Bandwidth Is Important

A lack of on-line bandwidth continues to underscore the challenges of turning networks into computer utilities

P> Don't underestimate the power of bandwidth.

As Internet content and service provider offerings blur, companies' competitive strategies will focus more on price and faster access. Although prices can, as in any business, be raised and lowered to reflect the current market, on-line speed involves choosing technologies, testing them and in some cases developing them from scratch. Bandwidth, as perhaps illustrated best by the current cable modem frenzy, has become a market of its own.

Yet, the problem of bandwidth also goes beyond markets. In a very real sense, bandwidth is an enabling technology. Computers these days are too powerful for the networks they're attached to. And the networks themselves are presently incapable of adequately supporting the many newfangled applications most would associate with the information superhighway -- browsing the World Wide Web, using an Internet phone, watching on-line video, or constructing intranets.

Since Intel invented the microprocessor in 1969, the power of a given processor has roughly doubled every 18 months -- a phenomenon named Moore's law, after Intel's legendary founder Gordon Moore. This steady development has decreased the price of computing power to the point where a computer on the desktop costing a few thousand dollars contains more processing power than multimillion dollar computers from 1969. Few computer companies were able to survive this transition, but those who did now participate in a much larger market.

The challenge for telecommunications providers and related computer companies is to drive a similar phenomenon in bandwidth -- without putting themselves out of business.

The slowest link

Most people think the Internet takes too long to access, to download, to see anything other than a simple image. Even if a cable modem can download data at a rate of millions of bits per second, the speed of the entire system will still depend on other transfer points in the network that do not operate so speedily. As a result, the network is only as fast as its slowest link, which is often far slower than most telecommunications carriers or Internet service providers would care to admit. "People are tired of having to wait to see graphics," said Emily Green, senior analyst at Forrester Research, Cambridge, Mass. Bandwidth is the solution to that problem, and maybe more importantly, the next step in the Internet business.

Accordingly, the bandwidth business proposition, for both service providers and equipment manufacturers, is finding all the weakest links in the network and strengthening them. A recent alliance of Sun Microsystems and Motorola, for instance, will offer an entire package of devices to target most of the weakest links -- networking software, superfast modems, servers and related services. They will supply the package to phone and cable companies.

It is but one of dozens of similar and brewing alliances, most of which will probably not get far beyond news-release stage. But in the next decade, some will succeed in an end game where bandwidth, like computer memory, can simply be ordered, purchased and inserted into the system. Customers will know what bandwidth would be appropriate for which applications -- just as today an astute PC buyer can determine what level of processor and memory is needed to run certain spreadsheets, word processors or communications software.

Creating an economic model where bandwidth can be ordered on demand won't be easy, but the rewards for companies who try to forge this model are almost certainly worth the risks. The eventual winners will be to bandwidth what Intel is today by virtue of its dominance of microprocessors: One of the most powerful players in the infotech industry.

Bandwidth-on-demand: A menu of services

"Bandwidth will be the key to move cyberspace forward," said Brian Ek, vice president of government affairs for Prodigy, White Plains, N.Y.

To that end, Internet bandwidth today has taken over the importance of the power of the personal computer chip. The better the technology -- faster, more powerful -- the more people will use it and spend money. "We need something that will catapult the industry like the Pentium," said Ek. "That's the kind of leapfrog broadband communication we need."

Others agree. "How fast things happen on your PC will drive the content of the Internet," said Pete Higgins, group vice president of applications and content, Microsoft Corp., Redmond, Wash.

So what's the magic solution? Right now, there are four main choices when it comes to Internet access (see chart). Most people now have a dial-up connection, which uses traditional phone lines and connects them at fairly low speeds.

Cable modems, which can work at speeds up to 1,000 times faster than traditional dial-up modems, might be that leapfrog. The technology allows a file that would take two hours to download using phone lines to be accessed in 12 seconds.

In the past few months, cable companies have announced many deals with modem manufacturers that suggest the ball is rolling. Analysts think that ball will roll fast. Forrester predicts that by 2000, cable companies' aggressiveness and focus will amount to 7 percent of U.S. Internet subscribers receiving data over cable. That's almost 7 million on-line accounts, or $1.3 billion in revenue.

One of the largest players, Tele-Communications Inc., Englewood, Colo., announced last month it will form two new units to handle business and marketing of high-speed data delivery. TCI Internet Services Inc. will develop and manage on-line services for TCI's cable systems, including its @Home venture. The other TCI unit, Area Business Marketing, will concentrate on commercial sales of communication products for intra-company networks. Such an enormous chunk of the market will mean all-out war between the telephone and cable companies. Power companies and satellite businesses will also enter the fray.

Most say the cable companies will come out ahead because of the superior speed of cable modems. "The cable and power companies are the great contenders for the bandwidth throne," said Ek, although he thinks the cable companies ultimately will come out the winners. Rich D'Amato, a spokesman for the National Cable Television Association, Washington, D.C., said the industry is capable of becoming a telecom leader.

Rivals are even banding together to devise interoperable cable modems. "The broadband link team," which counts such illustrious members as AT&T Network Systems; Hewlett-Packard Co., Palo Alto, Calif.; Intel Corp., Santa Clara, Calif.; and Hybrid Networks, Cupertino, Calif., is working to develop an across-the-board standard.

Modem manufacturers have a lot to gain or lose in this market. This spring, TCI, Time Warner and Comcast will begin a deployment of 350,000 modems from Motorola Inc., Schaumburg, Ill., and Philadelphia-based Comcast said it would roll out service through 150,000 more from Hewlett-Packard.

But cable is not the only technology to offer faster speed. For years, telephone companies have been singing the praises of ISDN, or integrated services digital network, which sends information at a much faster speed over a digital system. ISDN is a well-understood and tested technology, and it is becoming more widespread. At the end of 1994, ISDN was available only on about half the lines in the United States; at the end of 1995, many Baby Bell regions are at 80 percent. ISDN is also becoming more affordable. Equipment can cost a consumer as little as $375, compared with about $500 in 1994. However, about 10 hours a month on-line can cost a user $65.

Although lower prices and wider availability would jump-start the ISDN market, the arrival of even faster technologies, such as cable modems, threatens to eclipse it. By 1998, ISDN's top speed will follow behind in market share to high-bandwidth cable, predicted Forrester.

If cable wins over ISDN, telcos are expected to promote another technology, asymmetrical digital subscriber line. ADSL is faster than ISDN but slower than cable. It was originally created for interactive television, but is a digital standard that can also be used for the Internet. One of ADSL's biggest selling points is that it runs on traditional telephone wires, so consumers would not need extra lines or have cables installed. "If telcos can roll out ADSL or buy cable companies, they'll stand strong in the game," predicted Ek.

And while cable looks like it's ahead now, anything could happen. Cable companies do have some weaknesses -- cable modems are now largely untested and unproven, and the cable industry itself has been criticized for not understanding the Internet market, or even being sure it has legs.

Regardless, when high bandwidth becomes the norm in Internet use, most expect it to have a far-reaching social and economic impact on everyday life. "The home will become a true place to live and work," said Ek. "I see a future of 50 million wired households."

Reader Comments

Please post your comments here. Comments are moderated, so they may not appear immediately after submitting. We will not post comments that we consider abusive or off-topic.

Please type the letters/numbers you see above

What is your e-mail address?

My e-mail address is:

Do you have a password?

Forgot your password? Click here

Washington Technology Daily

Sign up for our newsletter.

Terms and Privacy Policy consent

I agree to this site's Privacy Policy.


contracts DB