A few months ago when I was starting WWWiz Magazine I called EarthLink and talked to Bob Johnson. They had ten people and were busy as hell. At COMDEX 95 in Las Vegas, about four months later, I talked to Sky. He said they had 145 people. When I did this interview, two weeks afterward, he said they had 175 people.
When I got to the EarthLink office, I didn't even have to ask if they were still growing. The lobby was like a full elevator with people packed in like...well, people in an elevator. People passed through the room with difficulty, while two frantic receptionists worked the phones and the crowd. After I settled in and started to listen in on some of the conversations around me, I realized I was in the center of a hiring frenzy. Nerds everywhere! EarthLink is literally bursting at the seams with people and work.
How did EarthLink's energetic, bright, young CEO, barely out of high school, become the savvy, insider, part-of-the-big-five, fastest-growing business...entrepreneur of the year?
Sky, 24 years young, has built an ISP (Internet Service Provider) organization that is solidly in the national top five providers. If you want to get online there is no easier or better way than to insert an EarthLink disk for Mac or Windows and sign on. All the programs you need are there and you just answer a few simple questions like what do you want your online name to be and how do you want them to charge you for the service. Click on an icon, the program dials their 800 number, and zap! You're cruising the Web with the best of 'em.
I managed to squeeze into Sky's crowded office (the company is growing so fast that he shares an office with an assistant) and start at the beginning.
WWWiz: Where are you from?
Sky: I was born in New York City, 1971. My father was a sculptor and my mother was a dancer and a poet, in the Bowery area, Greenwich Village. My maternal grandfather was a big IBM engineer, one of their Fellows. Forty of their top guys, they'd just sit around with their feet up on their desks and think of new ideas and brainstorm. I don't think it's like that there any more, but he used to dream all day; it was pretty cool.
We moved to Los Angeles when I was very young. When I was growing up, I spent a lot of time with my grandfather on my mother's side of the family. My mother was pretty much the only pure artist; the rest of the family were pretty technical—physicists and so on. I believe that influenced me a lot.
WWWiz: How did you get interested in computers?
Sky: I had a very early obsession with computers. I thought about the world around us and what computers could do to improve the standard of living. Of course, back then I wasn't thinking in economic or sociological terms. I just thought they were so cool.
I remember when I was about nine or ten; I got a Timex Sinclair, and I got it preconfigured with 1k of RAM, and I was stylin'. I was writing games and stuff, and the games I was writing needed more RAM, so for Christmas I got a 16k memory add-on, which was a box that had its own power supply. [laughs] This was state-of-the-art. I'd hook it up to my dad's TV, and if I wanted to play games, I didn't have a lot of money, so I'd write games and record them onto a tape recorder and so on.
Through high school I started really getting interested in computers; I recognized that the ultimate purpose of a computer was to communicate. You know, like in the '80s Microsoft came up with things like, "Information at your fingertips," and it was the "Information Age" and all that stuff, but really, information is completely worthless unless it's communicated, so ultimately it's all about communication. I believed that computers could facilitate communication and I looked for ways to do that. One of the ways that sort of popped out was imaging and mass communication—advertising, stuff like that. Using computers to do what you used to do by hand—things like desktop publishing, for example, computer animation.
I got out of high school in 1988 and I was looking around at opportunities, and I wasn't interested in computer technology for the sake of technology. I was interested in the application, ultimately. What would you be doing with it? Communications was obviously where I was moving. I really wanted to get into it right away, and I had actually already been into it through high school and done some really interesting things— worked on Michael Jackson videos, you know, some cool stuff.
I looked around at the curriculum of some colleges, and they weren't teaching what I wanted to learn. What I wanted to learn hadn't really been invented; it was just starting to evolve. I figured that colleges were always going to be at least three or four years behind. I wasn't willing to step back that far. Also, I had a really good classical education from high school, a lot of humanities and stuff, and I felt that I had a strong enough foundation to move forward—good writing skills, you know, basic stuff.
WWWiz: Where did you start?
Sky: I went and got a job, traditional moving-up-in-the-world kind of thing, in the stat room of an advertising agency, and after about a month-and-a-half, they brought in the Macs—they'd had no computers before—they brought in the Macs, they put them in a room, and they wondered what the hell they were going to do with them. They'd heard that you couldn't possibly replace all this traditional junk with computers, so I took the office manager out to lunch. I said, "I know all about those things; I know how to run them. I could convert this company; I can get this thing set up." She bought it; I was transferred that day, and that night and for the next month I was up all night reading Quark 1.0 manuals and figuring out what the heck the thing did.
Within about three months, I'd established the computer graphics department of this advertising agency in Burbank, and hired people and started managing this thing. It was kind of cool; it was like a separate business, almost. I loved the technology; ultimately I was using the computer to facilitate communication. I was there for about a year, then I got offered a job at a much bigger company that had the CBS Television account, went there, and did the exact same thing. I converted their traditional media over to digital media. I was 17 years old and I was kind of despised by a lot of people. [laughs] But it was fun, technologically, to do that and make some kind of headway toward this conversion.
WWWiz: Where did you go from there?
Sky: In 1990 I got really fascinated with, well, disappointed with, the lack of what I consider culture in Los Angeles. We were here in this great melting pot of racial and cultural diversity, but it's not like New York or Boston, where you have history or centralization, people getting together to communicate. Everything was dispersed and so spread out here, and there was really nowhere to go where you could hang out with people and talk without getting intoxicated. Really the only place to do it was at a bar. Otherwise, it just didn't happen. So I decided to open a coffeehouse, and I got together with a friend of mine; we got some investments from family and friends and opened up a coffeehouse over on Melrose Avenue.
To me it was communication on an analog level, but it was fun, business-wise, to play with it. We had discussion groups where people would come and just talk about stuff. It just didn't happen anywhere else. Within about eight or nine months we had a cafe that was really happening; we had MTV coming in doing "House of Style" shoots in there, we had Vogue and Details writing us up, we had writers coming in with their laptops during the day, screenwriters would come in, Julia Roberts was a customer. Anyway, it was really neat; we had floor-to-ceiling bookshelves, church pews for chairs, couches and a pool table. It was a lot of fun, both business-wise and culturally.
WWWiz: But you're not doing that any more...
Sky: It didn't hold me for very long, so I turned it over to my partner, who ran it. This time I wanted to really control my destiny, so I started a company to do digital imagery—computer graphics—in Studio City. We were pretty much an overnight success because of the connections I had in the entertainment industry; most of the stuff I was doing was entertainment industry-related. We were creating computer graphic advertising for Disney and Fox, divisions of Disney; it was a lot of fun. Once again, I got on the cutting edge to really push the media to its limit.
Around '93, I had known about the Internet and what you could do, but what I considered to be missing was a standard to develop applications on, a standard to use to actually link computers together. In '93 it really hit me that TCP/IP was a standard. It had been around for a long time, but I hadn't really seen the evolution it went through, and I realized, when I really started looking that the Internet was the perfect combination of all the things I had been looking for: culture, communication, technology. It was, I believe, the next major evolution in computing. We had taken these computers and put them on people's desks; that was a huge revolution in itself. We'd given them information, but information that's static is an nth of a time less powerful than information that is moving. So we were going to take these computers all over the world and globally connect them to each other, and the power of your computer would be multiplied by every other computer on the network. It hit me like a brick.
I decided the first thing I was going to do was to write a software application that would allow people to get on the Net quickly. I realized that the stupid, simple issue was that this power was there, but the on-ramp to get onto it—and I'm not using that as an analogy to any superhighway crap—the runway was too long, and people would fail. They would try to configure their computers, they would try to get an account with one of those companies that was really a guy working out of his garage, and there's no software, and you had to download it from a PPS—it was just hell. It was like, "Here's this magic kingdom, and nobody's charted a path to it," so I figured I could do that through software.
So I developed a software application for Macintosh computers I called Internet Navigator, and I wrote the design spec; I did about 90% of the work developing this thing, and defined the product. It had SLIP and PPP connections, email, news, ftp, Gopher, Web, an Archie search, IRC, a glossary, a built-in Help library, and it was seamless. It was like an OS for the Net. This was what I wanted to create. I sort of got to the point where I [developed] the prototype, EarthLink Software—that's my Web browser. Things like shortcuts and searching, and taking away all the stuff like—you know, like to this day, search engines tend to confuse people.
WWWiz: Are you a programmer?
Sky: Not by traditional definition. I mean, programming is pretty simple. I'm not going to fool anybody and say that I would go in myself and write all the C code to do this. I wouldn't be able to do that. My talent is really understanding the user, the user interface, what they're trying to do, and this is 90% of writing a software application, defining the specifications for the way all the various parts interact and so on. My strategy was to do this thing and then go and write a business proposal and start a company, get funding, produce this, market it full-out.
It was around late '93 that I recognized that there was still a missing piece, that my application would not work the way I wanted it to, because prior to having software you have to have a connection, and there were no companies at the time offering any type of access that was user-friendly. The guys at the time were UNIX weenies 100%. If you didn't know UNIX, forget it, 'cause you had to know UNIX to get a SLIP account. I knew that UNIX was a dead end; I mean, it's a great operating system, but that's for the people who deliver the service to worry about. We need something that is a direct connection to the network. Your computer can run client server applications and really ultimately use TCP/IP. Are you familiar with the OSI model? Open Systems Interconnect—the seven layers? Well, if you look at it from the perspective of layers, telephone companies have never lived above the bottom two layers: the physical layer and the datalink layer. Datalink layer: dial tone. There are five more layers on top of that...ultimately on top you have the communication layer. It was going to take a long time for that evolution to occur. Also, similarly, the Internet providers at the time really dwelled in the bottom four layers. They didn't usually get above TCP/IP. It's a far cry from what we're doing today.
So, I recognized that nobody was stepping up to the plate to handle that, and also if I were to go and attach step one, that either I could produce the software once I had step one handled, or I could watch on the horizon to see who was going to produce the next standard software, and align myself with them. So I basically went and attacked step number one. That was to create a company to really concentrate in one week and just sort of build the model and then blow out all over the world, and I named the company EarthLink Network, and I got funding from private investors, from high-tech and venture capital people who were looking for an Internet.
WWWiz: How did you find these people?
Sky: Just through contacts I'd made. I was lucky, in terms of finding really good, smart people.
WWWiz: So did they help you actually develop the company or just fund you?
Sky: Mostly they funded me. One of them has a lot of business experience and started huge companies in the past, companies that do close to a billion a year. So I got a lot of guidance and advice and that's been very helpful.
We got started in early '94. I built a network, I started lean—myself and one employee—and basically rolled out access—July 1, 1994 was our first account.
I said I took care of step number one; now, step number two still had to exist. We had to have some software for the guy to get on. Now that was my whole play, right? That I was going to make it easy for people. So I did; I made it easier than anyone else was. I made a deal with Macmillan; a friend of mine had written a book, published it, and it was the Internet surrogate. But it was the best that we could find—it's still a damned good product—and it had a disk in the back. We wrote some quick start guides, some basic material so you didn't have to read this thing to get it to work. You put the disk in, you went through the configuration procedure, you were online. That was the beginning. Once I had the company up and running, customer support working and so on, I went to step two, and I'd seen on the horizon where things were moving. I saw Netscape, I saw Eudora, which had been around for a long time, but that was the standard. I said I'm going to bring these things together, I'm going to write the code I need to write sort of a shell around it to do all the configuration, get the accounts set up automatically and get the person online and into Netscape as fast as possible. So, that was where Total Access came about; that was released in May 1995, and it's been a rocket ride ever since.
WWWiz: How did you arrange to combine all those products on your disk? Was that a problem?
Sky: Well, it was just a matter of negotiating deals, distribution and so on. I did that very early; we were one of the first deals that Netscape made on this level, and we were able to get a very favorable position in terms of creating this package. But my target has been just to develop something that gets somebody on the Net as fast as possible, to the point where it is a useful application to them. I mean, it's one thing to have them connected and have nothing to do. My mission doesn't stop once they're connected; they have to get on and it has to become a useful and enjoyable part of their life. That's the ultimate product of the company.
So after we produced the one for Windows 3.1, we produced one for Windows 95, and we produced—and we have the only one—for Macintosh. We're in a really interesting position right now, in terms of the power of the software and what it does. We have a good R&D team here; they have fun jobs. Now they're writing games in Java for our subscribers. An interesting future lies ahead there.
WWWiz: So are you making money now?
Sky: We're making a lot of money. The business is not in a profitable position, but we're not really interested in that right now. Right now we're interested in growth and building what we believe will be an awesome future. We're standing on the edge of a huge, wild frontier, and it's yet to be conquered, and we're working our way through. We haven't even gotten to the Mississippi yet, you know what I mean? Crossing from the East.
WWWiz: How many disks have you shipped so far?
Sky: It's in the millions.
WWWiz: And how many people so you have signed up, or is that a state secret?
Sky: State secret. We're growing right now at about 10% a week. It's crazy.
WWWiz: How about your size in the scheme of things, compared to NETCOM and the others?
Sky: We're pretty darned big. DataQuest wanted to rank us, like, number three. I don't know if that's true. We're probably in the top five national companies. We have about 170 on staff.
WWWiz: You had only 145 at the time of COMDEX.
Sky: I know. We've hired about 25. We're bursting. I mean, next month we'll be in a building in Pasadena that's ten times the size of this one.
WWWiz: What do you see for the company five years from now?
Sky: I look at it in terms of the big picture. What has occurred in the last two years, or in the last year- and-a-half, is that the global communications infrastructure that's in place, and that includes telephone, broadcast media, cable, and coastal mail—the people that run those, the people at the very top who are sort of guiding these things are moving, all realized, at pretty much the same time, that the Internet was the future of global communications. Suddenly we had a protocol that allowed us to take existing infrastructure, like existing fiber, existing broadband, all that stuff, and retool it to serve millions of people on a two-way communications level.
So that one fiber strand, running from here to New York, that AT&T paid a hundred million dollars to lay, can now handle let's say 10,000 calls at once. All those calls are all circuits, they're all physical pipes that have to exist at the time of the call. If I call my grandma in New York, I tie up one of those circuits. I own that circuit for the length of the call. There's a limited capacity—there's a huge expense to deploy it, and that cost ultimately gets laid off on the consumer. What we're going to do with the Internet is move from a circuit switch network to a packet switch network, which means that we no longer will use a circuit from A to B, we'll now use packets from A to B, all in one big pipe.
It's kind of like we built this highway from here to New York, we put four lanes on it, and we allow only one car in one lane at one time. Imagine the waste. It's just ridiculous. We're having a complete shift in network technology that's allowing us to put multiple cars in multiple lanes at the same time. In fact, there are no lanes—just a bunch of cars randomly going to their separate destinations, and going different routes to get to the same destination. The efficiency that you get with this is unbelievable. So they kind of realized what was happening. Now, there's nothing you can do to stop it because the TCP/IP platform is public domain, and any company is geared to develop applications for it, and the press caught onto it, and the right things happened at the right time, and in the next five years I think your phone will be connected to your computer, your fax will be connected to your computer, your TV may be your computer or may be connected to your computer.
Your computer itself will be much smaller, much cheaper, much more of a consumer-oriented product. It'll be 500 bucks instead of 2,000 bucks. The software you use will be on the Net. Microsoft is going to have a hell of a time catching up to what's going on here. This is the final step that's needed to make the computer a real mass-consumer product, just the television is. We want to be there, developing applications, providing the network services, helping people get on the Net however we can, as long as we need to. That's really where our focus remains. If that means we have tens of millions, hundreds of millions of subscribers, who knows? I want to position the company to develop applications and provide services that facilitate those applications.
WWWiz: Do you see yourself as a cross between a telephone company and a television station, and all that in a few years, where you're providing the full range of entertainment and communication?
Sky: I believe in the Internet. The Internet's doing something to your standard company model. It's going to create more companies, smaller. Let me explain how that works. I go out on the Web, and I go to a Web site and it's Disney's home page—a 500-billion-dollar company, trillions of dollars just rolling out from all over the place, they control the world. They've got a Web site. I click next, and I see Joe's Fishing Emporium. A guy sitting up in his house in Vermont, he's created a Web page. Although the content may be richer and more surveyed and so on, more colors on the Disney site, their medium that they're using to deliver is exactly the same as the medium that Joe has up in Vermont. So ultimately it's sort of equalizing the ability of people to communicate. That creates a lot of chaos right away because all communication is not created equal—there's noise and bandwidth. That will play out by itself in a normal market condition. You'll have services that will arrive just to filter out noise for people.
That's an informational service they're providing—a Web page. It's static information, usually static. We're moving to the next level now, which is applicational communication. So this company here is Microsoft, and I'm looking at a big, fat Web page, all this money popping out of their ears everywhere. They're creating these huge applications, and they're all over the place, and it cost them a billion dollars to launch Windows 95, whatever, and then you've got Joe, who's an awesome C programmer and has these great ideas, like he's going to create this home accounting package that will go and open up spreadsheets of all your balances, and at the same time will go out to all your banks all over the Net and collect all your balances and dynamically update and do all this stuff on the Net. He can create that application and put it up on the Net the same way Microsoft could. So he needs a much smaller organization. He needs much less money. He can leverage the network, the communications medium, to go directly from source to target audience much faster than traditionally.
Now if all these things evolve—the computers—the way believe that they will, and the bandwidth is there and everything happens, we will no longer need an operating system the way we see it now. A computer will consist of a chip, some memory—the memory may not even be on the computer, it could be on the network somewhere—a network OS, a power supply, and a monitor and some input device, you know, the keyboard mouse, etc. Instead of having a local hard drive and a local CD-ROM, you'll go to Acme Terabyte—they bought a terabyte of disk space for ten cents a meg. Whereas if you went and bought a 700-megabyte hard drive, you're going to have to pay a dollar a meg for it. They bought it for ten cents—they can sell it cheaper, they sell it to you for ten bucks a month for five gigabytes. You've got way more flexibility, they back it up every night, they have full redundancy, it's fast, and then your storage is remote.
You go to Joe's Desktop Publishing Software Emporium out on the Net, and you subscribe to a desktop publishing package three, or whatever, and it's got Quark, it's got PhotoShop, it's got Illustrator, and it costs you five bucks a month. And you say, "Okay, I'm going to run PhotoShop now." Point your Net OS at that code, the code comes from that site, you get the latest code 'cause they always just updated one copy of it, and the code is streamed over the Net. You're using a bunch of big pictures and you run out of disk space. What do you do? Do you go down to Fry's and buy another one? No. You go to a Web site, and you type in, "I need more space, and I'll take the ten-gigabyte option, add another five bucks a month onto my bill." Bam! You've got more space, instantly adding to your flexibility and what you can do. And your computer is much smaller, much easier to make, and much less expensive. No moving parts.
WWWiz: And less power...
Sky: And less power, right. Better conservation.
WWWiz: Works on an airplane for a full ten-hour flight instead of three minutes.
Sky: That's right. In an airplane, you plug into the Net. Look at the ramifications of that. You didn't have to buy $2,000 in software; the software is on the Net. It's impossible to pirate. You realize that? It will be impossible, because even if you ran the software locally, it would still require you to be connected and it would go out and check your license every time you opened it. The cost of developing and making money from software will go way down. You realize that? Not only that, they'll be able to offer it on a subscription basis. You won't have to lay out 2,000 bucks for those three programs. You can just pay as you go. Fifteen bucks a month, whatever. The company that produced it will consist of the developers, maybe a marketing guy, a CEO, some finance people. No huge marketing expenses; advertising will occur on the Net. The promotion, the feedback that they get—I mean, it goes on and on. One day they can say, "Hey—let's do an add-on. You get this add-on for an extra dollar a month. All these filters." And they can offer all these packages. I mean you may not pay by the month. I'm not saying that's going to be the model, but it could be advertiser-supported, it could be transaction-supported, it could be whatever.
So EarthLink really wants to be an integral part of all of that, wherever we position ourselves. We really want to be in the network and facilitating, being the people who talk to the end user. That takes a lot of talent to do that.
WWWiz: I'm definitely impressed with what you guys have done. Now, you just went national. Are you trying to expand that still? What about Canada...?
Sky: We're bringing Canada online now, and we're looking at international as the next obvious geographic step. The Pacific Rim is looking interesting.
WWWiz: So you've set up licensing agreements for the Pacific Rim?
Sky: Well, not completely. We're still evolving that strategy. We'll be there probably in half a year.
WWWiz: I know you have a lot of customers that are businesses that are putting their presence up on your system. What percent of your customers are businesses, and what percent are private connections?
Sky: Sixty percent of our subscribers say that they use the Net for business. It's really amazing. A lot of people with home offices, a lot of entrepreneurial people. It's really cool. They're able to work at home or in a small office, they're able to leverage this communications medium. It's a mix, a really good mix, but the majority of our subscribers do use it in some way for business.
WWWiz: Do you know all of the people who work for you?
Sky: I haven't met all of them yet. I actually had that as a standard thing—every time somebody was hired I would meet them right away, but we got so swamped that we got a little behind.
WWWiz: I got lost in your lobby among people applying for work.
Sky: Yeah...we have a really cool bunch of people. I was talking to a guy I met in the bathroom. He's in tech support. He has his own BBS. He's going, "This a great company. I've got this BBS. This is what I've been wanting to do all my life."
WWWiz: Do you have a degree?
Sky: No. I've lectured at UCLA and CalTech recently I gave a keynote address there—but no. I've seen people with Ph.D.'s who are complete failures, and...I mean, I believe you can learn a lot, but there's no better school than here.