The man who was appointed chief technology officer for Nortel Networks in June came from Broadcom, but before that he occupied a similar role at Cabletron. In his tenure there, however, Cabletron split off its enterprise business into Enterasys, where Roese became much more focused.
He joined Nortel, he says, because of its scale in six sets of business: wireless, wireline, carrier, enterprise, applications and infrastructure.
“Those are the characteristics of a next-generation communication provider,” he says. “If you don’t have all six of those, you better find a way to get ’em, because without them, you are going have a disadvantage in solving problems such as making the extended enterprise work.”
Roese spoke to Computing Canada during a recent stop in Toronto to help officially open a new Nortel headquarters.
Computing Canada: A lot of your career has been spent on the enterprise side of things. How much convergence is there between the needs of corporate businesses and the carriers that Nortel has traditionally focused on?
John Roese: Historically, not a lot. They have very different views, they operate based on different paradigms. One of the things we’re banking on at Nortel is, quite frankly, the problem spaces that we’re beginning to see emerge are actually more likely to occur at the intersection between carrier and enterprise than less likely.
If you think about the Wi-Fi enabled cell phone, it’s a very tactical, tangible thing. I can have this wireless access on a high-speed data network, it’s a local area network. Now that’s very appealing, right? The issue with it is that the real usage of that system if it can provide some level of seamless communication so that it doesn’t care what network it runs on. When it operates in a Wi-Fi environment, many people assume the Wi-Fi network it’s going to attach to a hotspot will be operated by a carrier. Now, we would argue that’s actually the place it’s least likely to end up.
CC: Why is that?
JR: If you are in a Starbucks, the odds of you having cell coverage are pretty high. Why would you jump onto a hotspot just for the purposes of making a call if you can already do it on a cellular network? The economics aren’t any better. The real advantage is if you can enter the enterprise infrastructure – if you can walk into your facility and the cell phone says, “Oh, I now know I’m in my office, let me stop using the cellular infrastructure and connect over the Wi-Fi network and let me do IP telephony and take advantage of all the enterprise services” – the four or five-digit dialing, access to your local voice mail. It sounds great in theory but the problem making that work is that the two worlds are in entirely different worlds in terms of concepts like service, quality of service and accounting.
CC: Why is that?
JR: If you think about communication security, one of the core principles is this concept of “triple A” – authentication, authorization and accounting. Authentication is who you are, authorization is what you are, and accounting is how do I charge you for what you’ve done.
Those three As, when you look at them from a carrier perspective, carriers only care about two of them. They care about authorization so that they know you are on their network and they care about accounting so that they can bill you, but they don’t really care if I hand you my cell phone and you make a phone call. That’s okay in the carrier world. If you jump into the enterprise world, they only care about two As as well, but they’re a different two. They care about authentication, because they want to make sure that you are using your computer, and they want to make sure you are using your computer the right way, but they’re rarely going to bill you on a per-bit basis. The technology, in order for it to be successful, means that we have to come to some sort of reconciliation, we have to have computer security models and transport models that are compatible with both sides.
CC: Is that happening?
JR: The maker of the cell phone is not going to be able to figure that out necessarily. It is typically going to be the infrastructure providers, the operators and the technology providers that actually build the framework and the security architecture to accommodate this, and there are very, very few out there right now that understand both enterprise and carrier from a wireless perspective.
CC: And Nortel?
JR: When we look at Nortel – I’m one of the new guys so I can say these things – sometimes we’re smart and sometimes we’re lucky. I think we may have been a bit of both, and maybe more lucky than smart, in that the company hadn’t changed structure dramatically over the last five years in terms of the markets we were pursing. We didn’t concentrate in one market. Five years ago that might have been the smart thing to do, but the results would have been different today.
The net result is Nortel today has a multi-billion dollar enterprise business, a multi-billion optical business and a multi-billion carrier wireless business. We have multiple assets and capabilities in all the domains where the problems are actually originating and coming together, and lot of our strategy is about trying to unify them and act as one company so that in these very complex problems that are in the intersections of these domains we are actually focused on innovation and technology that bridges that gap.