Web Masters Episode #32: Ira Fuchs

You can’t tell the story of the Internet without talking about a different computer network that came before it called BITNET. Hear why on┬áthe latest episode of Web Masters


BITNET - The History of Domains

Aaron Dinin:

The internet has become so ubiquitous that it’s hard to remember it didn’t just spring into being, but of course, that’s not the case. Our collective journey toward having internet everywhere, or I should say mostly everywhere was a slow process that took decades.

Ira Fuchs:

You have to also put yourself in the mindset at 1984, today, if I say to you, “Would email be a good idea?” You’d look at me strangely, “What are you asking? I don’t even understand. Of course, email is a great thing, we all use it, or we all use instant messages, whatever.” But back then it wasn’t so obvious, because you have to remember, there was no internet, of course, there wasn’t even yet an NSF net, which precedes the internet, and ARPANET, which had started in 1969, 1970 was highly restrictive in terms of who could get access to it. It was institutions with DoD contracts, and ARP is part of the DoD. It was not meant to be used by just anyone.

If you were a history or an English professor, or you were a student at an institution, you were not getting access to ARPANET. Moreover, there were very few institutions actually connected to it. This is also even before there was a TCP/IP, which are the protocols that undergird the internet. It was also expensive to connect to a network like ARPANET because it required special hardware and so on, it wasn’t just built into your IBM mainframe to do that.

I ended up filling up a couple of passports, traveling around Western Europe, and everyone wanted to connect.

Aaron Dinin:

The network everyone wanted to connect to at the time wasn’t the internet, it was a popular precursor to the internet called BITNET.

Ira Fuchs:

BITNET certainly was dominant. From ’81, ’82 until probably ’87 or so, I joke and say that, in a way, BITNET was the warm up, the decade long warm up act for the internet. It got people to understand and to use electronic communication.

Aaron Dinin:

Sure, when the public Internet and World Wide Web came along with concepts like email and instant messaging in the early 1990s, the adoption was rapid. But a big part of the reason the internet grew so quickly it was because lots of people already understood concepts like email and instant messaging, thanks to BITNET. On this episode of Web Masters, we’re going to hear from Ira Fuchs, the man who started it. Are you ready to hear the story? Then let’s get dialed in.

[INTRO]

Aaron Dinin:

Hi there, and welcome to Web Masters. This is the podcast that explores entrepreneurship and innovation by talking with the Internet’s earliest pioneers, or in the case of this episode, we might even say the pre-Internet’s earliest pioneers, because that’s what we’ll be discussing when we talk about BITNET.

My name is Aaron Dinin. I’m a serial entrepreneur. I teach innovation and entrepreneurship at Duke University, and I research the history of the internet, because I believe it gives incredible insights into the entrepreneurial process. But before we get to those insights in this episode, I want to take a minute to thank the company helping me bring them to you.

Support for Web Masters comes from Latona’s. Latona’s is a boutique mergers and acquisitions broker that helps internet entrepreneurs and business people buy and sell cashflow positive digital assets. Things like eCommerce stores, Amazon FBAs, content websites, domain portfolios, SAS apps, and any other type of online work from anywhere internet businesses. If you’ve got something like that, and are thinking about selling it, be sure to reach out to Latona’s, where their team of experts can help you get a great return on your years of hard work and investment. Or if you’re hoping to buy an internet business, visit the Latona’s website where you’ll find lots of businesses listed for sale, and that list is constantly changing with new incredible opportunities. That website is latonas.com L-A-T-O-N-A-S.com.

BITNET has a fascinating place in digital history, mostly forgotten now, it was an academic digital network that was so popular at universities during the early days of computer networking, that there was a time in internet history when people describe themselves as using BITNET, when they were actually using the internet.

Ira Fuchs:

Even after internet became a thing at institutions, in the late ’80s, beginning of the ’90s even, there were many faculty who, in writing their papers and so on and talking about how they communicated, they always said they were using BITNET, even though they weren’t using it anymore. It was so ingrained in their memory and I guess because it changed the way they did their work that they credited BITNET, even though they were no longer even using BITNET, or even using its protocols.

Aaron Dinin:

In that sense, it’s kind of like how someone might use the word band aid to describe any adhesive skin bandage or Kleenex to describe any tissue. BITNET was, for a time, a generic term for describing how computers were connected to each other, specifically across universities. But BITNET and the internet are, from a technology standpoint, two different things.

Since I’m guessing, a lot of people listening don’t know much about BITNET, let’s start our exploration of BITNET by letting Ira Fuchs, BITNET’s founder, explain the key differences between the two networks.

Ira Fuchs:

I suppose there are a lot of differences. But one difference is that BITNET was all about store and forward of files. The files being email or binary files, but they were file. Basically, you send an entire file to the next node in the path that gets you to where you want to get to. If you wanted to get from CUNY to Duke, and I wanted to send you a paper, a document that was obviously on my computer in a Word processor, I would address it to Aaron at Duke, and the software then would take that entire file, and send it to the next step in the path.

That next step could be, I don’t know, it could be NYU. It’s possible, and NYU would send it to Delaware, and the entire file would be sent, each step along the way. It would get in queued at the next place, and then eventually will be sent along the way. Now, if the lines weren’t too overly taxed, that would be a matter of seconds to get from CUNY to Duke.

But if it was very busy, that file would sit there for a little while. Not a long time, typically, but it might sit there for minutes, but you didn’t notice because you didn’t have the exact feedback that you get today. Eventually, it would get there and certainly be faster than dealing with sending it through the mail. Instant messages are an exception to that in that they do go store and forward, but they take the highest priority. Even though there are files being sent, the interactive message can be sent at the same time, and will get there essentially, instantly.

The internet is based on packet switching. So, every message is broken into a relatively small number of bits. The message that I’m standing, take the file that I’m sending between CUNY and Duke, that file is broken up into potentially thousands or millions of packets, and those packets are sent to Duke, but each packet may take a different path.

One packet may go to NYU and in fact to go through various nodal locations in internet and eventually get to Duke. But it all gets reconstructed at Duke before it gets handed to you. Now, that provides for a much better utilization of the network, because we’re sending packets every which way, it also comes back to what happens if one link is down, well, it’ll go some other way.

It opens up the possibility of other protocols for things like the web, that wouldn’t really been imagined using just BITNET. Now, people did try to create BITNET services, which allowed you to have menus that would come up on your interactive screen and so on, but it was just playing around at that time. The Internet is a completely different beast in terms of its routing, and in terms of the way it utilizes these lines, and I’m sure I could come up with a dozen other major differences, but the BITNET protocol was designed mostly for moving a file from point A to point z, where you’d have to go through a bunch of intermediate points.

That worked well for what it was used for. But it could not have supported the protocols and the services that are common today on the internet.

Aaron Dinin:

To summarize, BITNET was basically a tool for sending files and messages from computer to computer. Sure, it’s only a fraction of what people do with the internet today, but again, digital connectivity had to start somewhere. Before people could get comfortable buying things online and posting silly videos of their cats, they had to get comfortable talking to each other. That’s a big part of what BITNET was able to accomplish, despite the immaturity of computer networking.

Ira Fuchs:

BITNET was primarily 9,600 bits per second, which today sounds like impossibly slow, but we could do a lot with that. BITNET supported email, it looked the same as email today, really, there was no Domain Name System. You would have been, let’s say, Aaron@duke, without the rest of the domain name. It supported file transfer. If it was a binary file or a picture or something, you could send that over BITNET in the same way, and it was the first, I think, to support instant messages.

I could send, just as AOL used to have you send instant messages to people, and you can send instant messages now through iMessage, or something like that, BITNET had instant messages. I could send a message to Aaron@duke and it would just pop up on your screen.

I tried as BITNET started to grow from its first days, I would say good morning, to every one of the sites every morning. I would actually type in and say good morning.

Aaron Dinin:

Imagine an internet so small, you could literally say good morning to everyone on it each and every day. But of course, that’s what things were like in the earliest days of networked computers. The other important piece of context to understand about BITNET that you heard Ira alluding to, is that it was created for faculty, staff, and students at universities, and mostly restricted to them as well. In other words, like most computer networks at the time, BITNET wasn’t some open for anyone to use democratic gathering of the masses. It was scholars doing scholarly things, researchers doing research things, and students doing well, probably student-y things.

That shouldn’t surprise us because at the time, the majority of computer networking was affiliated with either governments or universities. That’s also where Ira was coming from.

Ira Fuchs:

I got interested in computers when I was in high school, and that was a long time ago. I remember going to Manhattan, I lived in Queens, I lived in Forest Hills, and we would go into Manhattan to get computer manuals at the IBM building on 57th Street. There was an IBM building there before the one that’s there now. But there was a place you could go, you could buy computer manuals for what was then mostly assembler language, and maybe the beginning of Fortran.

I remember being interested then, then I went to college, I went to Columbia University, and I majored in physics, but I was very, very interested in computing. I began taking all the courses that they offered in the, what was then combined computer science, electrical engineering department, actually, it was the other way around, it was electrical engineering and computer science. Computer science took second fiddle, I guess, and I really enjoyed it a lot.

In my senior year at Columbia, I got a, not full-time, but certainly a significant part-time job at the Computing Center as a systems programmer. I began to get involved with actually programming the mainframes that I’ve been using as a student.

Aaron Dinin:

Would you mind talking about what programming was like back in those days? For some reason, that always fascinates me, and I love hearing the stories that illustrate how far we’ve come.

Ira Fuchs:

Back in those days, everything was card decks. I was saying this to someone recently, how exciting it used to be to submit your card deck in the morning, and then you’d come back midday, and the output for your run was basically paper that was wrapped in general around your card deck and put into a bin and the bins were either by account number or name, I don’t remember. But you’d come back and you’d wish or hope or pray, especially if it was a class assignment, that the amount of paper that surrounded your deck was not too little and not too much. Because if it was very little paper that wrapped around it, that meant that your run failed, and it failed on some simple syntax error or something like that. If it was a lot of paper, it meant that there was a crash, a dump of the program, which meant it had also failed.

What you wanted was in between amount of paper, which meant that at least you had gotten some result.

Aaron Dinin:

I can’t imagine how much harder it must have been to code back then. When I write code, I expect things not to work the first time and the errors are what tell me what’s broken so that I can immediately fix them. Honestly, I don’t know what I’d do if my code had to be perfect the first time, I guess I’d get way more stressed and honestly probably be much more excited when it worked.

Ira Fuchs:

I was musing about too someone, the excitement that was associated with that. Now, think about that, you got maybe two or three runs a day in, if you were lucky for your program. Today, we have, obviously, everyone has the ability to get instantaneous reinforcement feedback. This works, this doesn’t work. Especially if you use languages like Python, or Ruby, or something that’s highly interactive, you don’t even have to compile the program to see the result, and even if you do, it takes a second to get a result, you get instantaneous feedback.

Obviously a much better world that we have now in terms of developing highly complex programs, but there’s something missing about that excitement about coming back and seeing your card deck and whether it had a lot of paper wrapped around it. Anyway, that’s how things started for me in computing. I then became a full-time staff member at the Computing Center when I graduated from Columbia as an undergraduate and I began work as a graduate student at Columbia in computer science.

Aaron Dinin:

Okay, you’re a computer science graduate student at Columbia, but I know BITNET launched out of CUNY, not Columbia, which I’m guessing means you didn’t launch BITNET while you were still in school. When and how did that transition to CUNY happen?

Ira Fuchs:

There’s a whole story that goes along with that, which I won’t bore any listeners with. But I was made the executive director of the university Computing Center of the City University of New York, which is, I guess, the largest urban university, perhaps in the world, certainly in the country. It had, at that time, 250,000 students on 22 campuses, and the university had made the decision to centralize all of its computing.

They created a budget and agreed to create a central facility, and my former boss at the Computing Center at Columbia, became the university dean at City University. He then decided that I was the right person, at 24 years of age to be the executive director of this new computing center, which was perhaps the largest university computing center anywhere at that time, one of the largest, if not the largest.

I had 150 people or whatever working there. It was interesting. I was learning a lot on the job. But my mentor, Ken King saw something in me that I didn’t maybe even realize was that maybe I could actually be good at that. Anyway, I ran computing there, and this was… Now, I came there in 1973, and we networked all the campuses. We created a network of IBM 360 computers. There was one at each campus, and we had a very large central facility, with much larger mainframes on 57th Street in Manhattan.

We connected all the campuses to the central mainframe. They had a certain amount of local computing, but it was also using the capability of the central mainframe. This was for academic, as well as administrative computing. This was very successful, this networking, and it was based on the software, which came with the IBM systems that we were running, and that software was called RSCS. I think that stands for Remote Spooling Communication Subsystem or something like that.

It is a relatively simple protocol that uses store and forward of files and messages. Basically, if you’re a node on an RSCS network, and you want to send information to one of the nodes in the network, there’s a certain topology that you have to adhere to. Let’s say, there are three nodes and you’re A and you want to send to C, you may have to send it through B.

Now, that all happens automatically. You don’t have to address it in any way other than, I want to send to Aaron@duke, let’s say, and it figures out to get to Duke, it has to go through Delaware, and it passes through the Delaware computer and gets to Duke.

Aaron Dinin:

But the original network you built was only inside of CUNY, right? How did you get beyond that?

Ira Fuchs:

At that time, the only way the software was really being used within academia was locally. Something like what City University had done. The largest network, though, probably in the world at that time, was a network called VNET, and VNET was the internal IBM network, which connected virtually every mainframe, every site within the IBM Corporation using this RSCS protocol.

As it turned out, one of my high school friends and college friends later was one of the people instrumental in creating VNET at IBM, and he was in IBM Yorktown Heights, and I’m not sure I can reconstruct this completely, but one day, we were sitting around talking, and I’ll give him credit, I don’t know if he was the one who thought of it and said at first, but he may have said something like, “You ought to do something like this for higher education.”

I thought that was a fantastic idea, although rather daunting, the idea being ultimately to connect all the scholars of the world. That was how ambitious we were thinking at the time, and I was pretty young, so it was easy to think that way.

Aaron Dinin:

That’s definitely true in most entrepreneurial activities. I always like to say the more naive you are, the easier it is to overlook the ambitiousness of what you’re trying to accomplish. But I guess it worked for you, you got connected to other universities, right?

Ira Fuchs:

The next step was I was at a meeting, what’s called SIGGUCs, Special Interest Group for University Computing, or something, because the IEEE oversaw the [inaudible 00:20:14] groups. But in any case, I was at a meeting, I don’t know where it was, Indianapolis or someplace, and I was with my colleague, my counterpart from Yale University, and I broached this idea with him.

I said, “What do you think of having a network that would connect universities in this way?” As his first test, I said, “Would you connect to CUNY, to start this network?” He thought it was a fine idea. So, we connected our two universities together. For the first time, there was a flow of data, of communication between Yale and CUNY. The faculty, the students, as well as the first users of it, were probably the systems programmers in their respective computing centers, to be able to share fixes to programming problems that we’re having, and so on.

Aaron Dinin:

The connection with Yale was successful, and then what? How did it spread to other universities?

Ira Fuchs:

I realized this might be the right time to go further. So, I wrote to something like 40 colleges and universities along the eastern seaboard. I still have the letter someplace, a very short and simple letter, and I described what we were doing, and would they be interested in being part of this? The response was overwhelmingly positive. There were no negative responses, but there were maybe some lukewarm or no response, but I got a lot of positive responses.

We came up with basically three requirements for this network to come into existence. The first was that if you wanted to join, you had to pay for a lease line between your institution and another institution that was already connected to the network. You had to agree that you would permit one or more other institutions to connect to you, that would allow the network to grow. Lastly, you agree to pass the traffic for those third parties, for those other universities downstream from you, at no cost. You wouldn’t be counting all the bytes that people sent through your institution and trying to recover the cost.

This was the basis for the network. There wasn’t a dime of government money, federal or other monies in this network, it was all being done by the institution. In a way it was, and people have written about this, since then, it was a grand social experiment, to see… Because everybody was dependent on everyone else. If you decided you were going to drop out of the network, you would destroy communication, you would kill the communication between other institutions, other than your own.

If you’re sitting in the middle, ABC and you’re B, and you say, “I don’t want to do this anymore.” Well, now, A and C have a problem talking to each other. It didn’t fall apart, it instead grew rapidly.

Aaron Dinin:

How did it get beyond the east coast where you launched it?

Ira Fuchs:

I guess one of the most important days or points in the growth of the network was when the institutions on the west coast, the University of California, and Cal State said, “We want to join this network.” The problem is, how do you get from California to another institution that was connected, when it was mostly in the East, at this point? We’re starting to move west, from the east, but there was no one in the West.

Two things happened, first of all, University of California decided to pay for a link back from San Francisco or LA, I forget which connected first, back to City University of New York. Because that link was a very expensive link, it’s a little hard to think about today, when you think about the cost of communications, but when you have to pay for a leased wire, essentially, it’s running 3,000 miles, give or take, you pay by the mile. These are not very fast lines, either. I might add. They were 9,600 bit per second links.

They paid for that link. I don’t remember what it cost, but it was certainly thousands of dollars a month, and we agreed to bend one of the rules of BITNET, and that was that the institution that was connecting, which was one of the campuses of UC, actually was Berkeley, now I remember, it was UC Berkeley, we agreed that they could charge a portion of that link back to the other institutions in California that would in turn connect to them. Because otherwise Berkeley would have to pay the whole bill.

When Stanford connected and the other UC campuses connected, they paid a portion of the expensive transcontinental link. That started to connect the West and then the West started to come east, the east started to come west. Essentially all of the major and not even major institutions in the US were connected to BITNET.

Aaron Dinin:

BITNET went international too, right? How did that happen?

Ira Fuchs:

I think ’84 or so, I got a visit from someone from IBM Research, and they made me an offer, one of those canonical offer you can’t afford to refuse, and they said, “Would you like for BITNET to expand to Europe?” I said, “Sure, that’d be great, because our goal is to connect the scholars of the world.” IBM said that they would be willing to pay for a link from CUNY to Rome, Rome, Italy, they would also pay for the inter country links from Rome to other major cities in Western Europe. Then it would be the responsibility within each country for universities to make their own connections.

I said, “Great, what do I have to do?” They said, “Well, we want you to go out and sell this idea to universities in Europe.” Which meant I would be traveling a lot and away a lot. It sounded like a good thing to me. CERN where the World Wide Web really started connected to BITNET, England, France, Germany, everybody. What happened there was also the Europeans said, “Well, we want to connect, but we want to have our own administrative oversight of what we do, in terms of the connections between our countries and so on.”

They created a separate legal entity, BITNET was now a corporation, a not for profit corporation, but they created something called EARN E-A-R-N, the European Academic Research Network, which is basically their BITNET Europe. Later, BITNET spread to Israel, into the Gulf states, and there was something called GULFNET. A little later, it made its way to Japan.

A lot of these connections were either to California or directly to New York. If you were to draw a picture of the network, it would look like a bunch of nodes with a lot of connections to those nodes. Sometimes the nodes near the end, of course, might only have one connection, sometimes only two or three, CUNY had the most, I don’t remember how many, but it was probably more than a dozen direct connections, because it was just easier sometimes to connect directly to CUNY. You’d have fewer hops to connect to other places, as a result.

Aaron Dinin:

What was the place you were most surprised to get connected to?

Ira Fuchs:

We managed an interesting coupe, we were the first to connect Russia as part of an academic network. We had to get permission from the Department of Commerce, the US Department of Commerce, they didn’t really understand what we were doing completely, but somehow they said yes to it. I traveled to Moscow, and gave a talk to the equivalent of our Academy of Science, the Russian Academy of Science. Which was interesting, it was the first time I’d ever given a talk where I was having simultaneous translation.

There was this little person in a glass box, at the other end of the room who was translating everything I was saying, so that the people in attendance could listen on their headphones, and Russia ended up connecting as well.

Aaron Dinin:

You mentioned getting connected into CERN where Tim Berners Lee would eventually start the World Wide Web. Do you think there was any relationship between BITNET and the eventual web?

Ira Fuchs:

The world changed dramatically, when Tim Berners Lee created hyperlinking, using a protocol that he designed, which was HTTP, and then it got expanded by Marc Andreessen and others, and they created Mosaic and we had a graphical web browser. I could tell that whole story too, but I don’t claim to be part of that, other than a user.

I was probably one of the first users of Tim’s HTTP protocol. He and I, we know each other, but I knew him because of the back and forth at CERN with connecting to BITNET. I had a NeXT Computer, which Steve Jobs had given me, and Tim also had a NeXT Computer. When I found out what he was doing, I tried it out immediately. Now, it was pretty primitive back then. You had a textual document, and certain words, you could tab to them, or maybe on the NeXT, you could click on them, and they would, using the HTTP protocol, take you to another document. The web was born, so as they say.

The idea that you would then get much more than a link to another text document, required other people to contribute, and a lot of people did. But everything you have today on the net, you could draw a straight line all the way from.

Aaron Dinin:

Reading between the lines here, a bit, I think the way Ira’s story trailed off here points to a combination of pride in the incredible foundation BITNET laid, and maybe a hint of regret about what it might have become.

Ira Fuchs:

People say to me, some version of, they don’t say it this way, but some version of, you’re so smart, why aren’t you so rich? If you were there so early, why are you not the Bill Gates of the networking world? It’s just not the way we thought of it at the time. It wasn’t, this was an opportunity to make a lot of money. I’ve worked in higher ed or in not for profit my whole life. So, I didn’t think about it in that way.

We joke, some of my friends, who were early on in the world of networking, or internet pioneers joke about how, if only we had simply grabbed the biggest and most prominent domain names, we would have made a fortune, just on that. Would not have to develop any technology at all. We were all there before there was a DNS, before there was a mcdonalds.com. It would have been pennies to take every name you could think of and then just make them available on the market, I’m joking. But I don’t go back and think about how gee, if I had only done that.

Aaron Dinin:

I even think Ira is being a bit too modest here. In our conversation, there was never a hint of fame or money seeking. In fact, he worked tirelessly to make sure BITNET was not a commercial enterprise.

Ira Fuchs:

One of our basic tenets of BITNET was that it be absolutely non-commercial. It sounds funny now, what’s the internet today, but commercial, but BITNET went to great lengths. We had rules that we wanted to enforce. If an institution dared to try to use the network for any commercial purpose. Now, there were those that argued that universities are commercial entities of a form also. They take in money, they spend money, they deliver a service. But, we didn’t want any companies connecting to sell things, and so on.

It was all about non-commercial use. We did allow some of the Federal laboratories to connect, and the only commercial organization I can recall to connect to BITNET was IBM, and it was only connecting to IBM Research, it wouldn’t allow a salesman to connect to a university to sell a computer, it was just IBM Research. Getting IBM to agree to that was no small feat. Even though they were delighted that this worldwide network was using their protocol, they did not love the idea of connecting their computers to a bunch of academic computers, even if it was using their software. But they finally did agree to it, although there were many hoops you had to go through to get yourself, your individual to get access to the gateway between IBM and BITNET, but they did allow it.

Aaron Dinin:

If BITNET wasn’t about making money, what was it for? Why do you think the world needed it?

Ira Fuchs:

The problem ultimately was how do you connect the scholars of the world? You also have to think back to what it was like for scholars to communicate back then. Essentially, you had three modes of communication; you had the telephone, you had fax machines, if you wanted to send a paper document, and third you had the US or the whatever mail system, physical delivery system was available. There wasn’t any other way to share information. Obviously, you did share information through peer reviewed journals, and so on, but there’s a long time lag, as you know, from writing an article, to getting it out to your colleagues.

This was a huge shift. Joining BITNET meant that you had to begin to network your campus, you had to figure out how you’re going to make this new resource available. Things that we take for granted today. It’d be like saying to someone, well, there was electricity all of a sudden, how did you get electricity out to people? Well, that’s kind of what we were doing.

Aaron Dinin:

How did you convince the scholars of the world or really anyone that they needed this new type of connection?

Ira Fuchs:

The first example that comes to mind, which was, I’d say, one of the driving forces in having worldwide networking was for the physicists, the high energy physicists. The people at CERN, at the Center in Geneva for Nuclear Research and High Energy Physics, and at SLAC, Stanford Linear Accelerator Center, these were physicists that were constantly traveling back and forth, to work in one group or another. But they were always in communication with one another.

Now, had they had to use the traditional forms that I mentioned before, US mail and fax machines and telephone, they didn’t require any convincing that something like BITNET would be of benefit to them. I don’t know how much communicating you do with your own colleagues, in your own field, but it wasn’t hard to convince historians and economists, and engineers and so on that there was a value in communicating with their colleagues.

Now, I won’t say that all the communication was highly academic. We weren’t supporting things like film. The significant use of the Internet today is probably for salacious purposes, shall we say, and that wasn’t really possible on BITNET. I suppose you could send still pictures, I don’t know if anyone did that, but I suppose that was possible. But you could just be communicating with your friends over BITNET, and people did.

But the scholars of the world, I think I used that phrase in something I wrote or a talk I gave, as the grand idea. It wasn’t just faculty, it was meant to be students as well, and administrators, and really they all used the network as they used the internet now.

Aaron Dinin:

From an entrepreneurial perspective, this part of Ira’s story is, I think, important for understanding the relationship between innovation and consumer behavior. Consumers don’t care much about what something is called, internet, BITNET, heck, most people use the phrase web and internet interchangeably these days, even though they’re actually two different things. Consumers also don’t tend to care about how things work at a technical level, what they care about is the value they get, and BITNET created an unprecedented value for people at universities. They could collaborate faster and easier than ever before, and that was revolutionary.

Aaron Dinin:

So revolutionary, in fact, that people wanted even more, and this is where the core architecture of BITNET just couldn’t keep up. So, other technologies ultimately supplanted it.

Ira Fuchs:

The Internet, I’ll just say appeared, but it of course, is a longer story to that too, which involves NSFNET and other things happening. But eventually, it was clear that there was no need to continue running BITNET as it existed at that time. We made the best of it, we also, for a while, this is the one time we had an infusion of any government funds into BITNET. I requested and got a grant from the NSF to write the software necessary to send the BITNET data over the high speed NSFNET lines.

We were actually using NSFNET to run BITNET, at least for the institutions that connected that way, if you understand what I’m saying. It meant that you could send a lot more data over BITNET. That was a temporary measure. It worked well, but it was not meant to be a long lasting thing.

Aaron Dinin:

Too slow, too inefficient, too expensive, the BITNET network slowly and mostly unnoticed by its users faded away from campuses and was replaced by the Internet. In this sense, the story of BITNET and its relationship to the Internet is a microcosm of every story of innovation. Technologies don’t spring up overnight, they slowly, sometimes very slowly transition from one thing to the next. Maybe the technology you create remains popular for centuries, or maybe it’s popular for only a few years. Either way, each new piece is critical for moving the entire process of innovation forward.

In the story of the Internet, BITNET was, without a doubt, a critical piece of technology that helped to spur the digital innovations we all use and benefit from today.

Ira Fuchs:

There’s no book about BITNET. The furthest we get is Wikipedia entries, I guess. As I like to joke about BITNET, my claim to fame for BITNET… Well, I was recognized by being inducted into the Internet Hall of Fame, largely for BITNET, but as I said, when I received the award, the most significant recognition of BITNET to that point was the fact that the Oxford English Dictionary decided that BITNET was a word and added it to the dictionary and gave me credit, or as they usually do with an etymology, they point out where it was first used, and it was first used when I used it to describe this network.

I’m happy that the Internet Society decided that it was something that they wanted to recognize. I say that because when the Internet Society decided to have an Internet Hall of Fame, you know, they really wanted restricted to things that were directly involved with Internet and with the protocols that Internet utilizes. A lot of the awards that have been given, or a lot of the people that have been inducted are involved with, if not TCP/IP, some other protocol, or some other use of Internet.

BITNET predates that. It would have been easy to forget that and say, it all started with Bob and Vint, and NCP, which was before TCP/IP, and then TCP/IP and HTTP for Tim, and so on. BITNET just used some commercial protocols that were there. Well, that’s true, but I think if you think about it a little longer, you realize that it had an impact.

Aaron Dinin:

BITNET definitely had a critical impact on the growth and adoption of the internet, and for that, we all can thank Ira Fuchs. Oh, and one more thing, we actually all owe Ira our thanks for another useful technology, he helped create LISTSERV. That’s right. If you’ve ever been part of a listserv, well, that was Ira’s work too. Naturally, I couldn’t let them out of our conversation without at least asking him to share that story.

Ira Fuchs:

Well, I was still at City University of New York, and I applied to IBM to get a grant for City University and EDUCOM, which was a consortium of universities interested in computing, to create, I think we called it the BITNIC and the BITDOC. That stood for the BIT from BITNET and the NIC, Network Information Center and the BITDOC for the documentation Center Development… I forgot. In any case, part of the grant would go to City University of New York, part of it went to EDUCOM.

As part of IBM giving this grant, they said that I had to be part time at EDUCOM to oversee their part of the grant. I became the Director of Networking Activities at EDUCOM, while I was also Vice Chancellor at City University of New York. One day I was down in the Princeton Office of EDUCOM, sitting around with two of my colleagues, and we were talking about the idea of how you could use email to communicate with a lot of people at the same time along some common topic.

We came up with the idea for LISTSERV, and it basically was, you write to an address, you create a name for the list, I don’t know, snow. I can’t think of anything, I’m looking outside, it’s snowing. You say snow@listserv. LISTSERV knows all the people who are interested in your topic of snow, and now when you write to it, it explodes it and sends it out to all these people, and we named it LISTSERV.

As far as I know, that was the first… Well, certainly first use of the name. But the first major use of that idea. I don’t claim that no one else had ever thought of having a mail exploder, I’m sure someone did, but I don’t know that anyone had implemented it, and had it widely used. We spelled it L-I-S-T-S-E-R-V. I don’t know if anyone ever wondered why, but we wouldn’t put an E on the end, and that was only because the way it got implemented was it was implemented in the IBM operating system, and the user IDs can only have eight characters. That’s why we had to drop the E.

But anyway, that’s how LISTSERV was created. It was a joint idea of Dan Oberst, who was a good friend of mine, and Ricky Hernandez, who was a programmer who worked at EDUCOM, and he basically wrote the code and LISTSERV was born. It grew very fast, to have a lot of groups, because I guess it was another one of those right idea at the right time.

It was expanded on greatly by Eric Thomas, who was at CERN, I believe at that time, he took the idea and created what he called LISTSERV enhanced, and even turned it into a commercial entity, which I think it’s still a commercial entity today. But there have been a lot of other implementations of the idea of creating mail exploders and that sort of thing. But that was the birth of LISTSERV was back then in ’84, ’85.

Aaron Dinin:

Well, when something as well known as LISTSERV is like a footnote in the story of what you’ve created, I’d say you’ve done some pretty incredible things, which, of course, is the case for Ira Fuchs. I’d like to thank him for taking the time to share his work with us. I hope you enjoyed it, too, and if you did, you know what I’m going to ask, go to your podcasting app of choice, and like the episode, leave us a great review, or whatever else you can do on there to help us spread the word about Web Masters.

While you’re at it, be sure to subscribe so you receive the next episode as soon as it’s released. If you have any questions, thoughts or feedback about this episode, you can reach us on Twitter, we’re @WebMastersPod. I’m on Twitter to @AaronDinin. That’s A-A-R-O-N-D-I-N-I-N. If you haven’t already been doing it, head on over to medium.com, where you’ll find all sorts of articles I’ve written about startups, entrepreneurship, sales, marketing, fundraising, and those kinds of things.

Thank you to Ryan Higgs, our audio engineer for his help piecing together all the sounds, and thank you to our sponsor, Latona’s for their support. If you’re interested in buying or selling an internet business, don’t forget to check out latonas.com, and if you’re interested in learning more about entrepreneurship and the history of the Internet, well, you’re in luck because we got another episode coming soon in just a few days. Until then, well, it’s time for me to sign off.

Leave a Reply

Your email address will not be published. Required fields are marked *