September 30, 2004 10:45 AM PDT
Newsmaker: The technologist who has Michael Powell's earSee all Newsmakers
How strictly the FCC decides to regulate emerging technology promises to have a lasting impact on areas as disparate as voice over Internet Protocol (VoIP), fiber to the home, instant messaging and even digital video recorders.
Robert Pepper is the FCC's chief of policy development, which requires him to be a kind of government futurist, advising Chairman Michael Powell on which regulations are wise and which would be harmful. He's also co-chairman of the FCC's Internet Policy Working Group. Previously, Pepper directed the Annenberg Washington Program in Communications Policy Studies.
CNET News.com spoke with Pepper about topics, including VoIP, broadband over power lines, wiretapping Internet phone calls and what would happen if John Kerry is elected.
Q. Tell me what you do at the FCC.
A. I'm chief of policy development for the commission, which is analogous to the chief economist. But what I do is focus on issues that are essentially the next set of issues. It's a broad policy perspective looking at the intersection of technology and policy (and) new technologies.
Does that make you a kind of chief futurist for the FCC?
I don't know if I'm the chief futurist, but I try to meld technology and policy and the financial markets, and look a little bit around the corner (at) what's next. I spend a pretty good amount of my time trying to understand what the financial community thinks we're up to, because we want to understand the financial community's perceptions. Right or wrong, their perceptions determine when they open a checkbook and invest.
Last month, the FCC ruled on wiretapping requirements for Internet telephone calls, indicating that VoIP services that did not touch the public switched telephone network would be immune from the Communications Assistance for Law Enforcement Act (CALEA). Is that your interpretation, too?
That is my interpretation; I agree. We had received a petition from the Department of Justice and the FBI that we now put up for comment, and we have a proceeding looking at the CALEA issues more broadly. What we said was, "Until we complete that proceeding, the voice over IP services that do not touch the (public phone network)--other people refer to them as the peer-to-peer services; Free World Dialup and so on--are not subject to CALEA. That might change in some respect, but we have not made a determination.
Didn't the commissioners make statements indicating that their mind has been made up?
It is complicated, because all the commissioners and I think most people agree that in today's world, there is a legitimate role for law enforcement to catch the bad guys...What is complicated when you move into the voice over IP world is that the ability to achieve that goal is more difficult and complicated as a technical matter. And therefore, the question is how you implement the goal without overwhelming the system, or how you do it in a way that is not counterproductive--and that is hard, technically.
You didn't give the Justice Department and the FBI everything that they had proposed. They had wanted at least CALEA to cover peer-to-peer VoIP networks. What made the FCC stop where it did?
We didn't think that actually applied under the statutory definition.
Can you see this opinion changing? Or would it take an act of Congress to alter CALEA for it to extend to peer-to-peer networks?
I am very reluctant to speak for my five bosses, the commissioners. (But given the way the law is written), I think it would be difficult to conclude that peer-to-peer VoIP services are subject to CALEA. If they are, then you would also, I believe, be bringing in things like Xbox Live and Yahoo's instant messaging with voice. There are lots of peer-to-peer communications like services, IM and games that in fact today are not subject to CALEA. Personally, I don't think it makes sense to apply CALEA.
I don't want to prejudge what the commissioners will conclude, but I think that the burden is on those who want to make all of those peer-to-peer services or peer-to-peer activities come under CALEA. By the way, just because these things are not subject to CALEA, it doesn't mean that law enforcement is not capable of getting access under court order to information that they need.
Police have been able to conduct Internet wiretaps for at least a decade. So if the FBI wants to eavesdrop on someone, why don't they just serve a wiretap order on that person's Internet provider instead of lobbying for CALEA regulations?
Where they do it depends upon the particulars of the service and configuration and architecture. But the intent of CALEA, back when Congress passed it, was to essentially enlist the traditional carriers to make it technically easier to get the information that law enforcement needs when it has the court order to do so.
Back in those days, there was far less of the kind of electronic communications than we have today. Going to an ISP and getting an e-mail record is very different than being able to listen in on a conversation where (at) the end of the conversation, there is no record...In the early days, yes, it could be done, but it was difficult. Today, it's even more difficult.
Isn't that why the FBI developed Carnivore? Could it be upgraded to handle VoIP protocols in addition to ones it can already understand, like HTTP, FTP, SMTP and so on?
That could be one of the ways in which you can achieve the goal. But this is one of the issues that people have raised--clearly, the law enforcement community believes that it is not sufficient to be able to do what they need to do for very legitimate law enforcement purposes.
Let's talk about the broadcast flag for a moment, and the FCC's recent ruling saying TiVo could offer its forthcoming service. Do you foresee a world where the FCC becomes the arbiter of what technologies may or may not exist?
Only as it applies to the broadcast world.
I recall that in the FCC's broadcast flag ruling last fall, the commission didn't answer whether software-defined radio should be covered. Should it?
If you fundamentally believe in copyright protection, and I do, then whether the device is bent iron and wires or hardware combined with software really doesn't make any difference. So, in the future, as more devices incorporate software, you are going to see that as part of the consideration. TiVo and a lot of these other devices that we approved already incorporate a lot of software as part of the device.
Might the FCC ever be in the position of saying a software program that is mirrored on thousands of sites around the world is no longer legal to distribute in the United States because it might violate our broadcast flag rules?
I haven't thought about that, but I think it's hypothetical. I don't see us getting into that. But at the moment, that's not the case.
Let's talk about VoIP again for a moment. There's a lot of action in Congress--what do you think is going to happen?
I leave that to congressional watchers. I think that the important point is that voice over Internet Protocol is a fundamental change in the architecture, the service and the costs underlying traditional telephony. As a result, you have to question whether or not the traditional regulation and regulatory structure applies. So voice over IP really puts a nail in the notion that we have a distance-sensitive network. What does that mean for things like federal jurisdiction or state jurisdiction, interstate versus intrastate? What does that mean (for) the whole superstructure of taxes, fees, subsidies that are built around the old legacy network?
Do the same universal service assumptions make sense today for VoIP?
In a world where your local calling area is the United States, I don't think so. I think that we have to find new mechanisms to support the goal. The goal does not change. We want affordable phone service for everybody.
What might these new mechanisms be?
There are several that already have been proposed here at the FCC...Of the two proposals that are on the table, No. 1 is to base a contribution or a fee based on your connection. So, if you have a dial-up connection, you will pay one fee. If you have a higher-capacity connection such as a T1 or DS3 or whatever, you pay a higher fee. It's not based upon whether it's local or long-distance. It's based upon the size of the pipe.
A second proposal is based upon phone numbers. If you have a phone number, you pay a fee, and the concept there is technology-neutral. So if you are a traditional phone company or you get service from a traditional phone company, you have a phone number, you pay. If you have a voice over IP service from your cable company, you pay. If you have a wireless service, but you have a phone number, you pay.
Is broadband over power lines going to hit regulatory walls from state and local officials?
I don't know, but I have seen a number of the broadband over power line experiments, and the technology appears to work. There are obviously people who are concerned about interference. What people have to understand is that there is already radio interference that comes from existing power lines. The broadband over power line technologies that I have seen actually limit the unwanted emissions from the power line. If there is interference of particular channels, they can be filtered out. Our Office of Engineering and Technology is working to make sure that broadband over power line does not create unacceptable radio interference issues.
I've talked to some state regulators who are very bullish on this, because they see it as a way to reduce the cost of operation, begin to cap cost increases and begin to manage the need for new electric generation plans to meet spikes in demand. When we were in California in July, we took a look at the trial that AT&T and PG&E are running. Some of the PG&E people were talking about possibly doing real-time pricing down to neighborhoods if they have broadband over power line...If your dishwasher had an IP address and you could set it so that when the price of electricity drops--if they have a sale--it drops by 60 percent or 80 percent. That's when you run your dishwasher. Maybe that happens at 3:30 in the morning.
Is there anything that the FCC needs to do to help local wireless neighborhood or regional wireless Internet service providers along?
Wireless ISPs--WISPs--are some of the most exciting companies and developments that I've seen in a long time. You have a lot of little companies--we estimate somewhere between 2,000 and 2,500. They are providing broadband service in urban and also rural areas without subsidy. They are being deployed very rapidly at a low cost. They break even with relatively low penetration rates. They can operate on mountaintops. They can operate in inner cities and neighborhoods. This is very exciting.
What, if anything, can we do to help? We have had several workshops, forums, roundtables. We have a wireless broadband task force that's looking precisely at that question. They are working on a report that I expect is going to be released sometime this fall. A couple of things, however, we already know. First, we do need more spectrum available--both licensed and unlicensed--for these operators. Second, we already have a proceeding asking the question of whether there is something that we can do, in terms of our power limitations and interference rules in rural areas, that will enable these operators to go longer distances in order to reach even lower-density areas.
Should cable companies offering voice communications be subject to the same rules as traditional telephone companies?
In some ways, they are today. What Cox Communications, Time Warner Cable and CableVision are doing is certifying that when they provide telephone service, they are doing so as competitive local carriers. The question is: If we have a world where you know everybody is migrating to an IP-based voice service and there is a lot more competition, what rules, if any, should apply?
The wireless industry provides a very nice model for a light-touch regulatory regime. They are still subject to certain things: CALEA, disabilities access, Universal Service contribution and 911--the social policies. But they are not subject to the traditional economic regulation, and they are not subject to state certification, and that has worked really well.
Let's say that John Kerry is elected, and the FCC shifts to a Democratic majority with a Democratic chairman. How might that affect the commission's position on technology-related topics?
Well, that's just speculation that I don't really want to (explore). What I will say is that I have been serving essentially as policy adviser to the chairman since the end of 1989, for five chairmen in three administrations. All of the chairmen that I have worked for understand that technology is fundamentally changing the legacy world and that the old rules are no longer sustainable.
Out of the modern chairmen, Michael Powell is the most technologically sophisticated. He absolutely understands the power of technology. He is invested in technology at the FCC--we hired new engineers, we revitalized the technology side of the FCC.
If new technologies might make some of the FCC's regulatory responsibilities less vital, are there good arguments for either radically visiting the idea of the FCC or even abolishing it eventually? You're a smart guy who will always be able to find a job.
Civil servants should not be worried about jobs. We should be worried about what the right thing to do is. In an ideal world, we are going to have a very competitive market and communication services that are going to be over wires, wireless and all kinds of new technologies. If we can reach that point, then I think the need for a traditional regulatory agency is very different than what we have had in the past. So we have to ask not what the role of the FCC is, but rather what an appropriate role for government is, right?
I think that there will still need to be technical capability or spectrum management, in terms of understanding and identifying the technologies and the bands, and making sure that the rules are compatible with the new technologies. But we shouldn't be in the old command-and-control (style of) spectrum management.
1 commentJoin the conversation! Add your comment