Geert Lovink

Principles of Streaming Sovereignty

Introduction

Xchange is a network of tactical non-profit organizations and individuals involved in "streaming media" since the birth of the technology in 1996-97.1 This chapter focuses on streaming media,2 usually described as audio and/or video, "streamed" from a dedicated Internet server, either live or on demand. Besides live webcasts, streaming projects build up online archives and databases. Unlike similar cultural networks and lists, the "cloud" of streaming initiatives I describe in this chapter has not evolved much since its genesis, and the reason for this is one of the main topics I will investigate here. Those who are not interested in network stagnation may as well skip this chapter. As a subculture of experimentation, Xchange remained in the shadow of events and by and large survived waves of commercialization, insisting on the original drive to seek new-media models that go beyond traditional broadcasting. The Xchange mailing list has had consistently moderate traffic and, with a few exceptions, refrained from intellectual theorization of the field. Like many radio initiatives, Xchange is weak on discourse and lacks even basic elements of reflection and (self-)criticism. Unlike the Nettime list, the Xchange community did not have controversies over the issue of list moderation. And unlike Syndicate, Xchange successfully managed to ignore trolls and ASCII artists begging for attention and trying to monopolize the channel. I have decided not to tell the Xchange history in a chronological manner. Instead I have chosen to highlight patterns in the activities and focus on practical limitations for streaming media networks in terms of available and affordable Internet capacity.

I did not choose to cover Xchange because of its turbulent history. There has been none so far. I would propose to read Xchange's uncertain and somewhat flat state as a typical post-bubble allegory. Most of us would subscribe to the idea that "visual culture" has become dominant over the written word. But the Internet is an interesting exception. What happens if the technology is there (in this case streaming) but fails to be implemented? Why hasn't the Internet yet attained the immense importance of film, television and radio? What makes playing moving images and music on a computer network so uncool? Is it only the poor quality that keeps people away? Artificial life and virtual reality underwent a similar process after the hype ended. Mild stagnation also happens to certain advanced mobile phone applications. What could be missing here is the network sublime. Scrambled high resolution is by definition not cool low tech. In response, those looking for the instant satisfaction of utopian promises go away frustrated. In the case of streaming networks we are not just talking about "404 Not Founds" and broken streams. It is a larger, invisible infrastructure that simply does not deliver. Streaming is a field for pioneers with a long-term view. This chapter describes how cultural practitioners respond to such a state of sophisticated stagnation. Can creative and subversive concepts be freeze-dried and saved for another decade? What is the impact of working with advanced applications going nowhere? Technologies that have been sidetracked by history and placed in the waiting room for an indefinite period of time challenge dominant notions of speed and the boom-and-bust logic of information capitalism. It is not that the future is in ruins. Instead, streaming scapes, beyond excitement and disappointment, invite users to lie back, uphold their techno-expectations, stretch their time-space and enjoy the stumbling fuzziness on offer.

Over the first five years of Xchange's existence, issues such as bandwidth, standards and models for decentralized webcasting were remarkably stable. The immense popularity of MP3 file-swapping on Napster (around 2000) and other peer-to-peer networks such as Gnutella and KaZaa largely took place in a parallel universe, it seemed. Despite the worldwide growth in Internet users, who by now have installed media players on their PCs and increasingly also on PDAs and other wearable devices, this has not led to radical shifts in the independent arm of the streaming media industry. Streaming has remained very much a desktop PC experience. Despite the steady rollout of broadband and cable modem connections since the late 1990s, independent non-profit streaming media did not witness a breakthrough comparable to the boom in e-mail, chat rooms, webcams and MP3-swapping. Partially in response to the worldwide stagnation of broadband, the central thesis of this chapter is to interpret "minor" collaborations as "sovereign media". But first I will present my version of the first five years of streaming media from the perspective of the Xchange network.

The Beginnings of Streaming

Before going into the life and work of Xchange, I will quickly go through the early history of "streaming media." All beginnings are arbitrary, but one starting point could be the release in April 1995 of the first version of the RealAudio player by a company called Progressive Networks (later Real Networks), founded by former Microsoft employee Robert Glaser in Seattle. Early versions of the audio compression software provided only on-demand audio.3 The first live broadcast was radio coverage of a basketball game: the Seattle Mariners vs. the New York Yankees on September 5, 1995. The RealAudio player (later RealPlayer) was freely distributed to users who, at that time, were typically connected to the Internet via low-bandwidth modems. The player supported connection rates as low as 14.4 Kbps, which delivered audio quality comparable to the sound of a decent AM radio. In October 1996, the first stable version of the RealAudio server software went on sale, enabling users not just to receive live audio signals but also to stream out to the Internet. From then on, practically anyone could start his or her own radio station on the Net.

Another beginning important in this context could be B92's netcasts. The oppositional Belgrade radio station switched from air to Internet on December 3, 1996, after Slobodan Milosevic closed it down. For three days, B92 could only be heard via the Internet. The crucial lifeline was created by B92's in-house Internet provider, Opennet, set up by mathematician Drazen Pantic, who had heard of streaming media well before Progressive Networks launched its first player. Drazen, now based in New York at NYU and the LocationOne gallery, told me the story of early streaming software, from "multicast" to the very first versions of the RealPlayer. "Early experiments of Carl Malamud and the multicast group before 1996 were promising, but still out of the conceptual and infrastructural reach of many. I received all the announcements of the multicast group, including the announcement of their legendary live broadcast from the US Congress. Back then, one really had to be connected to what you could consider broadband in order to be able to receive a live stream. When Progressive Networks came out with their producer/server/player bundle it attracted a lot of attention. But early releases (up to version 3) were just not usable, even though compression software (called codecs) got better and basic stream could be delivered even to an ordinary 28.8 Kbps modem. However, the quality, delays and reliability remained bad for quite a while. A really interesting and progressive approach was that Real launched the version of their bundle for Linux as well as for Win and Mac. That fact kept many people focused on what Progressive Networks was doing."4

He went on to explain that early streaming-media networks were all about the "art of compression." The smaller the files were, the more people would be able to participate and the less the Internet at large would be congested. How many dropouts could one bear, having to re-open the connection to a streaming server somewhere on the other side of the world over and over again?

But let's go back to Belgrade. The closure of B92 came as a response to the massive student protests of late 1996. For weeks, hundreds of thousands of demonstrators protested the government's annulment of municipal elections won by Milosevic's opponents in Belgrade and 14 of Serbia's largest towns. Drazen Pantic: "When Milosevic banned Radio B92 in early December 1996, it was just one of those unimaginable synchronic situations that happen every once in a while. We immediately started distributing news clips in RealAudio format. We neither had the expertise, the bandwidth nor the software for live transmission on the Net. But XS4ALL and Progressive Networks jointly helped with bandwidth and server software. We got a server capable of 400 simultaneous connections, donated by Progressive Networks, installed at XS4ALL."5 Within a day or two there was a live RealAudio stream from Belgrade carrying B92's programming. That same stream has stayed up and running, except for a few interruptions during the Kosovo conflict in 1999, to the present day.

Principles of Streaming

I will leave out the rich prehistory of streaming media here.6 This chapter will instead focus on the post-1996 period, taking the Xchange network as an example of the stagnating independent and "tactical" Internet culture that emerged in the aftermath of the mid-1990s Internet hype (led by Wired magazine, and developing in 1998-2000, parallel to the dotcom hype). RealAudio technology made it possible to join a global network for the price of a local telephone call - and that made all the difference. The Xchange network was founded in late 1997 by three members of E-lab in Riga (Latvia), Rasa Smite, Raitis Smits and Janis Garancs who started an online audio project of the same name in July 1997. The three had participated in and helped organize the maelstrom of conferences, festivals and workshops during the European "short summer of the Internet" in 1996-97. This all led to the launch of Xchange.

Xchange is an example of the new-media cultural initiatives of Central and Eastern Europe that had started to flourish after the fall of the Berlin Wall.7 As a mailing list and website, Xchange was meant for "alternative, non-commercial Internet broadcasters and individual audio content providers" and aimed at setting up a "net.audio network community."8 The list was to provide its members with announcements of new radio links and timetables for collaborative live webcasts and texts. Since December 1997, E-lab's Ozone group has been doing live web sessions every Tuesday, sending out net.radio experiments, live music and mix-jam sessions, sometimes together with other net.radio servers located elsewhere. Ozone invites local musicians, poets and writers to present their work to the global online net.radio community.9 There is an IRC chat room, an important tool for net.radiocasters to find out who's listening, who's picking up the stream for rebroadcasting, and who has content on offer to be downloaded. In the latter cases, the Ozone group in Riga might then go to the URL and rebroadcast the signal, integrating the incoming sound into the program.

In her 1997 welcome statement, Rasa Smite, member of E-lab in Riga and co-founder of Xchange, mapped out the terrain Xchange was to cover.10 She described net.radio as a blend of different radio initiatives, some with a community/pirate background, others exclusively exploring webcasting. Net.radio is not one, it is many. It is this blend that makes the culture, not Internet technology as such. Some streaming initiatives seek to serve both local (FM) and global (Internet) audiences. Others do live transmissions from festivals, parties and conferences. Others netcast from clubs. Some, like the Budapest-based Pararadio, webcast for a specific local audience; others aim at the global Internet population. In a posting to the Xchange list, Slovenian net.radio pioneer Borut Savski summed up the different elements of free webcasting: 11

- (Live) real-time text;

- Audio and video transmission;

- Worldwide accessibility and international concepts;

- Synchronized broadcasting from multiple sources on the same platform (site);

- Atomized (international) production groups gathering as they wish;

- A differentiated (international) audience;

- Creation of no-copyright platforms of independent productions;

- Information banks (texts, interviews, music, archived live production);

- Individualized means of access to archived text, sound and vision files;

- No cost difference between local and international access; no repressive legislation (so far).

Over the coming years, the Xchange network would practice - and embody - these different elements. They define the Xchange network, keeping it separate from commercial currents. In the mid-1990s, non-profit net.radio pioneers enjoyed incredible freedom. The spirit was very much like that of pirate radio, an element the Amsterdam critic Josephine Bosma brought into the network through her writings, interviews and responses posted to the list.12 Xchange could be described as a global network of audionauts festively exploring virtual frontiers. The authorities were oblivious, as were mainstream media and the corporate world. Intellectual property rights were a non-issue for the early non-profit streamers. Even though some net.radio stations occasionally played mainstream pop CDs, for the most part they webcast independently produced music and soundscapes. Out there on the Net they found the freedom to be left alone, to experiment with the new medium, connect it to local radio, pick up sounds in techno clubs and tiny studios, send soundscapes out into the cyber-plains. By 2002, Xchange had about 450 subscribers, almost all content providers - contributors who participated in building the community.

Webcasting, Not Broadcasting

Despite the open and pluralistic approach, there was one question on everyone's mind: what makes streaming media so different from broadcast media? Rasa Smite: "Everything! It is not just because audio is streamed via the Internet, thereby reaching a global audience. Streaming also gives you a certain freedom. The distributed and decentralized structure of network radio is very encouraging. It offers inexperienced artists possibilities to participate, to be involved and to 'network' - equally for everyone on the Net, and in particular for those in remote places, individuals, and micro-scale initiatives."13 This is not just an idealistic statement. Small streaming-media initiatives active within Xchange have been operating in the way Rasa Smite describes. The practice, however, came at a cost, if you like: there was only a tiny audience for net.radio, despite the rapid growth of the Internet user base. But it was a reality most initiatives had no problem with. From very early on, the political economy of bandwidth defined the size of the net.radio projects. Non-profit projects simple could not afford the equipment and traffic costs to handle thousands of online listeners. Only a tiny fraction of Internet users had enough computing power, storage capacity and bandwidth to fully enjoy streaming technologies. On top of that, general interest in radio was limited anyway. Only those with stable and open connections and flat-rate prices would potentially be interested in tapping into streaming media. Berlin-based net critic, Nettime co-founder and streaming activist Pit Schultz: "The economies of streaming have to be put into consideration when the rather vague concepts are taken in. Narrowcasting is explainable when you look at the bandwidth costs. An average urban pirate station has more listeners than the biggest trance-streaming pipes on the Net, just for economic and technical reasons."14

Unlike academic IT researchers, the cultural sector does not have access to independent, non-commercial bandwidth. The only way to earn money with streaming media is to install banner and pop-up ads, ask for donations or require users to pay for content. Broadband is becoming available to a steadily growing (yet relatively stagnant) audience.15 However, large backbone providers such as KPNQuest have gone bankrupt, partly because the content industry will not start streaming before "digital rights management" and tougher copyright laws are in place. Bandwidth prices might not fall until decentralized peer-to-peer networks have been tamed.16 The overcapacity of bandwidth around 2000 was caused by a shortage of customers and content. Fights between telcos over last-mile access to households only made matters worse. Despite decent penetration of ADSL/broadband and cable, the streaming-media industry is still in its infancy. It might take years, even a decade, despite urgent calls from the technology sector for a new "killer app" that would create a new wave of global demand for IT products and drag everyone out of the 2001-03 economic malaise. Digital rights fees imposed in 2002 in the USA resulted in the closure of a number of online radio stations. While broadband users in the US doubled in 2001-02, the overall number of Internet users leveled off for the first time.17 Online payment systems for (music) royalties might be another long-term solution. However, such a distributed system might not work unless it is a grassroots initiative. Standards pushed by the (US) media entertainment industry will most certainly face resistance from young consumers.

Online streaming, accessible via MediaPlayer, RealPlayer and other applications, is different from MP3 files that can be downloaded and then played offline. In response to the inherent limitations of the medium, streaming-media producers developed an ambivalent attitude towards high tech and the overly optimistic forecasts of telecoms. While the rollout of fiber optics was welcomed, the daily online reality lagged behind the television advertisements promising frictionless speed. The enemy of the future was technofuturism. The future had become a glittering commodity, a merchandised myth unrelated to actual experiences. In response to ugly, baroque interfaces and obese 3-D files that took a lifetime to download, many independent Internet initiatives shifted to the low-tech ASCII art aesthetics of minimalist green on black. Streaming took an odd position in the bandwidth dilemma.18 Streaming initiatives needed seamless capacity, but at the same time tried to prove that smarter encryption software would make more efficient use of scarce and expensive bandwidth.

This leaves us with the question as to what net.radio could be in a strategic sense, in the understandable absence of a mass audience. Responding to a post by the Amsterdam net.art critic Josephine Bosma on the Xchange list, Pit Schultz summed up a few "vectors of wishful possibilities":

- connecting old and new media (net.audio connected to real radio);

- random access: producing live and for archive (audio on demand);

- stretched time: geographically dispersed small groups producing from home studios;

- public content: experimental DJ performances making non-profit copyright-free productions;

- global sprawl: representing regional styles and mixing them with global ones;

- soundscapes: deconstruction of the song via remixing, sampling, overdubbing, cutting;

- free press: direct information without censorship, small news channels, talk shows, net.chat.19

Within this range of possibilities, archiving was an exciting new option for radio, which so far did not have the content-on-demand option. Once a program aired it was gone, unless taped - vanished into frequency nirvana. The only other possibility was that, in theory, civilizations in neighboring galaxies might pick up the program, and store, archive and properly metatag it. Or you could use a time machine to go back and push the record button. Many radio makers would agree: computer storage combined with streaming media has changed the very nature of radio. What now seems the normal option of looking up a radio program on the website if you missed it on BBC World Service, for example, is in fact a revolution in terms of what radio is all about. Rasa Smite: "Some net radios do serious archiving of live sessions. Some of us have experienced that sometimes there have been more listeners of a recorded 'last session' than during the 'real' live broadcast. But many others don't pay attention to archiving (too boring?). Doubts sometimes appear in between the necessity of archiving and the viewpoint that live shows are more exciting than recordings."20

For networks such as Xchange the issue of archiving is closely tied to audience development. By nature online archives are universally available. Sydney-based net.radio artist Zina Kaye explains: "If you missed a live stream you could go to the archive and listen to it for a whole week. The streaming media database Orang in Berlin offered their services to Xchange members to archive radio shows and audio files, stored under categories chosen by the individual members." The Laudible server in Sydney wrote a piece of code that referenced the Orang database in order to give it a customized interface that displayed Australian and New Zealand content.21

Radio on demand had a great future. Now that information could be stored and spread all over the Net, the issue was how listeners could find content they liked. One did not easily find live netcasts through search engines or global portals. Announcements on lists and websites worked but reached a limited audience. The virtue of decentralization was also the problem. How could content be grouped in an archive, accessible through a web portal, using a central interface, without any claim of ownership and control? The fact that files remained on their own servers, accessed via links, made databases unreliable in the long run, as URLs changed. As time passed, the Xchange list started to specialize in announcements of live webcasts, instead of theoretical debates about net.radio. The list developed slowly according to the users' needs. Many were looking for a place to announce their webcasts or get pointers to others' sessions. As an online "radio guide," the Xchange list was used by hundreds of streaming initiatives and events.

Meetings and Webcasts

Like other list-based networks, Xchange grew in a short amount of time through a series of meetings and collective netcasts where members met in real life. Net.radio Days in Berlin in June 1998 was an exciting early event of the newly formed network, directly followed by the Art Servers Unlimited gathering of independent Internet art initiatives in London in July 1998.22 In the same month, there were live webcasts from the Polar Circuit workshop in Tornio, Finland. Perhaps the largest and longest Xchange project happened in September 1998, when around 20 members gathered in Linz, Austria, and performed Open-X, a live 56-hour webcast. The "webjam" included a long list of remote participants.23 In November 1998, members gathered in Riga for the Xchange Unlimited Festival.24 The next meeting took place in March 1999 during the third Next Five Minutes Festival in Amsterdam, with a special section devoted to streaming media.

The biggest festival/conference organized by and for the Xchange network was the Net.Congestion event in October 2000, sort of a micro-version of the Next Five Minutes tactical media festival, organized by (approximately) the same Amsterdam crew.25 Panels included "The Network Is the Narrative," "Bandwidth Aesthetics," "The Hybrid Media Show," "Target.Audience=0," "Web Documentaries," "Tactical Streams," "Protocols and Alternatives," "The Art of Making Money" and "The Doom Scenario" (about congestion and the impact of streaming on Internet infrastructure). The festival statement struggled with the notion of the electronic avant-garde, the community having lost its grip on the medium it was affecting to direct. The contrast between the 80 million copies of RealPlayer in circulation by 2000 and the quasi-voluntary isolation of streaming artists was growing by the day. Why hadn't Xchange grown at a similar pace? Why had streaming remained such an unknown phenomenon, even among new-media artists? As a solution to this discontent, a "visionary scenario" was offered in which the artist would be a "toolmaker, directly effecting the production and distribution of streaming media."26

What was being cast? Only a minority of streaming initiatives labeled their content explicitly by genre: techno, rap, reggae, drum 'n' bass or industrial. More commonly Xchange streams 'map' ambient environments rather than transmitting messages. Unlike the star DJs familiar from clubs, regular radio stations and recording labels, most streaming DJs are low-profile or anonymous. The dominant presentation form is the live mix fusing music, spoken word and sound.27 MP3 files or streams from the Net are often used. In Xchange's own streaming sessions every Tuesday, Raitis Smits explains, "Everyone can join with his or her RealAudio live stream. The simplest way is to mix your sound source with another. Each of the participants is doing one part of the live session (for example, one is streaming voice, another background music)."28 On an IRC chat channel running in parallel, participants exchange experiences and announce what's coming up.29 A technical complication is the 5- to 10-second delay of each stream; this demands a sense of discipline and anticipation from participants.

Playing with Loops

Raitis Smits describes the "loop" as another technique that uses delay. You take a stream, re-encode it and send it to the next participant; the sound goes round and round, creating multiple layers. Eventually the stream turns into noise.30 Another frequently used technique is sampling. Daniel Molnar: "Our generation grew up with information overflow; that's why we are into sampling. I'm just trying to sample the world, I ain't trying to synthesize any part of it, I'm just stealing the interesting pieces and putting them together."31 These are the techniques used during the live sessions. The webjams' main difference from previous sampling practices and radio broadcasts is this collaborative, interactive aspect. The offline craft of sampling, often done in solitude at night, can now be done in a networked context. Software and critical discourse are also important elements, but the actual streaming between servers should be regarded as the essence of streaming art.

In 1998 the Ozone group in Riga started a series of "mobility" experiments to explore how streaming could escape the stasis of the PC-cum-radio station glued to the desk. The Riga group tested the minimum bandwidth and equipment needed for streaming. A mobile streaming studio was set up - RealServer 5 on a laptop. It was used at Net.radio Days for encoding and running the server using a dual ISDN Internet connection. They did live streams from clubs, encoding via telephone line with a 28.8 Kbps modem.32

Later versions of the encoding software were unsuccessful at streaming through phone lines; they often produced noise. Obviously the software was made for higher bandwidth and could no longer compress audio for phone lines and 28.8 modems. Ozone also did a transmission from a train between Riga and Ventspils as a part of the Soros Contemporary Arts exhibition Ventspils Tranzit Terminal. Rasa: "We didn't try to encode signal via mobile phone (it had too narrow bandwidth, around 9 Kbps). Instead we used mobiles to transmit audio signals (sound, talk, music) from the train to the Ozone studio in E-lab in Riga, where it was received by another mobile phone. There it got encoded into RealAudio signal for further distribution on the Internet."33

Not all initiatives were minuscule by default. The choice of remaining invisible was open to every group and individual. Fashionable underground music genres such as techno, hip-hop, jungle and drum 'n' bass immediately drew online crowds, as did already established radio stations. By late 1997, the mailing list of the London streaming site Interface (started in January that year) had 1,400 members and reported 3 million hits on its site, a considerable number at the time, most coming from the club scene. When Josephine Bosma said only institutions had enough bandwidth to listen to net.radio, Interface member Eezee answered, "We at Interface have an average of 10,000 to 12,000 listeners on a daily basis now."34 In contrast, mostly state-funded electronic "art" music usually attracted small, fairly specialized highbrow audiences. A third category, beyond the pop/avant-garde opposition, were the autonomous audionauts, webcasting in the great digital nirvana free of any consciousness of an online Other. All three models were to be found in the independent streaming scene: commercial pop culture, experimental sound art and autonomous "sovereign" webcasters.

Narrowcasts and Archives

The explicit aim of the Xchange participants was to set up temporary streaming exchanges, not to rebroadcast radio content. This contrasts with most broadcasting officials' belief that streaming was an ideal supplement to the conventional distribution channels of radio, recorded music, television and film. Remarkably absent on Xchange is the usual debate around commercialism, the clash between the not-so-secret aspirations of some to become big versus the determination of others to stay small, to avoid selling out. Japanese media theorist and experimental "mini FM" radio producer Tetsuo Kogawa wrote: "The point is not the stronger power of the transmitter. As long as it is alternative (later + native), it must be different from usual broadcasting. Forgetting 'broad'-casting, we insist on 'narrow'-casting. In my understanding, the more creative or positive function of the Web is not 'casting' but weaving. Unfortunately, the Internet is used as a new type of casting, though."35

The community aspect of Xchange remained small and pragmatic. In this context, Tetsuo Kogawa's one-watt transmitter could be seen as a good example of the Xchange approach. He wrote: "The coverage is proper for a community within walking distance and the technique is cheap and easy. In my workshop I built a set within an hour. In my workshop, I built a transmitter, showed something of radio art and invited an audience to the process: radio party. More aggressively than in Europe, the community culture in Japan has been destroyed. That's why we have few community radio stations in Japan. But this situation might be good for web radio because the 'listeners' are separated and have no physical/geographical 'community' anymore. The web may [reunite] them in cyberspace at least for the time that web radio works."36 The one-watt metaphor can easily be transported into new media. What makes the Net unique is not its ability to become one big metamarket but the potential for millions of exchange nodes to grow, an aspect that has yet to be fully understood.

Localization

Canada's Radio 90 is a good example of "localization" of net radio. It is a local station with an easy-to-use public streaming interface. UK net.artist and activist Heath Bunting founded the project. Zina Kaye explained, "Heath wrote a web-based scheduler that steers the content of Radio 90. Xchange members would input the time of their shows and they would be heard via a one-watt transmitter, installed at the Banff Center for the Arts in Canada. The people of Banff no longer needed to have a computer in order to access streams that typically had no name or brand or advertising. No doubt the Radio 90 scheduler gave streaming media initiatives more listeners."37 Similarly, an FM transmitter installed at the Society for Old and New Media in Amsterdam rebroadcast net.radio streams, including B92's signal when the station was taken off the air in April-May 1999, to the Nieuwmarkt district on an irregular basis.

Another aspect of Xchange-style streaming culture is the link with "real" space, in most cases clubs. The Extended Life Radio project from Berlin emphasizes the link with locality. "Physical space is most important for us, and it doesn't need to be connected to the Net. The connection via Internet of two or more physical spaces gives the possibility to synchronize those spaces at least partly and for a certain time. It's an image, located in real time and real space, for and about information, experience, network, and communication. Translation. Inside and outside. Crossing and melting borders. For any activity in public space it's very important to create a certain atmosphere, an 'interface' which reflects what it is about. It's about this translation of (in this case) sound, which comes out of a machine without any body or human traces, into something you can experience which creates an atmosphere."38

Klubradio, also in Berlin, is based on the idea of users worldwide tapping into groovy underground clubs and listening to live DJs. With a good connection and a bit of luck users could, for instance, plug the stream into an ordinary amplifier and have their own Berlin techno party. This is all done at little or no cost, without complicated satellite connections or the interference of events agencies or telecoms or record companies.

Xchange is one of many "adagio" networks. Instead of picking up speed riding on techno storms, fired up by innovation and commerce, they place emphasis on slowly performed works that stretch time and space. The unknown and yet-to-be-defined "otherness" of streaming technology implies an outsider position. "Cyberspace is our land," as the slogan of Frankfurt-based webcasting artist group Station Rose goes. Analogous to "off-the-radar" free radio and club mixes that loop for hours and hours, streaming events reach out into the vast darkness of cyberspace. The streams promise to open up other dimensions of time and consciousness. Not interested in the size and mood of its audience, streaming media focuses on maximum interactivity among equals.

The different aspects of streaming listed here spring from technological circumstances: lack of bandwidth combined with a chaos of standards. Netcasters have learned to redefine these limitations as virtues. Make no mistake; every streaming artist would love to operate in a situation of bandwidth abundance. Outside the corporate and academic IT worlds, scarcity rules. It is next to impossible to tell whether low tech is a passionate belief system or a necessity. Instead of a culture of complaint, there have been attempts to utilize streaming media's "micro" status in the best possible ways. Centralized experiments such as Web TV have so far failed. Berlin-based activist/artist Micz Flor has worked on net.radio projects in the UK, Central and Eastern Europe and Asia. He explains: "We are all still waiting for the new front end, the browser of the next generation, where all these media outlets come together at the screen and speakers and what else of the user, listener, or whatever you would want to call the next generation receiver. The ideal client 'solution' is not there yet. And that's a good thing. So far, not even multinational lobbies such as Microsoft or AOL managed to prune the Internet into the shape they would dream of. In fact, every attempt to shape the multitude of formats, players and codecs has only put strength to alternative solutions. Peer-to-peer distribution channels such as Gnutella are one example; alternative audio video formats such as Ogg or DivX are another."39

In the end it was all about playing with the limits of new technologies. Lack of bandwidth was countered with an abundance of imagination. Still, some of the borders were very real. In some instances streaming could become unpleasantly expensive. Matthew Smith, working for the Ars Electronica Center and the Austrian broadcaster ORF in 1998, discussed the cost of streaming and argued for the use of existing media. He wrote to the Xchange list: "If you want to find 'new' ways of providing content in a setting such as the Internet, it is not very efficient to clog up the net with high-bandwidth audio. The logistics of the net are not made for it, and who can really afford the necessary bandwidth to be able to serve 1,000 high-quality RealAudio streams, even after shelling out $5,000 for server software and about the same amount for a server. I don't believe that anything in that price class is for free, meaning distribution - to place your content on a setup like that will eventually cost the same as buying time on a 'classic' broadcast medium."40

Pit Schultz gave another example. "Our Klubradio server used about 1.5 terabytes over May 2002 and that is not even much compared to large streaming sites such as live365. A regular provider would ask about 4,000 euro a month for this type of streaming traffic. Who is willing to pay that kind of money? Canalweb, our provider in Paris, closed down, like the other ones we previously used. After Canalweb disappeared we went down from 4,000 to 1,000 visitors a day. Our server capacity shrank from 2,000 concurrent users to 25."41

Xchange initiatives often used free demo versions of the RealAudio software that had a limited capacity (well under 1,000 streams) and an expiration date. Others got software donated by the Real Corporation itself. The potentially high bills of Internet access providers had an effect too. Without sponsorship and voluntary limits to capacity, streaming networks such as Xchange could not have flourished. Necessarily, experimentation would have made way for dotcom business models. As this did not happen because of self-imposed limits, Xchange partners still exist, whereas most dotcoms do not. In particular, those who were betting on a possible Web TV revolution have been badly burnt.

One of the problems Xchange successfully tackled was the question of how to find net.radio streams without a centralized portal.42 The Xchange homepage offered links. Heath Bunting of the Radio 90 project in Banff, Canada, came up with a program schedule. Radioqualia (Adelaide, Australia) developed a similar idea, a global mini FM network, which it called Frequency Clock. The founders explained their project on the Xchange list: "A geographically dispersed independent network of net.radio stations, broadcasting on autonomously owned FM transmitters, could strengthen challenges to centralized institutions that are predominantly associated with FM radio, encouraging a rethinking of existing broadcast paradigms, and the incorporation of more open systems for determining content. In such models there is space to develop radically open-ended systems of content management, allowing for abatement of centralized program administration."43 Do-it-yourself programming was seen as an effective answer to the "portalization" of the Web.

In November 2002 Radioqualia, which had moved from Adelaide to Amsterdam and London, released the 1.0 version of its Frequency Clock Free Media System. It is a shared resource for building streaming channels, open source software with a program database, a timetabling system and a customized streaming media player. In the age of broadband and cable modems the "always on" mode is an important feature for streaming media. Users can schedule audio or video programs from the database in specific time slots, creating a continuous and ongoing channel, or alternatively, a channel that broadcasts only at special times. Producers can also instruct the timetabling system to play default audio or video when a time slot has nothing scheduled. This means audiences will always have something to see or hear.44

The Network Is Not the Organization

On November 10, 1999, at the height of dotcom mania, Adam Hyde and Zina Kaye posted a proposal to the list to upgrade the Xchange website to a dynamic portal. "Right now, we are at a stage where as a group we have the advantage, because we have been around for a long time and have a good relationship with each other. But the entertainment industry is catching up with us, and we will lose our lead and maybe our unique identity if we don't quickly distinguish ourselves from other mainstream streaming portal websites."45 Three years after Xchange's founding, the Riga-based E-lab was still the only one really taking care of its activities. The website had not changed much and was nothing more than a list of links to the participating net.radio sites.

This proposal came at a time when projects such as the Frequency Clock, irational.org's World Service, Radio Internationale Stadt and TM Selector began to offer streaming radio guides, each in its own way. Riga, however, lacked the resources to bring the network to another level and turn the website into a lively hub. It proved hard, if not impossible, for a network with modest affinity among its participants to set up a decentralized working group to delegate technical and content-related tasks. The issue here was the true limits of non-profit and e-mail-based networks. Lacking formal organization, neither an NGO nor a dotcom, Xchange seemed to get stuck at the mailing list level. Nonetheless, a few months later, xchange.x-i.net was launched, but the portal initiative never really took off.46 Xchange remained an announcement list with occasional short dialogues.47

Xchange chose not to formalize the network and turn it into an NGO or lobby group. Instead individual members moved their focus towards collective development of software, the material the Net is made of. According to Adam Hyde (Radioqualia), there were Jaromil (dyne.org), August Black (Kunstradio), Thomax Kaulmann (RIS, OVA, OMA), Pit Schultz (Bootlab), Micz Flor (LowLive), Drazen Pantic (Open Source Streaming Alliance), Alexander Baratsits (Radio Fro) and Heath Bunting (World Service); all were heavily interested in streaming software, as either developers, researchers, organizers or commentators.48 The number of actual software developers within Xchange has so far remained relatively small. Streaming is not a traditional area for hackers and geeks - the proprietary nature of (mainly Real) software may be one reason. A cultural explanation could be the fact that music and video can only be dealt with on a higher application level. It boils down to the central question: why would you want to use the Net to rebroadcast old-media material? Streaming, therefore, was left to new-media artists plus some non-tech cultural types (read: those who remain on the easy-to-use desktop level and do not produce code). This could explain why there has been such a delay in the development of open source streaming-media software, compared to, for instance, the Linux operating system and the Apache server software.

Others involved in Xchange such as Rachel Baker, Lisa Haskel, Walter van der Cruijsen, Mr. Snow and Superchannel battle the technocracy by training others to use streaming software. Programmer and streaming art project organizer Adam Hyde explains, "These individual networkers within Xchange are very involved in issues surrounding software, these issues may not surface in discourse through the list (it's not a very "threaded' list) but certainly individuals within Xchange do their own work individually or collaboratively and then post the results."49

The politics of proprietary code as applied to streaming (e.g. proprietary codecs) is a well-known issue within Xchange but is not a focus for debate. Some members post news articles or links about the topic but it does not turn into a thread. Adam Hyde: "I think this is a very embryonic debate everywhere. The whole MP3 phenomenon as highlighted by Napster did not settle into debates on how the Frauenhofer Institute and Thompson (who own the MP3 standard) could close down anyone using an unlicensed MP3 algorithm; instead the hot ticket was how wonderful peer-to-peer technologies are. Proprietary media technologies (MP4/Ogg Vorbis/DivX) are just about to heat up and then it will be interesting to see if this groundswell will prompt Xchange into more political discourse."50 After the introduction of Microsoft's MediaPlayer, Real gradually lost ground, though not as severe as the demise of Netscape in the face of the near-monopoly of Microsoft Explorer. Apple's QuickTime (mainly installed on Apple's own machines) is a viable third player.51 The role of open source players is so far almost zero.52

Open-Source Streaming

In mid-2001 the Open Streaming Alliance (OSA) was announced. If Xchange had failed to set up a common portal/weblog or proper NGO, perhaps it could at least contribute to streaming software and test alternative network architectures. This shift in emphasis from collaborative webcasting towards software, driven by initiatives such as Radioqualia, had become visible during the Net.Congestion conference in Amsterdam in October 2000. In an e-mail Drazen Pantic mentioned scalability of capacity and platform independence as the two main aims of the alliance. From the beginning the proprietary nature of Real software had been a problem. There was little to say about the rise of Microsoft's MediaPlayer. The monopolistic marketing policies of Bill Gates were well known. But what about alleged alternatives such as Real and QuickTime? By 2000, open-source streaming software started to become available but wasn't widely used. Although Linux had gained a strong position in the server sector, desktop open-source software had not (yet) managed to reach the average consumer - not even avant-garde early-adopter Xchange artists. OSA planned to enable free and open-source tools for encoding and serving QuickTime, Real Media and Mbone streams, producing streaming content in one run, through just one encoding process, which obviously would save time, equipment and resources. Drazen Pantic: "Corporate software vendors try to monopolize streaming-media standards, using proprietary and closed code for encoders, players and servers. RealMedia, for example, started its operation with a noble idea to help independent broadcasters, but in the course of corporate battle - mostly with Microsoft - they sealed their code and became an opponent of creativity and innovation themselves. Closed code, and especially proprietary codecs, alienate content from the producers and enable control over distribution."53

Simultaneously, progress was made on the archiving front. With the motto "You don't have to know everything, you just have to know the reference," Berlin-based Orang Orang (Thomax Kaulmann and Frank Kunkel) launched its Open Meta Archive software. This open-source "context management system" was able to "categorize and publish rich media documents including text, photo, audio and video in RealMedia, QuickTime and MP3." Finally a variety of multimedia content could be stored on one database.54 The future of community networks would be "hardwired" or, to be more precise, "softcoded" in software that would define decentralized (peer-to-peer) network architecture. The openness of software and the ability to use a variety of standards was going to be decisive.

Despite its low profile, the Xchange network could not escape the sea change in the general atmosphere. The Net was no longer a cozy family but a big and most of all anonymous place. In December 2001 the entire Orang Orang multimedia archive, used by many Xchange members to store their streaming files, was deleted by a hacker after several harmful intrusions. "There is nothing like 100% security. Sadly, this digital vandalism hits a site which always supported a community of free exchange and free access."55 According to Pit Schultz, the result of such hacks is that "small providers or self-run co-locations, public access sites of universities and libraries move from a policy of the free digital commons to a strategy of paranoid enclosure, while the security experts and service industry prospers. In their midst, former hackers who still perform their sport like innocent boy scouts praised by the net culture, discourse as role models."56 The attack was a rude wake-up call for a "minor" online community that for five years had successfully operated in the shadow of turbulent Internet events.

Xchange and Riga

Informal, decentralized networks such as Xchange may be indifferent to commercial interests, yet at the same time they are unable to represent the interests of their members in, for instance, negotiations about storage capacity, production of broadband content and financing of Internet traffic. Xchange suffered from the traditional stereotype of "organization" being "bad" because of the fear it results in bureaucracy and abuse of power. In this case, the lack of organization stalled the network. There was no moderation group to manage the direction of the list. Because of total reliance on contributions by individual members, the Xchange list eventually entered a state of regression (which pragmatists are perfectly happy with). The lack of organization resulted in the implicit expectation that "Riga" would take care of everything. But resources in Riga were scarce. Although Xchange had no official legal status, it was unofficially owned by E-lab. If Riga took no initiatives to improve the site or the lists, no one else did either. As a result, much like the Spectre list (the successor to Syndicate), Xchange ended up as a low-volume announcement list, with few personal messages, let alone debates.

The atmosphere on the list had not always been that friendly, and was slightly hostile at times. The "coziness" of assembling in cyberspace was missing here. Members had their common interests, but their moods were not always in sync. Postings were often written in a peculiar, impersonal style. By 1998 everyone knew all too well what the pitfalls of a list community were. The short summer of the Internet was over. This was the age of infowar and spam, of hacktivism and trolls. There was hardly any room left for naïveteacute. In October 1998 organizer Rasa posted a message to Xchange reflecting the growing unease. "There is confusion about what exactly we were/still are looking for. We are complaining and blaming and provoking each other or whatever (really funny, isn't it). We are dreaming about open spaces, but are we paying attention enough to the importance of personal relationships - understanding and respecting each other?"57 As a response to growing tensions and the lack of real outcomes beyond limited collaborations, the Riga group began to lose interest in the list and focus on badly needed improvements to its own situation, both in terms of space and resources.

Over the years the E-lab group had been slowly shifting its net.radio focus towards the broader issues of "acoustic spaces." Looking back, Rasa Smite says she would no longer limit net.radio to just streaming media. According to her net.radio first of all means networked audio communications - and that is potentially an infinite field. Rasa: "Internet radio can, for instance, provide access to publicly inaccessible technologies such as secret military objects, or follow developments in the field of satellite networks. One could think about acoustic GPS space or combine wired and wireless, global and insular technologies."58

Riga shifted its attention to other projects that were close to Internet radio. I will mention two of them because they illustrate how networks try to escape impasses they have little control over. Acoustic Space Lab was a project initiated by E-lab aimed at obtaining new experiences beyond the usual Xchange webcasts, looking at what other shapes a streaming network could take.59 It took place in August 2001 at the Irbene Radio Telescope in the Latvian forest, utilizing a former Soviet antenna with a diameter of 32 meters. It was a cooperation between scientists from VIRAC (Ventspils International Radio Astronomy Center) and an international team of 30 sound artists, net and community radio activists and radio amateurs, who experimented with the antenna, recording sounds and data from planetary observations, communication satellites and the surrounding environment.60

There was also the issue of work space. Since 1998 there had been talk of a center for digital culture in Riga. So far E-Lab had undertaken all its activities from a tiny attic room surrounded by artists' studios, in a grey government building housing a variety of cultural institutions. The view over the Daugava river was magnificent, but there was hardly adequate work space for the expanding group and the ever-growing number of PCs. Besides local initiatives, the Riga group had quickly focused on building links within the Baltic/Scandinavian region. In May 2000, the Riga Center for New Media Culture RIXC was founded.61 The media space was to be located in the former sculpture studio, which needed serious renovation. In March 2002, an international architects' workshop met to develop conceptual guidelines for the design of RIXC. The next project was a festival devoted to media architecture and the interconnection of urban geographies and information networks, and how the social dynamics of "virtual networks" can be applied to physical conditions and facilitate the expansion of public space.62

Beyond Remediation

After presenting this version of the history of the Xchange network, I would like to bring together some elements of an independent streaming network philosophy. If the Internet was going be something truly new, as all the visionaries claimed, then streaming media would be a prime example of how to supersede the old, one-to-many model of broadcasting media. From their infancy, the net.radio initiatives featured here have tried to prove that decentralized networks are not just a weird idea but a viable practice. The hyped-up dotcom cycle, from startup to sellout and bankruptcy, was not inevitable. The "clouds" of webcasters and online audio archives took up the challenge of proving that Marshall McLuhan and contemporary "remediation" theorists such as Bolter and Grusin were not always right.63 According to Bolter and Grusin, "remediation is a defining characteristic of the new digital media." (p.45) "Each act of mediation depends on other forms of mediation. Media are continually commenting on, reproducing and replacing each other, and this process is integral to media. Media need each other in order to function as media at all." (p.55)

Bolter and Grusin's remediation concept is common sense within media theory. Remediation may be the default option, but at least temporarily, in the shadow of corporate capitalism, it should possible to unfold other practices - that is the claim critical Internet culture is making. McLuhan's law, which holds that the content of a new medium is by definition sourced from previous media, is not false and can easily become a self-fulfilling prophecy. But new media open up possibilities for other forms of narration and aesthetics. They are not just tools to tell the same old story over and over again. It is up to new-media practitioners to seize the opportunities and discover the language of new media, liberated from depressing laws of techno-determinism. If content and interfaces are merely special effects of hardware and software, then why bother in the first place?

Rebroadcasting existing audio, be it live or prerecorded, is not what (independent) streaming is primarily about. And, contrary to Bolter and Grusin's statement, streaming media networks do not express "our desire for immediacy" (p. 35). They embody the desire to network, to link and stream. Long-term collaborations are much more characteristic than celebrations of the short-lived "live" effect. Communication does not need to be "real." In the post-hype period the aim of streaming networks such as Xchange is not necessarily higher image resolution or better sound quality. The bandwidth is not available to such civil networks anyway. But there is a general issue here, beyond the bull and bear market for telecom stocks. Networks in general do not attempt to gain higher levels of "reality" (in the sense of immediacy), as Bolter and Grusin claim. The genuine wish for faster machines and connections should not be confused with the desire for "more reality." The issue, rather, is: does the technology (in this case streaming software) give users access to information and each other? What fuels the imagination? Streaming media explore new network conditions and do not seek to rebuild the old audiovisual world into the virtual. Bolter and Grusin limit new media to the MP3 level of non-interactive customers interested solely in downloading their favorite "remediated" Metallica songs (soon available in Dolby quality), uninterested in contributing to the peer-to-peer networks they use.

Former B92 sysadmin Drazen Pantic reads the remediation issue as a misunderstanding. "Conceptually, streaming media is rarely understood as media per se, but instead as an extension or replacement of the corresponding classical media. So streaming video is taken as poor man's TV, while streaming audio for a while was considered as a replacement for radio. Neither of those either/or alternatives are actually realistic - both streaming video and audio are different media than their corresponding counterparts, with their own codes and structural rules. But this misconception has caused people to expect easy plug-and-play delivery and seamless broadcast quality delivery through ordinary telephone lines."64

In Minima Memoranda, a short but rich collection of aphorisms, Tetsuo Kogawa investigates possible meanings of the streaming concept. Instead of using the obvious reference to water and nature (panta rhei), Kogawa investigates the line metaphor in a phenomenological manner. "Lines relate to binding, weaving and streaming. They can bind audiences into a tightly integrated "network," a marionette-like circuit. However, lines are not always tight but loose. Loose lines weave webs. In the weaving-weaved web, the signal does not cast itself but streams by itself. Casting is a one-way process while streaming is interactive: streaming in and back."65 Streaming resists remediation by its very definition. We can only speak of streaming media where there are open feedback channels.

Minor Media

Even though links have been made between Xchange as a "minor medium" and Deleuze and Guattari's concept of "minor literature,"66 some would hate to see their network explicitly linked to the Parisian philosophers. Whereas some find useful concepts in the works of "D&G," others detest the academic fashion and theory hype that surround worn-out labels such as "rhizome." That's the danger when theory operates within the zone of popular (media) culture. Despite such reservations, I will look into the minor media notion as part of my search for independent streaming-media concepts. It was the German theorist and curator Andreas Broeckmann who placed this concept in the new-media context. Minor literature, he wrote, is a "literature of a minority that makes use of a major language, a literature which deterritorializes that language and interconnects meanings of the most disparate levels, inseparably mixing and implicating poetic, psychological, social and political issues with each other."67 Strategies of "being minor" are intensification, refunctionalization, estrangement and transgression. In the context of media art, for Broeckmann "becoming media" is "a strategy of turning major technologies into minor machines."68

However, the usefulness of such statements within down-to-earth circles like Xchange remains more undiscussed than disputed. Whereas pragmatists hate to see such academism overruling actual practices, others see a limited role for theory as one of many alternative ways of storytelling. But what does it mean, in terms of social capital, to label your network project "rhizome"? Are "minor media" really proud to see themselves as such, despite the positive-productive meaning Deleuze and Guattari give to the term? Who wants to be minor? The strategy of independence may be a choice, but techno-cultural networks often do strive for more power and resources. The term "heterogeneous practices" sounds less pedantic. There is a wide consensus that networks such as Xchange are, in principle, based on mutual respect for difference, grown out of a process of "resingularization" to become ever more different.69 Creating a nice, safe new-media ghetto can mean a one-way street; a situation in which growth and transformation are no longer options. If size doesn't matter, there should be no difference between becoming major and minor.

Nonetheless, in the case of Xchange, media freedom was created by the lucky circumstance that the mainstream ignored what was happening. If, for instance, there was a parallel between radio in the 1920s and streaming media in the 1990s, was the eventual outcome (state-sponsored corporate domination) likely to be the same? No, there was no such defeat. This is the point at which historical parallels and history as such could backfire on those who act. Networks were sparks of change. They either ignited processes or remained sparks in the dark. Though the streamers had little illusions about their actual power, the utopian promise was alive and well. Dancing nodes like Xchange seemed possible. Another possible world was embodied in software and a lively decentralized network practice, which were supposed to spread like cultural viruses. If nothing worked out, at least the participants were having a good time, while turning their backs on the system inside a self-created "temporary autonomous zone."70 Such zones can be big, can fill up the universe - at least for a day.

The "minor" practices of Xchange questioned the eternal recurrence of the same (content). The cynical path from underground outlaw via fashion to sellout and mainstream market player could be avoided. While technology was a precondition for independent streaming networks, it did not dictate the form that "the social" would take. The technology was challenged not to determine the streaming politics and aesthetics. The content of new media is not by definition yesterday's papers. Both the content and form of new media can be radically different from those of previous media as long as network participants are aware of media laws and willing to negate and transcend them. This may not sound revolutionary but if a growing network of passionate media and art producers take the newness of digital media seriously, a lot can happen.

According to Erik Davis, DJ and author of TechGnosis, Internet radio is not part of the regulated and commodified spectrum. Comparable in this respect to early radio, it is a "space of openness, of indetermination, of the effects of the unknown."71 Internet radio cannot be merely radio received via the Internet. Rather than emphasizing the convergence of media, for developers' communities it is more interesting to search for the radical and unconditional "autopoiesis" of new media. In this case: what is the unique quality of streaming and how can the self-referential dynamics be strengthened? How can the proclaimed autonomy of cyberspace be defended against the vested interests of film, radio, television and the recording industry?

For the Xchange network and numerous other streaming-media initiatives (including commercial ones) the proclaimed victory over old media comes at a price of voluntary marginalization. The media industry has been betting on a combination of technological convergence and syndication of content run by conglomerates. The response to this concentration of power in a few hands has been radical fragmentation. "Faced with the ubiquity of a zillion portals, channels, live-streams and file formats all screaming for attention, what's a net audio selector to do? Go niche. Go uumlberniche."72 Instead of fighting the mainstream or claiming territory within established channels using "pop" strategies, a multitude of parallel worlds were created.

The thesis under debate here touches the very essence of new media: its claim to be different from previous communication tools. I am explicitly presenting an "idealistic" viewpoint here (centered around an idea). Streaming media have the technical potential to question the iron necessity of the return of the "one-to-many" broadcasting model because everyone who is interested can install a streaming server and start webcasting. This ability to both stream out and receive streams has the potential to fragment the "mass" audience into dispersed user/producer groups. The technical peer-to-peer approach (as opposed to the centralized client-server model) may be obvious for some, but its consequences are far-reaching. In contrast to broadcasting, we may define streaming media as channels that make audio and visual material available on the Internet. That may sound pretty dry and straightforward.

With the porn industry as its avant-garde, state and commercial radio and television almost immediately started to dominate streaming media with their repackaged content. During dotcom mania, as some companies tried to define "web-specific content," many jobs were lost and many internal Internet departments closed or lost staff after the dotcom bust. Remediation of existing material is a threat to independent streaming cultures, as it reduces the new medium to a secondary rebroadcaster. It is therefore of strategic importance to further investigate streaming models that go beyond repackaging others' content. This could also imply a critique of existing peer-to-peer networks, as their users hardly create and upload any of their own material and mainly download mainstream content.

Only a limited number of sites webcast live from clubs or events. The depressing reality is that nearly all mainstream content remains one-to-many rebroadcasted material. Xchange's collaborative netcasting techniques casting remain unknown, even to those in the field. The additional function of streaming technologies for mainstream media organizations, then, would be the ability to access material after its original broadcast (the on-demand feature). The value of streaming for existing media organizations is found in the storage and retrieval capacity of the Internet, and not so much in the "live" aspect. Independent streaming, on the other hand, stresses the importance of networked webcasting and, most notably, does not retransmit already existing signals. These initiatives provide the Net with new, as yet unknown content and forms of subjectivity. Becoming minor, in this context, can be described as the already mentioned strategy of "turning major technologies into minor machines."73 Against the mass media, a heterogeneous network of networks could flourish. This is not mere theory. The listener-as-producer, submerged in immersive space, designing a unique, personal mix of up- and downstream data. Audiospace theorist Erik Davis: "Electro-acoustic spaces aren"t simply a genre of music or a backdrop of good VR - they are interfaces with the machine."74

No More Audiences

In his text "Media without an Audience," Dutch media theorist Eric Kluitenberg, organizer of the Net.Congestion streaming media festival, argues that the networked environment should be seen as a social space: "The active sender and the passive audience/receiver have been replaced by a multitude of unguided transmissions that lack a designated receiver." 75 Beyond broadcast hegemony, he traces the emergence of "intimate media," which have a high degree of feedback. Media without an audience were first described in a 1992 text about "sovereign media" written by the Adilkno group (of which I am a member).76 Eric Kluitenberg further developed the idea, as did Joanne Richardson. Kluitenberg makes historical references to Bertolt Brecht's 1932 radio theory77 and George Bataille's text "The Accursed Share," in which Bataille writes: "life beyond utility is the domain of sovereignty."78 According to Eric Kluitenberg sovereign media should be understood as media beyond use. "They should not be understood as 'useless' but rather as 'without use.' Sovereign media have emancipated themselves from the demands of functionality or usefulness to exist in their own right."79 Erik Kluitenberg lists the Xchange network several times as an example of sovereign or intimate media.

The concept of sovereign media shows similarities with Andreas Broeckmann's reading of Deleuze and Guattari's idea of minor literature. Both emphasize the productive aspect of mediation. The difference, however, is that sovereign media no longer feel obligated to make references to the mainstream "majority." The act of declaring sovereignty over one's own mediacasting leaves behind dialectical polarities such as big/small, major/minor, broad/narrow, alternative/mainstream, and pop/elite. Sovereign media have long stated their declaration of independence and are not even indirectly focused on the "average user," "normal people" or "the Johnsons." The only function of mass media is to produce raw material, data garbage that sovereign media producers then freely use and reinterpret in their cut-ups. Remixing is not remediation. Mixes and cut-ups create entirely new art works and cannot be reduced to the nature of this or that source material. During the making of a mix there are no attempts to reach the higher plane of "immediacy." Instead of transplanting content from one platform to the next, sovereign media are getting serious about deconstruction. All meaning, all images and sounds, must be taken apart. Sovereign media no longer need the support or solidarity of the public (which the minority concept still appeals to). They have emancipated themselves from any potential imaginary audience. Live Internet radio often has few or no listeners. But this in no way bothers the streaming artists. That is true media freedom.

Tetsuo Kogawa speaks in this context of "polymorphous radio" or "polymedia." For him, communication is a "structural coupling". "The separation between transmitter and receiver is merely a political operation. Technologically, there is no separation between them."80 In the same context, Lev Manovich theorizes about "micro-media," pointing at the growing importance of tiny wearable devices. The terms vary, but the overall direction is the same. If streaming networks are serious about their intention to overcome the broadcast paradigm, they will have to free themselves from the public as a database filled with subjects. Adilkno: "Sovereign media do not approach their audience as a moldable market segment but offer it the "royal space" the other deserves."81 Certainly there is a historical connection between the democratization (availability) of media and the miniaturization of technology (portability). It is now time to reflect on the unavoidable trend of becoming micro. Does the proliferation of media technologies imply a solution of the "media question"? Scarcity often leads to speculation. Absence fuels the imagination. Will the universal ubiquity of networked devices foster a diverse climate of digital creativity and discontent, or unleash a culture of indifference?

Towards a Theory of Humble Networks

Unlike the dotcoms with their promises of unlimited market growth, networks such as Xchange have high scalability awareness. "Think Big" was the dominant leitmotif of the cyber age. The liberating spirit of mega was usually associated with the "tiger" economies of southeast Asia (before their 1997 economic meltdown).82 In Wired Bruce Sterling wrote a paean to the "overwhelming urge to be tall." "My beat is Jules Verne's idea of Big, the Prestigious Big - mega projects that exist because they exceed humanity's previous limits and break all the expected scales. Prestige mega projects are not big simply for functional reasons. They are not about the economic bottom line. Mega projects are about the top line - the transcendent, the beautiful, and the sublime. They are built for the purpose of inspiring sheer, heart-thumping awe - not unmixed with lip-gnawing envy from the competition. Mega is a very special conceptual world, a territory of fierce engineering ambition, of madly brash technical self-assertion. Mega is a realm that abolishes the squalid everyday limits of lesser beings."83

At the same time Dutch architect Rem Koolhaas made the rounds with slide shows and exhibitions featuring the growth-without-planning in Lagos, Nigeria, and the southern Chinese mega-cities designed overnight with the help of ordinary PCs. For Koolhaas, the "XL" strategy has been a liberating move away from petit-bourgeois politics and its bureaucratic regulatory regimes. The bold bigness of generic cities, their transurbanism, their mass-engineered towers, reflect the urgency of - and desire for - an anonymous mutated modernity.84 In a swift move the metaphorical bulldozers destroyed the dusty microcosms of decades. The techno-imagination of the New Era was anything but viral. It took a while for the promoters of Big to realize that large-scale projects were solely driven by speculative financial setups. Bigness could easily collapse if financial resources were withdrawn and economic recession set in. In that sense Koolhaas" XL approach is a product of the roaring nineties, the extraction of value from the post-1989 peace dividend. The gigantism of the Clinton era proved to be a special effect of short-term bubble policies, not a long-term trend. As The Economist formulated diplomatically, "the IT industry is becoming less of a growth story and more like a standard cyclical business. Traditionally, vendors have driven most big IT markets. But IT buyers are increasingly reluctant to play this one-sided game."85 Paradoxically, less growth also leads to fewer players. In a stagnating market the Big becomes even bigger.

Minoritarian practitioners, working within the spirit of Deleuze and Guattari or not, do not seek open confrontations. The humble streamers, passionately tinkering and hard to distract, hide in the shadow of the Big gestures, ignoring the zeitgeist, perhaps secretly hoping that the techno-cultural memes will one day burst out into society. They do not account for their tiny activities to the authorities and their "popular culture." Many never had revolutionary dreams in the first place. After the total disaster of existing socialism, leftist infighting snarled up long marches; it simply wasn't the time to Think Big. It was not a coincidence that the Xchange network was administered from a former Soviet republic. Let's conclude by saying that petite networks, cute or not, are here to stay. Time and again, pocket-sized nodes are proving immune to the fast-paced fluctuations of global capitalism.

Streaming Futures

Five years after its founding the Xchange network has found a modest, pragmatic way of operating. Despite the fact that a common Web portal-cum-net.radio scheduler has not yet emerged, collaborations do happen. The network regularly meets and puts out print publications. Pit Schultz (Klubradio, Berlin) stresses that, despite the loose ties, projects have emerged out of the Xchange network that do attract audiences. "I can only talk about the numbers I know. Sites such as Betalounge, Groovetech and Klubradio have thousands of visitors a day, not a gigantic number but certainly more than the average new-media institution with perhaps a 1,000-times-higher budget."86 No matter how fragile independent streaming networks may be, the Xchange example shows that valuable collaborations are the result, perhaps not visible to the outside world, but so much more sustainable than the defunct dotcoms.

A critical streaming discourse is still in its infancy. Caught between the established 20th-century discourses on radio, (pop) music and sound art, streaming is still off the radar of most critics and curators. This is true even in the new-media arts system itself, with its recently opened centers and annual festivals. University departments and cultural institutions with their own dedicated streaming servers are still a rarity, even though streaming from live events is on an upward trend. System administrators do not like bandwidth-eating streaming servers. Yet the streams are silently streaming. Pit Schultz: "The role of 'sound' is really important, and what that means in a geographically diverse Internet context, is providing a platform for non-textual exchange. There is what one could call the 'Nordic element' in Xchange. Not talking much, but saying a lot. Much of this (invisible, silent) work is done on the local level, in developing nodes, interconnecting them in a loose way. Even during the most active times of Xchange, it would be difficult to describe where the fascination manifests itself - in the in-between, the actuality of the live element, the process of exchange, meetings and relationships."87

The broken dreams of Web TV still echo through the Net. Limited by underutilized broadband capacities ("dark fiber"), Xchange is setting out to explore what sound means beyond downloading MP3 files. Their message is a simple but challenging one: streaming is more than radio or television on your computer screen. Like peer-to-peer networks, independent streaming networks put on the table the question of what users have to contribute once they are confronted with the wide range of technical possibilities the Internet has to offer.

Footnotes

1. This chapter was based on research into the few thousand or so postings to the Xchange list from December 1997 to July 2002, Xchange members' related websites, and relevant publications. I would like to thank Joanne Richardson, Rasa Smite, Pit Schultz and Adam Hyde for their valuable comments.

2. A random "googled" definition of streaming media states: "Streaming media is a way to enhance World Wide Web sites by integrating video and audio with a site's text-based content. Unlike downloading a video or audio file separately and then playing it back later, streaming media plays when a Web page (with the embedded streaming video or audio) downloads on a user's browser. To include streaming video or audio, the server hosting the Web site must support the application. To play streaming media embedded in a Web page, the user must have a helper application such as a viewer or plug-in. Viewers are typically offered free to users" (members.tripod.com/Lori_Sylvia/stream.htm).

3. A technical explanation: "Uncompressed CD-quality WAV and AIFF files are too large to stream over the Internet for playback in real time. Instead, lighter-weight compressed formats such as MP3 and RealAudio are employed for streaming network audio. These formats use what are called 'lossy' compression schemes, reducing file sizes by eliminating certain inaudible data from the original files without too significantly degrading playback sound quality. MP3 and RealAudio are excellent streaming formats, achieving performance factors great enough to allow real-time encoding/decoding over current network bandwidth conditions while delivering satisfying audio quality." From: linux.oreillynet.com/pub/a/linux/2001/03/23/streaming_media.html?page=2.

4. E-mail interview, January 17, 2002.

5. Ibid.

6. There had been early networked sound-arts experiments in the early-mid 1990s, before the Internet, using BBS systems and direct telephone connections. Heidi Grundmann of Kunstradio (ORF) in Vienna was one of the pioneers. See the interview with Heidi Grundmann by Josephine Bosma, Nettime, July 15, 1997. URL: www.kunstradio.at/. We could also mention Gerfried Stocker, whose Horizontal Radio project in June 1995 connected 36 radio stations for a 24-hour program in which listeners were able to control the audio mix via a Web interface. URL: gewi.kfunigraz.ac.at/~gerfried/horrad/.

7. We could mention here the first and second Art+Communication festivals organized by E-lab in Riga in November 1996 and 1997; a workshop during the second Interstanding conference in Tallinn, Estonia, in November 1997; the radio workshop at the Hybrid Workspace during Documenta X (organized by Kunstradio, Vienna) and the weekly radio shows from Kassel; and the network (hosting) activities of Radio Internationale Stadt (Berlin) and XS4ALL (Amsterdam).

8. Introduction, Xchange, December 2, 1997. URL of the list archive: www.xchange.re-lab.net\m\index.html.

9. During the first live session, on December 2, 1997 Peteris Kimelis was interviewed about his work 3 Frequencies. "He has made a sound just from 3 frequencies -> 3000Hz , 5000Hz , 8000Hz. These are natural frequencies of the human ear but what you actually hear is beeeep and it's hard to listen to this sound for a long time." Raitis, Xchange, December 3, 1997. The Xchange live web sessions are archived at ozone.re-lab.net/sessions.html.

10. For more on Pararadio see the interview by Josephine Bosma, Xchange, December 11, 1997.

11. List adapted from Borut Savski's posting to Xchange, December 15, 1997.

12. See, for instance, Josephine Bosma, Interview with Pararadio, Xchange, December 11, 1997; net, 'radio' & physical space: E-lab/RE-lab/Xchange/OZOne, Nettime, January 7, 1998; From net.art to net.radio and back again, written for the 1998 Ars Electronica catalogue, posted on Xchange, July 11, 1998.

13. E-mail exchange with Rasa Smite, January 13, 2002.

14. Pit Schultz, in an e-mail, June 13, 2002.

15. See www.ecommercetimes.com/perl/story/21537.html. On May 18, 2003, the Pew Internet Project published the research paper "Broadband Adaptation at Home": "In the survey Pew found that 57 percent of dial-up users had no interest in upgrading to faster access, even in areas where such service was already available."

16. On June 25, 2002, US Representative Howard L. Berman proposed a bill aimed at the "unbridled" piracy taking place on decentralized peer-to-peer file-sharing networks by introducing legislation that lets copyright holders employ tools to prevent illegal trading. Berman said copyright holders were at a disadvantage against P-to-P pirates under existing legislation. "While P-to-P technology is free to innovate new and more efficient methods of distribution that further exacerbate the piracy problem, copyright owners are not equally free to craft technological responses." Berman's bill came on the heels of a slew of legislative efforts to curb piracy in the digital realm. Another bill, proposed by Senator Ernest Hollings, sought to incorporate digital rights management technologies in all consumer electronic devices. Source: PC World, June 26, 2002.

17. Source: Arbitron/Edison Media Research: Internet 9, "The Media and Entertainment World of Online Consumers," July 2002. URL: www.edisonresearch.com/I9_FinalSummary.pdf. By mid-2002, seven in 10 US Americans had access to the Internet. In early 2001, 13% of those with Internet access had had a broadband connection; as of July 2002, this number had grown to 28%. The inquiry only asked if users had had streaming-media experiences. Users are defined here as consumers of media products, not as potential producers of content and social networks.

18. The dilemma for individual artists between high and low bandwidth was based on the growing paradox between promises and technical capacities on the one hand and the actual availability of affordable high-speed connections on the other. George Gilder in August 2000: "Since the commercial advent of wave division multiplexing in 1996, bandwidth has been increasing at least four times as fast as Moore's Law, if not faster, and promises to continue to do so for at least the next several years." (www.forbes.com/asap/00/0821/097.htm). The reality, however, was one of rising costs and a stagnation in Internet usage due to telcos withholding "dark fiber" (unused capacity) from clients for profit reasons. See: Geert Lovink, The Bandwidth Dilemma, posted on Nettime, March 26, 2001 (originally published in German and English in Telepolis online magazine (www.heise.de/tp).

19. Pit Schultz, re: A Stimulus to Make the Most Productive Use of Radio, Xchange, May 19, 1998.

20. Rasa Smite, Xchange, December 9, 1997

21. Interview with Zina Kaye, Sydney, May 31, 2002.

22. Report of the Art Servers Unlimited meeting can be found in Acoustic Space 2, Riga: E-lab, 1999, pp. 14-24.

23. xchange.re-lab.net/56k and www.openx.aec.at/xchange/. See also report in Acoustic Space 2, Riga: E-lab, 1999, pp. 34-40, and Zina Kaye's post on Xchange, September 7, 1998, which includes the names of the participants and a program.

24. For the program of the webjam see Rasa Smite's posting on Xchange, November 29, 1998.

25. First international Net.Congestion streaming-media festival, Amsterdam, October 6-8, 2000, net.congestion.org. The website contains an archive of all the panels. The initial concept was developed by Adam Hyde. The first announcement was posted on May 3, 2000, by Honor Harger.

26. Conceptual Background, Net.Congestion Reader, Amsterdam: De Balie, 2000, p. 4.

27. More on the philosophy of mixing: Geert Lovink, "An Inventory of Amsterdam Free Radio Techniques," in Neil Straus (Ed.), Radiotext(e), Brooklyn: Semiotext(e), 1992.

28. Raitis Smits, "Xchange Open Channel: Co-broadcast Experiments in the Net," in Nettime (Ed.), Readme! Brooklyn: Autonomedia, 1999, p. 349-350.

29. An IRC chat at the #xchange channel, dated March 31, 1998, was printed in the Nettime reader Readme! Brooklyn: Autonomedia, 1999, pp. 343-350. An example of the city-to-city cast: "Listen to RA radio aura today 8PM(GST): active.llaky.fi/aura/aura.ram, music/sounds/and/ambience by >scott scott >radio Banff >pk >martins ratniks >sera furneaux >matti adolfsen.. sydney((o))tornio((o))stockholm((o))banff((o))scotland((o))riga." Xchange, August 27, 1998.

30. As described in Raitis Smits, Xchange Open Channel: Co-broadcast Experiments in the Net, in: Nettime (Ed.), Readme! Brooklyn: Autonomedia, 1999, pp. 349-350.

31. Daniel Molnar, Xchange, January 4, 1998.

32. E-mail exchange with Rasa Smite, January 13, 2002.

33. URLs for the Pulkvedis club event in May 1998, with participation of D:U:M:B, hardcore techno DJs and VJs from Rotterdam: ozone.re-lab.net/archive/dumb.ram and ozone.re-lab.net/archive/dumb2.ram. URL of the train transmission: ozone.re-lab.net/archive/vtt/train.ram. Information provided by Rasa Smite via e-mail, June 4, 2002.

34. Josephine Bosma and Eezee E, Xchange, August 17, 1998.

35. Tetsuo Kogawa, "The Other Aspects of Radio," Acoustic Space 3, Riga: E-Lab, 2000, pp. 26-28. Also: "Minima Memoranda," Next Five Minutes 3 Workbook, Amsterdam: De Balie, 1999, pp. 103-104.

36. Xchange, February 25, 1998.

37. Interview with Zina Kaye, Sydney, May 31, 2002.

38. Extended Life Radio, Xchange, February 12, 1998.

39. For the entire interview with Micz Flor by Geert Lovink, see Nettime, Tactics of Streaming, April 26, 2002.

40. Matthew Smith, re: G2, Xchange, June 2, 1998.

41. Ibid.

42. "Portal" is a late-1990s term for a website that brings together Web resources for ordinary users on a central page. A portal usually imitates the layout of a newspaper front page. At the left and right are small menu bars for links. In the centre are the headlines of the main stories. A definition of a portal, found with the Google search engine, reads: "A portal is an entry point or starting site for the World Wide Web, combining a mixture of content and services and attempting to provide a personalized "home base" for its audience with features like customizable start pages to guide users easily through the Web, filterable e-mail, a wide variety of chat rooms and message boards, personalized news and sports headlines options, gaming channels, shopping capabilities, advanced search engines and personal homepage construction kits." URL: http://www.auburn.edu/helpdesk/glossary/portal.html.

43. Radioqualia, Xchange, October 13, 1998. URL: www.radioqualia.va.com.au.

44. See: Radioqualia, streaming media software for arts released, Nettime, November 7, 2002.

45. An Open Letter from Adam and Zina, Xchange, November 10, 1999. See also Adam Hyde's report of the Streaming Media Europe Conference, November 22, 1999, which contains interesting comparisons between Xchange and commercial streaming portals. "During the conference it quickly became apparent to me that the Xchange community is very advanced in its use and thinking about streaming media. In fact many of the ideas I heard coming from businesspeople working with streaming media did not (generally) reflect that same quality of ideas and breadth of understanding about the medium that I know to exist in Xchange. I found this very surprising and exhilarating as I personally was beginning to think that we (Xchange) were being left behind by industry."

46. Rasa Smite, Next steps for the Xchange Website, Xchange, February 20, 2000.

47. An exception which should be mentioned here is Xchange's involvement in streaming projects related to the Kosovo conflict in 1999, sparked by the closure of the Belgrade independent radio station and streaming pioneer B92 (see Chapter Three on the history of the Syndicate list).

48. This part of the essay draws heavily on an e-mail exchange with Adam Hyde in May 2002. The quotes are from the same exchange.

49. E-mail interview with Adam Hyde, May 20, 2002.

50. E-mail interview with Adam Hyde, May 20, 2002.

51. Here are some statistics compiled by Nielsen/NetRatings: in April 2002, RealMedia reached 17 million at-home viewers, compared with Windows Media at 15.1 million and QuickTime at 7.3 million. At work, Windows Media drew about 12.2 million unique viewers, compared with RealMedia at 11.6 million and QuickTime at 5 million. Nielsen/NetRatings' last multimedia report, for December 2001, showed that RealNetworks reached some 32 million users at home and 16.3 million at work; Windows Media hit 14.6 million at home and 9.9 million at work; and QuickTime reached about 7.4 million people at home and 5.5 million people at work. Source: .

52. "Linux Media Player" (linmp), a solo project so far, has been in development since late 2001. See: sourceforge.net/projects/linmp/.

53. E-mail interview with Drazen Pantic, January 17, 2002.

54. Pit Schultz, OMA alpha release party, Xchange, June 21 2001.

55. Pit Schultz, orang.orang.org is getting reconstructed, Xchange, December 31, 2001.

56. Pit Schultz, digital hooliganism, Nettime, March 16, 2002.

57. Rasa Smite, Xchange, October 17, 1998.

58. E-mail exchange with Rasa Smite, January 13, 2002.

59. Derek Holzer, Call for Inputs: Acoustic Space Labs!! Xchange, June 28, July 2, and August 3, 2001, and e-mails from other Xchange members in the same period. For a report see Mukul, The Wire, September 2001, posted on Xchange, September 3, 2001.

60. Rasa Smite, e-mail interview, January 13, 2002.

61. RIXC is a joint effort by a number of independent local cultural groups based in Riga, Liepaja and other cities in Latvia working in the fields of new media, art, film, music, youth culture and the social field. The founders of RIXC are E-lab, the film and TV production studio Locomotive, and the Baltic Centre, an NGO for alternative education and social projects. URL: rixc.lv.

62. Jay David Bolter and Richard Grusin, Remediation: Understanding New Media, Cambridge, Mass.: MIT Press, 1999. "Remediation is a defining characteristic of the new digital media" (p. 45). "Each act of mediation depends on other forms of mediation. Media are continually commenting on, reproducing and replacing each other, and this process is integral to media. Media need each other in order to function as media at all" (p. 55).

63. E-mail interview with Drazen Pantic, January 17, 2002.

64. Tetsuo Kogawa, Minima Memoranda, in Next Five Minutes 3 Workbook, Amsterdam, 1999, p. 104.

65. See: Gilles Deleuze and Feacutelix Guattari, Kafka: Towards a Minor Literature, Minneapolis: University of Minneapolis Press, 1986.

66. Andreas Broeckmann, "Minor Media - Heterogenic Machines," in Acoustic Space 2, Riga: E-Lab, 1999, p. 78. URL: english.uq.edu.au/mc/9909/minor.html.

67. Ibid.

68. Feacutelix Guattari, quoted in Andreas Broeckmann, "Minor Media ñ Heterogenic Machines," in Acoustic Space 2, Riga: E-Lab, 1999, p. 82.

69. Reference to Hakim Bey, Temporary Autonomous Zone, Brooklyn: Autonomedia, 1991. In the early 1990s the anarchist TAZ concept became part of the cyber-libertarian Wired generation vocabulary. Some claimed the entire Internet to be an autonomous zone. This could more accurately be said of certain virtual communities such as Xchange.

70. Erik Davis, Acoustic Cyberspace, Xchange, December 29, 1997. Also in: Acoustic Space 1, E-Lab, Riga, 1998, p. 23.

71. TMs in TM.SELECTOR#2, a connoisseur's guide to net listening published by the www.irational.org group (October 2001).

72. Andreas Broeckmann, "Minor Media - Heterogenic Machines," in Acoustic Space 2, Riga: E-Lab, 1999, p. 78. Staying close to Feacutelix Guattari's terminology, Broeckmann features the Xchange network as an example of "heterogenesis" and "molecular revolution." "Xchange is a distributed group, a connective, that builds creative cooperation in live-audio streaming on the communication channels that connect them. They explore the Net as a soundscape with particular qualities regarding data transmission, delay, feedback, and open, distributed collaborations. Moreover, they connect the network with a variety of other fields. Instead of defining an 'authentic' place of their artistic work, they play in the transversal post-medial zone of media labs in different countries, mailing lists, netcasting and FM broadcasting, clubs, magazines, stickers, etc., in which 'real' spaces and media continuously overlap and fuse." See also Andreas Broeckmann, "Konnektive entwerfen! Minoritaere Medien und vernetzte Kunstpraxis," in Stefan Muenker and Alexander Roesler (Eds.), Praxis Internet, Frankfurt am Main: Suhrkamp, 2002 (posted on the Rohrpost list, May 15, 2002).

73. Erik Davis, Acoustic Cyberspace, Xchange, December 29, 1997. Also in: Acoustic Space 1, E-Lab, Riga, 1998, p. 24.

74. Eric Kluitenberg, "Media without an Audience," in Acoustic Space 3, Riga, 2000, p. 7.

75. Bilwet, Media archief, Amsterdam: Ravijn, 1992. English translation: Adilkno, Media Archive, Brooklyn: Autonomedia, 1998, pp. 12-15. This essay has been updated and commented on in 2001 by Geert Lovink and Joanne Richardson, Notes on Sovereign Media, Nettime, November 14, 2001, and published on the Web in Subsol magazine: subsol.c3.hu.

76. Bertolt Brecht, "Der Rundfunk als Kommunikationsapparat. Rede uumlber die Funktion des Rundfunks" (1932), in Bertolt Brecht Werke, Schriften I (1914-1933), Band 21, Frankfurt am Main: Suhrkamp Verlag, 1992, pp. 552-557.

77. Quoted in Eric Kluitenberg, "Media without an Audience," Acoustic Space 3, Riga, 2000, p. 7, posted on the Xchange list, October 19, 2000. George Bataille, The Accursed Share, Cambridge, Mass.: The MIT Press, 1991.

78. Kluitenberg, p. 7.

79. Tetsuo Kogawa, "Other Aspects of Radio, From Mini FM to Polymorphous Radio," in Acoustic Space 3, Riga, 2000, p. 27. See also: Tetsuo Kogawa, "Minima Memoranda," Next Five Minutes 3 Workbook, Amsterdam, 1999, pp. 103-104.

80. Adilkno, Media Archive, Brooklyn: Autonomedia, 1998, p. 14.

81. See Jee Greenwald, Thinking Big, Wired 5.08, August 1997, on Malaysia's Multimedia Super Corridor. After the 1997 monetary crisis and the 2000 NASDAQ crash, "big equals good",ideas became less prominent.

82. Bruce Sterling, "The Spirit of Mega," Wired 6.7, July 1998. URL: www.wired.com/wired/6.07/mega_pr.html.

83. See: Rem Koolhaas and Bruce Mau, OMA: S M L XL, Rotterdam: 010 Publishers, 1995, and Lars Spuybroek meets Rem Koolhaas, "Africa Comes First," in: Joke Brouwer et al. (Eds.), Transurbanism, Rotterdam: V2_Publishers, 2002, pp. 160-193.

84. Reprinted in The Australian, August 27, 2002, p. 31.

85. Pit Schultz in a personal e-mail to the author, June 13, 2002.

86. Ibid., June 14, 2002.