Chapter III: The Aesthetics of Bits and Pieces and Other Web Based Art

Amy Alexander: The Multi-Cultural Recycler
Various Artists: Loose Ends/Connections and Feedback
John Neilson: Radio Stare
So Where Does Bits and Pieces Fit In?

The Internet is a globally based, non-centralized computer network that is designed to enable computers to communicate with each other through a number of standard protocols. The most basic of Internet connections is available to anyone with a computer, modem and phone line. Further up the scale are Internet nodes and portals, which are companies or educational institutions that control access to the Internet, carry Internet content, and route Internet data to other nodes and portals. America Online is an example of an Internet portal. The content of the Internet includes, among other things, e-mail, chat rooms, newsgroups and the World Wide Web. The Web itself is only part of the Internet. However, it is the part that has garnered the most public and corporate attention. This is because the Web involves the use of text, graphics, and sound. It provides a visual, sometimes audible, and always interactive interface to the Internet.

The field of Web based art has come into existence in the past seven or so years. The first graphical Web browser, Mosaic, was released in 1993 (Cailliau). Previous to its existence, artists had been working with multi-media and computer based art, but not within a medium that allowed the virtually instantaneous distribution of that art to anyone in the world with an Internet connection. One of the earliest Internet based musical compositions was a piece performed by the San Francisco computer network band, The Hub. The piece, entitled Hub Renga (1989), was based on messages sent to them by members of the use-net group, The Well. The members sent in lines of poetry, and The Hub responded musically to their messages as the performance was broadcast live over the radio. This early example illustrates some of the potential interests in creating art on the Web. It made use of the possibilities offered for audience interaction. It also made use of the Internet’s potential as a source for live interaction on a global scale. Nevertheless, the piece was written at a time when the term Internet had not yet entered into the vocabulary of popular culture, and the World Wide Web did not exist.

The growth in the number of Web sites over the past several years has been exponential. The number of existing Web sites currently doubles approximately every six months. From June 1993 to January 1997, the number of Web sites increased from 130 to approximately 650,000. Although current statistics are not yet available, one can assume, using the doubling rule, that the number of sites may now reach well into the millions. Furthermore, the percentage of corporate sites between those dates rose from 1.5% to 62.6% (Gray).

From an artist’s perspective, the growth in content on the Web combined with a rapidly increasing number of users, greater speed, and growing technological capabilities, mixed with an ever expanding corporate presence, provides an almost unlimited source for the mining of artistic material. It also provides a virtual canvas and a gallery in which to display work, as well as a potentially enormous audience. David Ross, director of the San Francisco Museum of Modern Art, describes Web art as a revolutionary form:

Ross later describes how the Internet has created a new context for the art that makes use of it:

Currently, the number of Web art sites and pieces is growing at a phenomenal rate. Consequently, it is impossible to keep up with all of them. The pieces chosen for discussion here could probably be replaced by other ones that are equally relevant, but that I just never found. The rest of this chapter will take a closer look at some Web art works, followed by a discussion of the aesthetic ideas behind Bits and Pieces as they relate to the current state of art on the Web.

Amy Alexander: The Multi-Cultural Recycler

The Multi-Cultural Recycler is a graphics based Web piece that has some conceptual similarities to Bits and Pieces. The Multi-Cultural Recycler uses sampled images from Web cameras, commonly referred to as Webcams, to create instant visual art whenever the Website is loaded into a browser. Webcams are small video cameras that can be connected to a Web server and used to transmit live images over the Web. The piece, created by California Institute of the Arts (CalArts) faculty member, Amy Alexander, samples live Web images, then mixes and performs algorithmic processes on them. Unlike Bits and Pieces, the Multi-Cultural Recycler does not search the Web for its images. Rather, it has a preset number of Webcams, currently 45, that it randomly chooses from.

A primary difference between the sampling in the Multi-Cultural Recycler and Bits and Pieces, is that the Multi-Cultural Recycler's samples are taken from mostly live images (although some of the Webcams are only updated every several minutes or hours). This means that the images can change by the day or even by the minute, depending on the location of the Webcam. Bits and Pieces samples pre-existing content in the form of sound files, rather than live sound broadcasts.

When a user loads the page, the Recycling program randomly picks two or three Webcams to sample, depending on the algorithm chosen. A still image is captured from each Webcam, and analyzed for the presence of certain features determined by the chosen process. A collage is then made of the images. The outcome can vary from unrecognizable patterns to surprising and unintended juxtapositions of recognizable subjects. The meaning of unintended here must be clarified. The choices made at the Recycler’s moment of composition are random and without respect to the subjects of the individual Webcams. However, Alexander’s choices of which Webcams to have her work sample was very much intended. The Webcams have been chosen so that these strange and sometimes funny juxtapositions will occur. The random selection of material within strict parameters to achieve a desired result is similar to Cage’s approach in his radio works. For example, there are several Webcams of famous places such as the Wailing Wall in Jerusalem and Potsdamer Platz in Berlin. There are also Webcams of a guinea pig, a cat litter box and a toilet. Thus, these juxtapositions are bound to occur on a fairly regular basis.

Once a new image has been produced, the viewer may look at the source images. The viewer may also click on those images to link to the source pages for the Webcams. Viewers may also choose Webcams for processing if they’d like to see a particular combination of images. Furthermore, they can submit their work to be displayed in the Recycler’s gallery.

Like Bits and Pieces, the Multi-Cultural Recycler deals with the sampling and random recontextualization of a potentially exhibitionistic and voyeuristic subject matter. Many Webcams are owned and operated by people interested in displaying themselves and their lives on the Web. Some also display inanimate objects or pets. The personal reasons for this vary, but the end result is the same. Thousands, and perhaps millions, of Web surfers tune in to look at other people’s lives:

Because the Web is such a new medium, artists must develop new ways of dealing with it. Its limitations can be numerous, so many Web artists have taken those limitations and turned them into artistic material. For example, Alexander is limited by both the quality and the content of the Webcams. The images they provide are small, poor quality, and often give little more than a shallow and removed perspective of their subject:

Alexander also views the browser as a current limitation to Web art:

Figure 6. An example of Amy Alexander’s Multi-Cultural Recycler. The top image is a synthesis of the two bottom images. Viewers may also choose the Webcams for the Recycler to process if they wish.

My approach in Bits and Pieces is somewhat similar to the Multi-Cultural Recycler in that I view the use of randomly sampled sounds from the Web as potentially voyeuristic. Like the Webcam images, many of these sounds are disembodied or decontextualized, and most of the time only "narrowly representative" of the sources. By this, it is meant that most of the sounds are so short, or of such low quality, that they only give us a glimpse or suggestion as to the nature of their original source. The contrasts and juxtapositions of these sounds, however, can create meaning where little or none was present, inadvertently conjuring up cultural or social contrasts (i.e. an opera singer mixed with a rap artist).

Various Artists: Loose Ends/Connections and Feedback

The following two pieces were created by a group of artists collaborating from different locations across the US. The performances, which occurred on specific dates, but are now permanently archived on a host site, explored the use of sound in a collaboration of eight composers/musicians. Each work took approximately three weeks of preparation for the live broadcast. Both works lasted about 50 minutes.

The first piece, Loose Ends/Connections, was performed live over the Web on September 19, 1998 at 8 pm, EST. It involved transmitting audio from two locations to a third location that mixed the audio and streamed it over the Web. The first location was Mills College in Oakland, CA. There, composer/performers Pauline Oliveros, Maggi Payne, and Brenda Hutchinson performed with both acoustic and electronic instruments while their audio/Internet engineers transferred the recording to RealAudio and streamed it to the Morton Street Studio in New York City. The second location was at Harvestworks, Inc. in NYC. There, composer/performers Beth Coleman and Zeena Parkins added their input to the performance, which was also streamed via RealAudio to the Morton Street Studio. Finally, three more composers at the Morton Street Studio, Scott Rosenberg, Helen Thorington and Jesse Gilbert, added their input to the mix. The final product was then converted to RealAudio and streamed live over the Web. On the site, the audience could listen to either a high or low bandwidth version. Furthermore, the main page ran a slide show of video stills from Mary Lucier's 1998 Summer, or Grief with text from "A Conversation" by Allen Grossman. The slide show was not synchronized with the audio, and thus audience members could move around to different points within it.

The piece itself included the use of acoustic and electronic sounds. Insects, howling wolves, distorted voices, water drops, drum beats, saxophones, and clarinets all appeared within the mix. All performers had access to the sound coming from the other locations. However, there was an eight second lag time between locations due to network speed.

The piece had a meditative feel, changing very gradually. It also appeared to draw parallels and contrasts between technology and nature. Natural sounds such as animals, waves, and wind combined or juxtaposed with their electronically processed counterparts has been a common theme in electro-acoustic music for many years.

The second piece, Feedback, was performed a few months later on December 6, 1998 at 5 pm, EST. Like Loose Ends/Connections, this piece also explored the uses of sound in networked collaborations. Similarly, the music was streamed from the same three locations, although some of the artists were different. This time Mills composer/performers Pauline Oliveros, Maggi Payne, and Brenda Hutchinson played long tubes, conch shells, flute, and voice. Harvestworks had Shelley Hirsch and Jim Pugliese performing voice and percussion, and the Morton Street Studio had Jesse Gilbert and Helen Thorington mixing and adding prerecorded music. This time, instead of using a slide show, an interactive painting tool was created by poet and animator Neil Zusman. The painting tool does not interact with the audio (both use different browser plug-ins), but audience members are encouraged to paint to the music. The objects being painted change shapes and colors, making the experience unpredictable, and at times also meditative.

The music is similar stylistically to the first piece. It is a mixture of prerecorded electronic sounds, human voices, and instruments. One gets the feeling that these pieces were improvised, with only a simple framework given to shape the overall work.

Taken just as music, these pieces are nothing new aesthetically. Taken as a whole, however, they represent a new movement in electronic arts and electronic music. What they attempt to do technically is facilitate live interaction between musicians in different places. They also add graphical elements. Furthermore, the output is potentially audible to anyone on earth with an Internet connection. This piece, although hosted on a server used specifically for these types of works, could have been hosted on anyone’s private server.

The potential to be able to stream content from a personal server anywhere in the world is revolutionary from an artistic perspective. It presents the possibility of removing middlemen such as record companies, radio stations, museums, and other entities which may either charge for the right to transmit the work, or attempt to profit from the transmission itself. Collage artist and founder of the New York Correspondence School, Ray Johnson, pioneered this approach in the early 1960’s when he began using the postal service to mail his small collages and drawings to a personal network of friends and artists. He further encouraged them to participate in the works by adding to them and sending them back to him or on to others. He created an art movement that, for many years, existed outside of the galleries, and was thus completely in the hands of the artist and the audience (Longhauser). With the Web, the ability to keep the work in the artists’ hands and to display or transmit that art to the largest possible audience at relatively little cost means that more artists whose work would otherwise not receive exposure can have access to the global Internet community. It is doubtful that most commercial American radio stations (and even many public radio stations) would support or fund works like these because of their lack of commercial viability.

The medium also provides the ability to add more than just sound to the work. As shown here, these pieces incorporate graphics and, to some degree, interaction. The interactive potential of the Web (although not really utilized in Bits and Pieces), makes it very exciting and attractive to multimedia artists.

John Neilson: Radio Stare

Radio Stare is a Web piece that utilizes three different tools to create a strangely relaxing visual and sonic experience. Its central motive is the use of live police scanners as sound sources. There are several sites on the Web that stream live police scanner audio through RealAudio, and John Neilson utilizes them in a new and highly inventive way.

Radio Stare begins with a graphic of a radio tower against the night sky. A few lone beeps are heard. The view pans upward until the tower disappears. Another screen then opens with a first person perspective animation of a drive down a straight, lonely road in the middle of the night. Occasionally, one passes a radio tower. Simultaneously with this opening, the browser downloads a Quicktime MIDI file. Quicktime MIDI Instruments are a group of generic instrument sounds that can be played by most. The tune combines a slow, steady drum rhythm with mellow synth pads. It is intentionally bland and repetitive, with only the occasional pause and drum fill. Simultaneously, and perhaps most importantly, the browser opens up the RealAudio Player to receive a live stream from a police scanner site. It picks the stream from several available sites. The user then adjusts the volume on the RealAudio player to an appropriate level in relation to the Quicktime sound. Instructions on the page tell the listener/viewer to turn off the lights and relax to the experience:

I found this to be one of the more interesting Web art pieces I’ve experienced to date. Conceptually and technically it is very simple. It does not have live musicians communicating over a network, nor does it sample and manipulate Web content. Rather, it appropriates a readily available sound source into the art work without changing it in any way. It adds a separate sound and graphical layer to frame the radio source. The final result varies from undramatic and repetitive, to soothing, to strangely unsettling. The graphical and Quicktime MIDI aspects of the piece remain constants throughout. It is the police scanner output that really determines the feel of the work.

The rhythmic backing seems to accent the distorted and disembodied voices of police officers performing their mostly uneventful routines. In a true Cagian sense, the police scanner provides audio content that is completely beyond the composer’s control. Furthermore, it is beyond the police officers’ control. Radio programs and stations usually have a format and a schedule, so that even if the output, within the scope of Cage’s radio based works, is unpredictable, the output with the respect to those working at the radio station is almost completely predictable. Unpredictability and the possibility of surprise are taken to new levels with the use of police scanners. At any moment a situation may arise that causes a flood of communication over the police network. The communication, itself a collage of voices transmitted from around the precinct, is only marginally controlled by the dispatcher. Thus, completely unpredictable events (from the perspective of the police) such as robberies, shootings, and accidents, determine in large part the overall output of the scanner. It is these surprises, and furthermore the fact that they are live, that make Radio Stare such a simple yet engaging work:

It is also interesting to consider the network relationships that pervade this piece. The individual police officers each comprise a node attached to the central hub, which is the dispatcher. As they move about the area, they each report back their experiences and actions. Their sonic output is mixed in a collage of communication that each officer hears. That collage is then sent out over the Web through a scanner site, appropriated in a work of art, and sent back out over the Web to a new group of nodes, the audience.

There is also a strongly voyeuristic quality to the piece. There is something unsettling about listening to live radio transmissions of police officers performing official business without the knowledge that their experiences, however banal or life threatening, are being appropriated in a work of art. Radio Stare is successful because it uses the Web’s communication potential in such a simple but profoundly thought provoking way. Through its use of the police scanners, it becomes a work about the communication network it utilizes.

So Where Does Bits and Pieces Fit In?

Although this paper has discussed several sites that make extensive use of sound, the number of Web art sites based on sound is disproportionately smaller than those based on graphics. I feel, as a composer, that it is important to exploit the artistic potential of this medium in as many ways as possible. This being my first foray into Web art, I have seen many Web based pieces that are truly on the cutting edge of electronic art. Still, as far as I could tell, none dealt with the random sampling, processing, and broadcasting of Web sound. From my perspective then, Bits and Pieces is a Web first.

On a deeper level, Bits and Pieces incorporates the collage aesthetic in several ways. It is a mix of randomly chosen sounds, but it is also a mix of styles and aesthetics. I have attempted to take the chance elements present in Cage’s music, such as randomly sampled sounds, and mix them with controlled or algorithmic processes, such as pulse gates and phasing, present in composer Steve Reich’s early minimalist works. Other processes use predefined structures where certain variables are only determined at runtime, and the sonic output of which can be greatly varied from the content of the samples. In doing this, I am attempting to skirt the line between control and lack thereof. This creates an overall effect in which each piece is recognizable as a member of the set, but contains its own unique parameters and thus a unique identity.

Unlike Imaginary Landscape No. 4, in which the score, and thus the overall shape of the piece, is predetermined, Bits and Pieces does not have a definite score or overall shape. The fact that the process itself is randomly chosen every fifteen minutes creates one more layer of collage on top of those existing within the pieces. Since a new process is called every fifteen minutes, and users may only listen to the ten most recent pieces, a whole new group of pieces exists every two and a half hours. If the ten sections of Bits and Pieces are considered as a whole, then the shape of that whole is altered every fifteen minutes, and completely different every two and a half hours.

In choosing WAVE and AIFF files as the sources for Bits and Pieces, I also imposed a limitation on the work. WAVE and AIFF files generally share similar characteristics on the Web in that they are usually under one minute long and their sample rates and bit depths are less than CD quality. Another currently more abundant sound file format, is that of MPEG-3. This compression format allows entire songs to be compressed into only a few megabytes with little noticeable reduction in sound quality. Since a special MPEG-3 decoder is required for playback, the sound files would need to be decoded and converted to WAVE files before their use in Bits and Pieces. Because many MPEG-3 files are entire songs or pieces, the resulting decoded files would probably be several minutes in length, making them too large for use in Bits and Pieces. Furthermore, MPEG-3 was not used as the compression scheme to deliver Bits and Pieces, because it currently does not deliver higher quality streaming than RealAudio. Users usually download MPEG-3 files before playing them. This can take a long time for users with slow connections. RealAudio, besides being the most popular format for audio streaming, allows virtually instant playback of the installation.

There are also the questions of exhibitionism and voyeurism. The growing number of people displaying themselves on the Web, both graphically and sonically, has lead to a culture of anonymous voyeurism. Everyone is curious about how other people live, and what they do in their most private moments. There are people who don’t display themselves on the Web through cameras, but rather put sound files of their music on the Web (as I have done). They may also put up sound files of their favorite songs, lines from TV shows and movies, poems, or animal sounds. Through this display, they put some aspect of themselves onto the Web. It is as if they seek a form of digital immortality by committing some highly personal aspect of themselves to a global network. Bits and Pieces is most engaging when samples of this nature are retrieved. From these, it can become an unintentional narrative about the need to present oneself and the search for individuality in a crowded and often all too homogenous space. Similar to Radio Stare, it appropriates personal sound for artistic purposes. Some of the pieces will be bland and uninteresting, but occasionally a chance grouping of certain sounds will yield surprising and meaningful results. This, to me, is where Bits and Pieces succeeds.

In deciding on the processes that are used in Bits and Pieces, I have intentionally left the structure of the installation modular so that I can add or remove processes as I wish. At the time of this writing, there are five different processes available for the installation to choose from. Some play the samples out in their entirety, with just simple layering as the underlying structure. Others cut the samples up in various ways and transform them over time. A user listening to the ten most recent processes in a row will probably hear at least one of each type. Thus, since only 25 samples are used each day, they will hear certain samples several times over the course of the ten pieces, but treated in different ways. This is how the individual pieces are aesthetically tied together into a larger structure.

One thing I have come to know about my own aesthetic is that it can change very rapidly (sometimes on a weekly basis). Thus, the sound generating processes that are currently used in Bits and Pieces may be replaced very easily if I feel the need to hear something different. Thus, in a short amount of time, the sound of the work can be drastically changed. In this way, I have left the work open ended. The only permanent aspect being the downloading and using of samples from the Web.

On a final note, I would like to comment on ephemerality as relating to Bits and Pieces. None of the pieces produced are retained for very long. Listeners may hear the ten most recent works in the order that the installation has produced them. Each time a new piece is added, the oldest piece is erased. In this sense, Bits and Pieces, like most everything else on the Web, is in a constant state of flux. Its pieces, like so much music and art today, are ephemeral. They are not intended to have lasting power, but rather to come into being for a short time, perhaps never to be heard, and finally disappear from history without a trace.