Technologies for e-content
Leonardo Chiariglione - CSELT, Italy
Summary
From the earliest times, humans have produced artistic expressions of their minds that will be called here "content". They have done this sometimes for their own enjoyment, more often trying to get an economic remuneration from their efforts. Countless technologies have been invented and put to use for the purpose of providing ever more effective ways of expressing the human mind or of extending the reach of a given piece of content to more people farther away or to people who would be in existence years or decades or centuries to come. Most of those technologies have required the intervention of people or organisations capable to operate them. As technologies have become more numerous and sophisticated, a huge host of middlemen have become necessary.
A new technology coming to the fore has often meant a shake-up of the affected industry. The shake up of our age is the result of the combined effect of three main technologies: signal processing, that transforms analogue signals into digital ones, information technology, that allows the processing of data in digital form, and telecommunication, that allows the instantaneous transfer of digital data from anywhere to anywhere in the world. Content that is digital and reaches the end user in that form will be called e-content in this paper.
The first samples of the effects of the shake up is frightening for the industries involved: content becomes immaterial, can be indefinitely copied without degradation, can be offered from producers directly to consumers, apparently without the need of intermediaries. On the other half of the coin we see the emergence of completely new intermediaries providing directory services, profiling users, protecting content against illegal copying, supporting electronic transactions etc. This, however, is just the beginning, because the evolution of the three technology areas is unrelenting. More shake ups are in the making.
This paper has been written with the intention of
It is hoped that the technologies to come, some of which are the result of the activities of the author, will have a more evolutionary than revolutionary effect on e-content compared to the past and that their impact on society will be managed proactively more than defensively.
Table of contents
In primitive societies every able individual was capable of making himself all that was needed for a living: bows, arrows, shoes, tents, pottery etc., but the social organisation, even though the individual had to rely on himself, was far from uniform. Specialisation of functions already took place: the stronger became warriors, those good at making traps or hunting game were sent to get food, the wise ones became advisors to the tribe's chief and those cunning were sent to deal with other tribes. As society grew more complex, however, specialisation of functions affected the individuals in a deeper way. While still capable of making himself most of the tools for daily life, those in need of a sword had better go to a blacksmith or they risked finding themselves at a disadvantage in the time of need. To impress the beloved one it was advisable to go to a craftsman and buy a good necklace, should otherwise one risk one's chances. If a visit from an important person was expected, hiring a good cook to offer the visitor a sumptuous meal was a necessity so that one could get the much sought-after contract.
From the oldest times and under a variety of civilisations and social orders, people have happily parted with part of their wealth for the purpose of laying their hands on some physical objects produced by the skill of some craftsman to achieve some concrete goal or even for their own personal enjoyment.
But humans are ready to part from portions of their wealth for other, less physical, but for them no less important, matters. In case of illness the medicine man would offer potions. Sometimes, however, magic formulas, dances and prayers were expected to be even more effective. In case of drought the shaman would be offered remuneration if only he could get rain. The same would be done to obtain abundant game or a good harvest. The making of these "miracles" required a compensation to the person who would effect them.
With the progress of civilisation other types of crafts developed. The skilful use of words could give considerable benefits because something could be got done by somebody else who would otherwise never think of doing it, or would do it differently, were it not because of the ability to persuade him. The person able to give good advice to others would be considered a wise man and he could get several benefits by simply exploiting this skill. The ability to build arguments would convince people to request such services to defend themselves in front of the elders of the tribe. Great ideas about society would trigger the head of a social organisation to enlist the holder of such ideas so that the status quo in society would be preserved. The same could happen to a man trying to become the head of that social organisation by aggressively promoting a new social order. Or, if the holder of these ideas was smart enough, he could become himself the leader of that social organisation. Demosthenes earned a living by offering counsel services to Athenians in trouble with the law but failed to rouse Athenians against Alexander the Great. Cicero attacked Catiline and exploited the case for his fame and social standing. Confucius preached his philosophy in the hope that the ruler of one of the Chinese kingdoms of his time would hire him. Moses became the leader of his people within the reformed religion while Mohammed became the founder of a new religion.
From antiquity, possibly even before speech actually took shape, humans used songs, often accompanied by rhythmic movements of the body or by some musical instrument, to convey a message that could otherwise not be expressed, or expressed poorly or ineffectively, with words. This way of communication is effective because rhythm heightens the effect of words as it allows the performers to convey them with a passion that words alone often cannot. Authorities, civil as well as religious, were quick to get control of those social moments when people forget their daily troubles in those opportunities of social relaxation called festivals. The authorities often lavishly remunerated those involved in the musical and scenic organisation because of the role that these moments have always played in preserving and fostering social cohesion and because of their functions in maintaining the essential elements of individuality of a social grouping.
But alongside the lucky ones, those - groups of people or individuals - who could not get the support of some authority or simply wanted to follow their own artistic inspiration without compromise, could possibly meet with a very different fate. Artists and performers can be very good at playing a musical instrument or dancing or reciting a poem and people would even come from far away just to watch them or listen to them. The artists and performers may personally be very fond of being surrounded by a crowd of people enthused by their ability. Those around, if they feel like that, might even happily give some coins, just to show you how much they like the way performers sing, play, dance or recite. But others would give nothing thinking that the very fact that they have been watching or listening to the artists and performers was a sufficient reward. All this does not really matter if the artists and performers were wealthy persons and their only purpose in life was to express their creativity but, if they were poor or average persons, possibly with a family to raise, the border between an artist, receiving a reward for his creativity from the good will of onlookers, and a straightforward beggar blurs.
This description of artists and performers is not something recalled from past antiquity, it still applies to the cities of today. The sad fact is that while people take it for granted that they must pay for the work of a blacksmith or an attorney, they consider it an option to pay for the performance of a spontaneous singer. The artist has always felt the need for somebody else to take care of the promotion of his work. Of course the really excellent artists in general have a better life but it is a sad fact that the excellence of too many artists is recognised only after they are dead.
A further advancement in society - and ingenuity in developing appropriate technologies - created the written work, i.e. the textual part of what originally must have been complete multimedia works with recitations, music, songs, dances, choruses etc. Greek tragedies were composed of parts recited by actors (introduced into dramas by Thespis as late as the 6th century BC in Athens) and parts sung by the chorus, with the accompaniment of music. Troubadours sang and played violas. Kalevala - the Finnish epic poem - was recited by two persons who sat on a trunk holding each other while moving in a rhythmic fashion. Today Italian opera is still sung with the accompaniment of an orchestra and so are musicals in Broadway, New York.
It is a common fact of all civilisations, though, that while a means to represent words was developed and so allowed ancient literary works to he handed down to us, the same did not happen to the other components. It was only in the Middle Ages that music was represented in a way that could reach us. Glimpses of another component - dance - can be obtained from Greek amphorae and temple bas-reliefs and similar remains from other civilisations.
The harsh facts of life in antiquity did not really allow much space for those inclined to writing. Their motivations, in fact, were regularly personal satisfaction, associated with considerable personal wealth, remuneration by a patron who found pleasure in giving endowed people the possibility to manifest their talents or concrete support by some authority for works that fitted their plans of preservation, promotion or expansion of a social order. Under these conditions no one making a copy of a manuscript felt like he had to remunerate the author - that was, at most, for the amanuensis hired for that purpose. More likely the person making a copy thought that he was doing a service to the author because his work could reach farther and generally to society because he was propagating culture. And this was really needed, if one reflects that all of pre-Gutenberg Europe held just a few tens of thousands of books.
This does not mean that in antiquity some forms of commercial interest behind books did not exist. In Egypt a copy of the Book of the Dead was required at each funeral. Some surviving Babylonian tablets bear directions for getting a "book" from the library. In Rome Titus Pomponius Atticus employed a large number of trained slaves to copy and produce books with retail branches of his business in Rome as well as in the provinces. In 12th century China a publisher's account for a book of 1,300 pages indicated a 3 to 1 ratio of selling price to production costs. But, save for some periods of well developed and stable social order, it was hard for anybody to put in place a mechanism that would guarantee an economic return to an organisation that would look for and select manuscripts, produce a number of copies of a given manuscript, create a distribution channel and collect the proceeds of the sales.
From very old times public authorities saw it as their duty to collect books. The royal libraries of the Assyrian kings were probably for the exclusive use of the sovereign, but the Great Library of Alexandria was established by Ptolemy II Soter because the great body of knowledge on which rested the Hellenistic world was scattered - literally in pieces - in thousands of different places across the Mediterranean and learned men needed help to find old works. Qin Shi Huang-Ti, who ordered the burning of all works by Chinese philosophers in 213 BC, the Spaniards, who did the same in Mexico in 1520 and, Hitler in 1933 had a similar understanding of the role of collecting book and applied the same ideas as Ptolomy II Soter in the negative, with the goal of annihilating cultures that were alien to them. Keeping track of books published in a country is still a role played by many public authorities as in many countries a copy of each printed book must be deposited with the National Library. In France a copy of all audio-visual works published must be deposited with the Institut National de l'Audiovisuel (INA).
The invention of printing in the mid 15th century changed radically the economic basis of written words. A new style of entrepreneur, almost dormant until then, was born, whose job was to select and edit manuscripts, design the material, arrange its production and distribution, and bear the financial risk or the responsibility for the whole operation. And so in barely 50 years after the invention of printing the number of printed books in Europe amounted to 9 millions.
Printing, by offering books at a price orders of magnitude less than the equivalent for a manuscript, enabled broader layers of population to have access to culture and favoured the adoption of vernacular languages in Europe. The market in books, therefore, was from the very beginning largely national and, until recently, national markets have been rather impervious to one another. This excludes the scientific environment, where the market has been largely global since the very beginning of printing. Therefore, even though billions of books have been printed over the centuries, until quite recently it was easy to keep oneself abreast of everything new through the use of catalogues published by the individual publishers. Public libraries, often set up even in small communities, or their equivalent in the larger cities, gave everybody access to the entire spectrum of culture.
More than 300 years - until the appearance of photography - elapsed before new inventions extended the ability of man to fixate on physical carriers other types of media or to transmit messages across the globe at a speed never seen before. In the last 170 years photography, telegraphy, facsimile, telephony, phonography, cinematography, radio, television, audio and video recording etc. have provided humans with ever new abilities to communicate across time and space.
The invention of printing and of other initiatial technologies had the effect of permitting the dissemination of content, originally multimedia, to monomedia. Before movable type, text and images were combined as illuminated manuscripts. With the introduction of movable type, the mass distirbution of text became feasible, but at the expense of forfeiting the images of illuminated manscripts. It was not until the integration of text and engravings became economic that the expressiveness of illuminated manuscripts was approached. The invention of the phonograph reduced music to monomedia in that the visual experience of the presenters was lost, but made mass distribution of music economic. The invention of cinematography enabled the fixation of moving pictures, but eliminated the audio component of it for several decades.
A major feature of these new communication means is that their use, at least for a certain period of time after their introduction, is not "free". As the use of an invention is needed to store or transmit the information and then convert it back to a form that is consumable by human senses, the inventor is given the right to exploit the invention. In the first half of the 15th century Venice had already reached such a level of sophistication that granting "letter patents" was already a common practice, but in Mainz, where Gutenberg developed his invention, the Archbishop Elector did not and therefore a lost lawsuit could throw the inventor of printing by movable types into extreme poverty. More than 300 years later, however, L.-J. Daguerre, S. Morse, A. G. Bell, T. A. Edison, A. and L. Lumiere, G. Marconi and a host of other inventors took care to register their inventions with their respective Patent Offices and they and/or their heirs gained conspicuous benefits from those inventions. The legal setting of patents is the recognition of an invention as monopoly rights that are granted, as the Constitution of the United States of America states, "to promote the progress of the useful arts, by securing for limited times to inventors the exclusive rights to their discoveries".
Most consumers are unaware that a portion of the money they give the music shop for their favourite Compact Discs goes to the two employers of the inventors, Philips and Sony, and not just to the record companies and the artists. This benefit, however, will not last many more years as the Compact Disc was invented at the beginning of the 1980s. The actual number of years will depend on the specific country and the date the patents was filed in that country.
Printing, and the consequent appearance of the printer/publisher entrepreneur, brought to light an issue that had been staying underground for centuries, i.e. the "rights" associated with a literary or artistic work. Martial, a Latin poet of the first century AD, is credited with the invention of the word "plagiarius", to indicate a person who presented the literary work of somebody else as his. This was obviously considered a reprehensible practice as it affected the right of the author to have a literary work recognised as his. On the other hand copying a manuscript, as we said before, was not necessarily considered an infringement of somebody else's rights.
The invention of printing did not abruptly change the habit of making copies of literary works. Simply, what was done before by amateurs - copying manuscripts because "I need a copy for my own needs" and, by the way, "while I do this I spread culture" - could be done with the tools offered by the new technology, admittedly with some economic rewards for those involved in the process. Some entrepreneurs started doing exactly that, printing and selling books printed by others. The financially sensitive Britons reacted strongly when the Dutch started reprinting and selling books printed in England. The first copyright law, known as the Queen Anne's Act of 1709, was issued to stimulate the flow of knowledge "for the encouragement of the learned Men to compose and write useful books", but the motivation was that some printers and publishers "have of late frequently taken the liberty of printing, reprinting and republishing books without the consent of Authors or Proprietors of such Books". In the Anglo-Saxon world this Act started a process that led to securing the position of printers/publishers in the value chain because copyright aims to protect the economic investment of entrepreneurs. The author is also protected, but in an indirect way.
At the instigation of legislation enacted during the French Revolution, continental Europe has seen instead the prevalence of "droit d'auteur" (author's rights). This protects the rights of authors to receive a remuneration for their intellectual investment. The legal setting of droit d'auteur is twofold. On the one hand it confers to the author monopoly rights of economic exploitation, on the other it also confers moral rights, because the author is invested with the right to have his name, his qualification as an author and his work's integrity respected. This is exactly what Martial already implied with the word "plagiarius", if you consider that the word means "abductor". But because, as was recalled before, artists are not necessarily good managers of their material and immaterial assets, in most countries Authors Societies have been established to protect authors' rights. Next to authors' rights there also are so-called "neighbouring rights", referring to rights of performers. As a consequence Performers Societies have also been established in most countries.
The concept of "rights" associated to a literary work was gradually extended to artistic works "fixated" or imprinted on other carriers. The way rights of the works fixated on the different media are determined obeys a multiplicity of rules and conventions that depend on the way different communities have managed their businesses. For instance, in a sound recording rights belong to the composers of the music; the authors of the lyrics; the musicians for their performance; the designer of the record cover or sleeve; the author of the text on the sleeve; the photographer or artist if the sleeve contains artistic works etc. In a movie the complexity of rights may even be greater, because of the generally much larger budgets involved in the production of the movie that forces the originators of the idea to share the risks with other individuals and the large number of people involved - usually quite special characters (not that people in the sound recording business are not). Common to all physical carriers is the understanding that with the purchase of a carrier imprinted with a literary or artistic work fixated on it the buyer acquires the right to consume privately the work but does not "own" the work.
Books, records and movies needed distribution channels and works flowed slowly from the starting point to the consumer. As a result of the different transactions a portion of the proceeds of sales moves back to the upper portion of the value chain. With the start of radio broadcasting in the early 1920s millions of listeners could receive audio works in no time and with the start of television broadcasting in the late 1930s viewers could also receive audio-visual works. Thus physical carriers ceased to be the only way to deliver literary and artistic works to prospective consumers, with all the limitations imposed by physical delivery. The problem created, however, was how rights holders could get remuneration from this new communication medium, because this acted both as "killer" and "creator" of paying customers. Killer because some customers would not buy any more a work because they had already consumed it via the broadcast channel and creator because some others would buy works because they had been made aware of them.
Most countries took the approach that this medium, often firmly in the hands of the state, was a tool for the promotion of the masses. Part of the license fees that most countries apply to this "public service" were used to pay for the right to broadcast literary and artistic works. For the countries that took the road of "commercial service", broadcasters paid rights holders using the revenues from those posting advertisements on their networks.
With the price of domestic recording appliances - first audio and then also video - reaching levels affordable by the general consumers, anybody could make extended private libraries of audio and audio-visual works that had been received via the broadcast channel. Here, too, the problem already encountered with the appearance of brodcasting arose. The result has been that the ability to make copies of audio and video cassettes of broadcast works did provide consumers with the means to avoid buying the originals on tape from a shop but did also create a completely new channel to deliver audio and video works. In general it is fair to say that these new devices have found a more than decent place in the audio-visual business landscape by creating new outlets for the consumption of works. The compact cassette and the VHS cassette have become very important components of the music and video markets, respectively.
Most European countries grant consumers the right to make private copies, based on the principle that these are not likely to compete with, and so reduce the market for, the original works. At the same time they apply a levy on recording equipment, including blank tapes. The proceeds of this levy go to rights holders in order to compensate them for the loss of revenue by such private copying. In the United States the Audio Home Recording Act grants consumer the ability to make private copies of broadcast music.
The US Copyright Law has adopted the notion of "fair use", whose existence is assessed on the basis of four parameters:
From the beginning "rights" were defined by a country and applied to works in that country The Berne Convention for the Protection of Literary and Artistic Works of 1886 was the first attempt at creating a set of rules with a validity extending beyond national borders. Besides giving a broad definition of "Literary and artistic works", that applies to every production in the literary, scientific and artistic domain using a variety of expressions (Art. 2.1), the 1979 revision of the Convention addresses several key points:
Other international treaties relevant to rights of literary and artistic works are the Universal Copyright Convention of 1952, the International Convention for the Protection of Performing artists, Producers of Phonograms and Broadcasting Organisations (Rome Convention) of 1961 and the Convention for the Protection of Producers of Phonograms Against Unauthorised Duplication of their Phonograms of 1971.
The English Copyright Act of 1709 and other similar laws in other countries were triggered by rampant piracy of books, but with the invention of other carriers holding artistic works the problem of piracy extended to these other media. Until the arrival of the home recorder, however, piracy could only happen at the professional level because of the sophistication and cost of equipment required. Even at the professional level, however, the quality of pirated copies of records and films was much worse than the original. At the consumer level the home recorder became a potential tool for mass piracy. However, the substantial technology limitations of this type of equipment were such that an audio or video cassette would become pretty useless after a few generations of copies.
While broadcasting consolidated its position as the major channel to distribute artistic works just to see recording becoming a formidable competitor, a completely unrelated field - electronic computing - was undergoing a rapid evolution. Originally the computer was as a very large piece of hardware, sometimes housed in an air-conditioned room occupying hundreds of square metres, costing tens and sometimes hundreds of millions of US$. The use of this "big piece of iron" ranged from repetitive down-to-earth data processing to abstruse scientific calculations.
Electronic computing relied on some hardware performing the calculations, an operating system (OS) collecting the basic functionality of a computer, and application programs. The OS was a piece of intellectual property protected by copyright, but it usually came bundled with the underlying hardware as it was made very dependent on it. Therefore it did not make much sense to "copy" an OS. The application software was also protected by copyright and was basically of two types: one supplied by the computer vendor, again tightly connected with the OS, and the other customer-specific, not likely to be of interest to other users or, if it was, the owner of that software had all interest to keep it securely in his safe as it was likely a costly competitive tool. All this in a context of a population of users composed of very large corporations and a wide variety of computer makers, operating systems and programming languages. This landscape did not change substantially even with the arrival of the minicomputer.
The Atari, VIC-20. Commodore 64 and Amiga home computers - names that used to resonate loudly in peoples' minds - did change the landscape. These devices were extremely compact and would not be out of place in many private houses and could run a wide variety of content, mostly video games, produced by small software houses. The target of that content was a population of users who were general consumers counted by the hundreds of thousand if not by the millions. Content was on disks or cassettes and was usually protected because file or disk copying was within the reach of even inexperienced users. Protection was achieved by means of special artefacts added to the distribution media but soon "locksmith" programs began to spread that would break most protection mechanisms. This problem of rampant content piracy found a natural solution with the virtual disappearance of the home computer because of its transformation into video game platforms (Sega, Nintendo) that used hardware modules plugged into the console. On the other hand the Personal Computer (PC) covered a professional, easier-to-control, community of users. It is also fair to say that even amid a high level of piracy, the high growth rates of this industry created ample opportunities for remuneration to the players involved.
The first interactive multimedia applications developed in that professional context with the need to juxtapose pictures, possibly with some background sound or accompanying speech, next to text, until then the dominant component of PC content. This was different from the old models of content consumption because pieces of content developed in different contexts were required to come together to compose a new "multimedia" work. A further departure from traditional consumption models was that the ways the applications unfolded were determined by the way the user interacted with the applications. The problem of acquiring the rights for the different components of the application, each of them typically produced in different contexts by different individuals with the idea of having them consumed individually in separate environments has since plagued the multimedia industry. Some authors societies have developed "single stop shops" where all the rights for a multimedia application can be acquired, but the multimedia industry, by definition with a worldwide scale, can hardly be satisfied with such a piecemeal solution,.
In the early 1980s Philips and Sony introduced a new type of musical disc - the Compact Disc or CD or CD-Audio - that used sophisticated opto-electronic technologies to yield "studio quality" sound with no fluctuations of quality with time. The CD has been an extremely successful device as the number of CD players in the world is several hundreds million pieces. Unlike the compact cassette the average consumer could not copy the CD bit-by-bit, but could only copy the analogue output.
In the first half of the 1990s, the CD-ROM, a computer peripheral utilising the same digital basis of the CD, became a popular means to store computer data and programs. By mid 1990s the CD-ROM had become a standard peripheral for the PC and, soon after, with the diffusion of writeable CDs, it became possible for millions of users to make copies of their CDs preserving the crisp clear sound of their originals or bit-by-bit copies of their CD-ROMs.
At about the same time the audio-visual compression technologies had achieved maturity and they had been standardised (JPEG for still pictures, and MPEG-1 and MPEG-2 for moving pictures and audio). Millions of Video CDs capable of playing an hour of movie on the same physical carrier as the CD, digital television set top boxes, and DVD, a new generation of CD capable of storing two hours of high quality movies, went into use much to the satisfaction of users who could have smaller devices with more functionality and much better pictures and sound.
In the mid 1990s the Internet, a network of networks all sharing the use of a basic transmission protocol called Internet Protocol (IP), achieved a level of usage that extended much beyond the original academic/research/defence environment. HTML, a page formatting standard with the possibility to define hyperlinks to other parts of the page or other pages located anywhere "on the net" created the basis of the so-called World Wide Web, where millions of "pages" of hyperlinked text with graphics and still pictures could be accessed by PCs and displayed using a "browser". The ease of use and immediacy of obtainable results prompted several companies to add other media, such as video and audio. However, the low bitrate available to the majority of users (a few tens of kbit/s) and the high bitrate required by standard-quality audio and video, could only allow very reduced-quality audio and video.
In 1997 the growing number-crunching ability of the PC reached the point where MPEG-1 streams could be decoded by software programs in real time. In a matter of months millions of PC users began to employ the part of the audio compression standard formally called MPEG-1 Audio Layer III and soon nicknamed MP3. With a good encoding software, subjectively transparent quality stereo audio encoded at 192 kbit/s could be obtained, with a reduction in storage capacity of about an order of magnitude. If lower bitrates and/or poorly designed encoders were used, a reduced, but still interesting, quality sound could be obtained.
Users began to rip off music tracks from their CDs, compress them and store them on their PCs. A music file of some tens of Mbytes could be compressed to a few Mbytes and enthusiasts could make "albums" to their own liking. More than that, these enthusiasts could send MP3 files to their friend as mail attachments using the Internet, even though it could take some tens of minutes to send a typical MP3 file over the Internet.
Other Web "enthusiasts" saw more lucrative uses of the technology. By ripping off thousands of music files from CDs, compressing them in MP3 format and posting them on their web sites, Internet visitors could be allowed to download the music files for free. While the file would be slowly downloaded the visitor would be exposed to advertisements posted on the music site. This advertisement space would be sold to interested companies.
Even artists, both well established and new, tried the chance of accessing directly their fans and posted their songs in MP3. Alas, one thing is to use the web as an advertisement medium and another as a place to post wares for free. Once a music file is posted on the web everybody can have it and there is no means to recover it.
The history of the web so far, even considering its current limitations, and the history of MP3 in particular, has taught some important lessons. Artists, apparently, like the idea of being able to access their public directly. Consumers, apparently, like the idea of being able to find on the web the songs of their interest, download them and listen to them when and how they like. Some intermediaries - the lazy ones, I mean - fear the disintermediation brought by the web because they see it as leading to the annihilation of their role. Other intermediaries - the smart ones, I mean - know very well that web disintermediation is only an urban legend, because new forms of intermediation are simply taking over the old ones.
However, moving from a world where the normal way of doing business based on physical things is replaced by such immaterial things as bits, is by no means straightforward. Broadcast digital television is an example of how this can be done successfully. By subscribing to a pay-TV operator one gets a special set top box that gives access to content. The problem is that all the flexibility that exists today in the physical world where one can get a "quantity" discount (say, take three pay two) or one is offered good deals because one is known to be a fan of a particular artist or performer, is lost. This not to mention the limitation of being forced to stay within the offers of a given service provider, unless one is willing to stack a score of set top boxes from different service providers.
In Geneva, in December 1996, the WIPO (World Intellectual Property Organisation) Diplomatic Conference on Certain Copyright and Neighbouring Rights Questions adopted two Treaties, namely the WIPO Copyright Treaty and the WIPO Performances and Phonograms Treaty.
Both Treaties extend some provisions of the Berne treaty and add provisions which offer responses to the challenges brought about by information and communication technologies:
The process of converting the treaties into national law is under way. This includes the US Digital Millennium Copyright Act (DMCA) of 1998 and the draft Directive on Copyright and Related Rights in the Information Society and the draft Directive on Electronic Commerce proposed by the Commission of the European Communities.
It is an exceedingly difficult task to give a general description of the process by means of which a literary or artistic work is conceived, produced, fixated, promoted, handed over to a wholesaler, further transefred to retailers and lastly consumed. This not to mention the repurposing that most content, creayed for a certain purpose, undergoes in its lifetime. There are simply too many types of media, too many physical carriers, too many channels to get the work through the different intermediaries, too many different users of content and too many ways different intermediaries use to conduct business between them, including the way financial transactions are carried out. Here are a few examples:
Business models are in constant evolution, sometimes enabled and sometimes rendered impractical by the evolution of technology. One example is provided by subscription service television. Until the mid 1980s all content that reached the consumer was either directly consumable (books or magazines) or required a standard piece of equipment that everybody could acquire in a shop (radio, TV set, cassette recorders, CD players etc.). The business models that were applied to radio and television broadcasting were limited to public service and commercial service. Subscription services were not possible because anybody could tap an over-the-air broadcast. In the mid-1980s scrambling technologies matured and enabled subscription services to analogue broadcast television. A special type of equipment - a descrambler - was needed to recover the signal in a form that could be interpreted by a standard television receiver. The use of digital television, with its ability to better exploit the transmission bandwidth to carry between 5 to 10 TV programs where only one analogue program was transmitted before, gave further impetus to subscription services.
A second example is provided by a new type of video recorder that uses high-capacity hard disks to record TV programs either on line or off line. In the former case the use will be able to to take personal interruptions without losing the program, while in the latter the receiver will be able to record those programs that are interesting to the user based on his preferences. This new choice changes substantially some fundamental assumptions of the public-service or commercial-service television, e.g. the fact that so far the rights holder granted a license to use his content on the basis of the time the work is broadcast or the fact that those making advertisements may see the number of people viewing their ads dramatically reduced and will therefore become less inclined to use the commercial television channel for their advertisements.
The reduced potential of the Internet of today to carry digital media has so far allowed the implementation of only a very narrow range of business models. A typical use of the web today is for advertisement of products or services on offer. The web page itself is usually "copyrighted" in the sense that the author does not allow the page, as a literary and artistic work, to be used by others. On the other hand the person who commissioned the page is more than happy if the page is copied as widely as possible so as to maximise the reach of his message.
The web has created interesting opportunities to diversify the way different businesses are implemented. One of the latest attempts has used the narrow-band interactivity of the web with the wide-band broadcasting of satellite. The web is used as a means to post textual descriptions of videos on sale to professional buyers of video programs subscribed to a service. When a video is selected, the request is transmitted to the web server. This triggers the transmission of the selected video via satellite in encrypted form at lower-than-original quality so that the prospective buyer can browse through the video. If the video meets the needs of the buyer, a video tape is sent to him by regular mail.
The basic nature of the web eliminates the "where" for the source of content. The content is, to all but the lowest strata of protocol stacks, just an abstract reference, that could reside on a site across the world, or a mirror site closer to the receiver, or on the receiver itself, after being received over night. It also eliminates the "where" of the target. The receiver might be a mobile device, or an enterprise network that spans nations. This frustrates current law, even challenges the concept of sovereign nations, as current law requires the concept of jurisdiction that is "where" for either the producer or consumer of content.
The web also blurs "when". The first generation of receivers with hard discs allow one to time shift when one experiences broadcasts. A Service Provider (who also hosts the user's web page) can collect broadcasts that reflect the user's personal program guide.
The figure below depicts graphically the actors in the value chain.
Fig. 1 - The value chain
The central part of the figure highlights the three principal actors: the provider of content, the (set of) retailer(s) and the consumer device. Adding to these are four major actors:
In general it can be said that between every two boxes there is a flow of content and data associated with the content, a flow of usage rights associated with the content, a flow of technology rights that enable the carriage or consumption of the content and a financial flow that settles the transfer of content and associated rights. Further, regulation often plays a major role in enabling or disabling some of the functionality of some of the boxes.
Over the centuries the way content has been produced, fixated, moved around, sold and consumed has been dictated by technology. That this is not an observation limited to this technology-driven era we live can be seen from the way writing, traditionally done in China and Japan with a brush, is different from the pen-based writing of Western Europe.
Contrary to the past when most content-related technologies used to originate within the content companies or within industries with strong traditional ties with them, today technologies tend to originate from disparate business segments but may have an impact on industries that are often unrelated with the originating industry, an effect of the so-called technology convergence. This has created a situation where ignorance of technologies originated in other industries may turn into a loss because good opportunities may have been missed, In the worse case, the very existence of some businesses is threatened because a new unexpected technology may be used in a hostile way by some other party. This is not a transient situation, because in the next few years the accelerating page of technology evolution will continue to affect the business of content, often in an unpredictable way.
Time has come to make life more normal to everybody and remove the more excessive examples of this technology brownian motion. An effort must be made to create "the big picture" by assessing the range of technologies that are already being deployed by the different industries and their likely evolution, and those that are still brewing in the labs.
The purpose of this section is to make a first attempt at this assessment and show how technologies can be harmonised to accelerate the advent of a more rational world of e-content, that is capable of providing more more powerful ways of expressing content, more opportunities to its economic players, and more rewarding experiences for its end users.
Digitisation of the carriers that can be used for content has been a process moving in an unpredictable, certainly discontinuous fashion. Today the range of delivery media is impressive:
Most businesses are vertically organised with physical delivery used as the technology on top of which they build vertical offerings. The coming abundance of alternative delivery media, however, is likely to render this foundation ineffective, because delivery will become a commodity for all practical purposes. Users will select a carrier (or may be not even select a it as this aspect will be hidden to them) for cost versus features advantages, not because it is part of a vertical offer. It is time for those businesses to face this new situation and exploit delivery in its function of a bit pipe carrying content.
A technology supporting this trend is provided by part 6 of MPEG-4 (DMIF - Delivery Multimedia Integration Framework). This provides an abstraction layer between the applications and three main types of delivery: interactive, broadcast and local disk.
MPEG-2 Transport Stream (MPEG-2 TS) has been designed for the delivery of MPEG-2 Video and Audio data and does a good job of carrying these data over broadcast channels. The proof is in the hundreds of digital television service providers using MPEG-2 TS to serve some 25 million digital television receivers, a huge population if one considers that the enabling technology - the MPEG-2 standard - was delivered in November 1994. Other transport protocols have been designed for other applications, such for DAB (Digital Audio Broadcasting). Lastly ATM (Asynchronous Transport Protocol) and its ATM Adaptation Layer 5 (AAL 5) also provide a transport for digital data in a telecommunication network.
The three protocols mentioned above were designed with the goal of enabling the carriage of data with a real-time constraint. The Internet Protocol (IP), instead, was designed as a means to move digital data across a network of computers for such applications as file transfer or electronic mail that do not require real time operation. IP provides a good solution for the applications for which it was designed as proved by some 100 million computers around the world making use of it. TCP provides flow control functionality and UDP is a good way to send information without flow control, in such environments where multicasting of information is needed.
But important components of e-content - video and audio - do have real-time constraints. And while people have become accustomed to wait for minutes to get their selected songs via file download on the web of today or a few seconds over an ADSL link, they may not always be willing to wait two hours to get their selected movie even if they used an ADSL link. Protocols that can take care of prioritising data depending on some negotiated quality of service over IP are being developed, such as RTP (Real Time Protocol) and RSVP (Reservation Protocol).
In the mobile environment WDP (Wireless Data Protocol) and WAP (Wireless Application Protocol) have been adopted by the WAP Forum.
Even if in the long term a convergence towards a single transport protocol is desired by some industries, the wide variety of protocols in use today requires the application layer to be agnostic of the transport protocol. It is time for the businesses engaged in different transport technologies to face the fact that transport may not be unique and that this can provide interesting service differerentiation.
The MPEG-4 DMIF standard provides an abstraction between the application and the specific transport used.
Broadcasting, the point-to-multipoint delivery of information, is a successful way to deliver content today because interactive networks bring to the general user a bandwidth that is lower than broadcasting's by three orders of magnitude. But, as described in point 4.1, this is going to change and the future abundance of offers will make the specifics of delivery completely opaque to the user. The consumer could not care less if the bits he wants to get (actually he does not even care if they are bits of whatever, he cares about content) come from a cable network or from the telephone wires or via the ether or they happen to be on a local storage device, be it transient or permanent. It is, as in most matters of everyday life, a matter of balance between getting something here and now at a certain price or later and possibly at a different place but at a reduced price.
A wealth of choice does not necessarily make things simpler if we bear in mind that users are not expected to be proficient in technical issues. They should not be expected to switch connectors from wires to cable or to an aerial, nor should they be asked to click the right button of the mouse on icon A in window B or forced to close all files, reboot the system and start a new application with another user name. Users must be given simple and natural means to negotiate the terms of delivery, called "quality of service" (QoS). The technical choices may be "information now or delayed", "streaming or file transfer", "high bitrate or lower bitrate", "interactive or non-interactive", "guaranteed delivery or best effort", "no-error or error tolerant" etc. But these are still technical parameters. The user should not even see these alternatives.
Currently there is no such mapping between the QoS perceived by the end-user, the technical parameters that affect the delivery and the effect that they have on the individual media. Work in this direction has already started in some standards bodies.
By specifying the semantics of the interface to access delivery, the MPEG-4 DMIF standard provides the framework within which the mapping, when defined, can be implemented.
Devices to consume content used to be dull devices. The first phonographs were essentially mechanical transducers from the disc to the membrane and only later they were supplemented with an electronic amplifier. The radio receiver needed more functionality to search for the information the user requested from the dial and had to apply sophisticated - by those times - techniques to extract the information and to convert it to a consumable form. The complex structure of television made television receivers even more sophisticated.
In spite of the word "digital", a digital television receiver is conceptually not much different from an analogue one. But the digital nature of the data processed makes it easier to respond to the clear demand from users to provide more intelligence and better interfaces to consumer devices. And the range of possibilities is almost infinite: the ability to provide a good user interface, information about programs, picture in picture etc. These requests are best satisfied by adding electronic computer functionality to consumer devices.
This is where the mythical convergence between entertainment and computer technologies becomes a clash. In the early days of electronic computers, operating systems were just a collection of frequently-used routines that were shrewdly used to tie users to a given computer platform. This was OK as long as computers were closed system. Even the PC of today, used by many to "communicate", is a closed system where communication between people is possible because everybody uses the same type of processor, the same type of OS and the same applications from the same software company. This is not the kind of model that is going to be accepted by the consumer electronic companies whose customers have been listening for decades to radio programs on any radio or viewing television programs on any television set. In the digital television and multimedia world to come, where there will likely be several tens of processors and OSs, users devices will need to be able to process not just video and audio data but also executable code, without displaying the warning "unknown file format".
Since the mid 1990s Java has been promoted as the technology to enable computers to process code coming from other computers. This requires that every device must have a Java Virtual Machine and the necessary Java classes so that code coming from another computer platform can be properly executed. But in the mean time the Windows platform, in its different versions, finds new adopters every day. The Symbian initiative is developing an OS for the mobile or portable device world.
Whatever the outcome, the definition of an environment where both descriptive and programmatic content can be played back is fundamental to the future of e-content. Users should not be concerned by the physical delivery mechanisms, the transport protocol and the OS environment of their receiving devices. Content and a selected quality of service for its delivery, that ultimately is reflected in its price, and consumption is what matters to them.
In the 1990s coded representation of audio and video has made considerable progress. Such standards as the widely adopted MPEG-1 and MPEG-2 give the possibility to represent high-quality audio and video streams with a number of bits/s that is a good match for many delivery systems in practical use. Further MPEG-4, the most recent element of the MPEG family of standards, supports the encoding of the individual objects in a scene and the compositing of the resulting objects in a virtual space for presentation on physical audio and video presentation devices. By extending VRML (Virtual Reality Modelling Language) to support efficient file compression and composition of both synthetic and natural objects with real time characteristics, MPEG-4 can be used to create content that represents traditional audio and video scenes as well as virtual spaces populated with synthetic humans and other general synthetic audio and video objects.
A file on the web and even a specific book-marked point in a file can be uniquely identified by means of its Universal Resource Locator (URL), a powerful tool to locate resources on the web. Identification of generic audio-visual objects must be achieved, where objects can be both files and streams, be present on permanent severs or on given broadcast channels at given times or may have been copied off-line by the user device from a web server or a broadcast channel to a transient local file for future use.
Content representation presents another, longer-term and very pervasive, face. This is caused by the current evolution towards a world where the number and features of delivery systems and user devices keep on growing. Traditionally the creator of an application could know fairly accurately the type of delivery system that would be used to carry the application, the type of device used to consume it and often the environment in which content would be consumed. But today it is becoming more and more unrealistic for an author to expect to be able to design content knowing for sure which will be the delivery systems carrying the content, the terminal devices playing it and even less the environment of consumption. Scalability - i.e. a way of representing content that lets it scale smoothly on different delivery systems features - and graceful degradation - i.e. a way to represent content that lets it scale smoothly on computing platforms with different, possibly time-dependent, processing capabilities - will have to be designed in content representation algorithms. While in the past content representation algorithms could be considered as largely isolated from the rest of the system, it now becomes necessary to support these features by taking a system view of the problem.
Both features, scalability and computational graceful degradation, are already supported, although in a rather ad-hoc fashion, by the MPEG-4 visual standard. The Fine Granular Scalability extension of MPEG-4 Video currently under development in MPEG will provide a standard solution to this problem caused by the inability of the largely deployed IP protocol to guarantee delivery.
Describing a piece of content is a very general and undefined problem, because it depends on the intended use of the description. Authors and performers have an interest in making sure that their role is recognised, so that authors and performers can be traced back from a piece of content. Producers have an interest is keeping track of the roles played by the different contributors in the work and the corresponding rights. Application developers need descriptions to support their work of making new content out of scattered pieces of content. Retailers want to be able to offer content with all that descriptive information that can entice prospective users to access and consume the content. End users want to have access to all kinds of information that enable an informed choice such as who was the movie director, who were the actors, who was the historical figures they impersonate, when was the work produced etc. Correspondingly, search engines need information to create appropriate categories to respond to user searches.
Even if there will always be "big hits" that everybody will identify with a few words, it is a fact that the memory of these many hits quickly fades in people's memory and after some years describing them so that access becomes possible again may not be easy. In addition there will be many more types of content for which additional information is necessary in order to unravel them from a forest of other similar pieces of content.
As mentioned above, the MPEG-4 standard responds to the need to support modes of content consumption that cannot be planned in advance by its creators. Some users want to consume the content "as is", others may want or are forced to consume only a portion, others may want to consume some content in combination with content from other sources, still others may want to reuse a specific part of that content for their own needs. The expected wide range of content types and consumption modes will require a range of types of information describing the pieces of content.
Most of this information can be textual (title of movie, names of actors, place where the work was recorded etc.). But there are other types of description that are potentially of great interest to the user. An example is the description a person would make to a friend like "do you remember the scene where a robot with the appearance of a human rides a motorbike?" or the humming of an old song that a person looking for a record would make to the shopkeeper when the actual title, the main singer and other descriptions have been forgotten.
A technology providing an answer to this set of needs would therefore enable people to express queries that have to do with the semantic meaning of the audio or visual information. The result of these queries may very well be an object contained in an MPEG-2 movie or in a music file that will need to be extracted from the rest of the information, or an existing MPEG-4 object at a given URL.
MPEG is developing the MPEG-7 standard that will enable this kind of semantic searches. This will become an ISO/IEC International Standard in September 2001.
There is another "description" associated with content that deserves to be treated separately. In today's physical world a book can be read, resold and rented (for free froma public library), a CD can be played, usually not be rented and not resold, a video cassette is is for private consumption and may not be copied (and this is usually highlighted in the first few seconds of the playback). Usage of broadcast material can take many forms: public broadcast because a yearly licence fee has been paid, commercial broadcast because the user undertakes to watch commercials, pay TV broadcast because a monthly subscription has been paid, pay per view broadcast because the event has been paid for.
In the digital domain objects, groupings of objects making up a scene and groupings of scenes making up an application need equivalent "conditions of use", or usage rights, that are attached to the object/scene/application and drive users’ consumption modes. Such conditions of use would accompany the individual objects from the moment they are produced and the ownership is determined and be updated by every actor in the value chain for the subsequent actor, of course without contravening conditions set by previous actors. In moving through the delivery chain every object would undergo ownership and usage rights changes, it could be edited to yield new objects, it could be assembled with other objects to create a compound object or a scene etc. Eventually the rights to sell the application at certain conditions are acquired by the retailer who will put in place a certain strategy, in general devised to maximise some objective of the retailer (cash flow, profit, volume etc.). The consumer will then acquire the application with certain usage rights.
The existence of a standard way of expressing usage rights will greatly facilitate electronic commerce of content in the different points of the delivery chain.
Setting conditions of use without the means to enforce them is not very helpful. One element that enables enforcement is the ability to properly identify content. In the physical world there is always a means to recognise content: a book bears the name of the author and the publisher, a CD the name of the artists and the record company, a video cassette the name of the directors, the artists and the producer. But now we are talking about a world where content is a file or a stream without any means to trace authorship, ownership etc., if not by possibly going to all places where content is registered or making bit-by-bit comparison.
Identification methods exist for various types of physical content: ISBN (books), ISSN (periodicals), ISRC and ISMN (music), ISAN (audio-visual) etc. MPEG has developed identification methods for such types of content in digital form as MPEG-2 (audio and video) and MPEG-4 (multimedia objects). These standards already have provisions for adding information about the authority with which the work has been registered, followed by a number giving the specific registration number of work assigned by that authority. This is a first step towards a solution for content identification but not a fully satisfactory one, as the data kept by the different authorities is not homogeneous. Moreover the systems has the major weakness that people with malicious intentions can easily erase all these pieces of information.
Digital content needs technologies that enable identification of a given piece of content, be it audio, video, synthetic content, computer code etc., in such a way that the identification cannot be tampered with. These identification marks may be used to carry a range of information types. A case of particular importance is owner identification. This is possible because the owner may be the only one to possess the particular technology to mark content, or because the mark is standardised and the payload identifies the owners. Another example is provided by the need to alert the owner or its agents that a given piece of content has been altered. Yet another example is the implementation of usage rights enforcement, such as copy control, as defined by the group standardising DVD-Audio or to discriminate "new" content from "old" content as defined by the Secure Digital Music Initiative (SDMI).
Access to content becomes possible after the rules associated with it have been interpreted and executed. Execution of rules for content carried by traditional technologies can take many different forms. Some of them are implementable relying on customs, some on the good faith of a sufficiently high number of people involved, some on the relatively little advantage on the part of users in flouting the rules, and some because legislation has been enacted that places varying degrees of offence on those who contravene to the rules.
Over the last decades all countries have developed an armoury of legislation to regulate different aspects of this transfer of content from producers to users that we can generally call "content commerce". Today the fast evolution of technology has often removed the technical foundations of many such examples of legislation. Until a few years ago users could only copy the analogue output of a compact disc, but today devices to copy compact discs to a wrietable CD cost a few hundreds US$ and the writeable CDs may cost less than 2 US$. Users could copy movies from TV by programming their video recorders (those who could get through the complexity of the programming, I mean) but they still had to watch the advertisements. Now, with the availability of Personal Video Recorders (PVRs) capable of storing 15 or even 30 hours of TV programs (and many more hours soon), users can record off line a range of programs of their interest and skip advertisements instantaneously. PVRs are considered by some as the biggest destabilising factor ever in commercial television. The MP3 technology enables people to make personal collections of songs, send MP3 files to their friends and even create web sites where people can download their selected songs "for free".
Content protection has been adopted by some industries as the way to make sure that only those who are entitled to access content in digital form can indeed do so. On the physical distribution side, the DVD Forum has issued specifications that make use of encrypted audio and video content. Digital pay-TV services have done the same but so far technology limitations have been such that service providers had to hard-wire their protection systems on the set top box (STB) and give it away to their subscribers. The consequence is that a user wishing to access multiple service providers must get a STB from each provider, unless two of them decide to use the so-called "simulcrypt" system where subscribers to one provider are recognised by the other. To overcome this severe limitation of the user experience, the Digital Video Broadcasting (DVB) specifications have introduced the Common Interface (CI) that can be used, if present in the set top box, to plug in multiple service provider specific modules. Unfortunately the high cost of such modules has relegated STBs with the CI to a marginal role in the market.
The model of creating islands of users sharing a service-provider specific STB has worked in the pay-TV business even though at the cost of a very high initial investment needed to create a sufficiently large customer basis. Acquiring a subscriber may cost more than 1,000 US$, a major share of it being the cost of the STB. This model, however, is hardly exportable to the Internet where users are accustomed to access all sorts of content with just one device, with the added difficulty that so far they have done so without paying for content. Local access with a bitrate that is capable of providing high-quality content, such as that provided by ADSL, will require content protection, or the content will easily lose its value.
Technology evolution has made open access to protected content possible. This is achieved by injecting security at two levels: the first level is at the basic device hardware. One solution is to embed a device identity that is inaccessible to the user. On top of this sits a layer of software exposing secure APIs (Application Programming Interfaces). This security up to the APIs is guaranteed by the manufacturer of the user device, possibly supported by a security technology provider, but is otherwise independent of the service provider. The second level of security is applied when service provider-specific security devices are plugged in or application-specific protection schemes containing the rules to access content are downloaded to the device. Device identity becomes the key element for a variety of needs such as for revocation when devices have been compromised.
The Open Platform Initiative for Multimedia Access (OPIMA) has already defined several elements of an environment where this separation of security functions is realised.
In general, packaging of content in digital form requires the use of some patented technology. When content is purchased, the price usually comprises also some royalties to those holding technology rights, as in the case of the Compact Disc. The digital pay-TV case has more articulations:
The MPEG-2 AAC audio coding standard offers an interesting case study. If AAC is used in a hardware playback device, licensing of AAC technology rights can be obtained on a "per hardware unit" basis. If, however, the playback device takes the form of software given away for free, the rights holders of the AAC patents offer another licensing model, namely the possibility to pay a license fee "per content unit" basis. In this case it is clear that the complete bundling of content rights and technology rights is achieved.
This signals a trend that will become predominant in the multimedia world to come. We can expect to have an application with a plurality of individual pieces of content, each with associated content rights and technology rights, and each possibly protected with the technology selected by the content provider. A device to consume such an application, most likely a programmable device, must have the ability to meter the access and use of the individual pieces of content.
The Intellectual Property Management and Protection (IPMP) of MPEG-4 already provides a first solution to this problem, with the possibility offered to meter the use of both technology and content.
The delivery of content happens through different pieces of delivery systems. As long as content is carried by the so-called "public network" it can generally be assumed that content will not be tampered with by the operator of the network. When, however, content reaches the user premises in general there is no longer a guarantee that content will not be used in ways that are different from what has been specified in the usage rules. Therefore a level of security within the user premises must be provided that will withstand malicious attacks in order to overcome the usage rules associated with content.
Several industrial groups, some originated by the consumer electronic world, such as the Home Audio Video Interoperability (HAVi) and some by the information technology world, are addressing the development of specifications for secure home environments.
The web browser has become a ubiquitous means to access the world of information that is offered by the web. The web browser started as a program to view text pages encoded in Hypertext Markup Language (HTML) that included hyperlinks. HTML has undergone an evolution that has enabled it to include still pictures and applets. By using "plug-ins", i.e. special programs to view specially-encoded information, e.g. small video windows, the browser has become a means to consume a wide range of media. Notable among such programs is the VRML viewer to navigate 3D spaces. A fundamental limitation of the web browser, however, is its inability to manage time information, if not not at the level of plug-ins.
In contrast the world of television, from the beginning point-to-multipoint, has always dealt with time-dependent information, but the information to be consumed has always been prepackaged without any possibility to interact, if not at the level of switching from one channel to the other. Electronic Program Guides (EPGs) provide a natural albeit primitive way of navigating the offers of television programs from a service provider.
There are no solutions known to the general problem of navigating a fully-fledged audio-visual world - imagine a portal where a virtual reproduction of a physical world populated with synthetic humans capable of interacting with the visitor. Completely different interaction metaphors than the ones offered by a web browser or an EPG are needed.
In a world of a reduced number of service providers, where content is expensive and massified, it is easy to establish stereotyped relationships between the different actors of the chain. But if one extrapolates the now consolidated phenomenon of "personal web sites" and the more recent phenomenon of "webcams" one can easily derive a scenario where people will not necessarily be simple consumers but may decide to assume a multiplicity of roles, including those of authors or different players in the chain, like in the case of superdistribution, where a consumer may himself become a promoter of somebody else's content, possibly integrated with some of his own. The number of players in the chain will potentially reach the billions and they will take the shape of content providers, repackagers, value adders, service providers, consumers, resellers etc. The individual pieces of content will be counted by the trillions.
In a world where the number of actors will be very large, the number of pieces of content even larger, each piece of content may be at atomic level, conditions of use will be limitless and therefore the scope of negotiation infinite, the different players have their own individualities and transactions will be worth micro money, how will it be possible to find content and acquire the rights? Certainly not by human intervention! The only possible solution will be to endow each of these entities with intelligent behaviour that will try to emulate humans the best they can.
The Foundation for Intelligent Physical Agents (FIPA) is developing the necessary technologies, in particular the language and the software environment in which intelligent pieces of computer code, called agents, can talk to one another to, say, negotiate a deal or to accomplish a complex task together. This will allow billions of users to access vast content offers through a variety of delivery mechanisms with associated performance levels and costs. These intelligent entities will be able to make good guesses on their owners' wishes as they have observed their behaviour and correlated it with metadata associated with content.
Electronic transactions are developing driven by e-commerce. The complex world of e-content that will evolve out of the different technologies will have to interface with systems handling such transactions.
The last few years have demonstrated how important the role of technology has been in the evolution of the business of content. In spite of this, it would be naïve to think that the only variable affecting evolution is technology. The purpose of this section is to address some issues that, while being triggered by technology or having an important technological component, are under the control of other entities.
Public authorities have always have always considered part of their task to play a role in content. Today censorship is being abolished for most matters in westernised countries but in the past public authorities have supported certain types of content while banishing others. The Christian religion, newly raised to the level of state religion of the Roman Empire, tried hard to suppressed "pagan" festivals. The Muslim religion used to forbid the reproduction of human figures, while the Popes lavishly subsidised the visual artists of their times. The invention of printing triggered the establishment of the "imprimatur" system for books related to religious matters. The invention of radio prompted the public authorities of most countries to put the system directly under their protective wings, and this even in countries that prided themselves for their democratic credentials. Where this was not so blatantly executed, strict legal straightjackets were applied to broadcasters. Broadcasting using the VHF and UHF bands was under such closed supervision that for some time the USA government forbade the import of television receivers that used standards other than NTSC. Broadcasting via satellite had a strange mix of regulation and laissez-faire: in Europe a directive imposed the use of the MPEG-2 standard for TV satellite broadcasting, but remained silent on encryption systems. Other means to deliver content were simply left out of any consideration: cable television, audio and video recording, films etc. remained largely unregulated.
Nowadays the original overriding role that terrestrial broadcasting had played has succumbed to other unregulated delivery media. In many countries, satellite and cable television together have a higher audience than terrestrial television. Package media and pay-TV have evolved from a niche market to very important alternative means to consume television content.
While public authorities still tightly regulate terrestrial broadcasting, in both its analogue and digital versions, the rest of the television landscape has become a sort of Wild West land. No surprise that in all markets the number of satellite pay-TV operators can be counted with the fingers of one hand, if one considers that the investment needed to launch such services can be as high as several billion US$. Consumers, being locked by the STB technology, may have a hard time to move to other service providers. Only timid attempts at setting rules for free access are attempted. This is the case of the US law requiring that after the 1st of July 2000 consumers have free access to to cable television STBs that are service provider independent.
Throughout the history of content we have witnessed several times the appearance of a technology whose exploitation provided new ways of offering content followed by the passing of legislation to cope with the possible misuse of the technology beyond the original intentions. The English Queen Anne's Act of 1709 is probably the first example of the response by public authorities to the challenge that technology brought to the status quo of content. The law had the effect of reshaping the way business of content was done and content was consumed. The situation then remained static until the appearance of a new technology when a new cycle would start.
The awkwardness of analogue technologies tended to relegate the misuse of technology to the professional pirate. When such a misuse was accessible to the general consumer the result had an ephemeral effect as duplicated content tended to rapidly lose quality and therefore commercial value. Because of this, past legislation could consider separately the effect of technology misuse from the "right" of a citizen to access content for reasons other than personal enjoyment. This is, for instance, the case of content "fair use", as embodied in Section 107 of the U.S. Copyright Act, that exempts "limited" uses of materials from infringement liabilities.
Of course we know that, even in the most orderly countries, not everybody is fully law-abiding, that in using content not everybody has in mind the impact on the value of the work as a consequence of its use. Further we know that use sometimes is made not only for educational or non-profit purpose, for factual work and for a small portion of the work. But if analogue technologies are involved, awkwardness is a potent deterrent that helps honest people not to depart too much from honesty. Therefore, if the copy of a portion of a technical article in paper form may constitute fair use according to the legislation of certain countries, large-scale dissemination of the copy is impractical and the principle of "fair use" is automatically upheld.
CD-Audio presents the first exception. Contrary to what used to happen with Compact Cassettes, copying a CD-Audio on a writeable CD yields exactly the same quality as the original. Still, physically moving the CD from one point to another remains awkward and costly.
But other digital technologies are not awkward at all, particularly when no physical devices, such a writable CD, are involved. It may so then happen that the determination of the average individual to stay honest dithers. In the example of the technical article made above, it is very easy to copy a file to all subscribers on an email reflector that specialises on a particular technical area, with the result that all of a sudden the value of the article, which has an expected readership of a few thousands people, has been reduced to naught. If the Internet of today has a bandwidth that can take half an hour to transmit an MP3 file, that time can be cut to 5-10 minutes with ISDN. With telephone subscribers equipped with ADSL reaching the millions in a couple of years, a new hit song illegally posted on a web site could be downloaded by millions of users in a few hours before the site is closed down. At that time the value of the music may have been reduced to naught. Give very few more years and the same will happen to more bit-hungry content such as video.
Clearly something must be done. It is preposterous to stick to a paradigm that was justified by technology limitations that no longer apply to digital technologies. The way the audio and video content industries - DVD and SDMI - have decided to go is by encrypting content. This is the only way that prevent content from being duplicated without authorisation.
This solution will render obsolete another legacy of analogue technologies, the levy on recording devices. If content may no longer be copied unless so authorised, there is no longer a need to apply this levy on certain types of recording devices and media. The same is true of the draft EU directive on copyright. There is no a need to appoint network operators as the cops of the information superhighway if only encrypted, and therefore legitimate, content will travel over the network.
Content protection technologies now become the issue. If content protection will become the way to package content so that it can be traded in digital form, access to content protection technologies should not be discriminatory. The achievement of this high societal goal, however, must be balanced against the critical nature of these technologies. A rogue user may be given access to a technology because of the need not to discriminate him and as a result huge amounts of content may be compromised.
In different times the management of these technologies would have been entrusted to the state. In the 21st century there are probably better ways to solve the problem.
In antiquity the entire body of knowledge of a community was spread among its members and each individual or group of individuals could contribute directly to augment the common knowledge. The schooling system, either sponsored by public authorities or managed by private initiative, was devised to institutionalise and perfect what had been before a spontaneous process. Different grades were targeted to different layers of population. Libraries were devised as a means to facilitate access to the common body of knowledge when its size had exceeded the otherwise considerable memorisation capabilities of people of those times.
Scale excluded, the situation of today is not different from Babylon of 500 BC, Alexandria in Egypt of 100 BC or Rome of 300 AD. Public authorities still consider as part of their duties the provision of open and free access to content, i.e. by allowing free access to books and periodicals in a public library, by using revenues of general taxation. In many countries they do the same with audio and video, by applying license fees to those who own a radio or television receiver. The same happens in universities where most relevant books and periodicals of a given field are housed in a faculty library. The difference here is that this "service" is offered from the revenues a university gets from enrolment fees.
This kind of "free" access, ultimately, is not free at all. Someone has to pay for it, in a tortuous and not necessarily fair way. Protected content, possibly associated with means to meter content/technology use and electronic payment, has the potential to provide a more equitable way to allow access to the body of knowledge that society "owns" to the extent that it considers the access integral with the rights of citizens, while ensuring the rightful remuneration to those who have produced the content and enabled its representation and carriage.
How should public authorities implement that access? Should the richness of information today be considered so large that there is no longer a need for public authorities to play a role in content access, even if citizens must pay to access it? Should they get a lump sum from rights holders by granting them monopoly rights to exploit their works? Should they get a percentage of their current revenues to pay for a minimum free access to content to all citizens?
There is no need to give an answer to these questions because there needs not be a single answer. It should be left to each individual community to decide where to put the balance between the reward to individual ingenuity that originated new content and society that provided the environment in which that content could be produced.
Protected content has other unexpected consequences. Today the Web houses an amount of information, accessible to every Internet user, that has never been seen before in the history of mankind. Because information is in clear text and represented in a standard way, it is possible to create "search engines" that visit web sites, read the information and create indices of that information. Today the indices are the property of those who created them and they can decide to offer anybody access to them.
Again the way things have developed has been influenced by the features of the current generation of the Internet. Content is posted on the web as if it had no value, as anybody can access it, and consume it. As a straightforward extrapolation anybody may process it and extract that value added called "index". As long as the original information had "no value", it is OK for anybody to create that value added. But if content has value and an identified owner, it is legitimate to ask whether this value added processing can be left to anybody or whether the content owner should have its rights extended to this additional information.
If content, as we said, will be protected there will be, in general, no way for a third party to create indices. Then it will be the rights holders themselves who will develop the indices. This, however, has serious consequences on the openness of the information. The categories used by rights holders to classify the pieces of content are not necessarily the same as of a third party who may have completely different views of what makes a piece of content important. A creeping censorship may be put in place based on metadata.
The nature of media into this century has been that the source was known but the target anonymous. In most countries, one could purchase a newpaper at a newstand but the publisher would not know who the consumer was. One could receive a radio broadcast, but broadcasters would not know who their listeners were if not statistically. The transactions were anonymous. With digital media, the transactions, even fine grain patterns such as individual web pages, become observable.
The web has given the opportunity to monitor the behaviour of visitors to a web site. Every single click on a page can be recorded, This enables a web site owner to create huge databases of people profiles. As most of e-content will be traded by electronic means, the technical possibility exists for an organisation to build more and more accurate data about its customers or for a government to identify "anomalous" behavioural pattern within is citizens. Intelligent agent will render the creation of accurate representations of one's self from users' behaviours even more effective.
The use that individuals, companies and governments, or their intelligent agents, will do on such information is a high profile societal problem. This is not specific of protected content, whose consumption will require secure devices, whose security will be guaranteed by device identity.
Public authorities have often used standards to accomplish some goal. Television standards - and a lot of them exist - were often introduced at the instigation of public authorities for the purpose of either protecting the content industry of one country or to flood other countries with content generated within the country. Traditionally public authorities supported the approach of telecommunication standards to have a worldwide applicability because providing communication means between individuals, even across political boundaries, was considered a duty of public authorities. The consumer electronic industry has always been keen to achieve standards for its products because that would have increased the propensity of consumers to buy a particular product because of the added value of interoperability. The information technology industry has shunned standards in many cases: from the beginning IT products were conceived as stand alone pieces of equipment with basic hardware to perform computations, the OS assembling basic functionality but tied to the basic hardware, and applications, again designed to run on the specific OS.
Today, and more so tomorrow, technological convergence will allow any of the four industries to make products that can be used in any of the other industry domains. Some degree of standardisation will be required, but whose standard?
It is clear that a standard developed to satisfy the paradigm of one industry is likely to create an advantage for that industry. So it is easy to conclude that the only way of making standards is to have them at the lowest common denominator, so-called framework standards.
This conclusion is, instead, likely to create an advantage to one particular industry, the one that is least prone to accept standards: the IT industry. A better, and more equitable approach is to have, if so required, framework standards, where from time to time "reference technologies" are identified. These may be selected by those industries requiring concrete technologies that provide solutions to their specific ways of conducting business.
The requirement is that solutions both scale at a point in time (for example to available bandwidth) but also to scale over time. Frameworks anticipate these requirements, but interoperation requires a contradiction: a framework that creates a large solution space, but well known points in that solutions space that all implementations support. The implication is that standards require in concept at least two phases: the adoption of the framework, but then the separate question of which formats and media and transport populate the framework. The implication is that "interoperation" does not mean static designs; quite the opposite, it means evolvable designs.
Often use of reference technologies requires proprietary technologies. For decades it it has been the constant practice of standards bodies to allow such proprietary technologies in standards in exchange of a statement from the rights holders that they are ready to give free licence or to undertake to licence their technologies on fair and reasonable terms and non discriminatory conditions. Unfortunately the sectorial nature of standards bodies have left them behind in this age of technology convergence. In fact a company may be ready to licence its technology with non discriminatory conditions according to its way of doing business, but this may be discriminatory against other ways of doing business proper to other industries.
This paper has highlighted
This paper has benefitted from the thorough review and contributions of Geoffrey Morrison (BT Labs) and James Van Loo (Sun Microsystems).
The opinions expressed in this paper are of the author only and are not necessarily shared by the reviewers.
You are visitor no. since
Monday, 26-Jun-00