(January 5, 2004)
I have been using the Internet since the late 1980s. During that time, I have seen a great deal of change as the Net [as we used to call it] has grown from a small research network into the largest information system in the history of mankind. As a young computer scientist and writer, the growth of the Internet affected me profoundly.
The following essay is a personal history of the Internet: the story of how it came to exist, what made it flourish, and what it meant to me.
Since this is a personal story, I am also going to talk about my book Harley Hahn's Internet Yellow Pages, the best-selling Internet book of all time. I'll tell you the story of how I came to write the book, and a bit about how the book evolved over the years. (In fact, this essay first appeared, in a slightly different form, as the introduction to the 10th Anniversary  Edition of the Yellow Pages.)
Within this essay, you will find references to specific sections within my books.
On October 4, 1957, the USSR shocked the rest of the world by launching Sputnik, the first artificial satellite. For the first time in history, a man-made object was orbiting the Earth. A month later, on November 3, the Soviets launched Sputnik II, a much larger satellite that not only orbited the Earth, but carried a live dog (named Laika).
All of this happened during the Cold War, a time when the United States was very concerned with planning for a possible war with the USSR. The launch of the Soviet satellites caught the U.S. defense community completely off-guard. As a result, the U.S. government created a number of programs to develop American technology as quickly as possible, in order to regain the lead in the "space race".
One of the new programs was called ARPA, the Advanced Research Projects Agency, which was formed within the Department of Defense. ARPA's mandate was to sponsor research that would allow the U.S. to become the world leader in military-oriented science and technology.
Over the next several decades, the tentacles of ARPA stretched far and wide, insinuating themselves into many areas of the U.S. research community. By the late 1960s, ARPA had become the principal sponsor of advanced computer science research throughout the country. As such, ARPA had bought a number of computers for computer scientists to use in their research. The computers were installed at a handful of locations around the country. However, in order to use one of the machines, a researcher had to physically be at the computer site.
This was a serious limitation. However, in those days, computers were very large and very expensive, and there was no way ARPA could afford to buy and install a machine for every researcher who needed one. In order to save money, ARPA needed to find a way for researchers around the country to share resources. In pursuit of this goal, ARPA decided to fund the development of a network that would connect the computers and enable them to be accessed remotely.
On November 21, 1969, the first two computers were connected together: one in Los Angeles (at UCLA), the other in Menlo Park, California (at the Stanford Research Institute). On December 5, 1969, the Arpanet was officially established by connecting the machines in Los Angeles and Menlo Park to two more computers: one in Santa Barbara (at U.C. Santa Barbara) and one in Utah (at the University of Utah).
As you might expect, the software that was developed for the Arpanet evolved over the years. Eventually, this software became the basis upon which the Internet was founded. For this reason, many people consider that the Internet evolved from the Arpanet and, thus, the date December 5, 1969 should be recognized as the official birthday of the Internet.
Actually, the Internet did not really evolve from the Arpanet one step at a time. The real story, as you will see later, is far more complex and a lot more interesting.
In 1972, when the Arpanet was barely three years old, I was a student at the University of Waterloo, about 75 miles west of Toronto, a well-known math and computer school dedicated to educating Canadian nerds. I was studying a lot of mathematics and computer science, although I didn't have any specific plans. I loved math, and I found computer programming easy, so, like many intelligent people without a firm goal, I just did what came naturally.
At the end my first year, I took a vacation, during which I met a young woman who was to become my first girlfriend. It happened that she lived in Los Angeles and so, smitten with the intoxication of young love, I decided to take off a semester in the fall of 1972 and visit her in California.
My girlfriend was a student, and while she was in classes, time hung heavy on my hands. (I am one of those people who gets bored easily and suffers enormously when it happens.) To alleviate the monotony, I spent time visiting UCLA and, naturally, I gravitated towards the computer labs.
One day, while I was visiting, someone showed me a room with a couple of computers. One of these, he said, was connected to the Arpanet. He explained that it was possible to use this computer to connect to other computers around the country. In particular, this computer could connect to another computer at Stanford. This was the first time I had ever seen a long-distance computer network.
At the time, there was a well-known program called Eliza, named after the Eliza Doolittle character in the play "Pygmalion". Eliza was created in 1966 by Joseph Weizenbaum at MIT. His idea was to use the techniques of artificial intelligence to write a program that would act like a psychologist. Eliza had been installed on one of the UCLA computers, and it was possible to "talk" to it by typing comments on a typewriter-like device. Eliza would process the comments and type a reply.
I had an interest in such programs, and I knew of another one, called Parry, that had recently been created by a researcher at Stanford named Kenneth Colby. Like Eliza, you could talk to Parry by typing comments and reading the replies. However, where Eliza was designed to act like a psychologist, Parry was designed to act like a paranoid schizophrenic.
Well, I was an imaginative guy. Once I learned that it was possible to connect to the computer at Stanford, it was the work of a moment for me to run Eliza on one computer, and use the other one to connect to Stanford via the Arpanet and run Parry. I then allowed Eliza and Parry to talk to one another (by typing the responses from one program into the other).
Thus, it happened that my very first experience with a computer network, in 1972, was to use it to watch a psychologist program, running on a computer at UCLA, talk to a paranoid program, running on a computer at Stanford.
So what happened?
Eliza was a relatively simple program that used a technique called nondirective psychotherapy, a type of therapy that had been developed by the prominent psychologist Carl Rogers. In Rogerian treatment (which was quite popular in the 1960s), the therapist would repeat the patient's own remarks back to him in the form of a question. ("So what does your mother's need to clean the house so much mean to you?") The idea was to help the patient find answers within himself by having the therapist disappear, leaving the patient alone with his conscience.
Parry, however, simulated a paranoid who didn't have a conscience. Moreover, Parry was (for its time) a very complex program, much more sophisticated than Eliza. In fact, it was rumored that Parry was even able to fool real psychologists.
The results are as you might imagine: Parry completely dominated and confused Eliza (advancing the cause of paranoids everywhere).
By today's standards, using the Internet to run two programs in tandem might seem a bit lame. In 1972, however, this experiment seemed like nothing less than a modern miracle.
Still, in those days, computer science students didn't pay much attention to the Internet. We spent our time studying programming techniques, data structures, numerical analysis, and lots of mathematics.
By the late 1970s, I had become a computer science graduate student at U.C. San Diego. Even then, computer networks were not that important to me. To be sure, the Arpanet had grown to encompass virtually all of the major universities in the U.S., as well as many others around the world. However, the network was mostly used by professors and researchers, principally for email and for technical discussions. During my entire time as a computer science grad student (1976-1979), I don't think I even used the Arpanet at all.
After grad school, I decided to leave computer science to study medicine, which I did at the University of Toronto from 1980 to 1984. Towards the end of medical school, I decided that I did not want to be a practicing doctor, so I left the field and became a full-time professional writer. Since I had a graduate degree in computer science, I specialized in writing books about computers.
Even then, it wasn't until 1991 that I began to actually use the Internet with any regularity. Here is how it happened.
In 1991, the Internet had been around for 22 years and had become popular within universities, although not yet with the general public. The reason was simple: there was no easy way for a regular person to access the Net. If you worked or studied at a university, it was easy to use the Internet. All you had to do was get an account (that is, a user name and password) on a computer that was connected to the Net.
At the time, however, there were no commercial ISPs (Internet Service Providers). This meant that, if you didn't have a way to get a computer account at a university, you were effectively shut out from using the Internet.
In the spring of 1991, I was living in Santa Barbara. An editor from McGraw-Hill visited me to see if he could talk me into writing a textbook. The topic of the book was to be Unix, a complex operating system (master control program) that ran most of the computers connected to the Internet. Unix textbooks were important because, in those days, if you wanted to use the Internet, you first had to learn how to use Unix. Although Unix was easy to use, it was difficult to learn (as is the case with most tools for smart people).
In December 1991, I signed a contract to write the book and worked on it throughout 1992. In February 1993, the book was published under the title A Student's Guide to Unix. (Later, when I wrote the second edition of the book, the title was changed to Harley Hahn's Student's Guide to Unix.)
While I was researching the book, I realized how important the Internet was becoming. I came to the conclusion that explaining how to use the Net (which was not easy in those days) was something I needed to do for my readers. So, beginning in 1992, I started to use the Net in earnest.
I have never stopped.
When I first started to use the Internet, it was much different than it is now. In those days, there were four main services:
For two reasons, I'd like to take some time to talk about each of these services.
First, telnet, FTP, mail and Usenet formed the foundation upon which the Internet was created. In fact, these four services are so basic to the Internet, that they are still in wide use today.
Second, I want to introduce you to an idea that is fundamental to the world of computing: the importance of unintended uses.
When a well-designed computer tool is released to the world at large, smart people will modify and use that tool in ways that its creators never imagined. In many cases, this results in a burst of creativity that brings enduring benefits to a wide audience.
This phenomenon happened with all four of the services I mentioned above. In fact, it was the unintended uses to which these four services were put that allowed the Internet to grow so large in a relatively short amount of time.
I'll show you what I mean. Let's start with telnet.
Telnet is a service that allows you to log onto any Internet computer on which you have an account. Once you log in (by typing a user name and password) you can use that computer remotely.
When we talk about telnet, you will often see the word used as verb. For example, you might hear someone say, "In order to get the groatcake recipe you wanted, I had to telnet to a computer at the University of Syldavia."
(By the way, no one knows what the name "telnet" means. In all my research, including reading the original papers, I was never able to figure out how the name was chosen.)
By 1992, when I started work on my Unix textbook, telnet was over two decades old. It was, in fact, the very first service available on the Internet. As I explained earlier, the Arpanet was created so that researchers could share computing resources. They did so by using telnet. In fact, it was telnet that I used, back in 1972 at UCLA, to connect to the computer at Stanford.
When a computer tool such as telnet is first developed, it's hard to predict how important it really will be. Of course, every new computer tool is created to fill an immediate need, but the real test of a new tool is to wait and see what happens over the next few years to watch what people do with the tool.
This is a characteristic that is found throughout the world of computers. Because computer programs act as extensions of our minds, the very best tools end up being used for purposes that go far beyond what their inventors had anticipated. Such was the case with telnet.
Let's say that a programmer at a university wrote a program that provided a particularly useful service. Once the programmer showed the program to his colleagues, word would start to spread and people who had an account on the same computer would start to use the program.
However, what about other people, especially people at other universities? They would hear about the new service and want to use it, but how could that be arranged?
One solution would be to give everyone who wanted to use the new service an account on the computer that ran the program. That way, anyone could simply telnet to that computer, log in, and run the program.
To some extent, this solution would work, but not for very long. If the service proved to be popular, there would be requests from people all over the country asking for computer accounts. The administration alone would be horrendous, not to mention the security risk of allowing hundreds, or even thousands of people full access to the computer.
To solve this problem, it became necessary to figure out how to use telnet to provide services of general interest. A system had to be developed that was easy to administer without creating a security risk.
Before I can explain the solution, I will have to introduce you to a bit of technical jargon, so bear with me.
On the Internet, computers connect to one another in order to request services. Whenever one computer connects to another, the first computer must specify a "port number" to the second computer. The port number serves as an ID number that tells the second computer what type of service is being requested.
In the early days of the Net, it was decided that the basic telnet service would use port 23 to identify itself. Thus, whenever anyone used telnet to connect to a remote computer, the connection would be made using port number 23. Once this connection was made, the remote computer would require the user to log in by entering a valid user name and password.
However, there was nothing inherent in telnet that required a user name and password. All telnet did was connect you to a remote computer. What happened at the other end depended on how the remote computer was configured.
Thus, the solution to the problem of allowing general access to a service was to configure a computer so that it would accept telnet requests for a nonstandard port number, say, 4000. Whenever a person connected to the computer using port 4000, he would not be required to specify a user name and password. However, his privileges would be restricted so that all he could do would be to run one program, the one that provided the service.
Once the idea of nonstandard port numbers was invented, a large number of people began to create services and offer them, via telnet, to anyone with Internet access. The most interesting and enduring of these services were the imaginary role-playing communities that came to be known as "muds".
As I mentioned earlier, the only way to access the Internet in 1992 was by finagling a computer account at a local university (which I managed to do). Since I was working at home, I connected to the university computer by using a phone line and a modem. By today's standards, the connection was horribly slow, a mere 2,400 bits per second.
However, a slow connection wasn't that big a deal. All the programs we used ten years ago were text-based, which means they used only simple characters, such as letters, numbers and punctuation. Most of the time, a slow connection was just fine, because text-based information can be transmitted a lot more quickly than the pictures, icons, windows, and other graphical elements that we use today.
You might think it would be boring to use programs that could display only characters. Actually, this was not the case at all. Because communication on the Net was restricted to text, people spent much more of their time thinking about and talking about ideas, compared to creating and looking at pictures.
It is human nature to want to make images and tell stories. What I learned in the early days of the Net was that we create much more imaginatively with words than we do with pictures. At first, this idea may seem counterintuitive, especially when it comes to images. After all, wouldn't a visual idea be expressed better with pictures than with words?
Actually not. Just think of all the books you have enjoyed over the years, and how well the writers of those books were able to "create" pictures in your head. When you read a well-written description of something, your mind fills in the blanks in a way that is meaningful to you. When you look at a picture, it has less of an impact because, literally, it leaves nothing to the imagination. The reason I explain this in detail is that, during the time period I am describing, there arose a new type of text-based creative endeavor, called muds.
A mud is a complex, imaginary, computer-mediated role-playing environment, one that you can join (for free) and then visit again and again. To use a mud, all you need to do is telnet to a particular computer, using a specific port number. For example, I sponsor a mud called Zynna (which I will talk about in a moment). To use Zynna, you telnet to zynna.com using port number 4000.
The first mud was developed in 1978. It was modeled after the popular role-playing game Dungeons and Dragons. Indeed, the word "mud" was originally an acronym, MUD, meaning "Multi-User Dungeon".
Every mud has a theme, which gives it an overall atmosphere, usually fantasy or science fiction. Muds are highly participatory: you choose an imaginary identity for yourself often a non-human and then, whenever you visit the mud, you take on that particular persona. The mud software keeps track of all the players, and remembers their characteristics and identities.
All muds are oriented toward meeting people, talking, making friends and developing relationships. Some muds also allow you to have adventures, go on quests, and solve puzzles. These so-called "adventure muds" are very complex places, with an elaborate system of locations and inhabitants.
In many ways, muds are the epitome of what you can do with a large, worldwide computer network. When you connect to a mud, you enter a world that exists only in the minds of other people. There are many different kinds of muds, but what they all have in common is a degree of sophistication, creativity and imagination that exists nowhere else on the Net.
Muds have been around for well over twenty years, and they are still thriving. In spite of the fact that there is no shortage of highly visual video games and multi-person network games, there are literally thousands of text-based muds on the Internet today.
I first wrote about muds in 1992, when I was working on the first edition of my Unix textbook. I never did much mudding (as it is called), but six years later, I came to sponsor a mud of my own. Here is how it happened.
In 1998, I was living with my girlfriend of the time, and she was heavily into mudding. She and her friends would lament that, although they put a lot of work into their favorite mud, they didn't have control over it. They wished they could have a mud of their own, which they could customize and run the way they wanted. So, to please my girlfriend, I became the sponsor of a mud.
Although, officially, I was the sponsor, all I did was furnish the equipment and provide the Internet connection. I didn't really do anything to set up the mud; my girlfriend and her friends did everything themselves.
Eventually, after a few years, my girlfriend lost interest in the mud (and me), and moved on. However, even without the founders of the mud, I have been able to keep it alive by choosing smart, responsible people to run it.
When I need a job done, my philosophy has always been to choose the very best person I can find, then leave him alone to do the job as he sees fit. Over the years, four different people two women and two men have managed the mud, and they have all been excellent. Although I make it a point to never second-guess my managers, they know that I am always there should they need help with a serious problem. In all the time the mud has been in existence, this has only happened twice.
In the previous section, I observed that a well-designed computer tool will end up being used in ways that are much different than what the original programmers envisioned. This was certainly the case with telnet.
As you remember, telnet was the very first Internet service. It was created so that computer scientists would be able to access remote computers in order to do their research. What fascinates me is that this same service is now used to support elaborate, imaginary environments that are open to anyone on the Net who wants to participate.
Once telnet was established on the Net, the next important Internet service to be developed was FTP or File Transfer Protocol. (In computer terminology, a protocol is a set of technical specifications.) The original role of FTP was to allow researchers who used more than one computer to copy files from one computer to another.
As with telnet, the word "FTP" is often used as a verb. For example, "I'll be down for dinner in a moment. I just have to FTP a couple of substitute data files to the CIA computer."
FTP is like telnet in that it was designed to be used by people who had permission to use a remote computer. To start, you would use the FTP program to connect to a remote computer. Then (as with telnet) you would log in by entering a user name and password. Once you were logged in, you could enter commands to either upload (send) or download (receive) files. (If you have trouble remembering which is which, just imagine the remote computer floating above you in the sky. You send files up and receive files that are sent down.)
As with telnet, once FTP was established, it was not long before inventive Internet users started to put it to other uses, two of which are particularly important.
First, at the same time that the FTP specifications were being finalized, a number of programmers had started to think about how the Arpanet community might implement a system for electronic mail. At the request of one of these programmers, a rudimentary email facility was added to FTP.
The FTP-based email system wasn't much, and it was difficult to use. All it could do was pass one text message at a time from one computer to another. Still, it was email and, within a short time, the rudimentary FTP-based system jump-started the development of a full-fledged, Internet-wide email standard.
Even more important than FTP-based email was the idea of "anonymous FTP": a system that enabled people to set up "archives" of files that could be downloaded by anyone on the Internet.
In my opinion, anonymous FTP was the single most important service contributing to the growth of the Internet. In fact, in the first edition of my Unix textbook, I described anonymous FTP as "one of the most significant inventions in the history of mankind". Let me tell you briefly how it works.
To set up an anonymous FTP archive on a computer, the system administrator would create a directory (folder) containing all the files that were to be shared. He would then configure the computer so that anyone on the Net would be able to log in using the user name "anonymous", and without needing a password. Once someone had logged in as "anonymous", he or she would be able to download any of the files in the public directory, but nothing else.
For the first time in history, programmers could create programs and make them available to people all over the world. The reason this was so important is that anonymous FTP was the system that was used to distribute the very programs that built the Internet into a huge worldwide network. Whenever a programmer had a new program to share (or a new improved version of an old program), all he had to do was put it in an anonymous FTP archive. Then, anyone, anywhere in the world, could download the program and install it on his or her own computer.
In this way, anonymous FTP ensured that the software needed to grow the Internet was made freely accessible to anyone in the world.
The next Internet service, electronic mail, is something we all take for granted today. In the early days of the Net, however, relatively few people had even heard of email. Even as late as 1992, when I was working on my Unix textbook, the idea of an electronic mail system was new to many people in the university community. Still, email was quickly becoming crucial within many organizations, and I imagine that many of my readers were my readers because they were forced to learn Unix in order to use email.
In the early 1970s, email was new to almost everyone. However, it caught on quickly and, by 1973, email messages comprised three-quarters of all the data sent out over the Arpanet.
The original motivation for email was to allow researchers to work together by sending messages to one another. However, as with telnet and FTP, email hadn't been around long before people found a new way to use it: to form discussion groups.
In 1975, the first email-mediated discussion group was founded. Its name was MsgGroup ("Message Services Group"), and it was devoted to discussing technical issues related to the Internet.
At first, MsgGroup was administered manually by a person who took on the job of "moderator". The moderator's job was to maintain a master list containing the email addresses of everyone who wanted to participate in the discussion. Whenever anyone sent in a message, the moderator would forward the message to everyone in the group. As you can imagine, this was a lot of work. Eventually, software was written to automate the day-to-day details, and MsgGroup became the very first Internet mailing list.
Since the people using the Net in those days were mostly hard-core nerds, it makes sense that the first real Internet mailing list would be devoted to nerd-like topics. Still, it was not until September 1979 that the first non-technical mailing list was finally started. This was because the Arpanet was funded with government research money, and there was strong pressure not to use the Net for personal activities.
When the first non-technical mailing list was started, though, it quickly became a great success, attracting people from all over the world. The list was called "SF-Lovers", and was devoted to science fiction.
(How popular was SF-Lovers? Well, the very same mailing list is still alive today, over two decades later. If you want to check it out, take a look at the SF-Lovers Web site: www.sflovers.org.)
Today, there are tens of thousands of mailing lists on the Internet. In fact, no one knows how many there are, as anyone who has access to mailing list software can start his own list. A great many mailing lists are open to the general public, although there are some that are kept completely private. If you look through the Yellow Pages book, you will see a lot of mailing lists that my researchers and I have found. All of these lists are free, and open to anyone who wants to join.
About the same time that SF-Lovers began, back in 1979, another mailing list was started that also proved to be extremely popular and long-lived: Human-Nets. The purpose of Human-Nets was to discuss the effects of email and computer networks on human culture. As you might imagine, in an environment composed of highly educated and technically adept people, a discussion of how computer networks might affect our culture would inspire a great deal of spirited debate. Indeed, some old-timers who look back with nostalgia remember the Human-Nets mailing list as being home to the most inventive and vital discussions that have ever taken place on the Net.
What I find fascinating is that, long before most people had ever heard of electronic mail and computer networks, the people on Human-Nets were discussing the "network of the future". For a long time, the participants on Human-Nets discussed a brand new idea: Would it be possible to create a global network (which they referred to as "WorldNet")? If so, what would such a network mean to mankind?
Eventually, of course, the possibility was realized. The Internet came into being and did, indeed, become a global network.
However, the genesis of the Internet was not in the way you might have imagined. The Internet did not evolve from the Arpanet. What actually happened was much more interesting and, in retrospect, much more unexpected.
Now that we have discussed telnet, FTP and mail, I'd like to spend a few minutes talking about the fourth fundamental Internet service: Usenet.
Usenet is an independent, worldwide system of tens of thousands of discussion groups. In the early days of the Internet, Usenet was, by far, the most important discussion group system in the history of the world. For example, if you had looked through the various editions of my Yellow Pages book, you would have seen references to many different Usenet groups, covering every topic imaginable.
If you are interested in learning about Usenet (which still exists) the very best introduction is:)
By the early 1980s, Usenet was a large, complex and very important system. However, its roots were simple, and much less ambitious.
The story starts at Bell Labs, a renowned AT&T research facility in New Jersey. In 1969, a computer scientist at Bell Labs developed a small, streamlined operating system, which he called Unics.
(The name was actually a pun. The idea was to poke fun at a large, unwieldy system named Multics, which had been developed at MIT and which was full of problems.)
Within a short time, the name Unics was changed to "Unix" and, throughout the 1970s, while the rest of the country was listening to music, taking drugs, and protesting the War in Vietnam, the computer scientists at Bell Labs were raising their own consciousness by developing Unix into what was to become the best and most successful operating system in the world.
Because AT&T was a commercial entity, they put strong restrictions on how Unix might be used outside of the company. For this reason, a project was started within the Computer Science department at U.C. Berkeley to create a homegrown, unrestricted version of Unix called BSD (Berkeley System Distribution). The first version, called 1BSD, was released in March 1978.
Thus, it came to pass that, in the late 1970s, there were two really cool groups of Unix people in the country: those at Bell Labs (in New Jersey), and those at U.C. Berkeley (in California). Both groups were known for their camaraderie, academic freedom and highly innovative programming. (Although I never made it to Bell Labs, I did have a chance to visit Berkeley. I remember that people liked to bring their dogs to work and to spend a lot of time playing volleyball.)
In the fall of 1979, a graduate student from Duke University in North Carolina had just returned to school after spending the summer at Bell Labs. As you can imagine, after spending the summer in such a cool place, he felt a big letdown being back in school. What was especially hard was the feeling of being disconnected from the Unix community.
The grad student discussed his angst with a friend, another Duke grad student, and together they started to plan a system to pass news messages from one computer to another. If they could find a way to connect their computer to a computer at Bell Labs, such a system would allow them to rejoin the mainstream Unix culture.
They quickly recruited two more grad students, one from Duke, the other from the University of North Carolina (UNC) at Chapel Hill. Within a short time, the group had created a rudimentary news system, which they called Usenet (a contraction of "Users' Network"). In January 1980, they presented a paper at a Unix conference, introducing Usenet to the rest of the world.
Within a short time, a number of different programmers helped develop Usenet into a full-fledged system of discussion groups. For historical reasons, the groups were called "newsgroups" (which is still the case today), and Usenet itself was often referred to as Netnews. Similarly, the program that people used to access Usenet was called a "newsreader", and any computer that acted as a Usenet hub was called a "news server".
In the fall of 1979, the first Usenet connection was established between two Unix computers, one at UNC and one at Duke. By the end of the year, a third news server, also at Duke, was added to the system. By the summer of 1980, Usenet had expanded to five more news servers, two of which were at Bell Labs and one of which was at Berkeley. From that point, Usenet grew at an ever-increasing pace until it, literally, became an enormous global system. Today, there are tens of thousands of newsgroups and countless news servers around the world.
Today, all of the Usenet servers in the world are on the Internet. However, that was not always the case. Usenet was actually started on a completely different network, the UUCP network, and for years, UUCP was the glue that kept Usenet together. The UUCP networking facility was built into all Unix systems, which made it readily available. The actual name, UUCP, was taken from a tool called "uucp", which was the "Unix-to-Unix copy program".
By the late 1970s, Unix people around the country had started to use UUCP to connect their computers into a loose network. At the time, the Arpanet was strictly controlled, because it was supported by government money and was supposed to be used only for research.
The UUCP network was much more informal. For one thing, it had virtually no money. On the other hand, the Arpanet being funded by the Department of Defense had lots of money. This made for significant differences between the two networks.
For example, on the Arpanet, many of the connections were leased communication lines. Leased lines were expensive, but they were also fast and reliable. The UUCP network was created by programmers who used modems and phone lines to stitch together a low-budget network that spanned the entire country.
To do this, each local network had at least one UUCP computer that was programmed so that, at specified intervals, it would call one of its neighbors, establish a temporary connection and exchange data files. In this way, an email message or Usenet article could be passed from one computer to another until it reached its destination, which might be anywhere in the country.
I can still remember what it was like to use UUCP to send mail in the early days. The addresses could get tricky, because you had to specify the exact path you wanted your message to take, from one computer to another, as it traveled to its destination. For example, let's say you wanted to send a message to a friend whose user name was "nipper" on a computer named "delta". You might have to use the address:
This meant that the message should be sent to the computer named "alpha", which would send it to the computer named "beta", which would send it to "gamma", which would send it to "delta", where it would be delivered to the user named "nipper". Such addresses were called "bang-path" addresses. (In old-time typesetting, the exclamation mark character was referred to as "bang". This usage was picked up by the early Unix people.)
Eventually, UUCP addresses were simplified when programs were written to automate the routing. Then, you could simply send a message to:
A routing program would automatically figure out the best path.
Unfortunately, the automated system didn't always work and, even in the early 1990s, it was sometimes necessary to know how to construct a bang-path. Indeed, I explained how to do so in the first edition of my Unix textbook (1992) and in another book I wrote at the time, "The Internet Complete Reference" (1993).
As you might imagine, using a series of intermittent dial-up connections to send messages made for a slow system. For example, it would often take hours, or even a day or two, for a message sent over the UUCP network to travel from one side of the country to the other.
With the Arpanet, on the other hand, the connections were fast and permanent, and data moved quickly from one computer to another. Moreover, the Arpanet used the same modern address scheme that we use today, for example:
Thus, sending email over the Arpanet was a lot more convenient than using UUCP.
Still, the UUCP network worked. Moreover, it was cheap and there were no government regulations to dampen its operation. For these reasons, the UUCP network actually grew to be much larger than the Arpanet.
Once UUCP was established, a grad student at U.C. Berkeley set up one of their computers to act as a "gateway" that is, a connection point between the UUCP network and the Arpanet.
This was a big deal, because the Arpanet was tightly controlled by the Department of Defense and only approved sites were allowed to connect. Anyone with a computer, who knew what he was doing, could join the UUCP network. The coming of the gateway meant that, for the first time, UUCP users could send email to Arpanet users, and vice versa.
At the same time, the UUCP-Arpanet gateway was used to connect various Arpanet mailing lists with Usenet newsgroups. This significantly increased the number of people participating in the discussions, and had a lot to do with the growth of Usenet.
Within the next few years, other UUCP-Arpanet gateways were set up, to provide more points of connectivity between the two networks. In the eyes of the visionaries, all of this took the world one step closer to the holy grail of computer networking: the establishment of a single, large interconnected network spanning the globe.
I mentioned earlier that, in the late 1970s, I went to computer science graduate school at U.C. San Diego. Then, in the early 1980s, I studied medicine at the University of Toronto. I was very unhappy in medical school and, eventually, I got to the point where I felt just like Popeye when he is pushed beyond his limits ("I've had all I can stands, and I can't stands no more!") and I left.
As you can imagine, it was a big deal for me to leave medical school. Moreover, I was now faced with an important problem: What was I to do with my life?
At first, I just needed to calm down, so I used a bit of my savings to support myself and spent a few weeks in the university library, researching and making notes for a reference book about the English writer P. G. Wodehouse. (This is how people like me calm down.)
After about a month, I took a job working for a small computer company, saved my money, and moved back to California, where I settled in North San Diego County. However, I still had the same problem: What was I to do?
At the time, I was living with a girlfriend, and we had a limited amount of savings, so I had to find some way to make money. One day I was talking on the phone to someone who told me about a literary agent who had gotten her a job writing a computer book. This sounded like a good idea to me: I liked writing, I had a graduate degree in computer science, and I needed to earn money. In the fall of 1984, I met with the agent, and he arranged for me to write a series of computer books for a major publisher.
Eventually, I wrote three such books: one about PCs and DOS (the old operating system that was used before Windows); one about Xenix (a type of Unix for PCs); and one teaching Assembly Language, a very technical type of programming. I then went on to write other books, including the Unix textbook I discussed above.
Fast forward to 1991. The original three books had reached the end of their useful lifespan, and the publisher sent me a letter telling me that the books were officially out of print. This meant that all the rights to the books reverted back to me.
Not long afterward, as I was preparing to go to a computer trade show, I happened to talk to someone in the agent's office. He told me, "When you go to the show, be sure to drop in at the Osborne McGraw-Hill booth and say hello to the executive editor." I did so, and the editor and I got along well. As a result, we decided to reissue my Assembly Language book, which ended up doing quite well.
In the spring of 1992, the same editor called me and asked, "How would you like to write a book about the Internet?" I didn't realize it at the time, but this conversation was to be a major turning point in my life.
In 1992, the Internet was just starting to become popular. To explain why, I am going to have to get a bit technical for a moment, so bear with me.
On any computer network, there must be a set of rules called protocols that describe how data is be transmitted from one computer to another. In 1970, the Arpanet programmers developed a rudimentary protocol called NCP (Network Control Program). NCP lasted a long time, but by 1981, it was obvious that a newer, more robust system was needed. This new system was a family of protocols referred to as TCP/IP. (The name came from two members of the family, "Transmission Control Protocol" and "Internet Protocol".) By the middle of 1982, the Arpanet was running on TCP/IP.
Although the Arpanet was an important network, it was not the only one. There were many others, and quite a few of them had their own protocols. Here is a short list of the most important networks along with the years in which they were started:
The founder of FidoNet chose that name when a visitor referred to his computer as a "mongrel" (because it had been built with parts scavenged from a variety of different machines).
The networks I mentioned above were mostly for research and academic pursuits (except FidoNet, which was maintained by individual computer enthusiasts.) During the same time, a number of commercial networks were also started such as CompuServe (1979), The Source (1979), MCI Mail (1980), Delphi (1982), Genie (1985), and America Online (1990).
Eventually, gateways were established between the Arpanet and all of these networks, in order to exchange email and Usenet articles.
In 1982, when TCP/IP was adopted as the official Arpanet standard, computer scientists first discussed the idea of an "internet": a group of networks connected together by gateways. At the same time, a similar name was coined to refer to all the networks that were connected to one another via TCP/IP. This brand new network of networks was called the Internet (with a capital "I").
Over the next decade, more and more networks adopted TCP/IP and the Internet became larger and larger. As the Arpanet began to merge with so many other networks, the vision of a global network finally became a reality.
In 1990, the Internet was so large and important that the Department of Defense officially ended the Arpanet. By 1991, the Internet consisted of over 5,000 different networks in more than 35 countries. There were now over 700,000 computers, used by more than 4 million people.
So now you understand where the Internet came from. It was not created at a specific time, nor did it evolve from the Arpanet. The Internet developed over the better part of a decade (1981-1990), as more and more networks joined the global TCP/IP internet.
Today, the Internet has hundreds of millions of computers used by hundreds of millions of people. But, back in 1992, when I started to write my very first Internet book, "The Internet Complete Reference", the Internet was still new to most people in the country.
At the time I was working on "The Internet Complete Reference", there was no easy way to find Internet resources. The Web which was then called the "World Wide Web" was still brand new. There were very few Web sites and no modern search engines.
For the most part, Internet resources were accessed using the four basic services we discussed earlier:
In addition, there were also several other services:
Since then, Archie, Gopher and Wais, have vanished, having been rendered obsolete by the Web. IRC, however, is still used and has evolved into a global text-based chat system for instant messaging.
Because Internet resources were so hard to find, at the time, I decided that, at the end of The Internet Complete Reference, there should be a large catalog of resources. I got the idea from another book, The Whole Internet User's Guide & Catalog written by Ed Krol, which had been published earlier that year, in September 1992. As you can tell from the name, Krol's book had a catalog, and I decided that my book should have one too.
At the time, the Internet was just starting to become popular with the general public. However, learning how to use the Internet was difficult. Indeed, most people still used Unix to access the Net and, as I explained earlier, Unix is easy to use but hard to learn.
For this reason, Internet books were in great demand. Ed Krol's book sold in enormous numbers and, in my opinion, was the book that was responsible for opening up the Internet to the general public. My book, which came out in October of 1993, was also a best seller. In fact, it did so well, that the editor at Osborne McGraw-Hill asked me to create a brand new book, one that would contain nothing but Internet resources.
I agreed to do so, and the editor and I started to think about how to position the book. At first, we had thought of calling it something like The Internet Directory of Resources. However, the Osborne McGraw-Hill Publisher (the person in charge), whose name was Larry Levitsky, had an inspired idea. He suggested we print the book on yellow paper and model it after a phone book. We did and, in February 1994, The Internet Yellow Pages was published. It was my 11th book.
The original Internet Yellow Pages was much smaller and much less ambitious than the book you are holding in your hands. Still, from the beginning, there was something special about the book. There were two reasons for this.
First, I have always been a highly creative person, and if I was going to create a "Yellow Pages", I wanted to do more than compile a mere directory. Since no one had ever written such a book before, there were few specific expectations. This meant that I was able to do whatever I wanted, which allowed me to give free reign to my imagination. My editor respected my creativity, and allowed me to do what I wanted. (For the most part, this has been the case with all of my editors. They have always allowed me to do things my way, for which I am grateful.)
Second, as the McGraw-Hill artists worked on the raw material, it became obvious that if the pages were to be laid out evenly without breaking items into two columns, there would have to be a way to fill up the spaces.
To solve this problem, I created new visual elements that had never been seen before (at least in a computer book). First, I wrote a large number of fake advertisements, just for fun. The "ads" were supposed to be for specific Internet resources, but I used the opportunity to be as witty and irreverent as I could.
In addition to the fake ads, I created a number of "fillers", short sentences that were printed in small boxes whenever we needed to fill a small space. Again, I did my best to be witty, in order to give life to what might be an otherwise dull book. For example, "This is the first book of the rest of your life." (Never fear, I got better over the years.)
Writing the ads and the fillers was a lot of fun, and they proved to be popular, especially with the artists who were now able to make the pages bottom out, as it is called. In fact, I like the ads and fillers so much that I used them in all 10 editions of the book, including this one.
The first edition of The Internet Yellow Pages also had two other elements that have since been discontinued. First, I searched through many Usenet newsgroups, looking for interesting articles. I then edited them and placed them throughout the book under the heading "Look What I Found on the Net". We called these articles "excerpts", and I kept them in the book adding new ones each year through the 8th (2000) edition.
The second element that has since been discontinued consisted of pictures photos and graphics that were downloaded from the Internet (mostly with anonymous FTP). We discontinued them because we didn't want to have to worry about copyright problems.
When I wrote the first edition of this book, I had no idea how it would be received. In fact, I had no idea if there would even be a second edition. However, the book became more popular than anyone had anticipated and, every year, I was called upon to put out a new edition.
In the spring of 1996, the book (now called Harley Hahn's Internet Yellow Pages) became the first Internet book to have sold over 1,000,000 copies. To me, this was a landmark for two reasons. First, I had never had a million seller before, and I was proud of how well the Yellow Pages had done. In fact, when you add in the sales of my other books, the Yellow Pages makes me the best-selling Internet author in history.
Second, even more important, the sales of this book showed just how important the Internet was becoming to our culture.
Over the years, I have written many good books. However, when I look back, I see no correlation at all between how good a book was and how well it sold.
Like all authors, I'd like to believe that each best seller is an irrefutable testament to my skill as an artist. But when I force myself to be dispassionate, I am forced to admit that luck had far more to do with it than I would like to admit. Although I see myself as a good writer, it is all too clear that a lot of my success came from being the right guy in the right place at the right time.
The reason I mention this is because, as I look back, I see that the evolution of the Yellow Pages followed the evolution of the Internet itself.
In the mid-1990s, the Internet burst upon the popular culture like a supernova. This was not totally unexpected. As I mentioned, as far back as the 1980s, there were ongoing discussions regarding the creation of a global computer network. These discussions, however, were carried out on the Human-Nets mailing list and on Usenet, which meant that the participants were mostly academics who already had access to the Net.
In 1994, when the first edition of the Yellow Pages was published, few of the hundreds of millions of people who use the Net today had even heard of the Internet. They had no idea that, within a few years, they would own their own computer, and that their children would grow up in a world in which the Internet would play a dominant cultural role.
As the world changed, so did the Yellow Pages.
When I wrote the first edition, most of the work went into compiling resources and writing a short description of each one. At the time, it was so hard to locate Internet resources, that I put in everything my researchers and I could find. In fact, in the introduction to the first edition, I gave my readers an email address and asked them to let me know if they found any resources that they thought should be in the next edition. (I stopped doing this a long time ago.)
As the years passed, I changed the focus of the book from a mere compilation of Internet resources to a personal guide to culture, science, technology and the arts. I made this change for two reasons.
First, the Internet itself was changing. With the coming of the Web and the explosive growth of the Net, there were far too many resources to think about using even a small fraction of them. It was imperative that I refocus my efforts and those of my researchers. At one time, our job was to find anything we could. Now, our goal was to search the Net to find the very best resources for every topic in the book.
Second, perhaps more important, I began to grow as a writer. Spending so much time describing Web sites, mailing lists and Usenet groups just wasn't satisfying any longer. To be sure, creating the other parts of the book (such as the fake ads and the fillers) was fun. At heart, though, I am an explainer. If I am to be happy, I need to spend a lot of time researching, learning and explaining.
So I recast the book, changing it from a simple directory into a guide to life, annotated with Internet resources. Enhancing the book in this way afforded me the freedom to learn and write about anything I wanted which I consider to be the best job in the world.
For the ninth (2001) edition, I came up with a new idea. I wrote a large number of questions and answers that I called "Ask Harley". I scattered these short essays throughout the book. They didn't really have anything to do with the Internet. I just wanted to write them, and I thought my readers would enjoy reading them.
For tenth (2003) edition, I created a whole new set of questions and answers, in a slightly different format, which I called "Tidbits". Again, these were short essays I wrote, just for fun, because I thought my readers would enjoy them. Like the "Ask Harley" essays, they have nothing to do with the Internet.
I have since put both sets of questions and answers Ask Harley and Tidbits online. When you have a moment, take a look at them. My guess is you will find a lot to interest you.
As I look back and think about what I have seen, I can now realize why the Internet is so important to humanity. By connecting distant computers in so many homes and offices, a global network enables people everywhere to share the fruits of their labor, including their ideas. The Internet also allows people to be a part of something that is larger and more important than themselves, a necessary condition for human beings, if they are to have enduring fulfillment in their lives.
Earlier, I talked about the Unix operating system and how important it was. One reason why Unix was so popular was that every Unix system came with built-in networking tools. This meant that, if you could find a way to connect two Unix computers, it was easy to send data back and forth between them.
A programmer, talking in the argot of computer science, would say that Unix machines are able to "talk" to one another. Well, as someone who is well-versed in computer science, I can tell you that human beings find machines that can talk to one another infinitely more interesting than machines that work in isolation.
As someone who is well-versed in medicine and culture, I think I can also tell you why.
Human beings are social animals, driven by a strong biological need to be connected. As individuals, we are not happy unless we feel a connection with our family and our friends.
On a larger scale, our species is driven by a biological imperative, a force that has compelled us to build the Internet in order to connect ourselves into a large, supra-human organism. To me, the Net with all its computers, communication lines, information and people is nothing less than the next important step in the course of human evolution.
I admit that this is a strange concept. But think about how strange the idea of a beehive or an anthill would be to an individual bee or ant. Let us say that it were possible for you to talk to an individual bee or ant. How would you even begin to explain that the beehive or the anthill has a life of its own? That it exists because each separate insect does nothing more than follow his own individual biological destiny?
When I first thought about this idea, I believed that the Internet would bring only good to humanity. We would each participate in the Net in a way that makes sense for us as individuals. As a whole, we would, unconsciously, be creating something large and wonderful that, by its very nature, would bring us increased happiness and prosperity.
Now, I don't think it is that simple.
Since the mid-1990s, the world has changed a great deal, and the Internet has managed to insinuate itself into many aspects of our lives. To be sure, the Internet does bring out the best in us, encouraging us to create, share, think, learn and talk to one another. However, when I look around at the Net and how it influences us, I see also a great many problems: privacy violations, broken relationships, excessive advertising, a vast amount of misinformation, dishonest business practices, and so on.
For the rest of human existence certainly, for the rest of our lives there will never be a time without the Internet. For those of us who grew up before there were personal computers, this is a startling idea. Perhaps even more startling is the realization that the children who are growing up now will never remember a time without the Internet.
The implications of these ideas are far from obvious. In cultural terms, the Internet is still brand new, and it will be a long time maybe generations before we are able to come to terms with what the Net really means to our species.
Now that the Internet has become a permanent part of our everyday lives, we find ourselves influenced by strong forces that we don't yet understand and that we can't control. The Internet has become an extension of our minds. As such, it acts as an amplifier, enhancing both the good and bad aspects of our society.
If we are to understand what this means to us as individuals and as a society, we need to understand more than just computers and technology: we need to have a basic appreciation of ourselves, our culture and our institutions. We also need to understand psychology, history, philosophy, science, money, and relationships, and how these aspects of life fit into the new world in which we live.
In 2001, I wrote a book, Harley Hahn's Internet Insecurity (published by Prentice Hall PTR) in which I discuss these topics in depth, along with a lot of practical suggestions as to how to make the Internet work well for you. If these ideas intrigue you, I invite you to take a look at the book. It is, in my opinion, the best book I have ever written (not counting the Yellow Pages, of course, which is in a class of its own).
(By the way, the person who asked me to write this book was Jeff Pepper, the Publisher of Prentice Hall. He is the same person who, ten years earlier, when he worked at McGraw-Hill, asked me to write the Yellow Pages.)
When I think about the Internet and how much it has changed our world, the last ten years seem extraordinary beyond belief. I look around and see a new world, one that is filled with more creativity and communication than I ever thought possible. Truly, what I see today is much more extraordinary and much more profound than anything I would ever have anticipated when I first started to write about the Net.
Today, however, I look ahead with far more than a sense of wonder. Having written about the Internet for so many years, I know full well that none of us has any idea what the next decade has in store for us.
What a difference ten years makes.
© All contents Copyright 2024, Harley Hahn