CD Radio Promises Data and Audio Listening Relief

Ernest Rejman
Contributing Editor

If while driving on the interstate you cannot elude Hank Williams' laments about cheatin' love and his long-dead hound, or are trapped in New York, Chicago or San Francisco, irritated by bad reception when trying to listen to your favorite radio station, relief is on the way. It is coming in the form of digital audio broadcasting (DAB), popularly called CD radio, which, in its two incarnations -- terrestrial and satellite -- will not only provide some 100 channels of high quality fade- and static-free audio, but data as well. (Figure 1 shows the satellite version of CD radio; its terrestrial counterparts will use established commercial broadcasting facilities.)

There are several major players in the CD radio arena. Some, like Lucent and Delphi Delco, are well known; others, like XM, Sirius and DRE, are not, but will be. The parts needed to begin nationwide service are either in place or quickly coming together, and the point of entry will not be the home but the automobile.

According to Ben Benjamin, senior vice president of product management at Lucent Digital Radio (LDR), Warren, NJ, ever since the merger of LDR and USA Digital Radio (USADR) into iBiquity Digital, everything has been going very well. "The merger is good for us and the consumer, who we envision will benefit from this type of radio," he said, adding that the technology can now be brought to market faster and more cost-effectively: "Faster, because now a standard will be adopted in the US much earlier than it would have if we were competing. Cost-effectively, because we'll get to higher run rates and bring economies of scale to bear faster than if we were competing."

XM Satellite Radio, Washington, DC, is building a prototype digital satellite radio receiver. According to Richard Wormington, senior vice president of engineering and operation for XM, "We'll provide all this technology license-free to our partners to build AM/FM/XM radios so that when you want to listen to us in your car you just punch the XM button, and if you want to listen to an FM radio station you punch the FM button."

Ken Erickson, product director of North America Audio Business for Delphi Delco, Kokomo, IN, is confident receiver development and broadcast schedules will converge as expected. "Sirius already has one of its satellites up, and XM will get at least one up this fourth quarter, and the rest of their launch schedule is set up," he said. "We're at the point of getting first physical hardware on the receiver side in two months. Everything's reasonably on schedule, close to the timing stated by both XM and Sirius."


Delphi Delco is working with General Motors and "others" it is not yet prepared to name on a remote box that will feed the audio signals into the radio head. "From the user interface viewpoint it's transparent -- just another band on the dial," explained Erickson. Although this new technology's first application will be focused on music, there are a number of data applications. "Newer radios today typically have larger displays with more text capability, so messages will be shown there. If you had a station panel such as '80s Oldies, that could be on the first line, the song or artist on the second." In fact, depending on the service provider, there is a string of messages that could be shown on the display, including where to buy the CD that is playing.

Over the last few years the trend has been to increase display content, particularly in the case of cellular phones. In the navigation world, some displays are comparable to cathode ray tubes in their capabilities. This improvement has spilled into personal digital assistants (PDA) and, now, to radios. Ten years ago, a typical display had a 3.5-figure capability, showing time and frequency, and perhaps four icons at best. "Now you're seeing 8, 12, 2 x 12 with icons, dot matrix displays to enable increasingly more graphics and text capability," said Erickson.

This graphic capability is not cheap. "You pay a price," admitted Erickson. "The new base radios have more graphics than the old ones, and high end radios have significantly more graphics capabilities still, with sometimes a color LCD display, depending on the OEM." Although the new radios will be rolled into the base price of the car, disguising the cost, the fact remains that, depending on the content display capability, the cost of the radios can have as much as a 20-to-1 price spread.

Bringing 2.5 GHz into the automotive environment is a new field. Indications are that both Sirius and XM have done a fine job working out compression techniques and fitting considerable data into the bandwidth the FCC has regulated for them. The hardware is being built and tested as you read this. Obviously, beyond all the computer simulation, design reviews, peers reviews and progress made, the proof of the pudding should come out during the next quarter, when real-life testing begins. As Erickson put it, "Then the satellites get turned on and we start driving cars to see whether we have a saleable product, then do further development." There is still a list of minor action items being followed, such as getting the right type of decoding through software adjustment. However, as Erickson views it, "The proof is when I turn that switch and hear the music and get the data."


There were three proponents for in-band on-channel (IBOC): Digital Radio Express (DRE), Milpitas, CA; USADR (a consortium of broadcasters); and Lucent Digital Radio. The Lucent/USADR merger into iBiquity Digital has as investors 15 of the nation's top 20 radio broadcasters, including Infinity Broadcasting Corp. of Viacom Inc. and Clear Channel Communications Inc.

iBiquity will develop and commercialize digital broadcasting technology solutions, including Lucent's perceptual audio coder (PAC) audio compression technology, to enable radio broadcasters to send audio and data content via digital signals. These signals will allow existing radio stations to transmit CD-quality audio with crystal-clear reception plus wireless data for a variety of consumer applications, such as station and program content, stock and news information, local traffic and weather, and much more, over existing radio frequencies without degrading the transmission of current analog programming.

For a time, it appeared there would be a drawn-out systems battle. It was recognized that if IBOC was to become a reality in a reasonable timeframe, there had to be some form of a grand alliance.

According to Norm Miller, DRE's president, no one knows what the launch date is for IBOC. "This is the $64 question," he said, "when a system will be finalized, deployed and be production-ready. Particularly with the merging of Lucent and USADR, it means that there must be a melding of both technologies to a certain extent. With a consortium they can move forward quickly; at least there will be no opposition from the marketplace -- it'll be just like digital TV, which only became reality when a heavyweight consortium was formed. The USADR alliance with Lucent brings in the 600-pound gorilla that creates the de facto standard which then everyone follows."

Automobiles will be first in fielding the technology, at least at the receiver end. IBOC proponents do not propose subscription fees as part of their model, and conventional broadcasting will keep its current business model, which is commercial revenues driven. Once digital radio is established and there is a low cost receiver, the expectation is that the public will slip into digital radio almost without noticing it. The newer cars will have it, as will new stereos and portables.

IBOC allows digital to co-exist with analog. The listener still receives the standard analog AM/FM signal, so if he buys a new receiver and installs it in his car, there will be no difference as far as the analog stations are concerned -- he will get them as he always has. When he switches to the digital station format, audio quality will be improved but, more importantly, he will acquire the capability to receive a powerful data channel that can provide all sorts of customized information -- from real-time traffic to stock quotes and news. As it is defined now, the business model shows that the standard audio and news and conventional broadcasting will continue being free. Not so with the valuable, nationwide data channel that provides a wireless data link into your car as well as to hand-held devices.


Although the downlink is one way, it would be easy to couple it with an uplink using cellular technology. From a business model perspective this works well because users download data in a 10:1 ratio compared with what they upload; therefore, using the more expensive cellular infrastructure as an uplink would not be prohibitive. The view is that downloading more data than needed is expensive. For instance, the service supplier would pump out a flood of stock or traffic information, and the user would program his receiver, whether radio or PDA, to pick up the stocks of interest or the traffic zones he wants. Such a low cost, real-time data download like this is ideal and, if something more specific is required, the user can always reach for his cellular phone. The radio link for low cost downloading could be networked to work side-by-side with a cellular link for uplinking.


Satellite digital radio has two proponents: XM and Sirius. The FCC has granted them licenses to use the 2.5 GHz frequency. However, satellites still need to be orbited and earth stations set up for the system to work.

On the other hand, IBOC uses the current broadcast infrastructure -- approximately 14,000 radio stations in the US divided 50/50 between AM and FM. These stations already have towers and transmitters and are broadcasting signals. Converting over to IBOC is not an expensive proposition and, practically overnight, it enables these broadcasters to acquire a digital terrestrial network across the country. (Lucent Technology's approach to CD radio, shown in Figure 2, lies in adding an IBOC module to existing AM or FM transmitters in operating commercial radio broadcast facilities, providing them with digital capabilities for audio as well as a data channel.)

It is interesting to review the development of digital radio. At first, broadcasters fought satellite radio, fearing that it would ruin them. When the FCC finally ruled that it would grant licenses to XM and Sirius, things began looking less bleak, especially when it was realized that a shortcoming of satellite digital radio is its line-of-sight problem. In a metropolitan area, there is what is known as the "canyon effect." If you park under a shelter or drive through a skyscraper-lined street, you may not receive digital satellite radio.

A year after receiving their initial license, XM and Sirius realized that to enable themselves to offer a viable service, they needed terrestrial repeaters for the major cities and applied to the FCC for an additional license. Again, broadcasters did not like this, claiming that with repeaters they were no longer a satellite broadcaster but a terrestrial one. The FCC, however, granted XM and Sirius a license to operate low power terrestrial repeaters in metropolitan areas.

Thus, for satellite radio to work in metropolitan areas, the signal must be put through a repeater, providing the listener with a terrestrial signal. However, it should keep subscribers paying $9.95 a month for CD-quality music from being annoyed by a signal that keeps fading in and out while they are driving, as is sometimes the case with some FM stations. This addition increases cost but makes the system viable, assuming you are within range of a repeater. It also brings up the issue of how many repeaters are needed for cities like Los Angeles, New York, San Francisco, Detroit or Chicago.

IBOC's overwhelming advantage is its low implementation cost, and that makes possible a digital data network covering the US, capable of providing valuable wireless data links into PDAs and car radios. Here is the real advantage of true terrestrial digital radio.

For AM stations especially, the IBOC conversion cost is relatively low -- in the $10,000 range -- because most broadcasters have relatively modern transmitters. It just requires an IBOC generator that is plugged into the transmitter. The digital signal is fed through the box into the main transmitter as with subcarriers, where a subcarrier box is put into the transmitter, which already has a subcarrier feed. FM's case is somewhat more complicated because IBOC requires a linear amplifier, and many of the older FM amplifiers are extremely nonlinear and would have to be replaced. Therefore, on the high end the broadcaster may face a $250,000 investment; however, if the FM transmitter was designed during the last five years or so, conversion costs would run approximately $50,000.

The IBOC design requires that the receiver be digital-signal-processing (DSP)-based. Today's advanced DSP engines are more than capable of handling the kinds of data rates considered. Miller believes that, with a sufficiently high volume (tens of millions per year), the IC chip set for the entire IBOC design can be provided for less than $10. "Even if you double that," said Miller, "with today's technology, a $20 chip set is very achievable -- it's a matter of volume."

If this sounds too optimistic, consider that between 16 and 20 million car radios are consumed yearly in the US alone. If only car radios were switched over to digital, during the first year there would be a penetration of 10 percent, over the second 25 percent and in five years 100 percent. Approximately 40 million other radios are consumed in the US, ratcheting this number up to a 50 to 60 million yearly consumption. Clearly, some are cheap, giveaway radios, but with a market like car radios, boomboxes and stereo tuners, a US 40-million-radios-per-year market is reasonable. Obviously there is no question of volume -- it exists.

In addition, there will be low cost receiver chip sets implemented for PDAs, laptops and others as add-on modules. Those numbers are also significant. Consider the case of the PalmPilot, for instance, already a fairly powerful computational device with memory and processing power. Getting a small plug-in unit with an FM decoder that -- whether it is used for audio or not is immaterial -- provides a low cost data link capable of feeding information into it becomes very attractive. It would be possible to go a step further and add a Global Positioning System (GPS) chip to the module to prove location information as well, permitting the incoming datastream to be customized for the user's location, enabling him, for example, to find out when the next bus will arrive at the street corner where he is standing.

DRE has an alliance with Cue Networks, Los Angeles, CA, and this fall will bring in a digital carrier system based on its IBOC technology using the subcarrier spectrum, which can be taken to market now without any further FCC approvals. The focus will be on digital subcarriers, and DRE is designing with them a plug-in for the Handspring PDA, which is expected to be available before Christmas. The basic Handspring unit costs about $150 and the user can buy plug-in modules from independent suppliers that convert it into a GPS unit or a cellular phone. Some seven modules are available for it already. The module that Cue will deliver is a small FM radio with an IBOC-based digital subcarrier unit, and Cue will provide a host of datastreams for the unit.

Digital radio -- whether satellite or terrestrial -- will bring a robust wireless datalink to the mobile environment, as shown in Figure 3. Many of the problems with wireless are related to the mobile environment-particularly in the case of a vehicle traveling at 60 or 70 mph. Present systems work well if the user is stationary or moves slowly, but that is not a true mobile environment. A moving automobile is a very demanding environment, and few wireless datalinks are capable of feeding reasonable amounts of information into it. With digital radio using the broadcast infrastructure across the US, the technology with which to provide a robust datalink in the mobile environment will be available.

The wireless data link will also drastically reduce repetition. A problem with harsh environments is that if the bit error rate (BER) is too high, this must be met with redundancy, consuming bandwidth. If a data system has 128 kb available but redundancy becomes necessary, this is reduced to 64 kb -- half the bandwidth capability is lost.


The LDR/USADR merger came up with a unified design -- a system developed at Bell Labs that uses a high quality audio coder called perceptual audio coder (PAC). According to Lucent's Benjamin, PAC will be used because it allows efficient bandwidth use, always a challenge when attempting to communicate through digital means. PAC allows transparent CD quality at 120 kb. The channel consists of both the analog and digital signal in the sideband, which Lucent refers to as a hybrid IBOC implementation because it contains both the analog and digital signals. In that configuration, the digital signal is 25 dB down from the analog signal at the center of the channel. This provides an interesting challenge to receive the signal in the same coverage area as the analog signal requiring a modicum of processing, which must be accomplished while maintaining CD quality.

Although everyone involved in digital broadcasting is quick to point at partnerships with radio receiver manufacturers, the fact is that beyond those involved in supplying car radios, there does not appear to be much gearing up. All major manufacturers are looking at designing receivers for this technology, but not much seems to be happening. Sony is often mentioned; however, a call to the company revealed that although it is "closely following developments," it has no plans to go into production.

It appears no one is willing to commit or allocate resources until there is a standard in place and some sort of service has begun -- production lines are expensive things to keep idle. Receiver manufacturers will not get involved until after having identified the production time and the number of units that will be manufactured. All of this action will be triggered by the FCC implementing a standard. "Our expectation was that the standard would be adopted by year's end," said Benjamin. This may not happen due to purely political and bureaucratic reasons. One way or another, there is going to be a change of administration in Washington and, as an industry insider confided, "We're seeing a number of people from the FCC commission already beginning to look for, and finding, other jobs." A realistic appraisal indicates it would be surprising if a standard is in place before well into the first quarter of 2001.

Once the FCC adopts a standard, three major requirements must be fulfilled before service can begin on a commercial scale. Obviously, the first one is getting a number of digital stations up. Because of the relatively small costs involved, it is likely they will appear in sufficient numbers, at least in major markets, with large, readily available listening audiences. According to Lucent, this change should take place around the third or fourth quarter of 2001.

"We expect to have the major building blocks for receiver manufacturers to begin delivering product to the marketplace around the first quarter of 2002," said Benjamin. "A major element that must be provided is the IC that will allow them to build their radios in the compact form necessary for integration into a car's dashboard." This would meet the second requirement. The third most important requirement is exciting the prospective consumer about the technology through demonstrations at a number of shows and in markets in conjunction with broadcasts. Everyone involved has fresh in his mind what happened to stereo AM and there is no desire to repeat that debacle.

Lucent is working with "three of the major consumer IC manufacturers," which it declined to identify until that arrangement can be disclosed. "They will determine costs," said Benjamin. "Our assessment of the incremental cost for such a system is that it will fit within the construct of a $100 increment at the retail price level." (This means that, whatever its price is now, you will have to pay an additional C-note for the boombox or tuner of your choice.)

It is expected that eventually satellite and terrestrial broadcasting will merge. Because dashboard real estate is unlikely to increase, there will not be a second radio receiver there. Lucent is already providing some of the subsystems for the satellite broadcasters, who are using PAC in both the Sirius and XM satellites.


There are some countries, notably Great Britain and Germany, where digital broadcasting is already taking place. The European standard, Eureka 147, has been operational for three years. However, digital radio has not exactly set Europe on fire. One of the principal reasons is that receivers run about $1000.

Future service providers in the US are betting on IBOC for a different outcome. "First, US receivers will cost $100 or less in terms of incremental cost," said Benjamin. "Second, we know that broadcasters have spent considerable money and effort developing a brand following associated with their call signs. People know their favorite stations' call signs and with IBOC that won't change. They'll find the signal where it's always been. With Eureka 147, the European listener must migrate to an entirely different band and familiarize himself all over again with where to find the stations he wants."

While volume will certainly bring costs down in IBOC's case, the European receiver's design elements are different because it operates in S band. As one climbs up in frequency, equipment tends to be more expensive. "This is an advantage that has made the proposed US system's incremental cost smaller -- we're operating in the same frequency band," pointed out Benjamin.

It now appears that the European and American systems will be incompatible, but Lucent expects that once its system begins service in the US, there will be offshore applications. Lucent is presently dealing with European authorities, who, according to the company, are actively considering IBOC's adoption.


As mentioned previously, if a broadcaster has upgraded its equipment over the years, or the equipment is five years old or less, only a small modification is required. For an FM station, the upgrade cost would mostly be a high power combiner. The signal leaves the power amplifier and is delivered into the antenna to transmit; with today's analog power amplifier, the combiner is used to essentially sum the analog and digital paths, the latter being the second path that would be coming into the combiner. For an AM station, there are relatively inexpensive digital-ready exciter boxes for the transmitter. Because of linearity requirements, modern AM equipment is already designed to accept a digital system.

"Crucial data applications are linked to the more efficient use of time," said Benjamin. "There are situations related to traffic and downloadable maps. We see the system working as an adjunct to what is out there already, in terms of GPS-type guidance systems. We see in GPS that people need maps of very wide areas and often local municipalities need to redirect traffic due to construction or other factors such as traffic flow situations where a two-way road becomes a one-way road." This information would be automatically updated in the system as the driver enters the vicinity.

Today, as commuters crawl to work, traffic reports come in maybe every 15 minutes and emphasize heavily traveled roads that may not be of interest. A driver who might want to know about traffic that is beyond his or her area could program the receiver to profile his or her interests, and it would only pick up information specific to the route of interest, avoiding having to listen to all the other broadcasts. These functions would not necessarily be available as part of a subscription service, but rather be made available free by present broadcast service providers. However, other data such as sports scores and stock quotes that people want instantaneously wherever they may be, might be provided on a subscription basis.

These data could reach the user in audio or visual format. However, there are concerns linked to providing too much visual input. Work is under way with the Consumer Electronics Association to set boundaries on what to put on a dashboard without dangerously distracting the driver. A usage etiquette is being developed around how data can be displayed without encumbering the driver.


XM's development plan is typical of digital satellite broadcasting. "First you have to get a signal source programming center," said Wormington. "That is already under construction at our new facility in Washington, DC. Eighty-two studios, all modular, all digital, all tied together by fiber optics -- everything will be operational by year's end." (Figure 4 shows an example of a state-of-the-art facility under construction by XM.)

Next comes satellite transmission capability. In XM's case, this means 15 kW Hughes 702 satellites, the biggest, most powerful commercial communications satellites ever made. The first was orbited in the first quarter of this year, and the next one launches the first quarter of next year.

Then come the repeaters to cover urban canyons. "We have 1500, and their test deployment has begun," said Wormington. "There's a network set up in Pittsburgh that we're running tests on, and full deployment will begin by the first quarter of 2001. These will go in 70 cities to cover the top 70 Arbitron markets. The number of repeaters ranges from 200 in New York, down to four or five for a place like Albuquerque. It is directly dependent upon the urbanization level, and the amount of acreage that we need to cover."

All of this preparation is useless without fielded radio receivers. "We don't produce radios," said Wormington, "but have a number of partners -- people like Pioneer, Sony, Alpine, Delco, Sharp, Mitsubishi and others. We have a group in Florida whose sole purpose is building the hardware reference platform: antenna, tuner, an ASIC for channel decode and a DSP housing the software for the service layer decode, decompression and encryption."

XM is doing the engineering for a prototype radio capable of tuning into its system. The company has a working lab radio: satellite and repeater signals in, music out. It is now in what it describes as the "chip fabrication mode," leading to converting the lab radio prototype into chips. According to XM, these should be commercially available during the second quarter of 2001, with transmission beginning commensurate with receiver availability.


The signal is divided into an A and B side. The output is 4 Mb, which can be thought of as 2 Mb per side, plus padding and error-correction bits; thus, it is not a 4 Mb signal when it is sent, but the data content is 2 Mb. The A and B sides go to the satellites via an X-band uplink, returning as an S-band downlink. Both links are time division multiplex quadrature phase-shift keying (TDMQPSK)-modulated, going directly to the car. If the signal goes through a repeater, the repeater does an in-band receive and transmit with an amplification and change in the modulation scheme, from TDMQPSK to coded orthogonal frequency division multiplex (COFDM). The repeaters act as a single network, and the signal then is repeated out to the car.

The same information content comes over each of the two satellites and through the repeater. The radio can receive all three signals and is indifferent to which one it receives. The receiver is smart enough to select the best signal on a BER basis. This process is transparent to the user, who is unaware of which signal is being processed. In fact, the radio may switch back and forth rapidly, depending on signal quality.

The repeaters are mostly 2 kW, with approximately 10 percent of them slightly higher in power. In general, about 70 percent of the repeaters will be on a tower, monopole or building. Range is heavily dependent on environment. Placed on Mount Wilson in Los Angeles, a repeater might cover 20 miles. The same unit in New York City might reach five blocks. XM has already signed leases for 500 of the 1500 locations it needs, and the company indicates the remaining leases are coming in at a rate of 200 to 300 per month.

The 4-Mb datastream is divided into 8-kb primary channels, with every application working in some multiple of a primary channel. There is an overhead allocation of 2 kb for each channel, providing a way to do things like sending information on stocks in parallel with, and transparent to, listening to music. This capability is used as what is called "program-associated data" so that, for instance, the artist and song title of the piece being played appear on the display. Whatever can be digitized can be displayed.

A digital system has many capabilities, but just because they are available, it does not follow that they will be used. As Wormington puts it, "We're a broadcast system -- what we send, we send to everybody. We could use the system for point-to-point applications, but it's an inefficient way of using it, because it is like taking a message and broadcasting it over every radio station. Digitally it can do a lot of things; but from a business point of view there are only a few things we will concentrate on. Right now, we're putting out 4 Mb of data, and the bit doesn't care what it translates to. In this case, we're having it translate to music and talk. We can do a lot with it; the question is what we can make money on, and right now music and talk are it -- that is our business plan."

As for GPS or any other ancillary data source, XM is not considering it. "GM and Honda are very interested in telematics, but our system is designed to be a 4-Mb pipe to the car and that's it." Wormington added that a car radio manufacturer like Delphi Delco might choose to implement other data sources such as GPS or cellular, and integrate the systems. "We'll be a part of it, but aren't planning on doing the whole package. We don't make the radios, just facilitate their manufacture."

Like others, while admiring it, Wormington believes the value lies not in the technology, but in the improvement in programming. "People are going to hear radio like they think radio ought to be. Because we are nationwide, we don't have to appeal to a broad audience in a narrow geographic area and can focus on very narrow listener niches and provide the music that they love. Before, you could never have focused so exclusively on specific demographic sectors. People can make a business out of providing for narrowly focused musical tastes for $9.95 a month."

Of course, the big question is what the timeline is for an actual launch of services, and there does not yet appear to be a definitive timeframe. Experts like DRE's Miller believe that by the time systems are finally rationalized, field-tested, fine-tuned and taken to the receiver manufacturer (who then, and only then, will gear up for production), it may take as long as two years. Figure 5 shows a milestone chart for LDR's IBOC realization program.


The movie 2001: A Space Odyssey has one of the most inspired scenes in film history. In it, man's distant ape ancestor tosses into the air in triumph the bone tool that has given him supremacy over his environment and, as it spins skyward, it transmutes into a space vehicle in Earth orbit -- an elegant encapsulation of humanity's million-year arduous climb toward civilization.

The concept of digital radio brings to mind the culmination of that difficult evolution. While there is nothing technologically revolutionary about it, the ingenuity and application of existing technologies -- the marriage of satellites, high frequency, terrestrial repeating networks, large-scale integration on the IC side, decoding and encryption techniques -- to offer a capability that did not exist before is mind-boggling.

Some in the business wax poetical about it, like Delphi Delco's Erickson, "I frequently drive to the foothills in Tennessee and Kentucky, and go through several hours during which it's impossible to get a single radio station. Now, for the first time, people everywhere will be able to listen to the music and talk they want. They'll have more choices and reliability, as well as nationwide information services. The technology's promise is its message."

As Hamlet muttered, albeit in a somewhat different context, "'Tis a consummation devoutly to be wished." My satellite TV service boasts 220 channels. When channel-surfing, I am often awed by the incredible progress mankind has made since that long-gone ape creature tossed that bone on its way to the stars. We have acquired the ability to casually speed a radio signal off on a 46-thousand-mile journey, enabling me to receive anywhere on the planet and in the comfort of my own home, two channels that regale me with I Love Lucy in glorious black and white, as well as Roseanne and The Love Boat. It gives me three channels that try to sell me things I do not want or need, 20 that show me movies that were old before I was born, five that urge me to redecorate my budoir and prepare a gourmet meal, 16 that carry sports (often the same event) and so on, ad nauseam.

CD radio promises enormous potential, as did television and the Internet. However, if past performance of the use given to these resources is any indication, the promise should be taken with an extra hefty pinch of salt. *