In This Section
- Acoustics and Sound Reinforcement
- Archiving Restoration and Digital Libraries
- Audio for Games
- Audio for Telecommunications
- Audio Forensics
- Automotive Audio
- Coding of Audio Signals
- Fiber Optics for Audio
- Hearing and Hearing Loss Prevention
- High Resolution Audio
- Human Factors in Audio Systems
- Loudspeakers and Headphones
- Microphones and Applications
- Network Audio Systems
- Perception and Subjective Evaluation of Audio Signals
- Recording Technology and Practices
- Semantic Audio Analysis
- Signal Processing
- Sound for Digital Cinema and Television
- Spatial Audio
- Transmission and Broadcasting
AES 107th Convention Heyser Lecture: Roger Lagadec
Digital Audio and the Challenge of the Internet*
Roger Lagadec, AES Fellow
Having just entered the digital age, audio is about to face new challenges. Electronic technology is about to offer negligible-cost solutions to some key tasks in digital audio, such as coding and encoding, storing, transmission, and access. At the same time, the Internet is establishing itself as a practical system for the physical and commercial distribution of digital goods, with a massive impact on many aspects of future audio products. Recent developments in Internet and e-commerce solutions are useful in assessing which elements of forthcoming business systems may be susceptible to control. They are also valuable indicators of forthcoming challenges to be faced by the audio community in the areas of research, education, publishing, and standards.
|Known for his pioneering work in digital audio, Roger Lagadec now focuses on Internet commerce. He received his Ph.D. in digital signal processing for telecommunications. He has served as an AES governor.|
* Presented on 1999 September 25 at the AES 107th Convention in New York, the first in the Richard C. Heyser Memorial Lecture Series
When friends from my former existence as an audio engineer contacted me this summer, I was much surprised. I also felt deeply honored at being asked to inaugurate the Richard C. Heyser Memorial Lecture Series. I was aware of the risk of asking someone no longer in the mainstream of professional audio, or conversely, I saw the level of trust this represented. I will try to do justice to the honor bestowed upon me, and to gain your attention with thoughts on how the Internet will impact your industry. Proud as I am to look back on close to two dozen lectures at various AES conventions and conferences, I perceive the presentation I now engage in as a major challenge. I will return to the shining figure who is commemorated by this lecture in my conclusions. As I begin, I dedicate the occasion to the memory of the kind, incisive, ever-curious, pioneering, demanding, wise Richard C. Heyser, who left us early and gave our Society much.
For me, the last eight years or so have been with only superficial exposure to the main issues of professional audio. My family and I moved from Japan to Germany, then to Switzerland, where we have now finally settled. I switched from professional audio to data recording, to digital broadcasting, to conventional and electronic publishing, to the Internet, to e-commerce projects, to the world of telecom strategies, to mobile services, and to WAP. This profusion of experiences gathered outside the world of audio is what made me accept the invitation, hoping that they can provide a useful and relevant perspective on audio from outside the industry.
MEET THE INTERNET
Eight years ago, as today, Moore's law was valid, along with similar empirical laws, concerning the development of disk recording technology and the cost of transmission and switching. It is important to realize that Moore's law requires mass markets to sustain and finance its momentum. It remains hard to believe that Moore's Law may have a good ten years to go before it possibly runs its course. A good computer notebook today may have 6 gigabytes of hard disk space, but this is enough for 48 kHz, 48 channels, 20 minutes, and—forgive the obsolete number--16 bits of word length. As an illustration (and in no way as a format endorsement), an internal disk drive suited for 24 bits, 96 kHz, and a practical 20 to 30 minutes of recording time costs as much as an hour of good professional tape. The same disk drive can also accommodate my favorite 100-odd CDs on a notebook in the MP3 format—with space left for a few games and for the files of my mobile office. The notebook's tiny modems run at 56k and 128k--not that fast, but 6 to 12 times faster than those in common use eight years ago.
It is very clear that the professional audio industry is riding the coattails of the computer industry—not the consumer entertainment industry, as was generally expected a decade ago. To make this assessment a little less humbling, one may also recognize some influence from game technology and telecommunication. And from the Internet, not for its technology--it rides the same coattails you do—but because it is changing the way our society does business, an all-encompassing change that will unavoidably impact the audio industry.
The Internet did not really exist eight years ago. There were arcane systems, and I do remember how some of us belonged to the very early, pioneering adapters. The Internet I refer to here is a mass market, and the emergence of this mass market had not yet been predicted at the time. Indeed, five years ago, the buzz words in the industry of electronic media were still interactive television, video on demand, dedicated ATM networks, fiber to the home—visions that were never tested by the consumers, as the engineers could not put them to work.
We observe the continuous, bottom-up growth of a huge market, mostly in spite of all the inconvenience still to be encountered on the Internet. The market's growth is all the more extraordinary because the Internet is without dominating core products, without any dominating killer application, and financed by the consumers themselves: most Internet-enabling equipment—the computers and their periphery--would be bought anyway, and can be, as it were, reused for exploring the Internet. The Internet is the ultimate parasitic, opportunistic market; its actual killer application is the change it brings to our daily lives.
WHERE DOES THE INTERNET MEET PROFESSIONAL AUDIO?
In which basic ways can the Internet influence the audio industry?
The Internet is universal, costs very little, is a mass market, and its costs hardly vary with distance. The Internet's mass market customer base makes it a normative institution. A good bit rate for mass market use is one that fits the Internet state of the art. Formats with parameters in line with prevalent Internet use (and good marketing) will be adopted by the mass market. Formats overtaxing Internet users, their modems, their patience, their requirements for convenience, will not be adopted.
The Internet is a universal tool for exchanging personal information (e-mail) and structured documents (attachments). It makes business and technical exchanges easier, more casual. It allows for simpler cooperation. It is—in those areas that have embraced it--a very powerful tool for the dissemination of scientific information. This may look very important, but actually it is only mildly so.
Technically speaking, the Internet is an adequate vehicle for distributing good-quality audio in the MP3 format. This specific match of functionality and performance is of course important. Equally important, though, is the plain fact that the Internet remains essentially the domain of computers. The user does not relax on a couch, but sits on a chair. The computer is not in a listening room. The products the user gets from the Internet are best consumed and used on the spot—pictures, information, transactions, computer software--or delivered and used later—groceries, CDs, books, money transfers from a bank account. The Internet is not suited to delivering goodies to be enjoyed immediately under the shower, or to be listened to with full concentration via a stereo system. As a vehicle for distributing music for online, instant use, it is inconvenient--not only because it is too slow: more to the point, today's Internet is, in consumer audio terms, in the wrong room.
Equally important is the fact that the Internet gives one access to content, to information, that can be searched, found, ordered, distributed. This is especially relevant to the audio industry—which is of course only one sector of the electronic media industry--as it gives the customers direct, simple, and intuitive access to product and other information.
The Internet allows users to encrypt what they send so that only the final recipient will know what was sent. The technology is cheap and very powerful. When envisaging the future, universal use of unbreakable, no trapdoor encryption is a good assumption.
The Internet can be used—with still many limitations today--for broadcasting and for interconnection between systems. It is a limited, but economical interconnection channel.
The Internet allows electronic business to develop. You can look at catalogs, order, trace delivery, listen to samples, download software, initiate payments, publish advertising material. Today, ordering CDs via the Internet is commonplace. This is very important indeed, but the main impact of the Internet will not come from replacing the purchase of a CD, one-to-one, by its ordering via the Internet and its physical delivery via mail.
Most importantly, the Internet allows a direct dialogue between customers and suppliers—when selling physical goods and while selling electronic, virtual products. The threatening potential of the Internet to challenge the existing business system by enabling large-scale bypassing of both the existing distribution systems and their protection of copyrights—and conversely the opportunities it opens to redefine customer relationship and franchise, and to reinvent what is offered to audio customers--are the main topics of this paper.
THE INTERNET CULTURE
As a person closely involved in the main issues of Internet access, content, intermediation, one-to-one marketing, and e-commerce, let me give you my own view of this market's participants.
A faithful image of the society it belongs to, the emerging Internet economy is relentlessly opportunistic, pragmatic, and nonideological. I have observed time and again—among business people of faultless pedigree, not among ecstasy-crazed youngsters—the implicit attitude that whatever is possible is therefore obviously legal, or at least cannot be illegal. This presumption of legality is accompanied by a perceived gap between laws and economic reality, which I believe is also very real. There is finally the presumption that what is patently impossible to enforce--or what is commonly perceived as such— cannot really be relevant, and should not hamper business--indeed, should not slow down the gold rush.
Laws that are widely perceived to relate no longer to the vital sector of the society they should represent and police—here is a grave matter indeed. In Europe, copying books and magazine articles on a large scale is patently illegal, and photocopiers are cheap. There could be no clean, flawless resolution to this contradiction. The pragmatic approach was a levy on copying machines--surely no ideal example of the fair enforcement of a perfectly clear principle. We all can witness, a search engine's query away, the efforts of people who, out of enthusiasm for their favorite topic—medieval poetry; 18th-century engravings of small rodents; and, yes, current artists—will collect whatever they can find. They will lavish love and care on their Web site. As they commonly perceive it, they will not actually publish anything, but rather open their collection, so to speak, to the Internet community. In other words, fortunately or not, to the rest of the world. Copyright issues are, generally speaking, hardly ever perceived as such, even less so, certainly, than is the case with respect to home taping. It is not for me to issue any judgment on the perception of the average Web designer, computer nerd, average surfer, aficionado--but as we observe their behavior (and many Internet companies in Europe are extremely sensitive to the issue of copyrights, in view of their shareholders), we see no chance that they will ever change their attitude. They will buy what is convenient and carries a fair price. They will reject any trace of restriction in the use of what they have purchased. They will not bother with subtleties. They are enlightened consumers with no patience. They are the empowered economic heroes we sing in business school. They will not tolerate a flaw in the quality of the service they receive.
The Internet's image remains linked to sexually implicit content and pornography. There is some truth to the association—though one would not blame the video recorder, or print media, or the telephone for also carrying similar content. Europeans may indeed welcome the replacement of seedy shops by computer screens--usually at a healthy electronic distance from family content. Indeed, some of my colleagues observe with awe the relentless imagination of some sites not in the area of content—the product sells, we all know it--but in that of marketing a virtual product. Actually, in Europe again, we witness how a very strong consensus on banning specific content—fringe pornography, fringe politics--leads to successful coordinated efforts by institutions and individuals. The situation is vastly different where there is no such consensus.
We all know the saying that on the Internet, nobody knows you are a dog. On the Internet, solutions for commercial applications are being developed that must comply with existing laws seemingly contradicting the basic tenets of the Internet, say, by restricting participation in an Internet lottery to a particular country, although the Internet seems to ignore such boundaries. These are fascinating applications, in which the main challenges are with legal issues, payment, and perceived consumer trust. The dog (and the occasional human customer) has to be authenticated, located, his trust has to be gained. Establishing this relationship of trust—rather than attempting to impose one of limitation and control--is possible and indeed the only guarantee for sustainable success.
One lesson I learned from Internet projects is how business models matter, and how they are subject to change. A telecom giant may become a distributor. A portal may become a financial intermediary—not quite a bank, but far more than just a portal. A delivery company enters one-to-one marketing. We attempt to describe all the functions involved in electronic commerce with what we call the butterfly. Here it is.
In the butterfly diagram (which can be expanded to include platforms, infrastructures, ...) we can see all the functions allowing an electronic market to be created and operated, all the mechanisms linking what some people have to offer to what some people may want. One can represent advertising, image build-up campaigns, communities, push technology, one-to-one marketing, subscription to audio services.... One can also represent how new roles can be assumed—amazon.com selling CDs, consumers finding and ordering by themselves the CDs they would not necessarily find at the local music store, accessing the lyrics they would not find on the liner notes.
As it develops the new tools required to do business electronically, the Internet community is also learning to represent business models, to challenge them, to create alternatives, and to try them out. The Internet is making business models pure, and its interest in business model issues is very far from academic: it is developing a growing body of collective experience in successfully challenging the existing ones.
Exposure to advanced e-commerce projects brings other insights. Good projects have a way of raising fundamental legal questions—on taxation, product liability, content ownership. We are constantly reminded that the simplest and most ubiquitous things on the Internet--URLs, links, the location of a server, the click on a button, creating your own home page—are in the legal gray zone. Does a link to a Web site represent an endorsement? Does a search engine finding objectionable content perform an illegal act as it helps the user get there? If quoting from copyright-protected content is legal, can one take one song, cut it up into 50 quotes, and publish these with impunity? Can one publish a computer program that will just happen to render very faithfully a copyrighted picture or piece of music, although one cannot find a "copy" of the original in the code? This reverse engineering on content is possibly the most insidious of all challenges—not because of the Internet, which is not illegal per se--but because the Internet may create factual situations directly challenging the basic legal assumptions of a nonelectronic society.
In Internet circles, it is fair to say that attempts to prevent the free use of MP3 files are not very popular. Here is an area where the chasm between economic common sense (music programs are the property of the record companies) and daily practice (if it is a small data file, I can do with it what I want) is extreme. In Internet circles, as everywhere, we see the gray zone between home taping and piracy. The difference is that if you put one piece of music on your home page, for the enjoyment of your discerning and legally minded friends, you may find that you have suddenly many friends, not all of whom are discerning and legally minded.
For example, with all factual information carefully modified: a music lover in Hungary runs a Web site on which you find everything there is to know about a particular folk music group. Pictures, lyrics, scores, trivia; and also sound files. If you like the group and happen to make a search, the Internet will take you there. A good tool will transfer the Web site onto your notebook in no time. Copyrights are not mentioned. When you get to the site, you feel the serendipity of the find. You do not feel that you are being confronted with a (just possibly) criminal act. You certainly do not feel that you have possibly just committed one. You are on the Internet, acting like a carefree Cybernaut. Plus, if your search takes you there, chances are you already own the CDs, and thus—in your own perception--have already paid your dues. Or so you believe.
I have personally not a trace of a doubt in my mind that casual publishing will continue, out of the reach of control mechanisms which cannot be enforced—neither technically nor legally. I also believe that this will in no way ruin the music industry, for very practical reasons.
If there is one guiding principle in Internet applications, including e-commerce, it is that only convenience counts. Things must be simple, easy to understand, no-brainers. Not because the consumers have no brains—the e-mails in which they complain of the poor quality of services show astute minds at work. They just do not want to have to exercise their brains for reasons having nothing to do with their own interests. I see this as the toughest challenge to be met by, say, the Secure Digital Music Initiative: their proposals must be perceived by the customers as bringing them real benefits, not limitations.
A common prediction states that, when all is said and done, electronic books, music, all contents will only cost as much as it takes to copy them, with some allowance for the inconvenience of copying: buying recording media, spending time copying, labeling the copied product, the complexities of transferring, say, from PC to stereo set.
Strange to say, the Internet is proving precisely the opposite. Huge volumes of software are purchased on the Internet. As an informed user, this author is of course aware that copying a large piece of software costs almost nothing; nor can he claim that he purchases his software—pays for his downloads--out of constantly felt, deep respect for the owners' copyrights. He does so because he is given a good deal. Customers can try things out first—they could indeed go on trying the software forever, but that is inconvenient. Payment is easy. The cost is quite low--a few working minutes, much less than the time and money the software will help save. There is, most importantly, the essential promise of upgrades. Today's bugs will be fixed tomorrow, tomorrow's viruses will be recognized the day after tomorrow. Customers do not so much buy software as subscribe to it. Just buying software makes you a buyer. By subscribing to software, one becomes far more valuable—a customer--and both sides win. The cost-of-copying paradigm is fundamentally wrong. The right paradigm is that of value of subscription, with cost of copying as a healthy reminder that customers can respond to poor service quality wherever they want.
Some forms of Internet audio will therefore bring along lower prices. No distribution costs, no pressing costs, no returns, no out-of-stock items. All this must mean that all the songs on a CD will cost less than a physical CD—no wonder, all the songs of a CD being just files, without the added values (and limitations) of the physical product. Somewhere in the near future, though, there will be the full CD's true successor--the full CD, with its subscription, with information services, gossip, online forums, for those who are interested in buying more than just sound files, tailored to the user's needs and profile. Cobranded, comarketed, profile generating, interactive, customer driven, partly customer defined. The Internet and e-commerce show many practical examples of how such things are done, of how to make buyers become customers. Casual publishing—Internet home taping--will continue. Indeed, copyright protection will not easily resist the assault of challenged Internet users who love nothing more than another Microsoft they can ridicule with impunity. The software industry easily compares with the audio industry in its sales volume, vested interests, and, surely, technical expertise. In Internet circles, there is a perception that the only hacker-proof systems are for user-to-user encryption, and the only enforceable policy is one that focuses on very few people doing something universally accepted as extremely wrong.
I believe the final outcome will not be tighter control, but better products, transcending, so to speak, the simple audio files. The promise of better products rather than tighter controls is a good message. If this forward-looking strategy comes about, I see many new roles in the audio industry in creating value around the plain sound files: more information services, more impact from the world of pictures, more depth to the product, more processing functions. More publishing and subscribing. More value for the customer. More jobs. More music.
In the world where I now work, we say that "they" made the telephone, "they" made the Compact Disc digital audio system, "they" made the computers (indeed, "they" tried and failed to do many products), but that no one, no one, made the Internet. The Internet made itself. We are all making the Internet, we the modern citizens and actors of the electronic age. We will not accept an Internet with limitations we see as obsolete and based on fundamentalist arguments. Industries touched by the Internet will help make an Internet on which they will be forced, literally forced, to sell their cherished products the Internet way, in response to the customers' needs, and the customers' good money will reward them for providing what is wanted. Those among the content providers who capture first and who serve better the empowered Internet customers will thrive, based on quality and not on controls.
The abstract promised conclusions for research, education, publishing, and standards. Research should hardly be affected by the Internet's economic power—and a good thing too. Education will need to give greater care to enlightening students as to the economics of the business--including better audio enhancement for Web sites, which too often are very poor audiowise. Publishing is an obvious growth area, what we used to call multimedia, and the enrichment of sound files with lots of information, at the time of publishing and later. As for standards, surely the Internet will not dictate its own standards to audio, but the future producer will need to understand and use XML, and will need to support the many formats the Internet marketplace will create and make compete.
There will be, in response to the electronic marketplace we call the Internet, a massive change in the paradigms of the audio market. My fervent hope is for the audio industry to face this challenge in the spirit Richard C. Heyser displayed in the years he enriched the lives of the members of our Society. He identified an area where great improvements were necessary and possible. He found the right approach. He persevered. He argued. He worked very hard. He made great efforts to communicate his ideas convincingly and well. His publications are a great labor of love, witnesses of a grand search for truth. He wrote well. He was bold. He took risks. He embraced new ideas. He created new concepts, which will outlive us all. He was young of heart. Going through his anthology, you meet a man of very great value. May his spirit long inspire us, may his force be with us. And may the audio industry find, when facing the challenges of the Internet, women and men of a similar stature.
J. Audio Eng. Soc., Vol. 47, No. 11, 1999 November, page 1017