Digital aesthetics: artistic movement or oxymoron? Ex-MIT Media Labber tries to rise above "cyber-hip"

August  1997

Steven Holtzman is an MIT Media Lab alumnus and CEO of Perspecta Inc., an exciting Web information delivery startup company. In Digital Mosaics: The Aesthetics of Cyberspace, he claims that the digital world is reaching the point where it's possible to describe aesthetics of such cyber-phenomena as the Web, digital art and music, and interactive multimedia. Such aesthetics are bound to emerge, but at this point, we're in the primordial stages. Holtzman's book shows a variety of examples of cyber-art and discusses the aesthetic principles that guide creative people who use bits as their medium of expression. (2,500 words)

Mail this article to a friend

t's hard to find two people who can agree on a definition of the term "cyberspace." Yet Steven Holtzman, in his new book, Digital Mosaics: The Aesthetics of Cyberspace, wants to define its aesthetics. His book is, as he admits, a first step in a long process toward that definition. It's worthwhile reading for those who are interested in the artistic possibilities of digital media, but even as a starting point, it has its shortcomings.

Before I proceed, I should offer some disclaimers. Frankly, I may not be the right person to review this book. I am certainly no expert on aesthetics. I am vitally interested in the arts, yet I'm one of those reactionaries who feels that Andy Warhol was a fraud and that much of the art that exposes and glorifies "processes" (Jackson Pollock's splatter paintings; Brian Eno's music) is suspect. I am skeptical of certain aspects of the "cyber-cool" scene, as epitomized by Wired magazine and the book writer (and my college classmate) Douglas Rushkoff. I find Marshal McLuhan's famous quote, "The medium is the message," to be a cliche that has been used to justify all sorts of inept nonsense. The MIT Media Lab strikes me as a place where people do clever, creative things that have no relationship to reality. Bah, humbug.

These disclaimers are necessary because Steven Holtzman is a Wired contributor and an MIT Media Lab alumnus. He lionizes Warhol, Eno, and especially McLuhan in his book. (The obligatory quotes on the back of the sleeve come from an elite group of digi-hipperati -- Nicholas Negroponte of the Media Lab, Ann Winblad, who runs a powerful venture capital fund, and Wired founder Louis Rossetto.) And Holtzman does profess admiration for some of the socio-cultural aspects of the "cyber-cool" scene that I find trendy.

Yet Holtzman has a unique background, including graduate-level training in music composition. He is currently CEO of a very interesting startup company called Perspecta, which considers itself to be in the field of "information design." Most importantly, he has absorbed a prodigious amount of information about all sorts of things, and he uses this book as a framework for stepping back, seeing new types of relationships between different topics, and exercising some critical thinking. I decided to read this book because I wanted to see what lies behind the hype associated with the cyber-scene, and Holtzman seemed like a good source.

The book Digital Mosaics has two parts. The first is a summary of the existing creative technologies in the digital world; the second examines these developments in the light of aesthetic theory.

Advertisements

Digital worlds
Part one, "A Tour of Digital Worlds," covers some topics that are obvious to readers of this column, as well as others that are not as well-known. It begins with an explanation of the Web and hypertext. Then it covers virtual reality (VR), focusing on how it can be superimposed on the existing Web to provide a VR experience with lots of depth -- for example, taking a virtual tour of a city, finding a building that contains a restaurant, then clicking on that restaurant to find its own Web site or a review of it in a city guide.

Holtzman also talks about commercial applications for VR, including environment simulators for training, product design, and other purposes. One of his examples involves a cyber-shopping mall, designed by some people at IBM, where you can go to a music "store," preview and choose a selection of music tracks, and download them to a custom CD for playing on your car stereo. Oops...unfortunate example. The concept of proprietary cyber-malls is now considered to be a failure; the "make your own CD" concept was tried (by IBM and Blockbuster Entertainment) a few years ago, but it never got off the ground because they could not secure licensing agreements with record companies.

The next step beyond VR is multiple-user VR environments, which include MUD (multiuser dungeon or multiuser domain) games and 3D-VR chat rooms in which participants assume animated personas called "avatars." Just as chat rooms on America Online tend toward the banal, this section dives off the deep end into cyber-trendoidism, presenting sample dialogs that include people with cyber-nicknames like Hive-Mind and Starseed. Holtzman distances himself from these attitudes, but still he chooses to represent them in this book. He emphasizes the oft-lampooned idea that people can disguise their true natures when participating in online games or chat groups. (Why is this such a big deal? From the forged junior high school mash note to mail and phone fraud, people have been using traditional media to misrepresent themselves for centuries!)

The next important building block of cyber-aesthetics that Holtzman introduces is the fractal: the mathematical construct that enables computer animation systems to represent landscape features like mountains and shorelines in remarkably realistic ways. To introduce the fractal, Holtzman turns to a particular mathematical entity called a Mandelbrot Set (named after its inventor, the IBM mathematician Benoit Mandelbrot). Mandelbrot articulated in theory the concept of a fractal -- a structure that consists of infinitely many layers of self-replication -- and then, in the late 1970s, used IBM's computers to make fractals useful by visualizing them.

Since Mandelbrot's pioneering work, fractals have become very, very important tools in digital graphics. Fractals have made immense improvements possible, for example, in the realism of landscapes in sci-fi movies since the 1980s. Holtzman shows how several artists have used fractal technology to create new forms of visual art. His examples are fun and, unlike the original Mandelbrot Set, not well-known. But in concentrating on fractals, he misses an important part of the discussion on how digital tools have transformed visual art. Fractals should be seen in the context of other digital tools that artists have been using for the past several years already.

Adobe's Photoshop and Illustrator programs are the most important examples. It is no exaggeration to suggest that the impact of these tools (and more advanced packages, like those from MetaCreations) on graphic arts is equivalent to the impact of photography itself. Photoshop, for example, is an image manipulation tool. Originally, it was used as a digital equivalent of such traditional photographic retouching techniques as airbrushing, dodging and burning, and color and contrast correction. But since then, it has developed into a mini-industry of tools for doing all sorts of amazing things to digital images that are not replicable in the physical domain. In other words, it has gone from a relatively mundane tool for commercial artists to a true outlet for creativity. (An excellent place to view some of the possibilities of digital art that uses this technology is the SIGGRAPH show taking place this month in Los Angeles. Or if you miss that one, the Seybold Expo in San Francisco in late September includes a fine digital art gallery.)

Holtzman misses the opportunity to discuss fractals and other tools as extensions of the capabilities of Photoshop and its ilk. Photoshop is capable of "one-step" transformations of images, such as color and shape manipulation. Perhaps it's accurate to suggest that fractal technology takes those kinds of processes and makes them iterative and three-dimensional. It's also not unreasonable to imagine "fractalizer" plug-ins for Photoshop-type image manipulation programs that make this kind of process iteration explicit.

The rest of the first half of Digital Mosaics covers experiments in digital music and intelligent animation. Holtzman explores the creation of animated animal or plant-like animated objects that can do things like grow, move, achieve simple goals, and fly in flock formation. These are all examples of pre-existing artificial intelligence technology that has been extended a step further through visual representation.

One of his AI-derived examples is that of his own company, Perspecta. Perspecta starts with a semantic network of English-language terms. Then it reads text from news or other information sources, analyzes it, and populates the semantic network with the information items. Using a Java application, the user navigates through a 3D visual representation of the semantic network, zooming in through subject trees from the general to the specific, moving laterally to related subjects, and eventually finding news stories of interest. The result is a way of looking at information through a Web browser that is unique and a significant step beyond the current crop of news services that filter stories through keywords and present them to you on Web pages or in e-mail messages. Perspecta's Web site offers several sample information spaces that you can view with its downloadable client, including one on Digital Mosaics itself.

Digital music: Brain Operas and video games
The first half of the book concludes with a description of a cyberspace opera called Ghost Dance, with music by the contemporary composer Philip Glass. Ghost Dance seems like a precursor to a more ambitious project called Brain Opera, which was put on a year ago by MIT Media Lab people at New York's Lincoln Center and was accessible on the Web to anyone with a Java-capable browser. Brain Opera's "libretto" was based on MIT AI guru Marvin Minsky's book The Society of Mind.

Both Brain Opera and Ghost Dance are fascinating in concept: they are sound-plus-visual experiences in cyberspace in which the audience members encounter and interact with each other. The music? Ehhhh...I have not heard Ghost Dance, but as far as Brain Opera goes: I believe that, as Clifford Stoll implies in his controversial book Silicon Snake Oil, interactivity does not make for inherently interesting content. Talent is a more important requirement.

Holtzman's examples of digital music do not include analogs to his "intelligent" animated figures. In the mid-1980s, for example, the scientist and musician Peter Langston created a system that semi-intelligently composed music based on "licks," where a "lick" was a musical phrase (he borrowed the term from jazz musicians' argot). The system imposed a structure on the overall piece, chose some licks to fill the structure, and strung them together so that they sounded like a continuous piece. He demoed this system with a library of Mozartian licks; the system composed pseudo-Mozart. Indistinguishable from real Mozart? Nah, not really. Art? Hardly: the application was video games for his employer at the time, Lucasfilm Ltd. He wanted music that varied somewhat as players nuked the space aliens. Like image manipulation with Adobe Photoshop, this is another technology that started out in the mundane commercial world, and it will undoubtedly grow into a more creative tool.

Is the medium the message?
The second half of Digital Mosaics switches from a survey of existing digital media paradigms and experiments to a discussion of aesthetics and digital media's relationship to them. This is far more interesting than the first half; it's the book's payoff. Three major themes run through the discussion: nonlinearity, discontinuity, and autonomy.

Nonlinearity refers to the features of digital media, like the Web's hyperlinks, that let you skip around from one "place" to another at will, and therefore destroy the sanctity of the linear story line, narrative, or music composition. Discontinuity refers to digital technology's inherent "on" or "off" state. It makes certain things easy that are quite difficult in the analog world (e.g., the patterns obtainable with fractals), but also vice versa (e.g., representing geometrically complex visual textures, such as hair). Discontinuity has also been used as a stylistic element, as in Xerox's "pixellated X" logo; as Holtzman mentions this is analogous to modern art techniques that expose the medium -- there's that McLuhan quote looming over the horizon again -- like painter Frank Stella's "Black Canvases" (and, I would add, Le Corbusier's late-period beton brut architectural style).

Autonomy is the increasing ability of digital entities to do things on their own, like the digital plants and birds shown earlier in the book. A fourth theme is ease of iteration, which I mentioned above. Holtzman cites Andy Warhol's Twenty Marilyns as a precursor to this theme; in this work, Warhol used, and therefore commented on, mass production techniques.

Holtzman seems to be saying that nonlinearity, discontinuity, and autonomy are the key differentiators of digital media from the physical, analog world. Any notion of digital aesthetics must therefore derive from aesthetical possibilities of those three qualities. Since they haven't been present in the analog world, we can't really tell yet how these qualities will contribute to aesthetics.

He ends the book on a tantalizing note, saying that the aesthetic principles he discusses are new, and therefore artists haven't had time to assimilate them. But he insists that even these early principles are valuable: if McLuhan's quote is really true, then we should be able to deduce the messages, since we know what the media looks like now, and what it should look like in the near future. If you don't believe McLuhan, then you must (as I guess I do) believe in the idealistic notion of "pure art" whose messages transcend its media. In that case, the medium -- cyber or otherwise -- only becomes relevant in so far as it determines what limitations it imposes on expression.

Even so, we must allow that digital media also impose context on art: you have to be comfortable with how it's presented in order to receive the message. That's one reason why it's hard to call any existing digital artwork a "masterpiece." The art-consuming public and its hired guns, art critics, haven't assimilated the possibilities enough to be able to determine just what a masterpiece is. Just imagine a music critic in 1950 trying to judge the merits of Sergeant Pepper, Dark Side of the Moon, or any other modern music album you think is a masterpiece. Or, try to imagine a discussion of the merits of synthesized music from the late 1960s that uses something archaic like Switched-On Bach as its example. Without the tools or proper context, such a discussion just isn't very relevant.

In other words, it's debatable whether a book on the aesthetics of cyberspace -- with any claim on authority -- can legitimately be written now. That's one issue with Holtzman's book, although it's his intent to be provocative rather than authoritative. The other problem is that the examples he shows are often chosen for their "cyber-hip" value rather than their strength in illustrating an aesthetic principle. Hence he drags in the deconstructionist literary theorist Jacques Derrida, that patron saint of trendy academics. He could have included more mundane, less abstruse, examples to strengthen his points.

The final shortcoming of this book is, well, that it's a book. It's very difficult to illustrate digital media on paper with text and monochrome still images. This is kind of like a restaurant guide: it merely whets your appetite. But as an appetite-whetter, Digital Mosaics succeeds quite well.

Title: Digital Mosaics: The Aesthetics of Cyberspace
Author: Steven Holtzman
Publisher: Simon & Shuster
ISBN: 0684832070
List price: $25.00

Bill Rosenblatt is an enterprise IT architect at Sun Microsystems, where he specializes in media technology. Reach Bill at bill.rosenblatt@sunworld.com.

If you have technical problems with this magazine, contact webmaster@sunworld.com

URL: http://www.sunworld.com/swol-08-1997/swol-08-bookshelf.html

Last modified:

Resources

Digital Mosaics: The Aesthetics of Cyberspace, by Steven Holtzman http://www.amazon.com/exec/obidos/ISBN=0684832070/sunworldonlineA The Society of Mind, by Marvin Minsky http://www.amazon.com/exec/obidos/ISBN=0671657135/sunworldonlineA Silicon Snake Oil: Second Thoughts on the Information Highway, by Clifford Stoll http://www.amazon.com/exec/obidos/ISBN=0385419945/sunworldonlineA
• ACM SIGGRAPH 97 Conference & Exhibition
  http://www.siggraph.org/s97/
• Seybold San Francisco Exhibition
  http://www.seyboldseminars.com/
• Perspecta Inc.
  http://www.perspecta.com/
• The Brain Opera
  http://brainop.media.mit.edu/
• Peter Langston
  http://www.empire.net/~psl/
• Full listing of Bookshelf columns in SunWorld
  http://www.sun.com/sunworldonline/common/swol-backissues-columns.html#bookshelf
• Bill Rosenblatt's Client/Server columns in SunWorld
  http://www.sun.com/sunworldonline/common/swol-backissues-columns.html#cs

If you have technical problems with this magazine, contact webmaster@sunworld.com

URL: http://www.sunworld.com/swol-08-1997/swol-08-bookshelf.html
Last modified: