The sale was one of Cerent CEO Carl Russo's triumphant moments. He was responsible for the combination after he had sketched out his vision of where optical networks were going, and he saw making a Cerent-Cisco combination as the way for carving out a big chunk of the industry.

As for the dollar amount-Wall Street's been hot for anything with the word "optical" in it, said Russo, who irreverently told an online publication that even something called "Optical Bag of Doughnuts" could have a sellout IPO.

Russo says things like that. A jazz lover, he speaks in eloquent solos of logic and sarcasm, expounding on subjects such as his reasons for coming to Cisco, why it's not a handicap that Cisco doesn't own an all-optical switch, and why he thinks the company can catch up to Nortel in optical networking.

Now a group vice president, Russo's job is to define Cisco's strategy in optical networking. He blanches at the suggestion that he is responsible for Cisco's future, as he was introduced for his Opticon keynote in August. Still, his area is an important market that took Cisco by surprise and now threatens to leave the networking behemoth behind.

Optical networking is the Pokemon of communications, a financial fad that's drawn enormous stock valuations and torrents of venture-capital money-hence, the Bag of Doughnuts quip. But at the heart of it is a real need: as carriers rush to develop faster and denser networks based on fiber-optic transmissions, a host of new networking technologies has to be not only perfected and deployed, but invented.

"You get companies like Cerent getting started because there's clearly a schism between what the network's gonna have to look like and what it looks like today," Russo said.

The Cerent deal brought Russo into the Cisco fold, and his job has been expanded from running Cerent to running Cisco's entire optical-networking unit, a team composed of relatively recent Cisco acquirees. Their combined mission: define the optical networking schemes that carriers will use in the next generation of their networks. "Oh, and by the way," as Russo himself is fond of saying-do it better than Nortel can.

"We're trying to do things which I'm sure haven't been done before," Russo said. "At the end of the day, I think the investment community has sort of grown accustomed to Cisco, sort of doing things that they didn't expect us to do. Which is a bad thing. There's a lot of people that think if you're getting into optical networking you're gonna just win it, [and then] everybody [else] is going to leave. Nobody's leaving this space. So that's the bad side of it. The good side is that people do have an expectation that we'll perform. We're gonna do our best to fulfill that. If my hair doesn't fall out first."

Cisco-and the rest of the networking world-was goose-egged in the market for OC-192 (10-Gbit/second) optical networking switching. Nortel Networks made a concerted effort to move into that space at a time when other equipment providers thought it too difficult or too irrelevant.

"One of the biggest stories of the late '90s was Nortel whipping Lucent's rear end by seizing the moment when OC-192 was being implemented in green-field networks," said Pete Farmer, an analyst with Strategies Unlimited. "Nortel had the product to meet that need, and in doing so, Nortel has acquired a great deal of knowledge that it's going to be hard for anyone to replicate."

Russo agrees. "One, the management team at Nortel has done a good job. Number two, they've been doing optical networking for a long, long, long, long time. You can almost look at it and say they grew up in that space, and the space has sort of moved over on top of them. If you look at what's driving the growth of Nortel's business, it's only the optical networking business. That's the growth area of their business."

As part of its catch-up plan, Cisco has been playing the game it knows best-acquisitions. Along with Cerent, whose ONS 15454 system is selling well in metropolitan networks, Cisco bought Monterey Networks Inc., for its core optical switch; purchased a unit of Pirelli SpA for its dense wave-division multiplexing technology; and added Qeyton Systems AB, which was developing a system aimed at metropolitan area networks.

"I'm extremely comfortable with the road map. I think we have the strongest portfolio of products-and the strongest portfolio of products coming-in the metro space," Russo said. "And in the core space it's about staying focused on what you're trying to get done, which is build mesh networks, and that's where we want to compete."

"Why do we think we can compete with them?," he said. "It really comes down to taking the marketplace and splitting it up into the two networks that are being built. There's a metropolitan network that's being built, and there's a core network being built. If you want to think about it, the metropolitan network really connects buildings and other smaller networks into the optical infrastructure, and brings [all that traffic] back to a services point of presence (POP). And a core optical network interconnects all the services points of presence."

As an example of the kind of strategy Russo has to lay out, consider the network core and the "all-optical" buzzword. Boxes like the Wavelength Router from Monterey take an optical signal and convert it into electricity, which today remains the only way to interpret what's inside the data being sent. After taking its reading, the box converts the signal back to light and sends it to the next stage of the network.

Several companies have begun development on ways to avoid the opto-electronic conversions, allowing switching of the light itself in an all-optical box. And one of the acquisitions in that area was Nortel's purchase of Xros Inc., a startup designing a cross-connect that could be a key element of an all-optical switch.

Cisco doesn't have a Xros; no all-optical play has found its way into the company's acquisition net. But in Russo's vision of the optical network, that's perfectly fine, even if some others in the industry disagree with him.

"In the middle, once you get it into the optical domain, you're going to do everything you can to keep it in the optical domain. No question about it," Russo said. "But the services POPs: electronic. And in the metropolitan area, you're going to be much more packet-aware. This is not that hard-and yet, you see equipment vendors out there building stuff that's clearly being built to a very different picture than what their customers are telling them they're doing."

The reason is because data still has to be converted into electrical form in order to be read and interpreted. And certain points in the network, such as the services POP, will want to investigate packets in order to apply carrier services.

"I didn't come up with this," Russo notes. "I just walked around and talked to customers. I'm serious-I see multiple customers a week. I go to their facilities and just walk around. It is as plain as the nose on your face. You go to a services POP, and on one end of the building they have the core optical facilities.

"On the other end of the building they have the metropolitan optical facilities. And there ain't no fiber in between those two. There's a ton of racks with servers and routers and everything else. Take your time, figure it out. It's not that hard."

And that brings us to the game at hand: Russo and Cisco, trying to implement the plan he's set for metropolitan and core networks.

The vision concerning how those networks are being built is clear. Russo insists it's now a matter of simply being able to build them.

"It's golf: You're not playing against the other guy, you're playing against the course. The other guy may be very good, and you need to be aware of it, but this is not like tennis where you're hitting balls back and forth at each other," he said. "We just have to compete on merit. Against some pretty good competitors. With some pretty good markets, with a lot of revenue, a lot of market cap. It's going to be a pretty interesting challenge.

"Absent that challenge, I wouldn't be here. How can you miss that kind of a competition? If you like to play-I can't think of a better space to be in.

"I mean, if you want to compete in this industry, do you know a place where there's bigger dollars at stake?"