Q&A WITH GTE'S GEORGE CONRADES

In May, 1997, GTE Corp. made a surprisingly flashy move for an old-line telephone company: It bought BBN, which helped create the forerunner of the Internet and has remained in the vanguard of the industry. It also got BBN's chairman and CEO, George H. Conrades, an IBM veteran who had headed the Boston company since 1994.

Conrades is president of GTE Internetworking and executive vice-president of GTE Corp. Among his tasks: overseeing research and development and the integration of GTE's old Internet business with BBN's. GTE is based in Stamford, Conn. Business Week Connecticut Correspondent Susan Jackson recently spoke with Conrades. Here are excerpts of their conversation.

BW: Are data networks overtaking voice networks?

CONRADES: I think the answer is yes. Voice over data -- if you will, voice over IP is really more appropriate -- will replace circuit-switched. I think it's also a little bit like the checkless, cashless, paperless society. These other things will be around for a while as we go increasingly electronic.

The cost advantage and the opportunity to combine voice applications with data applications is so great, and the technology is not only present today, but will improve -- all of which will combine to cause this to happen at a fairly rapid clip. Over the next five years, we can certainly see the beginning of the impact on traditional voice.

BW: What has to happen before that?

CONRADES: Well, what I think has to happen is what's happening. The people are putting in high bandwidths, like ourselves, and others are upgrading their networks and beginning to announce IP telephony -- even AT&T.

BW: Why do you say even AT&T?

CONRADES: Well, their business is long-distance voice, and it's an imperative to stay up with technology. Rather than trying to protect the old ways, everybody is trying to get to it as quickly as they can.

BW: Can you give me an update on the BBS network overall?

CONRADES: When GTE acquired BBN last year, we really consisted of two major divisions, one of which was BBS Technologies. That's classic BBN, and that's where we do research and development in the Internet working arena, largely seeking contracts from ARPA, the Advance Research Projects Agency of the Defense Dept.

As you know, they funded the original manifestation of the Internet, the Arpanet, and the government has been a big pusher behind the evolution of IP networking ever since. The government is a very advanced user of IP -- way ahead of the commercial world. And so we continue to work with them because we're trying to advance the state of the art.

The other division we created to commercialize what we know about internetworking -- that became BBN Planet. And inside BBN Planet, we had the people -- the engineering and operations people -- who built and operated Planet's national network.

Today, within GTE we still have BBS Technologies as a stand-alone research and development unit -- all of this is in my area of responsibility. But it now is working even more closely with GTE Labs on new ways to improve communication voice and data.

In addition, there's the organization GTE started to go after the consumer marketplace. Planet has always been aimed solely at business and other organizations such as universities. We added GTE.NET service, and we call that whole thing GTE Internet Working Services.

And then we created a third unit, called the Global Network Infrastructure Unit. Its origins were in Planet, and we took them out of Planet and then combined that with the engineers GTE has who were working on building the network. So we're creating a global network infrastructure that runs the current Planet Network, and we'll eventually replace it with this brand-new, very high-speed, very reliable network.

The difference is that when we were acquired last year, we really had a T-3 speed national backbone [45 million bps, or 28,000 pages in 6.5 seconds]. At the end of 1997, we already were at OC-3 speed [155 million bps]. Today we're already at OC-12 speed in the national backbone [622 million bps]. That's really an upgrade of the original Planet network.

In the new network that we're lining up right now, we have several links operational at OC-48, and eventually, within a year, they'll be OC-192.

BW: Can you talk about multiplexing?

CONRADES: On a fiber, by optically breaking out the color spectrum and then sending light waves in those spectrums, they're only starting with eight channels, but some people think eventually you can get 100 channels on fiber, at OC-192. This is kind of unbelievable!

BW: That's like 100 TV channels. Tell me what else the consumer can expect from this. What does the average person see and hear as a result?

CONRADES: The average person today doesn't necessarily realize the enormous physical plant that's in place, just to handle circuit-switched voice calls. All they know is that it works, and very reliably. I think that will be the case with the Internet -- in the IP world. But one thing has to happen first, for the consumer to really appreciate this, and that's to have higher speed in and out of the home. There's a battle going on for high-speed access to the home -- the last mile, as some people call it. Other people even say the last 100 feet.

That, of course, is cable and the subscriber line technologies like ADSL that we're implementing because we can use our in-place copper. Then there's wireless, including satellite access in and out of the home. And even the electric utility companies are thinking about running information over the electric lines. Isn't that unbelievable? So there's a battle going on for the last mile. The good news is that, since so many people are interested in doing it, it will happen.

Now, let's assume that we do have higher speed, in and out of the home, by one or more of those media. The consumer then can expect to be able to have much more interactive relationships with information anyplace on the globe -- and to have it in a way that is easier to acquire. In other words, it's fast, it's present, it's always on. You know how your table TV is always on. You don't have to dial up. Well, that's the same way your access to the information on the Internet and to others will be -- always on.

That enables applications to be combined. Today if you're doing Internet access, we can call that a traditional Internet data application. But if you also want to speak over that same route, and connect and interact with somebody while you're manipulating data, then that's a combined application.

You know, that's kind of cool. You could pull up on your screen, and you know how you have your E-mail today? Well, you could click on one line, let's say, and there's your traditional E-mail. Click on another line, and there's a fax.

And there'll be people who'll say, I'll tell you what else I can do: I can also take that voice mail, and I'll just translate it into text and send it to somebody. You can come in a variety of ways into that system. If you're in the car, have your E-mail read to you. There are companies that provide software that do this today. So that's what I mean by large applications of the combination of voice and data.

There was a very interesting article written by Jerry Mulkowski that said maybe it isn't so much the contest for the consumer -- maybe it's helping the consumer interact in their daily lives with others. He said what about an application such as the "virtual refrigerator door"? I readily took to this example, because we've got Post-Its all over it. We have pictures of things, and we have reminders, and there's a calendar there, there are posted notes reminding our daughter to pick up something after school or that we'll pick her up at a certain time. There are dates to remember -- Saturday night is dinner with the so-and-sos.

And he says you can have virtual refrigerator door running all the time, where the family uses various accesses to the infrastructure -- PCs and soon, film-like devices that have screens where you can also read and see and hear all this, including conversations with your child and/or the teacher.

BW: When are we going to be seeing these kinds of whiz-bang services?

CONRADES: I think they're all beginning right now -- a lot of trials under way today. We're trialing Internet telephony -- you know, IP fax and voice over IP. And everybody's kind of getting down, working on it, trying to figure out how to make it useful. The technology is there and will improve.

The way the applications develop will distinguish the players. It won't be enough just to say, I can do IP voice or telephony over IP. There are a lot of words for this. People call it voice over data, or voice over IP, or IP telephony. These are all the same thing. It means being able to do both voice and traditional data applications over the same infrastructure, because it all gets digitized and then put in the IP format.

BW: What about the people's ability to apply the technology?

CONRADES: [The goal is] for the consumer to think it's really neat, it works, it's easy, and they don't even think about it. We pick up the TV clicker and turn on that television. All of us know how to manipulate a TV clicker, but nobody thinks about it. So for the applications to be useful, they are going to have to be extremely easy and almost invisible. I think that's hard to do, at first.

I'm sure you're a PC user, and I'm sure you've been on the Internet. But it takes a certain amount of understanding and training. I think there's a lot of work to be done yet to make these new applications easy to use.

BW: How long will that take, and what does that require? Who's going to have an edge in that?

CONRADES: Well, I think people like us will, because we have a strong base of customers to trial with, and I think we have outstanding engineering. I think we have some of the brightest Internet engineers of any company in the world.

BW: AT&T has a much stronger base of customers than you do.

CONRADES: They do. But.... they resell our Internet connection service, while they build their own. Meanwhile, we're off building a superseding network, the likes of which the world hasn't seen.

BW: When everyone else is working on this, how do you know yours is going to supersede?

CONRADES: Well, I think Qwest, we, and a couple of others have the advantage of going in with all the latest. We're not upgrading anything -- we're putting in brand-new from the get-go.

Technology is changing so fast. This fiber is going into the ground deeper, it's going in with more capacity. We're putting in our own optoelectronics, we

put in our own routers, and manage it. And we design it so that it's self-healing and so forth -- all of that is pure GTE. And I think we have the advantage in that network architective, to use the latest fiber strands, to use the latest optoelectronics, to use the latest routers.

We're installing today, the latest and fastest routers that Cisco can produce. And we get to go in with all new stuff and use all of our years and years of IP experience to make it right, to make it the best.

BW: Tell me a little bit about what else is going on in R&D. We have talked to you at some length about speech recognition, but what else can we expect?

CONRADES: I think speech recognition remains a key area. I think work in distributed, collaborative planning.

BW: What does that mean?

CONRADES: Well, envision a screen that has five or six windows open with people in them that you can see and hear. Let's say they're scattered around the world, and in the center of the screen is a distributed software object. It could be a spreadsheet, it could be a map, but it's software. In other words, it's not just a visual of it, you can manipulate it. And they're collaborating over this document. Maybe it's people doing financial planning over a spreadsheet, maybe it's people doing a construction project over engineering drawings and changing them -- that type of thing.

Another area is in wireless and especially broad-band wireless. We have a particular competence in broad-band wireless that's both radio and satellite. And I think also some technology in distance learning. I think distance learning is kind of a natural application for the Internet.

We have contracts with the military mobile networks. The personnel can have hand-held communication devices that allow mobile video conferencing. They can see and talk to one another, on a hand-held device that has a screen on it, while they're moving and the network is moving, going across a desert or something.

As devices become smaller and less expensive, we'll have multiple devices on our persons, and we'll be walking around and be able to enable things like, let's say, an automobile wreck and instant telemetry right into the hospital on how the person was doing before the wreck, during the wreck, after the wreck -- the doctors can see all kinds of information that will help determine, as they prepare for a person coming to the hospital, the likely condition of the individual.

BW: So who would be there taping this person?

CONRADES: No, you'd be transmitting it all the time. In other words, it's possible that in the future our bodies can have multiple IP addresses with a sensor, you know, to measure blood, your pulse rate, a sensor to measure your blood oxygen, a sensor to measure anxiety. That kind of thing. I mean, not everybody would wear them, but it may happen. And these could be constantly in touch, always on. Wireless, with the network. And if anything ever happened to you...

I know, everybody would say, "My God, isn't that intrusive. They'd always know where you are and how you're feeling." But my point is that you can see some of the medical advantages of that.

Then there's working with the satellites for broadband -- kind of like Internet in the sky.

BW: What is that?

CONRADES: That's just high-speed Internet working among satellites around the globe -- to complement high-speed fiber that's in the ground and to actually substitute for it in parts of the world where there's no fiber in the ground. There's a lot of fiber in the U.S., and in other countries, not so much. This will enable them to get high-speed Internet working globally through these satellites.

BW: So what are you guys doing on this?

CONRADES: We work on, really, kind of the algorithms that make these kinds of systems work. Once again, it's our Internet engineering that helps make these kinds of systems work. We have quite a history in satellite communications.

I'm sorry, I have to run. Thank you so much.

BW: Thank you.