Reading .MP3 Compression: Codec, the Network and the ‘New’ Music Industry

‘The Internet is neither a panacea nor a threat for the music industry - the threat is complacency. (…) the music industry cannot afford to take the success of the UK digital economy for granted. (…) [The government must act quickly to help] build a highly skilled and legislated digital music economy.’ Paul Brindley, New Musical Entrepreneurs (IPPR, 2000) [1]

‘Governments should recognize the unique qualities of the Internet. The genius and explosive success of the Internet can be attributed in part to its decentralized nature and to its tradition of bottom-up governance. These same characteristics pose significant logistical and technological challenges to existing regulatory models, and governments should tailor their policies accordingly. (...) Existing laws and regulations that may hinder electronic commerce should be reviewed and revised or eliminated to reflect the needs of the new electronic age’ The White House, ‘a Framework for Global Electronic Commerce’ (July 1997) [2] .

  On January 17th this year, popular tabloid The Mirror hailed the acronym ‘mp3’ as the Internet's most popular search term [3] , which they suggested was surely a sign of the Internet finally ‘growing up’ (The Mirror, 17th Jan: 13).  At varying times throughout the previous year, an ongoing race between ‘sex’ and ‘mp3’ for first place was charted by several publications [4] and the term itself became in many respects, subject to the close and dedicated monitoring of search engines [5] , who it seemed, began watching its every move by the minute.  The phenomenon of mp3, itself little more than an obscure compression codec from Germany [6] , has been subject to similar attention and exposure in online news as the fiercely debated notion of Internet and pornography, as covered by existing media.  Clearly both have been a source of intrigue, whilst producing moments of crisis for existing national laws.  Both events, perhaps by virtue of their controversial carrier medium, have hailed the need for significant changes in policy; changes that have seemingly allowed the technology to take the law into its own hands, pulling the wool over the eyes of the existing agencies who sought to protect a public for what they were about to see – before they saw it.  In both cases, the technology beat the policy makers by a clear distance and neither debate is near being closed.  It is my intention to examine a selection of the debates surrounding the .mp3 compression format, in an attempt to demonstrate an understanding of its role in the creation of a new market, the subsequent need for a reformatting of old legislation and its challenging of long-standing historical assumptions held within the music industry and by those who contribute to it: musicians themselves.

  Before this debate is moved any further, a few significant claims (or assumptions) must be cut into the grooves.  Firstly, I believe the network of the music industry to be chiefly concerned with the technologies (as techniques) of reproduction and distribution.  Here its reproduction functions par excellence involve the recording, production and mastering of sound signals onto a medium for mass-distribution.  Here, distribution works in terms of the availability of the product (in shops) and in its coverage (in print media, radio, film and television).  The Internet, I think, is precisely concerned with fast and efficient reproduction and distribution of bits over networked areas.  If we think of both the Internet and the music industry as striving to meet similar ends (the former in terms of mechanical mass production, the latter in terms of mass electronic reproduction), then the debate is mobilised over the degree to which latter developments had over the existence of the former.  Where the .mp3 [7] compression technique is concerned, however, the debate concerns file-size.  Quite simply, layer-3 compression, through a technique called ‘psychoacoustic masking’ [8] (New Scientist, 19-6-99:34), enables a significant step-down in the size of a stored digitised (or ‘sampled’) waveform (as in that from a CD); uncompressed a 3-minute CD track sampled at its correct frequency, takes up around 30 megabytes.  With layer-3 compression, the track is reduced to less than 3 megabytes (Ibid.) with minimal degradation in sound quality.  Through a combined study of elements of music industry history and some recent case studies surrounding the applications of layer-3 (mp3) compression, it is my intention to examine the some of the ruptures this compression technique has caused, in the wake of increased bandwidth, storage capacity and the general downsizing of technology, within the existing music industry network.

History: The Growth of the Music Industry

  In Performing Rites, Simon Frith (1996) offers a perspective on the ‘authority’ of technology over music.  His analysis effectively covers musical ‘production’ and ‘reproduction’, although Frith offers no detailed exposition of the significant recent changes in methods of distribution; perhaps this is because his work was written before the development of .mp3 compression and certainly before artists themselves had openly expressed interest in the virtues of computer networking as a viable means of reaching - and trading with - their fans [9] .  Frith does, however identify three distinct historical moments of organisation around technologies of musical ‘storage’ and ‘retrieval’.  In the first moment, Frith considers the ‘folk’ stage: where the music is stored in both body and instruments, only retrieved in rituals of performance (1996:226).  Here music was integrated into everyday social practices (as work-songs or lullabies) or cordoned off from the everyday as a ceremonial event (as a narrative story).  The second stage he identifies as the ‘art’ stage: here music is ‘stored’ through its notation.  Similarly, retrieval occurs in performance (Ibid.227), although a distinction is now postulated dividing the ‘ritual’ and the ‘cultural’: ‘the musical mind is thus elevated over the musical body’ (Ibid.).  Here Frith acknowledges a shift between the ‘mode’ of folk towards a structural ‘aesthetic’ appreciation of music: in his first model, stories are passed on through a medium of human observation; hence popular folk ballads are re-elaborated by each new performer’s take on the piece, perhaps often recalled from memory or extrapolated from earlier teachings. 

  Frith’s ‘art’ stage motions toward the integration of a standard grammar (or syntax) into conventions of storage: here Frith’s ‘art’ stage acknowledges a notion of authorship, where a composer - by virtue of the conventions of notation - produces a manuscript encapsulating a blue-print (or framework) within which their work can be interpreted.  In this stage, then, a re-elaboration of the folk ballad is replaced by an interpretive reading by the scholar with a surmounting knowledge of the theory of notation (and the essence of the piece is mapped for the performer by the mathematical [mechanical] principals of notation theory).  Although Frith makes no attempt to incorporate any historical rubric (or schemata) into his division of these stages, I use his ‘art’ stage to argue that notation brought with it the ‘composer/artist’; that the formulation of music into standard notation brought with it the recognition of the ‘author’ as the creator of the piece [10] .  In performance terms, however, the meaning of the notated ‘text’ is by no means fixed; becoming subject to the interpretation of the performer, who considers the notation in accordance with a host of other parameters: sympathy with the composer, control of the instrument and notational understanding.

  For Frith, the third stage - the ‘pop’ stage – saw technological processes of mechanical [re]-production and distribution as allowing for the mobility and availability of music.  This, I would argue is where conventions of copyright were first placed onto music [11] , alongside notions of ownership (the right for the artist to be recognised as ‘creator’ and for the publisher as ‘owner’ of its copyright).  In this stage, the ‘material experience of music’ (Ibid.) transformed the piece into a commodity (as object or possession).  Here music is captured (produced), stored on a mass-replicable medium (phonogram, disc or tape) for retrieval via a suitable decoding medium (stylus, laser or magnetic ‘head’).  Subsequently, the performance element (and the live spontaneity of the first two stages) is now subjected to the practices of technological production; for the artist, perfection (through takes, splicing etc.) can be now achieved in a studio, in a quest for the perfect cut.  Through citations of an interview with Glenn Gould in 1966, Frith remarks that new developments in technological reproduction ‘set aside the usual “hazards and compromises” of concert performance’ (Ibid.), and greater degrees of precision became part of studio technique.  Over time, increased studio use of magnetic tape articulated a shift from the ‘art’ stage of mastering (or capturing) a live ‘recording’ into a process of editing ‘records of ideal, not real, events (Frith 1987:65).

  If we are to adopt Frith’s three stages, then it is my suggestion that a new wave of digital technologies are articulating a space for a possible fourth stage; all processing is computer-aided from the stages of studio capture (with digital mixing, digital sample editing, sequencing and storage), through to performance (MIDI interfacing between digital instruments and ‘live’ sequencing of digital samples).  As Paul Brindley’s IPPR report suggests: in the wake of an imminent ‘fully networked environment’, the music industry must strive daily to achieve an ultimate end game where the ‘seamless delivery of music’ is fully incorporated into the process of music distribution (2000:i).  On both sides of the fence, general changes in operation within the industry are needed in order for the alleviation of the current fear of damage caused by piracy, itself resulting from changes both in hardware and software. The role of the musician/artist/composer will also be redefined in light of this: initial understandings of both instrument and notation will be accompanied by a surmounting knowledge of digital production practices and digital sequencing skills.  For an industry now primarily concerned with technologies of electronic [re]-production, both content and delivery are in a state of transition as I hope to demonstrate through case studies.  In consideration of new changes, it is vital to consider the ways in which the industry has developed; an historical study demonstrates a repetition of similar forms of upheaval and destabilisation; patterns of change have duly occurred in the contexts of musical content and delivery, however for the majority of this discussion, my emphasis will largely focus upon the latter aspect.

  In terms of ‘pop’ content, then, several writers have acknowledged the partial shift of music towards an amalgamation of the ‘familiar’ and the ‘new’ into the body of the song.  In Will Pop Eat Itself? Beadle (1993) considers the growth of the new waves of pop music, focusing his enquiry on the emergence of the artists of the ‘soundbite era’ who utilise digital sampling/editing to create ‘new’ structures from groups of ‘familiar’ sounds.  The development of digital audio capture and sequencing changed the parameters within which these musicians worked; here one of Beadle’s numerous examples is the growth of the 1980s band Pop Will Eat Itself and their use of the digital sampler to create music ‘collages’ [12] by cutting and pasting from the old and familiar.  This has lead to haphazard forms of music, whose origins can perhaps best be traced to what Baudrillard termed a ‘free floating mix of signifiers’ [13] , or, as Mark Poster’s ‘third stage’ of human history explains, the displacement of the representational character of language, which has led to a cutting (or a displacement) of the ‘referent’ from language, rendering the subject/self ‘decentred, dispersed, and multiplied in continuous instability’ (1990:5).  As the ‘pop’ landscape that Frith describes converges with digital technology, new representations of pop become available: digital music is accompanied by digital pop videos, A&R finds new terrains for talent scouting in home-brew demos and videos made available on the Internet.  Fan sites amass online and following in the footsteps of Lara Croft, virtual pop stars are being produced and tested [14] . 

  As Walter Benjamin explains, technology moves us ‘toward a certain form of art’ (1955/1992:242).  One of art’s foremost tasks, he argues, has always entailed a continual process of working towards the ‘creation of a demand which could be fully satisfied only later’ (Ibid.230).  We can see, perhaps without moving too briskly towards an exposition of the historical stages involved in this process [15] , that one of the most prevalent factors crystallising ‘the .mp3 debate’ is precisely that a new satisfaction of demand is being created: only one where a domestic technological development has temporarily authorised the bypassing of many of the music industry’s own channels, illuminating the need for it’s own (arguably perhaps imminent) re-evaluation [16] .  Before I progress, two significant points must be brought to light: firstly that the present debate reproduces an earlier moment in music history: a similar moment of crisis, occurring with the introduction of the C90 audiotape, which unleashed a subsequent fear throughout the industry that mass-music piracy was about to erupt.  Throughout its incubation period, vinyl albums were frequently printed with a logo of a cassette and crossbones, sporting the words ‘Home Taping is Killing Music’ [17] .  Secondly, that such repetition of effect can come into play within the industry, only to be re-iterated at a later stage by the next consumer ‘liberating’ technology prompts us to the question the nature of the music industry.  By nature, the music industry has always been chiefly concerned with technologies of reproduction and distribution; and as ‘new’ technologies (or new software) are developed (or made available) for the consumer market, the music industry – in order to survive – clearly must remain several steps ahead.  As Brindley’s proposal argues at outset, three new models of music distribution must be set into place, all to be implemented at different stages on the route towards a fully on-line music industry [18] (See Brindley, 2000)

  In Illuminations, Benjamin argued that film, as a new technology, was important in part because it involved ‘the liquidation of the traditional value of the cultural heritage’. (1955/1992:215).  The Internet, primarily a technology concerned with effective distribution of information [19] , is important as a new technology in part, precisely because it involves the liquidation of the traditional value of the ‘cultural heritage’ of music.  Essentially these remarks reflect Aragon and Sampaio’s (1998:153-4) argument that the new ‘mode of information’ produced by digital technology acts as a ‘reductive’ force, insofar as the shift from analogue to digital articulates a separation of the imitation of both ‘the [instrument] sound’ and a ‘sense of the recording moment’.  Digital mastery, by virtue of the re-sampling of an analogue tape (which was present at the time of ‘conception’) and re-mapping into a chain of binary digits, harbours a differing set of values for their audience; perhaps as a ‘third-step’ process of filtering from Benjamin’s unique art object (that with ‘aura’) to the mechanically reproduced object, to the digitally re-mastered mechanical product.  Although in many respects I find Aragon and Sampaio’s observation holds some weight, I believe there is a degree to which this is true – and they admit this – is there really that much difference between hearing the same song on vinyl, then on CD?  For most, I’d suspect, there would be little difference, however their arguments are framed within a discussion of gender differences in on-line ‘chat spaces’.  In relation to the physical presence or absence of a person (rather than placement of an analogue recording device in the studio), the division between ‘real space’ and ‘chat space’ (or between body and avatar) invokes real notions of difference [20] ; we could equally draw this comparison in relation to difference in perceptions of temporality between telephone and letter as culturally both mediums signify in different ways.

  MPEG Layer 3 [.mp3] compression, then, can be distinguished as a representative exemplifier, as an enabling factor for a process of liquidation (of the ‘traditional value’ of music) to take place.  As the cultural heritage of music is thus redefined (along digital lines), its’ notion of ‘traditional value’ is duly redefined, perhaps now ordered as a series of selective catalogues (in much the same way as the ‘classic’ radio stations, who serve their listeners with an historically-selective range of ‘oldie’ songs).  In this respect, individual songs - despite a uniformity of format (.mp3: 44Hz, 128Bps) - are unequal in terms of ‘server presence’ as some (the more famous) will be featured more than others.  On reflection, Benjamin uses mass-produced historical texts – and more importantly, their filmic ‘takes’ on historic moments – to best illustrate the process of liquidation [21] .  Regarding .mp3, then, this process seems to have curiously taken up a different form of liquidation: that of the ‘traditional value’ of the music industry itself.  Although I do believe that layer 3 compression allows us to think about music in truly new ways, I am less inclined to subscribe to the ‘doomsday scenario’ offered by Culture Secretary Chris Smith (see footnote #11); while there can be no doubt that his statistics offer a representation of the scale of .mp3 distribution, we should remember that thousands of CD players are still sold [22] , although this is complicated by the fact that CD-R media is becoming more commonplace and therefore a sharp rise in sales of CD ‘players’ could be attributed to the rise in CD-R units sold.  Also, as Hi-Fi ‘accessories’ wear, from experience, it seems as if the majority of CD players manufactured seem to break faster than any other component.  Although I have no statistical evidence to support this, I have noticed that popular Hi-Fi store Richer Sounds charges more for 3-year ‘super-care cover’ over these devices.  Finally, a new wave of Hi-Fi recording devices (CD-R, MiniDisc, DCC) are now available and these units often have a digital input for recording data straight from CD.  Subsequently, CD players without digital ‘outputs’ would be need to be replaced by home users wishing to ‘cash in’ on any of these new technologies.

CDs: The Death of Vinyl?

‘New media supplement, rather than replace, the old’ (Collins and Murroni, 1996:141)

  One of the key debates surrounding the CD was that it was smaller and tougher than vinyl.  However, this redefined vinyl’s role as the DJ’s medium and sales of technics turntables in the past five years, for example, have shown no signs of weakening [23] .  Indeed, Thornton (1995:47) remarks upon the growth of ‘disc-dance’ venues in the 1980s as being an exemplary indicator of trends in youth-culture.  With specific reference to the development of CDs, she discusses the generally-perceived receptions gleaned from the two formats: CDs on the one hand, were seen as ‘cold’, ‘clinical’ and ‘inhuman’ whereas Vinyl seemed so frequently to be diametrically opposed to this format: the vinyl DJ was often considered an ‘artist’ and a ‘musician’ working with a ‘warm’ and ‘real’ medium, as opposed to the ‘technician’ working with the laser and aluminium. (Ibid.64).

   Fascinatingly, then, the emergence of the hyper-piracy-friendly medium of .mp3, seems to chime in with a documented increase in sales of Compact Discs [24] .  Although this can be partially explained by Collins and Murroni’s IPPR report, it can also be attributed to economic factors and, perhaps primarily to a general growth in e-commerce ‘entertainment’ sales, with the upstart of companies such as Amazon.  The introduction of commercial CD-R units would equally warrant a growth in CD-R sales; can we draw a distinction between sale of the two formats (both of which are CDs) and how frequently is a distinction drawn between audio CDs and CD-ROMs?

Case Study I: Super Media? The Convergence of EMI, Time Warner and AOL

  In the wake of the merger between Universal and Polygram earlier last year, EMI announced their plans to merge with Time Warner just two weeks after the latter announced their merger with AOL.  The successful merger of both the British EMI with the US Time Warner, together worth an estimated $20bn [25] , would guarantee to create the world’s largest music company while reducing the ‘big five’ Keith Negus identified as owning 70 per cent of the recorded popular music sold in the world [26] into four key players.  Indeed the late 1980s and early 1990s were a time ‘of frenetic merger and acquisition activity in the music related industries’ (Negus, 1992:4), although in the late 1990s, what came to constitute ‘music related’ would also cover computer-related ‘interactive’ media companies.  In the US, online music sales are projected to grow ‘more than tenfold in the next five years, to more than $2 billion annually’ [27] .  Such a significant statistic has undoubtedly set the tone for a next-wave of mergers, as Negus observes, the term synergy ‘refers to a strategy of diversifying into directly related technologies and areas of entertainment and using the opportunities that this provides for extending the exposure of specific pieces of music and artists’ (Negus, 1992:5).  In part, synergies also function on a technological level; perhaps this merger proves this.  Convergence, it could be argued, is now reaching a stage where potential exists for computer software, video, discs and tapes to be combined and used on one multimedia self-contained home entertainment system; Time Warner, for example, set aside $500m last year to invest in new online efforts, before opting for the tried and tested techniques offered by the services gained from its merger with AOL [28] , in addition to an amalgamation of customer services and client details.

  Historically, such merging is not unusual for media companies.  In the case of the music industry, larger ‘umbrella’ companies with wider interests own many of the companies (See Negus, 1992:2).  For many companies, large or small, the developments articulated by new e-commerce markets lie high on business agendas.  In terms of the ‘online entertainment’ sector, the arrival of broadband technology lies at the heart of the e-commerce ‘revolution’.  With specific reference to digital audio, moving vast quantities of such content over the web will not be possible, likely or desirable while the majority of Internet users are dialling in on a 56Kb (or slower) modem. Perhaps this factor is what made the AOL-Time Warner merger such a perfect combination [29] .  EMI recently predicted that 10% of its sales would be generated via the Internet within five years and had already begun making investments in a series of web-based music firms [30] .  The combination of the three companies, it has been argued, ‘makes sense’ geographically: for its influence and coverage over both the US (Warner) and the UK (EMI), for a global market-share of 20% of the music industry [31] , and, most importantly, to enable the acceleration of the convergence digital distribution platforms across all media [32] .  The merger works on several levels and it is perhaps worthy of a brief discussion of the mutual benefits gained by the amalgamation (convergence) of these companies into a global corporation with a powerful ‘online’ presence.  The key is that EMI Warner are granted a huge new potential customer base from the 26 million people who subscribe to AOL [33] to have them (as the world's biggest internet service provider) guide them through cyberspace.  Combined, EMI Warner has a strong customer basis for its vast multinational portfolio of musicians and artists.  The potential for providing almost limitless entertainment via the Internet has perhaps been the topic of discussion for some time, but it seems this is the year when the business world will start to make it happen.

  However, it faces significant problems it must first overcome: issues concerning intellectual property online and, perhaps more significantly, security.  Both of these issues will be re-employed for my final case studies.  As Brindley suggests, the key driver for change in the digital economy is technological convergence (2000:13), a ‘process by which previously distinct information delivery systems appear seamless from the end user’s perspective’ (Ibid.). This combined merger provides further evidence that a digital economy is looming, although at present, MPEG layer-3 technology fails to provide adequate security measures to endorse this process; a digital economy is only worth what a consumer is willing to pay for it, and mp3, while liberating for consumers, semi-professional musicians and music hobbyists alike, lacks any security stringent enough to allow real e-commerce transactions to take place.  However, with the slow reactions from within industry, problems are beginning to occur.  There are several ‘secure’ formats in development at present [34] and a universal agenda - the Secure Digital Music Initiative (SDMI) - has been set into place by the ‘big five’.  Amendments have been made to the copyright act [35] , although these tend to look like they protect the record company’s interests rather than individual artists.  Of all the formats available, I have selected two which look like they may be feasible alternatives to the by now standard, mp3.  Firstly, Microsoft has a rival format – as an extension to their Windows Media Player 4 suite [36] - which they argue is secure, of higher quality and smaller than mp3.  By virtue of Microsoft’s ‘brand name’, its power and influence, this format is a possible future ‘standard’, one that can be utilised now.  However, Brindley (2000:14) mentions one other format for which limited technical specifications are already available [37] : MPEG7.  At the time of publication, little is known about the format, but as a member of the MPEG-family, there is equal promise for such a codec; both allegedly satisfy SDMI standards, although meeting ‘security specifications’ is not a marker of overall infallibility.  I will return to a discussion of both codecs in later case studies.

  Secure Online Music will be, I think, an acid test for the ‘future’ of digital distribution; the key issue at stake here is bandwidth.  As we move into higher-speed broadband Internet access later this year, the success of the mp3 format clearly demonstrates that music is particularly adaptable to this mode of delivery.  Firstly, it doesn't demand the bandwidth requirements that film or television will expect, so it will be used as a template – a kind of beta test – in the first-wave world of entertainment online.  There is no question that in a ‘technologically weighted’ Western sense, music is a global phenomenon, and one that can also be distributed into a broad array of devices.  Recently there has been much consolidation within the ‘telecommunications’ sector; many companies have combined forces and BT has just undergone a major shake-up to enable it to deal with service provision for the emerging new forms of Interactive media, many of which it will be ready to offer the existing network for.

Case Study II: The Rio Player – MP3 From Desktop to Walkman into PDAs

  As Paul Brindley explains, digitisation is most easily understood as the process of converting all information (text, audio, graphics and film) into a language that is understood by computers (Brindley, 2000:13).  The combination of digitisation with powerful compression techniques such as MPEG layer-3 have partially implied wide-ranging consequences for the industry as a whole; these have so far been represented in several legal battles; one such example can be found in the study of Diamond Multimedia’s release of its Rio layer-3 decoding walkman [38] .  Ithiel De Sola Pool (1984) described the then new-wave of digital media as representative of a potential shift in power, one that offered radical implications for the consumer: these were called ‘technologies of freedom’.  Indeed, one of the fundamental changes the Internet introduced was to make available inexpensive means for publishing and distribution on a global scale.  This means that composers and performers can produce music using home technologies [39] and that a distribution network is set in place, theoretically allowing their products to be distributed directly to their fans, effectively bypassing traditional music distribution channels completely.  On the other hand, the ‘liberating’ aspects of the software and hardware mean for those who believe that “information wants to be free” have a perfect tool for piracy and the dissemination of bootleg copies of songs.  Needless to say, the music industry – and for the time being, it’s artists – are concerned, and would like to see an end to all unauthorised mp3 activity.

  Before I launch into my case study I would like to consider some of the debates surrounding (or alternatives to) mp3’s and security.  Brindley is quick to remind us that ‘there are over 11 billion unprotected hard carrier sources in circulation at present – in the form of CDs’ (2000:37).  Every audio CD in circulation contains unencrypted, unprotected digital music that can be extracted or copied and converted into computer audio files or copied onto another disc [40] .  Little can be done about these CDs – the audio data pressed into them is completely susceptible to DAC (Digital Audio Code) ‘ripping’ [41] through a multitude of open-source, shareware and commercial digital audio readers [42] .  Once the DAC information is extracted, hardware CD-R or CD-RW ‘burners’ can be used along with almost any CD-Recorder software [43] to clone an ‘original’ audio CD, write the extracted DAC code onto a CD-R as audio, or store a quantity of .mp3 files as computer data.  If all this is taken into consideration, then there is, I think, a genuine scenario for the music industry to consider: where do these new and liberating technologies leave it now?  What is to be the role of traditional music publishers and distributors?  This phenomenon has been called ‘disintermediation’ (See: CSTB, 2000) [44] : the elimination of middlemen in transactions.  If authors, composers and performers alike are able to publish and distribute online themselves, surely publishers (as middlemen) are unnecessary?  Whenever we consider this debate, we are always drawn back to the fact that the music industry is a long-standing institution and functions in tandem with a variety of existing media.  Recently, film soundtracks seem to have articulated their own space within the Top 40; audiences watch the film then buy poster and CD.  If a radio station regularly plays a song over a number of weeks, this is likely to reflect in both its popularity and sales.  If anyone can create or publish, content will proliferate, producing a condition of information overload.  The problem for the consumer will not be to obtain, but to filter the deluge of information housed by the content-rich world.  Clearly in light of this, the music industry will need to redefine its own terms, although we can suspect many long-existing parameters to remain embedded; in this sense the job of an A&R representative will remain the same: they will still be listening to demo-songs, only perhaps online, rather than on tape or CD.  What is clear is that the emergence of a rogue digital compression format has caused a crisis in identity for the present music industry: it is therefore hardly surprising that battles are already emerging over the future character of the industry.

   In October 1998, the Recording Industry Association of America (RIAA) sued Diamond Multimedia, maker of the controversial Rio Walkman.  The suit purported to block the sale of the device.  The Rio, it argued, had violated the Audio Home Recording Act (AHRA); the device did not contain a serial copy management mechanism, meaning that copies, when made, could be moved from a player back to a computer with no security concerns.  The appeals court upheld the case on the grounds that the Rio was strictly a playback device (CSTB, 2000) [45] and there was no way of transferring information from it to another ‘walkman’ device.  The AHRA ensures the right of consumers to make audio recordings (either analogue copies or digital clones) for private, non-commercial use (Ibid.).  The case of the Rio solidified the first part of the legislative process for layer-3 digital walkmans, whilst also highlighting key issues at the core of the debates surrounding digital music.  Attacking the Rio for enabling easy .mp3 file transfer (and hence piracy) has proven to be a futile exercise; the device alone is unable to transfer files without the aid of a computer, which as a device, is conveniently configured outside the influential parameters of the AHRA [46] .  In the case of the Rio, a second loophole was also exposed by this case; that as a device it could be considered (from a consumer perspective) as a technology which makes portable, or ‘space-shifts’ files that already reside on a user’s hard disk: ‘Such copying is paradigmatic non-commercial personal use entirely consistent with the purposes of the act’ [47] .  What is however clear, is that while the RIAA’s attempt to ban Rio mp3 walkmans resembled an attempt to address the ‘problem’ of piracy, it did this in partial acknowledgement of a deeper failure to control piracy elsewhere [48] . 

  The Institutional Federation of Phonographic Industries (IFPI) claim that online piracy is ‘manageable’ (Brindley, 2000:39).  Yet, if online piracy is ‘manageable’ surely the music industry has nothing to fear?  If we consider this claim, we find several issues enmeshed within it.  Very little piracy occurs on the World Wide Web itself [49] .  As the web develops, so do newer generations of improved search applications [50] .  The American Society of Composers, Authors and Publishers (ASCAP) have their own tool (in the form of a ‘bot’ called ‘E-Z Seeker’) which when set to work, will scan through servers on an area of the web in search of ‘illegal’ mp3 files.  By their nature, then, the so-called ‘underground’ web-sites, are created to be found by ‘users’ (as casual pirates), although by this token they are thus traceable by ‘rights owners’ through precisely the same means.  On the surface, ‘bots’ can help keep the number of mp3s ‘online’ to proportions of a reasonably ‘manageable’ scale, although their real fears are not bound up with the web; as I hope to demonstrate shortly with my next case study, we are shifting from the realm of technologies of freedom to free technologies with utmost discretion encoded.  The RIAA’s failure to gain control of the Rio walkman has, if anything, taught the music industry a valuable lesson: that a void needs to be filled – ultimately constructing a division between the mechanics of commerce and the tools for hackers (or enthusiasts).  In this arena there are many noteworthy developments to discuss, we can consider the emergent SDMI-compliant formats such as Microsoft’s Audio codec and MPEG7.  All new waves of PDA notebooks come with mp3 decoding almost as standard; a decompression routine can be added to a device cheaply and with relative simplicity.  When all major WAP-phones and PDAs are likely to be fitted with decompression, where in light of the Rio case, these devices will sit comfortably outside existing RIAA guidelines, a plea for a secure format becomes of greater concern still. Dyson suggests that Playboy magazine began to watermark its images so that they could at least have some means of protecting them from re-publication (1997:139), although something considerably more ‘secure’ will be needed in order to police a future music format.

Case Study III: Napster and its GNU Cloned Friends – Enter ‘Killer’ Apps

‘Now we face a new situation: not only is it easy for individuals to make duplicates of many works or to re-use their content in new works, but the physical manifestation of content is almost irrelevant. Over the Net, any piece of electronically represented intellectual property can be almost instantly instantiated anywhere in the world’ (Dyson, in Wired, 30.7: July 1995) [51]

 ‘Internet piracy is no different than producing counterfeit compact discs. Theft of creative rights is neither fair nor legal, no matter how you do it, and we have been working aggressively to protect the rights of artists and recording companies on and offline’ (Frank Creighton, Director, RIAA) [52]

  It seems clear that complex issues stem from the divide between physical and non-physical space.  There is one rule for the space our bodies inhabit physically; standing on our own land we indeed know our rights.  In the spaces we reach through broadband connectivity, however, our rights suddenly seem different; the Internet muddies many of the basic democratic principles we expect in the spaces we inhabit ‘in real life’, as it were.  Online, even the question of where things happen seems to me a matter of ‘physical’, social and legal convention.  Indeed, the majority of ‘property rights’ debates in cyberspace are debates in rights concerning intellectual property [53] .  Legal scholars such as James Boyle [54] would point out that all property rights are socially created, limited in extent and qualified in relation both to certain types of actors and certain types of conduct.  Even with a house or a car, the solving-idea of property is indeed problematic, but with intellectual property its problems are just easier to see.  Boyle uses an example of the dangers of html piracy: on the side of his home page is a small .gif image with the message ‘click here’ and ‘to duplicate’ set above and below the image [55] .  Writer Jon Katz poses a similar question in his discussion of the recent Digital Millennium Copyright Act (DMCA), opening with a reminder that online transactions are virtual, hence not traditionally associated with property: ‘If you loan a CD to a friend who burns himself a copy, have you stolen from the music industry? What if your friend would never have purchased that particular CD anyhow?’ [56] .  Although different in approach, both writers hold the DMCA in dispute, similarly deriding it as an example of how ideas concerning physical law and culture can be ‘unthinkingly applied’ to a new type of reality.  Where the ‘work of art’ shifts from the age of mechanical reproduction towards an electronic age, the now aura-less ‘physical’ product begins to exist increasingly as a soft-structure of digital information on a carrier medium; precisely where and how its cultural ‘value’ in this form can be measured seems to have everyone baffled; including Dyson, Katz and Boyle. 

  Jean Baudrillard explains this shift as a process of inversion; one where the ‘medium’ becomes a determinate element in the exchange insofar as it begins to dominate over message, content and subject to a point of saturation within the field of the medium:

‘where what was a means of communication gains a kind of finality, and possibly a counter-finality, and then the strategists which revolve around the medium, the communications media, become more essential than the strategies which concern the contents’ (Gane, 1993:145).

  After being set into place eighteen months ago, the DMCA primarily wielded the power to shut down the controversial music-sharing Napster [57] software at colleges across the US.  Although on many campuses, the software’s ability to shift vast quantities of .mp3 data placed heavy strain upon network servers, causing congestion to the point of crashing; the program’s server-disabling factor offering reason enough in its own right.  In many respects, the software’s heavy demands on resources appearing to be the only significant concern to the US colleges.  Indiana University banned Napster for precisely the reason that it was sapping too many of its’ resources: accounting for up to 60 per cent of all traffic between the campus and the Internet by 12 February, when it was banned [58] .  What followed was a joint collaboration between the technicians both at the University and at Napster, which led to a significant breakthrough in curbing network congestion and an updated software client.  After the improvements had been set in place the ban was subsequently lifted; both parties involved felt an obligation to make their revelations public (open-source) and the code was submitted in application for its registration as an ‘Internet-standard’ specification (Ibid.).  Fascinatingly - although perhaps unsurprisingly – some of those students who were banned from using the software, considered it a blight on their constitutional ‘rights’ [59] , a debate which carries historical resonance within the music industry, especially in the US (See Appendix 1)

  As bandwidths increase, file-sharing software with functions similar to those of Napster are likely to become more commonplace; indeed the attention the software has received of late has led to production of an army of clones [60] on several platforms, especially in the GNU/GPL arena of ‘collective ownership’ (see Walsh, 1999:170), on open-source platforms, such as Linux.  Many of Napster’s concerns are to stem, as I will shortly explain, not primarily from the fact that it is a potentially damaging piece of software; it’s press tells us this much, but because it is a company and, as such, can be made accountable; Napster may have started out as a college programming project undertaken by a 19-year-old US college student, but it’s now a technology corporation ripe for investment in every aspect, aside perhaps from the chain of pending court cases it is accruing. 

  Although Napster has been the ‘killer app of the year’, gaining notoriety by artists, colleges, the RIAA and the music industry as a whole [61] , recent evidence supports the view that its six months of fame are coming to an end – a point which I will return to in light of the recent Metallica court case.  However, the implications for the software of this nature spread deeply across several debates.  One of many of Napster’s GPL open-source clones is to take the ‘piracy’ controversy a step further: Gnutella, which unlike Napster, doesn’t rely on a centralised directory for connecting users, meaning no business to invest in and no one to sue, except perhaps its users.  It was an application created by Nullsoft, the programmers of WinAMP (which is believed to be the most popular Windows .mp3 decoder around) [62] .  On March 14^th this year, Gnutella was released [63] , privately, and much to the dismay of Nullsoft’s owners: the AOL/Time-Warner conglomerate – also one of a long chain waiting to take Napster to court – who acted quickly to force the removal of the site (barely 24 hours after its official release).  Sadly for the multi-billion dollar corporation the dye had already been cast and those copies downloaded within this time were reposted almost as quickly as the official site was closed.  As an open-source project, its binaries were publicly available at outset, articulating a space for the development of similar software.

  Gnutella’s implications run deeper still, sparking a debate in line with another key fear surrounding Internet-based technologies: the pornography debate.  Previously I considered the debates surrounding a transition from technologies of freedom to free technologies with utmost discretion encoded; that the development of the web heightened debates surrounding fears of privacy and security is a well-trodden territory [64] .  MPEG Layer-3 files tend not to survive on the web because they can often be traced easily to a specific IP address or server, this argument can to an extent, I think, hold true for debates surrounding highly illegal images on the web, although insensitive single (.jpg or .bmp) images are undoubtedly less easy to identify (and more time-consuming to ban) than any files with a general .mp3 extension on ‘non-approved’ web-sites.  However, our fears of access to utmost online discretion, although in many cases perhaps desirable (as in reporting a crime) are, I think, of equal concern.  Not long after Gnutella’s release, I found an article which urged me to think about this next wave of highly discreet software in a new light; an Internet where physical anonymity exists even if one is still traceable seems liberating and much has been made of this in terms of theories surrounding race, gender, sexuality and disability.  But an Internet application where ‘near-perfect anonymity’ is offered could be an equally chilling prospect [65] : one striking at the very heart of the Internet’s most troubling issues.  As Sullivan’s report suggests, Gnutella is a unique application in that it can create virtual private networks, which can form and disappear without a trace, rendering it perfect for discrete transactions.  During one recent 15-minute session, a pornography investigator masquerading as ‘RedOne’ said he found 140 instances of child porn.  What was even more chilling were his comments: ‘paedophilia is traded in other ways, but it's just a matter of time before people are caught (...) [Gnutella] is a place where it could flourish. It can't be stopped’ (Ibid.).  Dozens of developers are continuing the Gnutella project in a Linux-like collaborative effort; clearly this new wave of GPL applications – for their virtues and their vices – are here to stay [66] .

  Perfect anonymity is a key strength of Gnutella over Napster because no government agency can watch what you search for and no marketing department can log your hits and target ads at you.  Study of a report on Metallica’s recent court case on May 10^th, indicates Napster’s obligation, when notified, to eject users trading copyrighted material off the system; Upon request, Metallica provided Napster with a list of 317,377 usernames monitored over two days, including file details for each user, among other non-personal details such as the time, date and the IP address of the Napster server to which the offending user was connected [67] . The report did not however, indicate whether the company were to face any fines for providing the software.  Perhaps this is because Napster’s lawyers are employing the ‘Xerox defence’ logic [68] ; in much the same way as Xerox is itself not liable for illegal photocopying made with their machines, Napster’s lawyers will argue that it is individual law-breakers and not the programmers themselves that should be held responsible for piracy.  As more cases are heard, Napster’s future will become clear; RIAA’s current suit seeks $10,000 in damages for each copyright-protected song exchanged (Ibid.), and in the light of Metallica’s progress, other artists will undoubtedly follow suit.

  When Napster is notified of specific users who are trading copyrighted material, it is able to remove them from the system; Gnutella doesn’t allow for this since there is no ‘system’ for users to be removed from.  As long as there are two users with Gnutella software, there will be a Gnutella network.  To stop it, the likelihood is that record companies will have to begin targeting individual users for prosecution.  According to Ian Hall-Beyer’s Gnutella website [69] , the program, much like the Internet itself, is designed to withstand a nuclear war, or more specifically, a frontal attack from record company lawyers.  Gnutella’s principle of peer-to-peer networking appears to be offering an entirely new architecture for efficient searching.  Unlike search-engines that create indexes which quickly go out of date, Gnutella - like Napster - searches machines connected to the net at that time, saving time for the user by alleviating almost any prospect of dead links.  Artists, the music industry and the RIAA, it seems, will not be resting too soon; we are, I believe, still on the eve of a significant change.

Towards a Conclusion

  In Out of Control, Kelly addresses issues concerning privacy, stating that encryption and digital signatures ‘are techniques to expand the dynamics of trust into a new territory (…) Pretty good privacy means pretty good society’ (:273).  In light of open-source developments we can certainly see a negative side to such technology; when a branch of privacy becomes ‘pretty good’ that transactions can be made in utmost discretion, there is a clear and present right to be alarmed.  On the flipside, the industry can equally make legitimate use of such tightening of security and privacy.  For example, let us imagine a version of Napster in which the user logs onto a secure server, with their credit-card details (or an e-money account) and was charged a feasible amount (say 50 pence) for every song they downloaded. If they wanted the newest music, they could be guaranteed a high-quality complete file with no ‘glitches’.  This download would be fully watermarked using SDMI standards and royalties, where due, would be credited instantly to the artists as soon as the transaction was made.  What’s more, the watermarked music could be encoded in such a way that it could be exchanged quite simply with the users’ own PDA, mobile phone or PC; this would be legal, in much the same way as it is apparently ‘legal’ to make a personal copy of a CD onto Tape, MP3 or MiniDisc.  A user is listening to a radio stream whilst driving and suddenly decides that they want to listen to a song, the radio station holds a huge archive of digital music on servers and will stream any track to a listener for a small fee (say 10 pence).  Radio technology is already in place in cars and the scenario I have described exists in the form of video-on-demand.  New formats (such as the secure MP7) incorporate more advanced search techniques than before, such as the ability to search by lyrics or through humming a melody.  A combination of all such existing technologies will undoubtedly articulate a space for new modes of music delivery and commerce.  The logical next stage is surely for the industry to develop a secure rival format to fill a void in the digital format market.  Increased failure to do this will gain the .mp3 format more notoriety, making it more of a de facto standard than it already is.

  Of course, within an industry undergoing such a significant set of changes, it is difficult to draw any solid conclusions.  However, as has perhaps always been the case within this industry, the market will be technologically divided between ‘professional’ and ‘non-professional’.  On the one hand there will be the hobbyist side who will, I suspect, continue to use MPEG layer-3 technology and the ‘search and capture’ routines that are already in place online.  On the other, commercial interests will be protected and supported by the interests of other existing media, such as film, radio and television.  For this side, an equivalent SDMI-conforming technology, such as Microsoft’s Windows Media Technologies-4 compression technique or MP7 will eventually become ‘industry standard’.  In many respects this divide mirrors quite exactly the continuing battle between the renegade open-source hacker ‘Linux’ culture, which believes that all information should be free, set against the corporate windows of influence exerted by Microsoft.  That the industry will continue to manage artists, composers and performers is I believe, inevitable; although existing structures of address and control will begin to transmogrify to incorporate the new arm of technological development.  The AOL conglomerate may take a while yet to find its feet, but much seems to be already in place.  A shift is imminent, although I suspect in a short space of time, many will find it familiar enough.