Published In October 2001
In a period marked by turmoil in high-tech businesses, voice, data, and video (VDV) cabling technology continues to progress toward better standards and higher bandwidth for commercial applications. Meanwhile, the home and multi-tenant unit (MTU) appears to have become an area of potential growth for VDV cabling. The meltdown of the dot-coms caused networking equipment suppliers, consultants, distributors, and contractors to lose many customers, because most of these startups went belly-up and many others lost a lot of money. While these startups may not have had any real customers themselves, they were very real customers for telephone and computer networks, including all the cabling necessary to support them. The subsequent troubles with the telecommunications companies and their suppliers were not at all like the dot-com debacle, as the problem was not startups vaporizing for lack of business but major old-line companies getting caught in a volatile market. Only a few percent of the hundreds of billions of dollars of fiber optic cables installed during this “boom period” are even in use. While one can only lament the detrimental effect on the economy as a whole, there may be very positive long-term effects, especially in the accelerated development of new technologies. Fiber optics The telecommunications boom was, of course, all about fiber optics. Just like the “land rushes” of the old west, dozens of companies, from well-established carriers to venture-funded startups, tried to install all the fiber they could as fast as possible to create nationwide (and worldwide) networks to supply enough bandwidth to meet the projected needs of information-starved companies and consumers. Many dot-coms blamed the lack of a broadband backbone for their inability to provide services that would gather enough customers and revenue for success. Unfortunately, the fiber installed appears to be 30 to 50 times more than was needed for long-distance backbone communications. While it obviously provides room for future growth, long-distance cabling represents only about 10 percent of the total network. Another 10 percent is used for local (i.e., city or county) cabling, while the final connection to the home represents about 80 percent of all cabling. At the same time, the local communications networks and the connection to the home was fairly stagnant. Only a few percent of consumers now have high-speed Internet connections. Fiber-to-the-home is still in the field trial stage and the attempt to upgrade old telephone lines with digital subscriber lines (DSL) came to a screeching halt in many areas due to the poor condition of aging copper lines. However, all the efforts to develop fiber optic technology during this period will not be lost. The glut of components caused by the downturn in demand will drive prices lower and make fiber-to-the-home and small office more attractive. New technologies like dense wavelength division multiplexing (DWDM) and optical switching will make local fiber optic networks more cost effective, driving fiber closer to the home and making the bandwidth we need more widely available. The computer networking needs of the dot-coms and Internet service providers (ISPs) helped speed along the development of Gigabit Ethernet and Fibre Channel. The advent of gigabit speeds in local area networks (LANs) led to greater penetration of fiber optics in premises networks and the development of new higher bandwidth fiber and high-speed laser links. Components gained a factor of 10 in speed with little additional cost and more is on the way, now driven by the next generation of links at 10 to 40 gigabits per second and more. Industry standards groups have been following all this technology development closely, updating old standards and adding new ones as necessary. New tests for high-bandwidth multimode fiber and laser sources, for example, have helped vendors and customers with predictable product performance. Perhaps the biggest breakthrough for fiber this year came in the EIA/TIA TR 42 committee that writes the “568” standard for commercial building cabling. After years of trying to force fiber to fit the copper cabling mold, they broke fiber out into its own standard, EIA/TIA 568 B.3. This new standard recognizes fiber’s uniqueness in network architectures, adds in new component designs, such as the small form factor (SFF) connectors and high-bandwidth fibers, including 50/125 fiber as an alternative to 62.5/125. Does this mean fiber is poised to take over network cabling? Not at all! Category 5 and its derivatives are holding their own in the cabling marketplace. Copper cabling Rumors of the death of copper cabling are greatly exaggerated. While most Gigabit Ethernet networks run over fiber optics, in truth, most computer networks are quite “happy” at less than 100 Mbps, since most PCs can only transmit or receive data at a fraction of that speed. Until recently, cabling standards were really copper cabling standards, and copper cabling, unlike fiber, had standard components. There is a big benefit to working where cable, connectors, and hardware always look the same. Most installers and end users also benefit from being familiar with copper. While Category 5 has been fading, replaced with Category 5e, Category 6 is still struggling to get through the standards process. Making cable to meet a 200 to 250 MHz performance spec is not as difficult as making connectors and hardware at that level. Cables have been reported to show cable-to-cable crosstalk (alien crosstalk) when bundled neatly, perhaps discouraging neatness when laying cables in trays. Most hardware manufacturers have Category 6 hardware available, but the fact that many use different electrical design schemes in plugs and jacks to meet the high-bandwidth specs means that one manufacturer’s plug may not be useable in another’s jack. In fact, it may have lowered that Category 5e performance. Nonetheless, Category 6 is poised for standards approval, with or without interoperability. But users will have to rely on manufacturers to provide information on what products their hardware works with properly, or go with a single-vendor solution. While the EIA/TIA committees struggle over getting a Category 6 standard finished, the Europeans are forging ahead with what the United States calls Category 7. Cable with 600 MHz bandwidth is a highly tweaked version of the classic modular eight-pin connector is the chosen style. Category 7 may make it to standards status, but no network using it is proposed, so it may end up being a standard without purpose. I’ve been known to say, “Category 7 will be fiber.” Applications Innovations in cabling designs do not seem to affect applications immediately. Although Category 5e is the current standard, plenty of Category 5 is still available and being used by cost-conscious end-users. One big application, of course, is home networking. Toss the standards out the window as homeowners buy Category 5 and hardware from their local home supply store and wire up their own networks. Based on questions we get on our Web site (www.CableU.net), there are a lot of DIY cabling installers who are learning as they go. Even vendors are getting in the act, as one major cabling supplier offers splitters to allow running voice and data or two data connections on one Category 5, in violation of the 568 standard. I doubt there is a lot of business in bailing out these homeowners, but contractors working on residential projects can probably find plenty of work installing Category 5 and Coax in new construction. Perhaps ¼ to ½ million homes will be built with cabling networks installed this year. A much larger market is the MTU. Apartment owners and condo developers are finding that their residents want fast Internet connections. Many are right out of college where they learned to expect high-speed Internet access. They are forcing landlords and condominium developers to offer high-speed access by simply choosing to live where it is available. Forecast Long ago, I learned forecasting from a seminar conducted by an industry “expert.” He earned my respect immediately when he started off by saying, “all forecasts are wrong from the moment they are made.” I do not have any great insight into the near future of cabling, other than to expect the slow evolution of the current products. “Category or ‘Cat’ whatever” four-pair unshielded twisted pair (UTP) cable will continue to be developed with a goal to become not only standards-compliant, but also interoperable. Fiber optics technology will progress, but slower than in the last few years due to a lack of investment. Multimode fiber may end up graded like UTP, as new variations for higher-speed networks become available. Component costs will continue to drop, faster if the economy falters. Overall, the slowing of the economy must be reflected in the market for networks and cabling and technological progress will slow from less investment, but continue nonetheless. HAYES is the founder of Fotec, the fiber optic test equipment company and the Cable U training programs. He can be contacted at Jh@jimhayes.com.