Ulrich SiebenA vegetarian Dog
“MP3” – the simple extension to a file name for an audio file – has made quite a name for itself in the in the meantime and even appeared on the front cover of Time magazine on one occasion when internet hype was still all the rage. Even though there was an exaggerated tendency to overrate the technical and business-development potential of the web at that time, the sobering legacy of the technical economy and new economy which many people are facing today is doubtless responsible for generating an exaggeratedly pessimistic view of the future at present.
The potential for knowledge, know-how and ability to cooperate that is stored in modern societies does not simply evaporate, however, when the stock-exchange index starts to crumble. A worldwide network of increasingly better-educated and more efficiently communicating people is constantly looking ahead from the basis of what has been achieved to date. Progress in both information technology and biotechnology is continuing to run at an accelerated pace, even if the markets would welcome a respite. People will not ignore the superior products of tomorrow but will want to buy them.
MP3 is not yet another invention but a whole complex of topics, many of which exemplify the progress being made in technological developments today and the impact that they are having. The development process does not take the form of a linear chain of effects but frequently constitutes an intricately interwoven network of mutual influences. The straightforward pattern of supplying consumers with industrial products – in this case MP3 music – has been broken through and, all of a sudden, the consumers themselves are playing an active part in the development process – in terms of the range of music on offer, the computer software required, the configuration of the internet as a communication network to suit their particular purposes, and even in the shaping of opinion on the future of copyright. And this is going hand in hand with the corresponding feedback to the consumer electronics industry, accompanied by a pressure to innovate. This new freedom that the audience is now enjoying is generating all kinds of unease, particularly since the constant further enhancement of computer performance and internet speed means that even the Hollywood film industry is now becoming involved too.
From the user’s point of view, the “MP3 revolution” may not even really have begun as yet. Although music is being downloaded in large quantities onto PCs worldwide, it is, after all, only on PCs that it is being stored so far. MP3 has still not reached the average consumer who, at most, may have received a music CD burned on a PC as a gift. Even experts do not generally have their PC infrastructure integrated with their home entertainment.
This new type of digital music is opening up new and greatly improved opportunities for the user, however, and the possibility of downloading music apparently free-of-charge from the internet, as is currently feasible, is not even the most important aspect here.
A normal CD has a limited playing duration and essentially only allows the user to play a sequence of the same few tracks. Users cannot put together programs to suit their needs and are forced to constantly trample the same old paths, being clearly limited in their choice.
The truly innovative feature of MP3 – and of the multimedia technologies as a whole – is that the information (the music) can be handled independently of a specific item of equipment or carrier (the CD). It can be stored on any type of digital storage medium, can be readily configured in different arrangements (the playlists) without being subject to time restrictions, and transferred between different equipment and different people via any desired digital channels. It is recombinant (1) with any other information and with machines that are still to be developed in future, such as new storage media and players, and also new user interfaces for the administration of music programs. It can be better transported through time, and no one will ever have to “lose” their record collection again simply because the LP has been superseded by the CD.
Now that music can be stored on hard disks, which are available in portable versions and in sizes smaller than a cigarette box at times, it is possible to record thousands of hours of music. With random access to archives of this type, people can today, for the very first time, compile their own individual music arrangements, thereby considerably enhancing the emotional impact of the music.
These archives, or Pocket Media Centres, which will soon contain more than just music, can be carried round and played wherever required, such as on the stereo at a friend’s party. And it’s not just music that is involved. A number of “early adopters” are already loading hours of spoken contributions from the radio onto storage media of this type, fully automatically, so that they can listen to them when they have the time, such as when travelling. Idle time can then be put to good use; listeners can readily skip over the contributions that don’t interest them.
Music from the internet also marks a new form of communication about music. Search engines provide people with ideas, and they come across previously unknown artists and find pieces of music that had almost been forgotten about again. On the internet, it is possible to browse through the archives of others with similar tastes and start chatting with them about music. And artists aside from the mainstream stand a better chance. Swapping and collecting, and talking about one’s collections are fundamental forms of human behaviour. When it comes to music from the web, it is not just a question of a download but, for some, the important factor is social networking with the “community”.
At the time when personal computers had just been introduced, it was possible to encounter “hackers” who had collected huge quantities of computer programs. Of the many hundreds of programs that they had at their disposal, they only made serious use of a few. The software producers regarded unauthorized copying as a major problem at the time and equipped their products with copy protection. As of the mid-Eighties, most of the “hard” copy-protection methods had disappeared from the market again, and the fact of not having copy protection of this type was seen to be advantageous for business.
The type of music collection that fits on a hard disk today cannot be listened to in its entirety on account of its sheer size, and it also cannot be stored piece by piece. Even now, the size of the available storage media is doubling every two to three years. People can, and will, engage in the collection of digital music. With portable storage media, entire collections can be passed on to others with just a single click of the mouse. Many of these pieces of music will perhaps never be listened to in their entirety, and just a few hundred will become part of the user’s constant listening repertoire.
With music infrastructure of this type, personal data security suddenly becomes a serious issue. In the event of a defect, i.e. a hard disk crashing, most users would lose all the data that they had compiled over a prolonged period of time. This constitutes a new situation. And this same defective hard disk would doubtless contain other data that would be irreplaceable in many cases – personal photos, scanned documents for one’s own “paperless” office, last year’s tax declaration, and perhaps all the pages from the web that someone had carefully selected and read… The questions of classifying, searching, accessing, separating what is important from the not so important, the aspects of quality assurance, security against data loss, replacement, and the synchronization of different data storage media that people may own, are all issues that are still waiting for solutions. Over the long term, the only people who will manage to cope with files that have been randomly copied from the web will be the minority who take the trouble to archive their files afterwards and who, indeed, have the know-how to do this. The majority require the appropriate software tools and support from outside before they can exploit these new, attractive opportunities. And, on this point, a solution will be found to the payment of authors’ rights for the majority of the public at large.
The development of MP3 technology in the internet has not followed a strategically linear course. Instead, there were a series of technical and social schemes that developed in parallel for a long period of time, which then suddenly recombined to form MP3 in 1995, without this ever having been a specific intention on the part of industry.
The performance of the semiconductor ICs, which constitute the substrate for information-processing products, has been undergoing exponential expansion for 30 years, increasing by a factor of one hundred to one thousand per decade. And this process will presumably not come to a halt over the next ten years either. This is a huge amount of progress: the speed difference between a car and a horse does not even amount to a factor of ten yet, despite this, it has changed the world. Back in the Eighties already, people were making the witty remark that, if aircraft had developed in the same way as ICs, a transatlantic flight would take only a few seconds, in a plane costing five deutschmarks.(2)
As the potential of IC technology increased, so did the complexity of the products that could be implemented on a single chip, and simplifications had to be found to enable this complexity to be handled. The switch to digital technology then made it possible to tap this complexity reserve. Digital ICs permit a particularly ingenious abstraction in that the know-how regarding the function of the IC is completely independent of the know-how relative to its production. It basically irrelevant whether a digital arithmetic unit is implemented electronically or pneumatically. It was this abstraction (which evidently functions much more effectively with freely programmable processors) that permitted the necessary division of labour for the development of the generally highly complex products of today.
Digital signals also have this same abstract property. They have no effect on the information that they carry. Pictures, sound and text can be transmitted together as required. An analogue television, by contrast, is a television signal, is a television signal, is a television signal … and that’s that.
In the light of these new opportunities, people began to seriously consider digital radio and digital television at the start of the Nineties. Although digital-reception technology was at least ten times more expensive than the analogue technology, the rapid developments in semiconductors meant it would only be a matter of time until prices fell.
The evident advantages of digital radio were too attractive to resist: employing data reduction methods together with cables and satellites, it would be possible to increase the number of TV channels hundreds of times over, and a digital system would also, it was thought, permit reliably encoded pay programs (sic!). A digital system would also make it possible to distribute interactive multimedia programs via radio and television channels.
IC manufacturers build the components for these products and had to follow these trends. Missing out on new key components is a highly risky business, since cost and quality considerations always tend to lead to integrated system solutions on a chip.
We (3) thus started to develop a new, efficient processor architecture for the planned digital radio – DAB – working as part of a broadly-configured European project to begin with. It very soon became clear, however, that digital radio would not be arriving as rapidly as had been planned. What should we do? The Fraunhofer Institute in Erlangen had an audio compression algorithm which made it possible to transmit music in CD quality via two ISDN lines. In view of the good relationship that Otto Witte, one of our concept engineers, had with those working at Erlangen and because we felt that this was a particularly interesting approach, we took the decision to program the MPEG1/2 Layer 3 audio algorithm, subsequently known as MP3, on our processors without having any special products in mind. It was by no means certain at that time that MP3 would become established, but we were able to gain experience with data compression in this way. Sometime later, a US company, WorldSpace, was looking for two semiconductor partners to develop a portable receiver for worldwide satellite radio. A comprehensive and precise market study was available showing that there was indeed a demand for such a receiver, together with secure funding for product development and the necessary satellites. We expressed our interest in the project, since it seemed plausible to us, and the know-how that was to be developed could also be used for other systems. It was presumably because we had got ahead with MP3 that we were awarded the job, and two ICs were created, one of which was an MP3 decoder.
Midway through the Nineties, Infineon attempted to create an alternative to the CD based on a semiconductor and developed a high-density ROM technology to this end. One of the ideas was to build a “micro-Walkman”, since this would not have required very much more space than the small button cell. The first prototype players of this kind used ICs from our company. Somehow, a Korean manufacturer got wind of this and used our ICs to build the first commercial MP3 player with flash memory that could be loaded with music via the PC. These were attractive items of equipment, much smaller than a Walkman and also much more robust in mechanical terms, although they were fairly expensive to begin with. Storage media for an hour of music cost more than 250 deutschmarks at the time, but since semiconductors constantly got cheaper (see above) a great deal of optimism prevailed.
Equipment of this type does not require any moving mechanical parts and is thus very easy to produce compared with a classic Walkman. In addition to this, the ICs are so small that the MP3 functions can make their way into other portable units, such as mobile phones and wristwatches, on a parasitic basis as it were. For the first two years, we were the market leaders and observed a rapidly-growing and confusing market with huge numbers of newly-developed products, which were not, however, sold in quantities of more than a few million per year.
At this time (1997), MP3 players were products for technology freaks – exciting but expensive – and it was difficult and a slow process to handle the software that came with them and also the music on the PC, encoding it, decoding it and downloading it onto the player, etc. This made MP3 players inappropriate for the average user.
The software for producing MP3-coded music that was “set free” in the course of this process, however, rapidly reproduced itself on the PCs in the internet, since it was free of charge. This meant that there was no stopping the global propagation and replication of digital MP3 music files. And the globally thinking and programming community has always found a way round the legal and technical means that have been implemented in a bid to halt its propagation to date. Perhaps there will always be a way out when things are so complex – otherwise biological evolution 500 million years ago would perhaps have come to a halt at a dead end.
Even Microsoft, exploiting its own market power, was not able to succeed in having its own competing “WMA”-Method accepted as the de facto standard for digitally-coded music instead of MP3.
In the meantime, the majority of MP3 players are not being built with semiconductor storage media any more but with PC hard disks. Hundreds of hours of music can now be stored on these, and the cost per hour is not that far removed from the cost of music cassettes. Some 400 music cassettes will fit on a portable hard disk today. Ergonomic solutions have been developed for archiving and handling music files.(4) PCs and other products for home use are starting to network with the infrastructure at home, and the speed of the internet is increasing rapidly. The true MP3 is still to come.
There is always an unavoidable “openness” to digital products in that they are linked to a world of programming. But programming is like performing magic. Although commands are issued in “C++”, “Java” and other artificial languages, instead of muttering away in Latin, the direct conversion of a wish into action still remains the same. This “Harry Potter effect” practiced by a large, inventive community changed the X-Box (5) back into a PC and will continue to produce miracles. The digital world is still expanding the possibilities open to it in inflationary proportions, and hundreds of thousands of engineers are wondering how technology and human behaviour can be linked together in the correct manner. The ten years to come will bring greater changes in the world than the past ten years – something that has, incidentally, been true ever since the start of the industrial age. And quite a few business models will become extinct en route.
I don’t think that it will be possible to avoid ending up placing expensive bets on erroneous trends by conducting increasingly refined market studies or engaging in more rapid product development. The primary requirement for remaining successful is to adopt a considered approach to the management of one’s options, have an organization that is capable of cooperation and ensure good access to know-how and capital. New-Economy Vers.2.0 cannot be excluded.
Only goods that are in short supply have a monetary value. As the digital media and the internet developed, the gripping question arose as to whether it would be possible to artificially restrict the availability of digital contents again, now that they can be reproduced anywhere without production costs or time outlay and can be distributed at will at the speed of light. To begin with, as digital television was being developed, people were very optimistic about this question but, under present conditions, parts of the industry start to fear that there will be no sustainable technical means of doing this.
It is very difficult to make a coding method technically secure if it is designed for the mass dissemination of contents which, additionally, have to be reproduced via loudspeakers and screens. Any music can be re-recorded digitally, for example, by “placing a microphone in front of the loudspeaker” – and the same will be true for films in future. Only a few digital copies need to be in circulation at the outset for propagation via the internet to be possible. A worldwide, technically-skilled, programming community will presumably always be able to circumvent copy-protection methods either in this way or by other means. Whether processes that are practicable in ergonomic terms develop in this way or not still remains to be seen.
Any powerful technology is Janus-faced. There is even another side to a broad-based “D igital Rights Management”, which is ultimately associated with the identification of the individual instead of anonymity and the curtailment of individual autonomy, and is proving to be an obstacle to customer acceptance in the first generation at least. In addition to this, DRM technology, depending on the way it is implemented, relates not only to music but to all types of digital information and could also be used for locking out people from information for purposes other than economic ones. At all events, it will be interesting to watch how the balance of power between consumers and industry changes if DRM technologies become more widespread.
How efficient are legal means for preventing the violation of copyright? There are currently considerations to make the removal of copy protection into a punishable offence, or even to ban research and development work on algorithms that could achieve this, and trials are underway. It will perhaps be difficult to achieve a comprehensive worldwide consensus on this, across all the different cultures (6), particularly it were to be non-commercial copies for private use that were to be banned.(7) A large number of countries have, however, started to revise their legislation.
Despite its efforts and partial success in stopping the average consumer from producing illegal copies of contents, the media industry will presumably have to expect there always to be a certain number (and probably not that small a number) of unlawful copies in circulation as long as the consumer does not see any convincing benefit in the use of DRM technologies.
There is nothing to say that the entertainment industry will not be able to cope with the loss of income it is currently experiencing, or will not be able to offset it through different business opportunities. Even today, entertainment is being sold via a wide range of distribution channels that have secure payment collection systems, e.g. television, radio, cinema, merchandising, live concerts, etc. And precisely in the case of entertainment products, it is also not always clear who the paying customer actually is – the person watching or the person wishing to make their products better known, who therefore has to attract people’s attention. There is thus much to be gained if an artist’s popularity can be increased rapidly and at low cost. Replication free-of charge in the internet is evidently highly suitable for this. If the replication rate is not on its optimum setting, then there is a great danger of being edged out.
Apart from this, large numbers of people are prepared to pay for services and particularly for ergonomics in the internet age anyway. The media suppliers will develop ways of meeting this demand in an innovative manner which also creates added value. And, as far as the speed of this process is concerned, it would perhaps not even be desirable for there to be an over-effective means of defence for protecting the status quo.
A further-reaching question arising from the developments that have been described relates to the social consensus as to how copyrights, know-how and innovation (which is becoming increasingly easy to exploit independently of special production processes) are to be appraised, remunerated and protected in the future, and how this protection is to be restricted in terms of space and time. This is because the risks associated with innovation will have been compensated at some stage, but further development requires a publicly accessible area of knowledge (8) and the free exchange of ideas. Legal positions, patents and copyrights, however much they serve legitimate interests, are actually more of a hindrance here. While copyright is respected in relationships between companies, it is not firmly anchored in the legal awareness of the population – and probably not without reason either.
In this balancing act, the lobbyists who are currently exerting a significant influence on the shape of the future legal situation in a number of countries naturally view the legal boundaries from the angle of their own particular interests. And the results will play a very considerable role in determining the opportunities that exist for economic trading and the distribution of (global) prosperity in the knowledge economies of tomorrow. The public should also pay sufficient attention to these developments too, so as to ensure that the general potential for development is not concentrated in the hands of too few people in future.
We have an interesting tendency to believe that the status quo will continue to exist and to regard the present state of affairs as marking the end of a historical development. Yet we are living in a world that is growing auto-catalytically in all directions, in an innovating world and in a world that is becoming increasingly efficient. Since the globe is not exactly overpopulated by altruists, it is not readily possible to bring this development to a halt – without catastrophic consequences.
It is only possible to speculate about what the future will bring in technological terms. It can be assumed that MP3 will gain further ground to begin with, but will then be superseded by better methods. It is not even sure that most of the music consumed tomorrow will be canned music. With faster computers, for example, it will be possible to input the score and synthesize the orchestra and voices. Apart from much more efficient data reduction, this would also have a great deal of advantages. A piece of music would not always have to sound the same, and users could adapt the interpretation of the piece to suit their personal taste through interacting with the computer. This would even involve a certain creative act. Synthetic voices could be developed in line with the audience’s taste. Since it is difficult to find an audience for music without prominent stars, it would perhaps be better to have a computer-simulated person appearing at the same time. Just how well this works is evident in the cinema already (9), and it will doubtless be possible to achieve this on portable equipment too soon, as well as in home cinemas, of course. But that is just one example, and things will doubtless work out differently in the end, or work out even worse.
In our future environment of high-performance computers, broadband networks, omnipresent screens, and excessive entertainment and information offerings and constraints, a number of people will still have a lot to learn before they can live happily and in a self-determined manner.(10)
(1) Biology: the selective combination of genetic inormation that has not so far been linked
(2) And, unfortunately, one in a hundred would crash ;-)
(3) This refers to Intermetall, which was subsequently acquired by Micronas
(4) Apple‘s “iTunes“ for example
(5) Microsoft’s game console, competing with Sony‘s Playstation.
(6) See Samuel Huntington’s: Der Kampf der Kulturen
(7) See §53, §95a, $108b of the draft Amendment of the German Copyright Law, 2002
(8) Richard Sietmann, c't 24/2002, p. 108: Software-Patente und Urheberrecht,
(9) e.g. the feature film “Final Fantasy”
(10) See: “Declaration of Independence (1776)”, http://www.whitehouse.gov/independenceday/declaration.html
Ulrich Sieben studied physics, mathematics and anthropology in Göttingen. He conducted scientific work at both the university and the Max Planck Society on aspects of signal processing by the hearing system. Midway through the Eighties he switched to ITT-Intermetall, a manufacturer of semiconductor chips which was starting to implement the first digital television sets at that time. He held several posts in the development sector of this organisation, which was acquired by the Swiss Micronas AG in 1997, including the position of overall head of development for almost a decade. Today, he is a manager of Micronas Holding GmbH. Micronas brought the world’s first MP3 decoder chip (the "brain" of an MP3 player) onto the market and, as an active member of international copy-protection committees, is in close dialogue with all the links in the music and equipment marketing chain.