Copper's Future/ Can It Compete With Cable?
Source: Inter@ctive Week
Inter@ctive Week via Individual Inc. : Who are you kidding?
Conventional phone lines can't compete with cable television wires
for delivering interactive services, such as access to the graphic
riches of the World Wide Web and other multimedia computing
services. Or can it? Only about 1 percent of the capacity of a
copper phone line is used up by a voice conversation. So telephone
companies are racing to figure out how to develop the other 99
percent fast enough that copper can compete with cable.
Digital Subscriber Line technology, the stuff that pumps up the
amount of data that can be transmitted across copper telephone
lines, often appears to be a confusing jumble of acronyms and
megabit rates. But to understand the current buzz on DSL, you really
need to know only one number: 560 million.
That's how many copper telephone wires will be in place around the
globe by the end of this year, according to industry estimates. And
it is the potentially overwhelming size of the market that is
driving the telecommunications industry to develop ways to stuff
those copper lines with ever more bits. Most carriers expect booming
demand for faster Internet access and telecommuting capability, with
its concurrent need for remote access to corporate computing
networks.
If telephone companies can reach that audience by investing in their
existing copper networks -- as opposed to building new fiber or
coaxial cable networks -- they can get reasonably priced services to
market more quickly. That in turn would allow them to compete
effectively with cable television companies, such as
Tele-Communications Inc. and Time Warner Inc., that plan to capture
a market for high-speed data services over cable and digital phone
lines that Forrester Research Inc. expects will be $2.5 billion in
2000.
The urgency to beat cable companies into this market is driving
telephone companies to examine more closely what they can do on
their existing networks.
"We realize that there is a gap between what ISDN can deliver and
what cable modems deliver," said Jerry Parrick, president of U S
West !nterprise Networking Services. "We're looking to ADSL to fill
that gap."
The competitive drive has led to renewed interest in a kind of
digital line that sends more bits to the customer than the customer
sends back. Every major telephone company in the country is
conducting lab tests on such Asymmetric Digital Subscriber Lines. At
least five -- Bell Atlantic Corp., BellSouth Corp., GTE Telephone
Operations, SBC Communications Inc. and U S West -- have announced
such ADSL field trials.
"ADSL is now a proven technology," said Rob Faw, president and chief
executive officer of Westell International and executive vice
president of Westell Technologies Inc., a company that pioneered use
of ADSL technology in its service access platforms. "We don't have
to convince anyone anymore that this stuff works. Now it's more a
matter of how quickly we can meet the price points and get it into
the network."
Danny Briere, president of TeleChoice Inc., a Verona, N.J.,
consultancy focusing on ADSL development, expects to see the first
commercial deployment of ADSL by late this year. "I think we'll see
orders for equipment this summer and then deployment will start by
the end of the year," said Briere. "I see Bell Atlantic and U S West
as neck and neck as to who gets there first."
In addition, the eight largest telephone companies (the seven
current Bells and GTE) recently galvanized around an approach to
creating data-to-the-home networks that uses ADSL to get customer
traffic to a central office, then aggregates traffic onto a
high-speed backbone using Asynchronous Transfer Mode, or ATM, which
moves voice, video and data in small chunks of a fixed size.
There are other clear signs that the push to boost copper is gaining
steam:
* Microsoft Corp., which announced plans to make its Windows NT
server compatible with cable modems, is now adding ADSL to that
list.
* Westell Inc., a leading manufacturer of access systems using ADSL,
announced last week it is working with Microsoft to make sure that
public network applications for Windows NT will operate over lines
equipped with ADSL.
* Major equipment vendors, including Ericsson Inc., Lucent
Technologies Inc. and Nortel Inc., are at last stepping into, or in
some cases, back into the ADSL arena. ADSL has been pioneered by
smaller vendors, such as Amati Inc., AT&T Paradyne (a unit that
Lucent plans to spin off), PairGain Technologies Inc. and Westell.
* Wall Street has discovered ADSL in a big way. Stocks of companies,
such as PairGain and Westell, have increased by four to six times
within the past year, while Amati's stock, which began trading a
year ago at just over $1, was valued at $21.875 on May 15.
TeleChoice has tracked these stocks, along with that of Adtran Inc.,
in a Copper Stocks rating that last week hit $275.88, up 30 percent
from the previous month.
What's the Deal?
DSL technologies leverage traditional modem and Integrated Services
Digital Networks, or ISDN, technology, to squeeze more information
onto a traditional phone line.
They use different forms of modulation, a technique that impresses
voice or data signals onto a copper line, to get more onto a copper
telephone wire than has been the case with standard voice phone
calls. Voice calls traditionally use just 4 kilohertz of bandwidth.
"We have copper wire into everybody's home, but we have only used 1
percent of that copper with traditional voice service," said Jeff
Waldhuter, executive director of access technologies lab for Nynex
Science and Technology. "Dial-up modems do a good job of maximizing
the 4-kilohertz bandwidth, but ADSL will make use of the remaining
500 kilohertz of bandwidth."
Commercial ADSL currently delivers 1.5 megabits per second
downstream and 64 kilobits per second upstream. ADSL products that
deliver 4 to 6 Mbps downstream and up to 640 Kbps upstream will be
out in late 1996.
Although developed with video delivery in mind, ADSL is actually
much more suitable for data transmission and has particular
advantages over the telephone companies' current best option for
residential data service, ISDN. Many of those advantages are due to
the fact that ADSL, unlike ISDN, would not be a service based on
existing telephone company switches. Therefore it doesn't require an
expensive software update to every switch and is much simpler to
configure and put into service. And because it isn't a switched
service, it does not use up switch capacity and wouldn't have to be
priced on a usage basis, as ISDN is today.
We definitely see ADSL as something we could use to offer a
flat-rate data service," said Mark Gallegos, product manager of new
business development for Pacific Bell. "That's part of its appeal."
The telephone companies have painfully discovered that the Internet
access community is accustomed to $19.95 "all you can eat" services
from Internet service providers and newcomers, such as AT&T
WorldNet, and are rebelling against the pay-as-you-go pricing
schemes that telcos set up for their ISDN offerings.
Finally, ADSL does not disrupt analog phone service -- high-speed
data services can coexist with traditional voice service on the same
copper lines.
ADSL also has one significant advantage over cable modems: The
bandwidth it delivers is not shared with other users. At first
glance, the 10-Mbps bandwidth of a cable network equipped with
modems seems significantly greater than the 4- to 6-Mbps capacity
that commercial ADSL is likely to deliver. But the ADSL bandwidth is
dedicated to a single user, while getting more than three households
online within a cable network node would quickly reduce the
bandwidth available.
How Fast Is Copper?
There are other digital phone line technologies in the running.
Telephone companies already use a kind of digital line known for its
high bit rate to provide 1.5-Mbps service over two pairs of copper
wires, without repeaters.
At least one High-Speed Digital Subscriber Line, or HDSL, vendor,
PairGain Technologies, is leveraging its high-speed technology in a
version designed for data access into the home. The Megabit Modem
runs over a single telephone line and can deliver 768-Kbps data
access in both directions over a distance of 7,500 feet from a
central office. That gives telecommuters the chance to send big
files, as well as receive them.
The idea, said PairGain Chairman and CEO Charles Strauch, is to get
into the market quickly with technology that is available and less
expensive.
"HDSL is more mature and has already come down the cost curve," said
Strauch, whose company's earliest megabit modems cost about $1,000.
"We'll see prices fall quickly on this as volumes pick up."
U S West is using HDSL technology from PairGain, as well as ADSL
systems from Westell, in its Boulder, Colo., trial this spring.
Looking further into the future, there is Very High Speed Digital
Subscriber Line, or VDSL, service, at speeds up to 52 Mbps -- five
times what cable operators promise for their high-speed data
services.
Orckit Communications, an Israeli manufacturer, demonstrated a VDSL
modem at the Telecom '95 exhibition in Geneva in October 1995.
"We have several of those modems spread around testing this VDSL
technology," said Dan Arazi, vice president of Orckit's marketing
and sales. "People never believe it works until they see it. But
they are testing it now on live loops, so that they can see how it
is affected by environmental conditions."
But Arazi, like PairGain's Strauch, doesn't see these very
high-speed products becoming commercial and cost-effective in the
near term. Orckit will introduce HDSL over a single telephone line
system in June at Supercomm '96 in Dallas and will launch its
commercial ADSL product, with downstream speeds of 8.2 Mbps, in the
third quarter of this year, Arazi said, at an initial end-to-end
price point of $3,000.
So instead of seeing VDSL delivering 52 Mbps over a long copper
telephone line, most industry experts expect to see a gradual
transition, in which carriers push fiber-optic connections closer
and closer to the customer and use one of the digital technologies
over the final copper link.
"This becomes a graceful migration for the copper plant," said
Robert Olshansky, manager of advanced service platforms for GTE.
So What's The Catch?
As with most new technologies, however, there are still hurdles that
high-speed digital phone technologies, beginning with ADSL, must
overcome.
Prices for the equipment that must be put in residences and
telephone offices are still two to five times higher than the
telephone companies' target prices, and they won't drop until volume
orders come in.
The target price we hear most often is $500 for both ends of the
ADSL system," said Orckit's Arazi. "$500 is definitely a realistic
price in the future, but I don't think it will happen in 1997."
There's also worry that ADSL modems won't perform up to snuff in
real-life networks, where the copper wire can be old, wet or perform
below expectations. "These are the kinds of issues we're working out
in the trial phase," said Sean Dalton, ADSL product manager for GTE.
In addition to ADSL modems, the carriers must refine the network
infrastructure that can support a widespread high-speed data access
service. Since ADSL is not a switched service, the traffic must be
routed into an existing backbone network, such as frame relay
service, where "cells" of data are moved about, or the carriers must
build some new backbone that will deliver and pick up packets of
digital data to and from customers' residences or offices. Then,
there's the matter of making it all work efficiently.
"We need to understand that this is not just the access platform
itself, but also the network management, addressing,
interoperability of routers and other systems," said Kamran
Sistanizadeh, director of network systems engineering for Bell
Atlantic.
"We're running a marathon here, and we're really just getting out of
the starting blocks," said Jim Bender, president of Aware, an ADSL
technology company.
Carol Wilson, Inter@ctive Week
[06-06-96 at 17:23 EDT, Copyright 1996, Ziff-Davis Wire] |
