NSF and the Birth of the Internet — Text-only | Flash Special Report
1990s


The World is Online and Caught in the Web
The 1990s will forever be remembered as the decade when the world came online. In the early years of the decade, the Internet was growing steadily, though few people had access to it. Still, people began to hear about the “Information Superhighway” that would change their lives. During this time, Tim Berners-Lee, a researcher at CERN, developed a new tool for sharing information on the Internet using hypertext that he called the World Wide Web. The first Web page was launched on Aug. 6, 1991. The Web soon made it possible for users to easily travel from place to place on the Internet, sharing text, images, and multimedia files. Soon the Web caught on in broader society. Hundreds of millions of people came online to surf this new digital frontier. Entire new industries sprang up (and in some cases crashed back down) as humanity embraced the biggest technological breakthrough since the Industrial Revolution. The Information Age had arrived, and the world would never be the same.

Video Transcript:

Karen Sandberg, Administrative Officer, National Science Foundation:
“A number of years ago, I was talking to a colleague, Gary Strong, I guess it was about ten years ago, and we were talking about the concern about the ‘have’ and ‘have nots’, the people who had computers versus the people who didn’t have computers, and that there was this big gulf, this big difference between the communities. And Gary Strong said ‘Don’t worry about it. This is going to change in a number of years.’ And I said ‘How so?’ and he said ‘Telephones.’ I said ‘what do you mean telephones?’ and he said ‘almost everyone has a telephone, and within eight to ten years, our telephones are going to have Internet capability, you’re going to be able to call and look at your calendar, you’re going to be able to pay your bills, you’re going to be able to use your phone as a weather forecaster,’ and it was just an idea that was just so totally astonishing to me.”

“So as the NSFNET was accomplished, I was in a meeting with George Strawn, and they were talking about how the NSFNET program had been basically successful in connecting many colleges and universities. And I was thinking, so now what is going to happen with the NSF networking program? I mean, mission has been accomplished. And I was really wondering if the entire division was going to go away. And in fact, as we know, networking continued, because the next job was to connect university libraries, small town libraries, K through 12 schools, and so this entire infrastructure that was built in a few short years, actually a decade, a little bit more, enabled those kinds of networking communications to also happen.”

“As the use of the capabilities enabled by the NSFNET, really skyrocketed, one of the solicitations that I looked at was for domain name registration. And the idea of domain names was used because before people were connected to long series of numbers, like telephone numbers, and of course that was getting very complicated with so many thousands and thousands of people. And so there was a domain name registry that was actually very simple, it was just a registration of a person’s or an organization’s name. NSF paid for it, and it was all free. Well, as time went on, at first the number of domain registrations was just a couple a week. And then a couple hundred a week. And very soon after, it was many thousands a month. The NSF awardee, Network Solutions, was really being pressed to have more and more people working on this, and this entire system was just like a wildfire out of control.”

“So, as the NSFNET continued, the idea of browsers came, the concept of browsers, and NSF had been supporting the University of Illinois, a man by the name of Marc Andreessen, at the University of Illinois, who developed a browser called the MOSAIC system. And we were so intrigued with MOSAIC, and as time went on, there were also a number of other organizations that also were doing work with browsers, because the Internet allowed so much information to be available and shared among researchers and other organizations, but it was really difficult to find the information that people were looking for. And so that’s what browsers were intended for.”

Video content credit: Cliff Braverman, Dena Headlee, Lauren Kitchen and Dana Cruikshank for National Science Foundation
Paul Wilson, MyMarketer.net
John Prusinski, Kathryn Prusinski, Michael Conlon and Joan Endres for S2N Media

Still Image Credit: © CERN

The Tools to Untangle the Web
The growth of the Web helped expand the Internet out of the academic arena and into the rest of society. The trouble was, as more and more people and sites came online, the amount of information online grew tremendously. How would people navigate this new and chaotic place?

To address these problems, NSF supported several projects to help untangle the web. One of the first was Mosaic, a breakthrough web browser that was the forerunner of virtually all modern browsers. Other initiatives included developing search engines to find information on the Web, as well as tools to keep Internet traffic flowing smoothly despite its rapid growth, and ways to keep networks secure from hackers, viruses and other threats.

Video Transcript:

(Voice of Eric Bina, Former Researcher, National Center for Supercomputing Applications [NCSA])
“In the beginning, let’s see, we were working at NCSA.  Mark was a student working there and I was a—they call it an academic professional—working there.  One evening, he came to me and told me that he really wanted to show me something that he was very excited about and he showed me this demo of the World Wide Web.  I think he showed me the web browser it was on was Midas.”

“He was very excited about this.  Mark is very good about getting other people excited about something that he’s excited about.  So he got me excited about it as well.  I went off and threw something together to give him a quick demo in a couple of days.  That’s where the first thing got started.  And this was all just kind of in our own time; project; after-hours thing for fun.  And that’s how the whole very first version of Mosaic came out.  It all moved very fast and it was a lot of fun and we really didn’t think at that time that we were doing anything earth-shaking.”

“It was fun for us to work on.  At the same time, it was going to be a really useful tool because we were all very conversant with the Internet at that time which many people didn’t know existed.  Back before the Web there was an Internet.  Everything we have now for the Web, most of it existed back then, you had to go through more archaic methods to use it.  So we knew we were going to make all of that easier.  Mark, in particular, felt that if we were going to make all of that easier, it would be a big change.  I wasn’t as much bought into that.  Like I said, Mark was good at getting you excited about it.  I was more interested in the challenges involved in getting the graphics to format and present right because I found that early on, once we started adding images that there wasn’t a lot out there that makes images and text in the way it had been done in paper media for years—in a very easy way.  So I was very excited with the technical parts of formatting the HTML and putting the text and the images up there and making it look nice.”

“Well the whole pictures thing, I must admit, was again Mark’s idea that I was against. Many people were against it.  I’ve heard stories later on from Mark—he didn’t tell me this at the time because he knew I was against putting pictures in—that Tim Berners-Lee was also against putting the pictures in.  And he told me that Tim was against it, because Tim said if you put pictures in there people are going to use it to display porn… and of course, he was right.  But by that point, it was obvious that everyone loved mixing pictures with text.”

“Actually, Larry Smarr, was—I think—at NSF and someone there had told him about this Mosaic thing that they had got off the Internet and he thought was really cool that was done at NCSA and was asking more about it.  Larry, at that time, wasn’t aware of the project and came back and talked to Joseph Hard who was in charge of our group at NCSA and that’s when it got to be all totally official and Larry came to me and said, ‘You guys need to really make this thing something because they’re all excited about it at NSF.’” 

“The very first release—actually we didn’t call it a ‘release’ because one of the other things (and I don’t know if Mark really pioneered this or not but I know we were one of the first people to pull it off well and we’re paying the penalty for it to this day) is having your users being your beta testers.  It used to be that you released software when it worked and unfortunately we began the process of releasing software before it worked.  So there were a number of beta versions put out early on.  We were putting new versions out on the web site every day or two.”

“It seems to me from my perspective that it’s often just taking something—like we did—just taking something that already exists and that may even be comfortable and familiar with but making it easy enough to use and sturdy enough to use that anyone can use without thinking about it; without having to work hard at it.  We made web pages really easy to make with the early html.  Children can do it.  We made the web browser really easy to use no matter what your platform was.  You downloaded a binary, you ran it.  That was it.  So I think the idea is if you really want to try to make a mark on something—which I wasn’t actually trying to do—but if that’s what they want to do, is to look at something that you think a lot of people are going to want to do that we do now that is difficult and make it easy”

“We’d frequently go to this coffee shop, Espresso Royale, because Mark wanted to get his coffee there.  I don’t actually drink coffee but I’d always go with him to this place and we’re sitting talking about how this is really taking off and one of us—I don’t know if it was Mark or me—commented, ‘You know, we might actually get to be a footnote in a book one day because of this.’  Which we were very excited about at the time.”

Video content credit: Courtesy of National Center for Supercomputing Applications (NCSA)
National Science Foundation: Cliff Braverman, Dena Headlee, Lauren Kitchen,
Dana Cruikshank
S2N Media: John Prusinski, Kathryn Sharar Prusinski, Michael Conlon, Joan Endres

Public-Private Partnership Pays Off
The decisions made by NSF back in the 1980s to shape how NSFNET would operate and grow were crucial to its growth in the 1990s. NSF made it possible for thousands of new users to join the Internet each month. Private companies did not need permission to link into the Internet—they just did it, and this allowed it to grow quickly, yet still operate smoothly.

Similarly, the structure laid down by NSF allowed networks in other countries to link into the Internet easily. The Internet became a world-wide phenomenon.

Video Transcript

Jane Caviness, Former Program Officer, National Science Foundation:  Initially, people didn’t even understand the possibility so we spent a lot of time on this is a possibility, that you’re geographically not disenfranchised if you’re in North Dakota or if you’re, you know, wherever, um, and that you can basically – this will enable all these things, but selling that in 1988 and 89 was – was still new.

Eric M. Aupperle, President Emeritus, MERIT Networks:  NSF contributed maybe, uh, somewhere on the order of $50 million toward the overall NSF Net program, and our partnership with our partners probably contributed another $50 million or thereabouts.  Um, $100 million sounds like a lot of money, but in the overall scheme of things, you’re thinking about the funding that goes to, uh, road construction or, uh, other infrastructure, it’s a drop in the bucket.

George O. Strawn, Chief Information Officer, NSF: At the time of 1995, the real rocket went off, and all of a sudden, the public became aware, through the fact that now that private sector was – were providing backbone networks and so forth, became aware of internet use and, uh, the usage skyrocketed and uh, the network bogged down again.

Jane Caviness, Former Program Officer, National Science Foundation:  Success was, you know, practically undid us.

George O. Strawn, Chief Information Officer, NSF:  There’s no way that the government could have or would have wanted to do it themselves when it hit an explosive growth phase and I think the, um, I think the entrepreneurs and the companies that provided backbone services all of a sudden found that they had uh, um, a much bigger job on their hands than they realized to be – to be satisfied much quicker, and they went into terrific factory mode of trying to scale up to handle the usage.  If the federal government had still been involved at that point.  There’s no way in the world we could’ve invested 1/10th or even 1/100th of the capital that was required to uh, to uh, transition to that huge growth in usage.  You may recall that the – that the initial telegraph line in the early to mid-1800s was funded by the federal government.  In one sense, the, uh, internet revolution recapitulated that, uh, previous century’s success where, uh, the government was responsible first under the uh, Arpanet development of the basic network protocols, then under the NSF Net, uh, expansion from a small two-dozen node network to a two-thousand node network that created, uh, enough impetus that the private sector said this is ready for, uh, this is ready for commercial use that all of a sudden we had something that was, of course, turns out to be even bigger than the telegraph network of the 19th Century had been.

Video content credit: National Science Foundation

New Industries, New World
With the NSFNET structure providing a stable platform and the World Wide Web providing an easy way to share data on it, the commercial side of the Internet grew in ways that were unimaginable just a few years earlier.

The Internet allowed computing power to transform the world in ways that would not have been possible if these computers remained unlinked. Businesses became much more efficient because information could be found and transmitted easily, products and payments could travel over the web. Every business soon had a website and was looking for ways to move its operations online.

People could shop online, talk to friends, research diseases and dozens of other innovations. News and entertainment media were transformed as consumers began getting their news, art and entertainment online.

Video Transcript

Stephen Wolff, Former Division Director, Networking and Communication Research and Infrastructure, NSF:  I had a, uh, a very close friend.  He retired from federal service in 1982 and tried to form, uh, an enterprise, that is, something to deliver to – to display groceries online and to deliver them to your door.  Uh, of course, he was, you know, quarter century ahead of his time, but it was clear from – he communicated to me the excitement and the promise of really ubiquitous and easy-to-use network in – in just making everyday life a lot easier.

Jane Caviness, Former Program Officer, National Science Foundation:  You know, the first few years I would – somebody on an airplane would say, “What do you do,” and it’s like, “How do I explain this to them?”  People didn’t even understand the possibility so we spent a lot of time on this is a possibility, that you’re geographically not disenfranchised if you’re in North Dakota.  We could talk about it but how fast it would happen, I mean it – the change was very fast.  The change from being able to get essentially no press in even tech publications to getting press everywhere, um, that happened in a, sort of a five-year spread.  You know, to going – people would ask me what I did and trying to explain it to them and then all of a sudden I didn’t have to explain it to them anymore.

Doug Gale, Former Program Officer, NSF, Former CIO, University of Nebraska:  It is one thing to talk about exponential growth in the abstract, and you know, I’ve talked about exponential growth in classes for years, but it’s another thing to really grasp in the sense of all of the complexity that it brings with it and so yeah, we knew we had to make it bigger and we made those plans, but we were always struggling, uh, to understand that, you know, in our heart of hearts I don’t know anyone that had any sense of YouTube, or any of the current phenomena that you see out there.

Jane Caviness, Former Program Officer, National Science Foundation:  I think the realization of when we had actually changed things was one time I was driving back to Washington, uh, the first time I heard a car company in Washington advertising something and they had in their ad something about their dot-com, and I thought, “We have arrived.”

Video content credit: National Science Foundation

NSF Decides (Some of) its Work Is Done
With the success of the World Wide Web and the explosive growth of the Internet, it was clear the Internet was here to stay. Keeping with its original vision, NSF decided that the Internet no longer needed public support to operate—private sector entities had found it profitable to build and expand the Internet’s infrastructure.

In 1997, NSFNET was decommissioned, and NSF awarded a contract to Network Solutions to assign domain names on the Internet.

Video Transcript:

George O. Strawn, Chief Information Officer, National Science Foundation: “Because the entrepreneurial and high tech community was interested in NSFNET by 1991, many of the most aggressive entrepreneurs said ‘oh, it’s time to get the government out of this business so we can step in and take over.’ And there were even several congressional hearings to that effect that I believe had been called for by entrepreneurs. And at that point, the point of the entrepreneurs was ‘ok, this technology is ready, it’s ready for the private sector to take over, the government should stand aside now.’”

Eric M. Aupperle, President Emeritus, MERIT Networks: “The commercial sector really wasn’t prepared to deal with the kind of traffic growth and capabilities that our partnership experienced.”

George O. Strawn, Chief Information Officer, National Science Foundation: We created network interconnection points for multiple private sector backbones to connect and take over the load of what the NSFNET had been doing. By the ’95 time frame, we were ready to retire the NSFNET backbone, and we did.”

Video Credit: Cliff Braverman, Dena Headlee, Lauren Kitchen and Dana Cruikshank for National Science Foundation
John Prusinski, Kathryn Prusinski, Michael Conlon and Joan Endres for S2N Media

Still Image Credit: : © 2007 Jupiter Images Corporation

NSF Looks Ahead
Although NSFNET was decommissioned in 1997, NSF continued to support technologies that helped make the Internet what it is today.

Image: Finger with computer chip.
Credit: Photodisc Blue, Getty Images