Internet media empires
Do you ever worry that huge media conglomerations will erase some or all of the equalizing influence of the Internet?
The preexisting power and economic structures will be influenced by this new world, the Internet, and [vice versa]. I think that's going to be a net positive, but I don't think it completely obliterates previous competitive advantages that were held by large media entities. But I do think that the Internet will look far more like the magazine rack at your local magazine shop than it will look like your cable TV dial. I do think that inherent in the openness of the infrastructure is the fact that there is plenty of shelf space for low-volume magazines, all the way up to specialist publications, all the way up to Time and Newsweek and Sports Illustrated. And will it be hard for start-ups to stay small and be Time and Newsweek? Absolutely. Not everybody has to be the People magazine. Not everybody has to be CNN. There are plenty of spaces in the food chain for Whole Earth Review or the Internet equivalent for a much wider spectrum of information or perspective.
In contrast, the big cable operators choose [to carry] only channels that they own a percentage of; current cable systems also have limited channel capacity. The Internet doesn't have that gatekeeper element or the channel limitation of cable. There will be quasi-gatekeepers in the sense that whoever controls the default home page that you go to has an influence. For instance, we saw that where Netscape decided that they wanted to charge [when] before they were giving it away. They were hosting Yahoo, and then nine months later they were charging Yahoo and four others $5 million a pop each to be on their search page. So there is a gatekeeping function, but it's a much softer gatekeeping function [than with cable].
I think that in the real world power does accumulate. In the car business, there are 15 major car companies on a global basis. That's way better than if there were one; it's not as good as if there were 500. But 15 is a lot better than the number of cable companies you can choose from, which is one, or telephone companies, which is one.
Not everybody can get their magazine at the newsstand; not every magazine that's popular sells to newsstands, but typically good newsstands in big urban areas have hundreds of magazines that you can choose from, and it feels like a much more diverse range of information, opinion, and interest types than what you get in other media.
So that's hopefully not a Pollyannaish perspective, but it is overall a more positive than negative perspective on what's going to happen.
Has your former employer, Microsoft, been good at allowing power to distribute throughout the broad spectrum of start-up companies?
I don't think the current nature of the Internet ecology is based on a conscious choice by any one company, including Netscape or Microsoft. It's been far too organic. I do think that Microsoft understands its own hierarchy incredibly well, and in their case (and I've talked to Bill Gates about this), owning the APIs (application programming interfaces) that are the popular APIs used by developers, and having the mind share with developers is themost important thing. So Microsoft's operating system franchise is fundamentally built on that, and many of Microsoft's businesses are built on or leveraged off of that position.
I think that Microsoft is using every weapon in its arsenal and is busy creating new ones to strengthen its position in the hearts and minds of developers. The particular PC real estate they're focusing on, above all others, is Web browser market share. I think it would be great for the industry if five years from now there were three to five players in the Web browser market, with each having between 15 and 30 percent share. I think as a practical reality that would keep [the Web] vibrant and open and prevent that particular influence point from being a choke point controlled by one company.
Do you think on the Internet we'll see any FCC regulation of things like RealAudio Netcasts?
I got a chance to meet FCC Commissioner Reed Hundt about a year ago to show him RealAudio and explain how it works. His attitude was that this is clearly not something that the Federal Communications Commission has appropriate jurisdiction. We're using the common carriage structure, there's no bandwidth limitation, there's no spectrum allocation, so from that standpoint, the FCC has no role.
Now the government, with this really terrible piece of legislation called the Communications Decency Act, has tried to regulate types of content. Fortunately, we have judges who read the Constitution, and now in two different cases [they] have thrown it out.
I do think there is a specter of content legislation. Those of us in the industry who care about keeping it an open, democratic place have to be very proactive to [implement] systems like the PICS filtering standard, which we were very involved in establishing into practice so that educators and parents can have the kind of ability to steer their kids away from content that they don't want them to see. If we all do that, hopefully the public policy folks will understand that the Net needs to be this open, vibrant place. That regulation, number one, inherently would stifle and censor, and number two, given the global reach of the Internet, it wouldn't work.
Are we overburdening the Internet's infrastructure the way that some doomsayers like Bob Metcalfe would say?
I talked to Bob about this last week. It's good to have a little bit of a Cassandra out there, because otherwise I don't think we would look at the aggregate impact of things. Still, I'd say no.
One of the features of our server is that it communicates with the client. So the client sends back to the originating server's quality-of-service information. We monitor that on our own site. We ask customers about it. And the quality of service has remained basically constant. The one caveat is that there will be storms from time to time that come up, where a given router that's pretty busy on the Net gets flooded or a given segment is overburdened. But the economic imperative is so great that, at this point, whenever that happens it's a business opportunity for someone else.
I can't say there won't be cases where we won't get some bottlenecks. I think the principal bottleneck is going to be the last mile. You need to collaborate with either the cable company or the phone company. Until it's trivial to compete on an end-to-end basis with those companies, which it's not today even with the regulation of telecom, we're going to have an artificial slowing down.
NEXT: The importance of being Rob