DSL has emerged as the primary broadband technology that people will use to drive the next wave of Internet applications. As the chairman of Dallas-based Efficient Networks Inc., Mark Floyd is helping to drive the adoption of DSL through a broad range of partnerships and alliances. In an interview with InfoWorld Editor in Chief Michael Vizard, Floyd makes the economic case for DSL and outlines the impact it will have on the enterprise as the technology continues to evolve.
Vizard: How widely adopted are your DSL products at this point?
Floyd: We have products in over 68 different networks, worldwide. About … half of those are actually in production networks, really starting to deploy DSL, and the other half are in some form of testing. Our largest customers in the United States are SBC Communications and Covad, so we have a major ILEC [incumbent local exchange carrier] and a major data CLEC as the top two customers here. Internationally, we’re allied with British Telecom and Telefonica Deutsche Telecom as well as all the major PTTs [postal, telegraph, and telephone companies] in Europe and Southeast Asia.
Vizard: What is the one big hurdle that needs to [be] overcome to make DSL proliferate?
Floyd: I think one of the biggest hurdles we’ve [crossed] is the self-installs, because these guys can’t roll trucks to every installation, especially the residential side. The numbers are so large. So self-installs is [the] way it’s going to go.
Vizard: What’s your take on the battle between DSL and cable?
Floyd: I think both technologies will co-exist. I don’t think one eliminates the other. In the business marketplace, DSL will be very dominant in that space because most cable networks don’t go around business parks or around the business areas. If you moved to residential broadband, that’s where the user has an alternative between telephony services, DSL for broadband, or cable. If you think about where DSL is, we have about 800 million copper loops worldwide. That’s our universe for targeting DSL. If you look at cable, they don’t have near the infrastructure globally.
Vizard: How big a business issue is this?
Floyd: I think it is relevant because if you look at an ILEC or even an AOL or some of the larger ISPs, if they lose a customer to a cable network, the cable network is not as open as a telephony network, and so they’re going to lose business as well. Once you get on a cable network, you have different ISPs and it’s pretty much a closed network.
Vizard: A lot of sites are limited in terms of the number of DSL lines that can be run through a neighborhood. Will this always be the case?
Floyd: As we go forward in the technology, that really should not be that big of an issue because the chipsets are getting [to] where they’ll carry more lines per chipset and more lines per line card. The physical space is becoming less of an issue as we go along, where historically you can only plug so many cards in. And I’ll give you an example. Historically, you only had two DSL lines per line card in a DLC, now we’re at eight. Pretty soon we’re going to be at 16, and then pretty soon after that – we’ll keep multiplying. So that issue will go away.
Vizard: How will DSL play out in the enterprise?
Floyd: From an efficiency point of view, we’ve sold to ILECs, we’ve sold to long-distance companies, we’ve sold to competitive carriers and to ISPs. Historically, we have not sold to the enterprise, meaning the Fortune 1000 companies. But we just did our first major transaction coupled with Covad and an ISP in Michigan called BigNet, where we have put a deal together for Ford so they can use DSL for their employees. I think they were going to try to get up to 178,000 of their employees to be telecommuting via DSL. As we go forward, I think you’re going to find enterprises will use leased line services for their mission-critical applications, and they’ll use DSL for their telecommuter branch offices. And then as DSL technology matures and we can offer service level agreements over DSL loops, we’ll start taking business away from some of the leased lines for mission-critical applications as well. Where we’re headed with DSL as an industry is into the enterprise.
Vizard: What role will virtual private networks (VPNs) play in driving DSL?
Floyd: Corporations won’t open their networks up to any kind of telecommuting if it’s not secure. So we’ll bundle products with routers and bridges with VPN and security applications.
Vizard: DSL is widely perceived as a data technology. Will it be applied to voice applications as well?
Floyd: There [are] really two major benefits for voice over DSL, and both of them are very economical. The cool thing about voice over DSL is that long-distance companies and alternative carriers or voice CLECs can come in and offer a broad range of services to small-and medium-[sized] businesses and take a bunch of traffic over one POTS [plain old telephone service] line that they can lease from the CLECs. If you’re a small business, and you have … seven or eight POTS lines and maybe one ISDN or a fractional T1 for data, you’re paying somewhere between US$25 and US$30 a month for each one of those POTS lines and somewhere between US$90 [and] US$200 a month for your data line. What we can [do is] take all that traffic over one twisted-pair line that you’d only have to pay US$20 to US$30 a month for, and just once.
Vizard: How does that help the telecommunications companies?
Floyd: If you look at it from the service provider [viewpoint], you could come in and offer wireless long distance and local data all in one package to the businesses. Then what happens is that if you leverage up a CLEC to get that one copper loop, you don’t have to pay that access charge – which is almost equivalent to 50 per cent of the rate – to the ILEC. The alternative carrier saves a lot of money by not having to pay access charges as much. The user saves a lot of money because he doesn’t have a lot of monthly reoccurring cost on multiple lines, plus his rates will probably be lower, and he gets a bundled service with everything on one bill. So it’s pretty competitive from the competitive carrier. And from the ILEC’s viewpoint, if you have an area [in which] you have copper exhaustion, such as Southeast Asia or Europe and some places in the U.S., and you have a fully deployed central office [CO] so [that] you can’t pull any more lines off that CO, you can use voice over DSL to offer multiple lines to customers over one physical copper pair.
Vizard: So what’s the next big challenge?
Floyd: We’re going to have to allow our service providers, like an SBC or a Covad or an IXC, [to] offer the same services to their customers over a DSL loop as they do in a leased-line loop. I think that’s one of the biggest next challenges we have. I also think you’re going to see this branch out into some other new technologies. I think you’re going to see wireless come into play, along with other advanced services. We’ve got to keep our eye on the ball of the market and stay ahead of it.