
FAXes as well.
SLIP technology is still rather new and somewhat experimental, so this
discussion will focus on the old reliable--method 2. How do you find a
service that will give the public access to the Internet?
Depending on where you live, you may have a provider you can call in
your local area code. If not, then most of your problem will be finding
the cheapest way to make a long distance phone call. This book does not
have a list of providers, since such a list will get out of date
rapidly. Instead, it gives you one cheap way to get on the Internet,
then gives you instructions on how to find out who the providers are and
what their rates are. I would rather teach you how to look out for
yourself than just give you some outdated advice. This method also has
the advantage that one set of instructions works for everybody. That
wouldn't be true if I listed 20 or 30 providers. Instructions are given
in "Gettin on the Internet Step by Step."
Types of Internet Providers
As I said, most of the problem of getting on the net occurs when you
live in an area that doesn't have a LOCAL provider. Basically there are
three kinds of providers and three ways to get to them:
(1) Providers of direct connections. If you are setting up a business
and need a high volume direct connection for your office Local Area
Network and can afford several thousand a year at least, you will want
to consider these high-end providers. They are not relevant to our
discussion. (But see Appendix C for more information).
(2) There are several regional networks and one national one that
specialize in low cost PC-to-host or SLIP connections. Costs range from
$20-40 a month to $2000 a year, depending on the services you need.
Performance Systems International (PSI) is a major provider of this kind
of service. Other networks offer services similar to PSI, although PSI
has the most extensive nationwide service at this time. This service
can be very competitive with BBS type service (see below) if you are a
frequent user of the net or need to send more than an occasional E-mail
message. It is definitely worth a look.
(3) Many computer bulletin board services offer E-mail or even
Internet connections for around $10-20 per month. Be very careful to
check out the connection charges. If you are not careful you could be
charged for using the bulletin board (per hour), using the Internet, the
long-distance connection, a surcharge for daytime use, and a per message
charge for E-mail! One of the purposes of this book is to explain the
minefield of charges so you don't get burned. The service recommended
bundles all the charges up front so there are no surprises.
Generally speaking, connecting to the Internet through a BBS is the best
method for the explorer. Once you've determined that you need the
Internet on a regular basis, one of the regional networks or PSI is
probably the most economical route.
Unless you are fortunate enough to live in an area where an Internet
provider is a local call away, you will have to contend with long
distance charges. Actually, these can be as low as $2 an hour and are
sometimes bundled in with the network connection charge.
Finding the Cheapest Long Distance Method
There are three basic methods of paying for long distance:
(1) You just pay for a call to another area code. This is very
expensive and not recommended.
(2) You use a provider with a toll free (800) number and pay for the
call in a higher connect charge. This is also very expensive.
(3) You use a Public Data Network after hours and pay around $2 an
hour (may be included).
Actually, the last method is the only workable one. There are a number
of PDN's.
CompuServe has a data network. You do not have to join CompuServe to
use it.
PSI has its own data network with many points of presence around the
US and abroad. These are divided into Class A and Class B, depending on
the level of service provided.
Tymenet and SprintNet are two other public data networks. You may
have heard of the SprintNet service PC Pursuit. For a monthly fee this
gives you many BBS nationwide as well as any computer that can be
reached by SprintNet.
In general, for a first experiment we recommend the DELPHI BBS and
SprintNet. DELPHI includes the SprintNet surcharge in its $13/month
bill (after hours use only--daytime is expensive everywhere). At this
writing you get 5 free hours the first month and 4 hours per month after
that. Additional hours are $4 each. There is also a 20 hours for $20
plan. Additional hours are $2 each with this plan. Detailed
instructions on how to sign up are given in "Connecting to the Internet
Step by Step."
Who Pays for the Internet?
All this talk of cost may be making you edgy. Eventually, everyone
using the Internet must face the fear--if I am calling up a computer in
Switzerland won't I be billed for the call? The marvelous thing about
the Internet is that although there is plenty of expense involved in
getting on it, there is no additional expense associated with what you
do after you are connected. THE WHOLE WORLD IS ON LOCAL. Thus, you may
have to pay for a $2 an hour call to Massachussetts, for your PC, your
modem and software, and a connect fee to your internet provider. But
you DO NOT HAVE TO PAY FOR EACH AND EVERY INTERNET CALL.
Since most people find it hard to believe that you can send mail
anywhere in the world or dial up a computer on the other side of the
globe without paying a special charge, I will spend some time explaining
who does pay for the Internet and how those costs are reflected back to
the user. One way or another you do pay for network usage, but these
payments are not in the form of a direct billing for each call.
First, you already know that there is no Internet, Inc. that monitors
all the calls and bill customers. Instead there are hundreds of smaller
networks that act as relays. Those networks *could* charge their
customers for each call, based on how much time it takes and where it
goes, but since no one is charging them, they have no real incentive to
pass on costs. Instead they charge a flat fee--usually based on connect
time, but for a 24 hour connection just a flat yearly fee.
Now network traffic does use up resources. Basically, the Internet
works like a potluck supper. Everyone with a direct connection allows
some of their system resources to be used by messages that are just
"passing through". They allow this because other systems allow them the
same privilege. Thus, it is in everyone's interest to allow some of
their resources to be consumed by other persons' messages, because
everyone comes out ahead.
So, sites with direct connections pay real costs in terms of lost
computing cycles, extra cabling, fancier equipment, and lost disk space.
These costs are passed on to their customers or shouldered by government
subsidy. But there are no direct charges associated with using the
system.
In the early days of the Internet, Government subsidy of the backbone
networks was crucial. The backbone was built with government funds and
it was government funds that paid for the extra equipment needed by the
universities and laboratories that carried more traffic than they
generated. With the development of commercial nets alternatives to the
Government-sponsored backbone arose. The Government subsidy is still
important, but becomming less so every year.
So, the short answer is that you pay for the network. You either pay
your provider a flat fee or you pay as a taxpayer for Government
subsidized network resources. Most of the cost you actually see will be
in your own equipment, the cost of placing a phone call, and whatever
your provider charges you.
Internet Basics
We've talked a lot about the Internet, but how do you actually use it?
There are three basic skills on which all Internet use is based:
Electronic Mail (E-mail)
File Transfer Protocol (FTP)
Telnet
All three commands rely on the Internet addressing scheme. An Internet
"telephone number" of another computer is its Internet Protocol (IP)
address, a number that looks like this:
225.225.12.38
This form, called "dotted decimal," is still required by some computers.
But, to make IP addresses easier for humans, this telephone number has
another form which is easier to remember:
hoople.usnd.edu
This means that computer ("node") named "hoople" is located at the
University of Southern North Dakota. The last component, ".edu", means
that the institution is in the educational domain. . Other domain
names look like this:
.edu : educational institutions
.gov : government (research laboratories and
.com : commercial businesses
.org : nonprofit organizations
.mil : military installations
In addition to these domain names, there are many two-letter country
codes, e.g.
.ca : Canada
.jp : Japan
.uk : United Kingdom
.us : United States
and many more.
To send electronic mail to someone over the Internet, all you need to
know is their "username", or "handle". This is followed by an at-sign,
the node name, and any domain names. Thus
joe@locoweed.chi.il.us
donotes a private citizen in Chicago, Illinois, in the U.S. Joe's
computer is called "locoweed".
As another (real) example,
president@whitehouse.gov
is the E-mail address of the U.S. President.
In the examples that follow we will give details for a typical
character-oriented computer. Windowing systems with menus, dialog
boxes, and so on will hide many of these details, but they are happening
behind the scenes. Also, once you are connected, you may be faced with
an old-fashioned command-line system.
Electronic Mail (E-mail)
The details of using the mail system depend on your system, but
basically it looks like this:
% mail
Welcome to Mail, Version 99.3 . . .
mail> send
To: president@whitehouse.gov
Cc: vice.president@whitehouse.gov
Subj: I'm on Internet
Type your message. Control-Z to exit
Bill--
I just got my Internet connection today. My address is
pdq@hoople.usnd.edu.
Give my best to Al,
PDQ
%
Telnet
In addition to using an Internet address to send E-mail, you can use it
to call a computer. This is rather like dialing up a computer with a
modem, except that the local computer (the one you called with *your*
modem) is calling up the remote computer:
your PC or Mac --> "local computer" --> "remote computer"
The example assumes that "home>" is the shell prompt given by your local
computer and that "%" is the prompt given by the remote computer (see
Section 2.2 on "shell prompts"). So remember, you don't type them.
home> telnet hoople.usnd.edu
Welcome to node HOOPLE. Now running Opus 2.0
username: pdq
password:
Last login 23:14:55 15-JUN-1752
You have mail.
%
% logoff
Session with hoople.usnd.gov terminated at 21:19.
home>
This method of connecting to another computer is called "telnetting".
In effect, you have used the local computer to telephone the remote
computer. You can now do anything on the remote computer (with certain
restrictions) you could do if you were "actually" logged on.
SPECIAL PROBLEMS WITH TELNETTING
Sometimes telnetting will put you into a menu-type program or even a
"screen oriented" program. A special problem here is getting the other
computer to recognize what type of screen you have. Since most
communications software "emulates a terminal", this amounts to telling
the other system what type of terminal your communications system is
emulating.
E.g., on a UNIX system you might type:
% set term vt100
to tell the other system that your communications software thinks it is
a VT100 terminal (a very common choice for emulation programs).
If you don't get this exactly right, your telnet session will "sort of
work". It's probably not worth spending a lot of time on this problem
for a brief contact with the other computer. If you are going to work
on the remote computer every day, however, you will want to get it
right. Most "flaky" behavior can be traced to this problem.
File Transfer Protocol (FTP)
It is easy to transfer files over the Internet using a special protocol
called FTP. FTP takes the place of programs like XModem or Kermit that
may be familiar to you if you use a bulletin board service. Now, you
might ask, if FTP transfers a file, what is the differnce between
sending E-mail and FTP; why prefer one over the other?
First of all, FTP avoids certain restrictions on the type of file sent;
in particular, you can send binary (non-text) files like programs and
pictures; and you are not restricted to any particular length, as is
often the case with mail programs.
Not only that, with FTP you can "browse" a directory of files before
choosing one. You can move up and down the directory hierarchy and list
files, looking for the one you want.
Finally, FTP lets you get places that you may not have an account (or a
friend to send you mail). The method for doing this, called "anonymous
FTP", is described in the next section. There are several hundred
"anonymous FTP sites" worldwide, with large collections of programs,
textfiles, and graphics. Anonymous FTP is the Internet equivalent of
publishing--a very important topic indeed.
Anonymous FTP
Further facility with Anon. FTP requires a little knowledge of the UNIX
operating system. See Appendix D for an introduction.
Getting on the Internet Step by Step
This course takes a different approach to . Rather than
In particular, you should contact the following for information.
STEP 1. SIGN UP WITH A PROVIDER (example DELPHI)
To get on the Internet you need:
1. a home computer,
2. a 1200 or 2400 baud modem that understands AT commands ("Hayes
Command Set Compatible"),
3. communications software, and
4. a credit card.
This example assumes that you are signing up with the bulletin board
service DELPHI. DELPHI has no commercial relationship with the author.
It was chosen for an example because its rates are typical of the low
end of the market, it offers nationwide service, it includes afterhours
access by SprintNet in its basic fee, and because at this time it offers
5 free hours of Internet access to new subscribers.
You should definitely consider other providers. You can get information
about other providers once you're on the Internet. In fact, that will
be one of your first steps below.
1. Set your modem to 2400 (or 1200) baud, 8 bits, NO parity, 1 stop
bit, full duplex, local echo and auto linefeeds off, XON/XOFF on, VT100
emulation. (see the manual for your modem, the manual for your
communications software, and Appendix A if you need help)
Don't worry if some of the options are missing. Usually, for example,
full duplex implies local echo off and vice versa, so your software may
not list these as separate options.
2. Use your communications software to issue (or type directly to
your modem) the command ATDT 1 800 365 4656. If you have a pulse-dial
phone your will use ATDP instead of ATDT.
3. DELPHI will give you explicit instructions, but for reference they
will look like this:
A. Make sure it is after 6 or 7 p.m. or on a weekend--unless you
want a $9 surcharge for daytime access.
B. Dial your local SprintNet access number (local call!): ATDT 123
4567.
C. After "CONNECT 2400", type @D ("D" must be capital)
D. After "terminal=" type , i.e. carriage return.
E. At the @-sign type
@C DELPHI
F. After Username: type
Username: YOURNAME
Password: OPENSESAME
Type your own username and password of course.
4. If at anytime you forget your Sprintnet access number, or if you
are out of town, you can look up SprintNet numbers as follows.
STEP 2. Your First E-mail
Get PDIAL, NIXPUB, INTERNET RESOURCE GUIDE, Info on PSI
Info on LISTSERV and signing up for E-mail
STEP 3. Your First Anonyomous FTP
Get surfing the internet
STEP 4. Your First Telnet
Getting Public Access UNIX and reading Usenet news
STEP 5. First E-mail
Getting Information on PSI
Getting Information on LISTSERV
You can do many, many things with the Internet. The information you
have retrieved, especially the Internet Guides, will give you other
ideas. We will go over some of the retrieval methods in Part III of
this course. For now, you should experiment and explore, using the
information you have as a starting point. The experience you gain will
be useful when we come to the more systematic study of Internet research
methods later.
Programs and Pictures
Evenually, as you gain experience finding and transfering information,
you will want to try you hand at transferring binary files containing
pictures or free software from one of the large FTP archives like
wustl.edu
sumex-aim.stanford.edu (Macintosh software)
simtel-20.mil (IBM and compatible software)
In fact, some of the first software you will want to get is software for
compressing and uncompressing files (see next Chapter), processing
graphics ("image files"), and perhaps some games.
First, what is a binary file? A binary file is one that you can't read.
Unlike text, which consists of groups of eight bit code letters
representing "a", "b", and so on, binary files have bits that are meant
to be read only by programs. You cannot "look" at them by typing them
out or with a word processor--unless the word processor understands the
format of the file. Expensive word processing programs--including the
most popular ones like Word Perfect and Microsoft Word--are increasingly
able to read a large variety of binary files.
Frequently encountered binary files include:
1. Files that use proprietary formats, such as word-processing
programs, spreadsheets, database programs, etc. These files contain
formatting (like italics, underlines, etc.) and perhaps graphics, and
other goodies beyond the simple text.
2. "Application Programs". These are programs written in (the
binary) machine language that your computer understands. They are
"compiled" from text files of "source code" written in a programming
language. Vendors almost never make their source code available--except
for free software, which you may have to compile yourself.
3. Text files that have been compressed to about half their size with
one of the popular compression programs. Compression makes texts files
binary. Compression doesn't do much for files that are already binary
unless the data they contain is very repetitive.
4. Files containing graphics like GIF, TIFF, PICT, or JPEG files.
More on this below.
Transfering binary files is as easy as transfering text files once you
understand the potential problems:
1. Most FTP programs start you out in TEXT mode. This means that
text files are *translated* when they go from computer to computer on
their way to you. This is fatal to binary files because their bit
pattern has nothing to do with the groups of eight bits that make up