 Mike Roberts, ICANN's first CEO, from 1998 to 2001. Mike, thanks for taking the time to talk to us. Peter Dengate Thrush said that ICANN was referred to as the ICANN experiment in the early days. Is that the way you saw it? I think most of us did in some form or another. You have to remember that the model for ICANN didn't exist. The blueprint was in the white paper and somewhat reluctantly people all over the world signed up to that, which didn't mean they loved it. So it was really up to the first board, the first nine members that had been recruited from around the globe and to some extent to the staff, to me and my helpers to make us credible and make that experiment turn into reality. When you say my helpers, how bad was the staff? You and who else? Well, I was extremely fortunate that our agenda, people ask, well, what were you supposed to do? In some sense, we were supposed to carry out the mandate of the white paper, but more directly, we were supposed to pick up several really nasty, intractable policy issues that the government managers hadn't been able to make any headway on. The trademark community was in a state of great upset. And as you know, because they're very closely tied to Hollywood and the big-time media people, their lobbyists were pounding on the Congress and on the White House all the time about cyber squatting and this awful damage that the Internet was doing to their business. So one of the reasons that Podesta and Clinton and the White House crowd wanted this off the table, off their table, was to get those people off their back. So one of the very first things we had to do was to take up this thorny issue of the relationship of trademarks to domain names and do something about the cyber squatting. What I'm hearing you say is they simply had a heavy lobbying influence with those governmental entities. Well, they did, and there was reality behind that because some of the nastier cyber-squatters were really doing damage to legitimate American brands. So how many people did you have working with you at the time? Molly Van Holling, who's gone on to a distinguished legal career and who was in a gap year before she clerked for Justice Souter on the Supreme Court, agreed to be my senior policy person. And she was a Harvard Law distinguished Harvard Law graduate who is now a distinguished member of the Berkeley Law Faculty. And by the way, a world champion biker, woman biker. Behind her came Louis Tutan, who at the beginning was at Jones Day, but subsequently came to ICANN as general counsel for a dollar a year. And then Andrew McLaughlin, who was another young Harvard Law graduate who had policy written all over him. A dollar a year. At the beginning, that's in addition to the IANA staff who had their own plate full of operational matters. That's all there was. So a very tiny group of people was the foundational structure of ICANN. That's true. What did you do for money? Money was a persistent issue. I'll give you where we ended up and where we ended up was a domain, an annual domain name fee which is assessed by the registrars and remitted to ICANN. But my understanding, and correct me if I'm wrong, my understanding was there was a time lag between the time that you guys started your work and when that actually occurred. It took over a year. So in that year's period of time, what were you doing for money? There was a collection taken up by a group of interested corporations in Washington that had been advisors to magazineer Roger Kuchetti from IBM, who was IBM's chief lobbyist in Washington, was very helpful. Vint Cerf, who was mostly senior technologist for his company. Vint has done everything in the Internet, as you know from your interview with him. But he also helped gather the forces and get some money flowing. And they committed the best part of a million dollars. And after quite a while we got some of that money. You have to remember that the way that charitable donations work in the United States is that the target organization has to have what's called a 501C3 letter from the government. And you don't get that right away. Ours took nine months. So in the meantime... Basically it means you're a non-profit. Yes, a charitable non-profit. So in the meantime, when the people directly trying to help us went to their corporate people to get a check, they said, what's their 501C3 number? And there wasn't any. They managed... And IBM helped a lot with a $100,000 unrestricted check. And others followed in. So finally by the spring of 1999, we had some money. But before that, nobody built us. Everybody put their bills in a accounts receivable basket. And my checking account and my credit card and some other things paid the bills. So wait a second. You're telling me that in ICANN's earliest days, it was sort of self-financed by a you person? Well, for instance, Molly and Andrew and I never got a paycheck for six months. And you're putting forth money of your own funds to get things going, right? That's right. We believed in it. We believed in the experiment. We believed that this... You have to remember that the internet bubble, we were right in the trailing edge of the bubble, but everything was in a very go-go atmosphere. And this was an opportunity to put some private sector public interest stamp on all this commercialization. Did you feel pressure to make it work? Oh, yeah. But I wasn't a newcomer to pressure. You know, I've been a university CIO, which is a very tough job. I'd been the designated guy for research university networking for 10 years in Washington. My impression is that, and correct me if I'm wrong, my impression is I can was very much a sort of the foundation going past the green papers and the white paper. All of that generated with Ira. The way that Vendis called the father of the internet. Yes. Is it a fair description to say that Ira Magazine or the Clinton aide was the father of I can? Yes and no. Ira did a great service by being the White House guy who chaired the committee that had all of these rough edged agency people in it, most of them from the research side of things, a few lawyers, a few NIST people. Ira essentially forced them to make peace over the green paper and the white paper, which had a lot of things in it that Ira alluded to that the research agency managers didn't think a lot of. For example? The whole idea of mandating things like competition in a domain name space, bottom-up management. This was not their cup of tea. Let's deal with that for a second, Mike. How do you get by in the typical corporation? Policy is set by the board and it filters down. Yes. You've got this funky new concept where you've got a bottom-up policy formation structure. Yes. Where the I can community is making policy and it's going up to the board. Was it difficult to get acceptance for that model because of its uniqueness? I have to digress a little bit. After the IAHC meeting in Geneva, which is a seminal moment, the white paper was out there. A lot of people who complained bitterly about the green paper stopped complaining and signed on. All the big players were in Geneva, even though it was this sort of ad hoc meeting. I was there. Becky was there. John was there. I was there. People like Larry Landweber from the academic networking community were there. By the end of that meeting, everybody said, well, we need to do this. What happened next was people started asking questions, saying, well, the white paper doesn't say how to do this. How are we going to do this? Of course, by throwing the ball in the court of a private sector nonprofit organization, that crossed up people who thought their private scheme, their own personal favorite scheme would go somewhere. One of the key moves was that Joe Sims from Jones Day offered pro bono to help John put it together. He sat down with John. He said, look, you're a research guy and you've been years on a research contract. That's not what this has to be. We have to put articles of incorporation together. We have to have bylaws. We have to find a board of directors of suitably senior people of a global diversity. So July and August of 1998 were really in my world because I was enlisted as a helper. I had just finished being a startup guy for the Internet to consortium, which needed many of the same things. So that's how we spent July and August, was making the concept in the white paper into a real organization. With the unexpected death of John Postel, how did that affect the direction, growth and structure of the organization, if at all? Yes. By the beginning of September, one of the other curves that we got thrown, which wasn't in the white paper, was that the government lawyers decided that the agreement with us could not be a sole source agreement. What does that mean exactly? Well, under the code of federal relations relating to contracts, the normal practice has to be demonstrated competition. This is sort of an article of faith. The government deserves a proper competitive bid on its work. So the white paper implied that once that the government, meaning NTIA really, came to agreement on the details that the selected organization would be off and running. So what the announcement about no sole source meant is that John's group had to throw a proposal over the transom at NTIA and any other interested party that wanted to compete for the business, if you will, the ICANN business, had to throw their proposal in the hopper too. That meant, of course, that everything had to be a lot more definite than was implied by the white paper. So it sounds like what I'm hearing is the white paper was sort of the frame, but you guys had to fill in everything. A lot of blanks, a lot of blanks. John's role ended up never being defined. I think Joe assumed, a lot of people assumed, I certainly assumed that John would end up being the CEO. But behind the scenes, it came out later that John had told Joe he didn't want to be the CEO. Joe wasn't very happy about that because he didn't think that the initial board was going to be very equipped to do an executive search. Why didn't John want to be the first CEO? He saw himself as a PhD research guy, not as an executive. And frankly, he didn't have a lot of admiration for the commercial side of the domain name business. Let me ask you this question. Postel is kind of universally accepted as a sort of nonconformist, if you will. Charismatic. Long hair, sandals, the whole thing. Which brings to mind the question, at the time that ICANN was born, it was coming on the heels of Watergate, a lot of distrust and authority. And a lot of you guys were children of the 60s. You were children of that. How much of that influenced the sort of structure and being that ICANN became? Ultimately, an organization like ICANN, especially technology nonprofit, is built on trust. You know, you can't write detailed enough bylaws to legislate all, especially anything having to do with the Internet. You know, an interesting statistic was there were 23 million domain names when we started. And today, almost 20 years later, we're headed towards 500 million. So how were you possibly going to write bylaws that scaled to that different world? We had to leave things as flexible as possible in, and we did. Someone mentioned to me that ICANN works because people believe it should work. Is that an oversimplification? Yes, although we took a lot of heat on a lot of different fronts from people that were not inclined to trust us. Because the model, you know, telecommunications worldwide, the divestiture of AT&T was only 12 years old. 82 to 96. And so there was a big hangover of the way you do these things is you pick a chosen corporation and put them under regulation and tell them how to run their business. That was, you know, we were 180 out from that model. But certainly inside the Beltway, there were lots of people that thought that ICANN should be run that way. Here you are, heading of your CEO, the first CEO of the Spunky Organization, the likes of which had seldom been seen. You don't have a hell of a lot of money to play with. Definitely. What kept you awake at night? The university community, where the internet was incubated, certainly Sirf and Kahn and Crocker and the early technical pioneers came up with a protocol on which it's founded. But in terms of making a market and demonstrating that this crazy open source suite of protocols would work, it was done in the university community with NSFnet. The internet went from having fewer than 500 nodes in 1985 at the end of the ARPANET period to more than 1,000 by 1989, almost all of them in the university community. So it was that community that really believed in this. To some extent, they believed in the initial board and they believed in people like Louis and Andrew and Molly and myself. But what did you fear? Did you fear, again, what kept you awake at night? I was very concerned that our legal vulnerability would lead to some sort of a loss of faith on the government side and it collapsed back into essentially a government-run corporation. That was always a possibility. In other words, if the ICANN experiment had failed, it's back to the U.S. government. People might have argued about that, but if you think about it, the Congress and the Bush administration by that time were not about to allow any other outcome. How much of a problem was it with the change in presidential administrations? We were very fortunate that by and large, and especially because of 9-11 and because of Iraq, the attention of the administration, that day-to-day the White House lived and died over that war and over the steps that were taken to strengthen our own homeland security. So wait a second. You're saying the plus was that you weren't on the radar. That's right. And we were also fortunate that Becky's successors Becky Burr as the director of NTI, Larry Irving, of course shared a Stanford background with me. We had a lot of good handshakes and he was very supportive of the ICANN model. And Becky was his second in command and really had the ball to work with us to make, I can't go. But to answer your question, it was her successors that maintained a light touch. As long as we sort of followed the game plan, paid attention to the bottom up, kept working on molding the support organizations into a wider model. And in effect, you know, in the beginning, I think Becky said this to you, we didn't talk about multi-stakeholderism in the beginning. It was sort of bottom up was the byword. But we did this lateral arabesque into a more generalized, supportive multi-stakeholder environment. John had the idea that the initial board would not be able to be trusted to do the right thing. You know, here's this researcher and the guardian of the protocols looking at recruiting effort that was grabbing evil hither and yarn from around the board. So in the bylaws, he made sure that Joe gave a very powerful role to these three support organizations, one for protocols, one for domain names, and one for numbers. It turned out that model didn't work. And in fact, it didn't even last a year before it was being subverted into a more generalized, more broadly supportive participation. And I definitely need to make the point that the reason that evolution worked was because of the willingness of so many people on a global basis to step in and do hundreds, thousands of hours of volunteer work. If you ask what made ICANN model work and the analysis in the latter-day analysis, it was all that volunteer help. Mike Roberts, thank you very much for taking the time to talk to us. You're very welcome.