Bruce Sterling
Designing Processes Rather Than Art

Workshop November 25–28 2008, by application
Public lecture November 25 at 6pm, open to all

blog.wired.com/sterling

Bruce Sterling, author, journalist, editor, and critic, was born in 1954.
Best known for his eight science fiction novels, he also writes short stories, book reviews, design criticism, opinion columns, and introductions for books ranging from Ernst Juenger to Jules Verne.
His nonfiction works include The Hacker Crackdown: Law and Disorder on The Electronic Frontier (1992), Tomorrow Now: Envisioning The Next Fifty Years (2003) and Shaping Things (2005).
He is a contributing editor of Wired magazine and a columnist for Make magazine.

He also writes a weblog. During 2005, he was the "Visionary in Residence" at Art Center College of Design in Pasadena. In 2008 he was the Guest Curator for the Share Festival of Digital Art and Culture in Torino, Italy.

He has appeared in ABC's Nightline, BBC's The Late Show, CBC's Morningside, on MTV and TechTV, and in Time, Newsweek, The Wall Street Journal, the New York Times, Fortune, Nature, I.D., Metropolis, Technology Review, Der Spiegel, La Repubblica, and many other venues.

The workshop results will inspire a critical essay by Marco Mancuso, founder and director of the DigiCult project - www.digicult.it, an Italian cultural project and platform involved in digital culture and electronic arts.

Workshop brief

Designing processes rather than art

(See HAECKELFORMS.jpg in Full brief)

Ernst Heinrich Haeckel's Artforms in Nature ("Kunstformen der Natur," 1904) is a legendary favorite of generative artists. Why? Because the "art" forms here have no human origin. They were all generated by processes -- in this case, by biological activities. By picking them out, curating them, and arranging them on the page, Haeckel framed them as "art."

Also, Haeckel's work has long been in the public domain, so it's free for use for any purpose. Generative artwork, for instance.

We'll design a simple "process" for creating graphic elements from Haeckel.

BEGIN.

1. Go to FlickR. Find the comprehensive scan of the entire work of Haeckel put up there by some freeware sympathizer. The online community certainly is supportive. Wow.
http://www.flickr.com/photos/origomi /sets/72157601323433758/

2. Pick out one image. "Bryozoans" should do.

3. Print out the image on paper. (See Haeckel-1.jpg in Full brief)

4. Get some scissors.

5. Cut the page in half. (See Haeckel2.jpg in Full brief)

6. Cut the halves in half. (See Haeckel3.jpg in Full brief)

7. Cut the halves in half. (See Haeckel4.jpg in Full brief)

8. Cut the halves in half. (See Haeckel5.jpg in Full brief)

10. Cut the halves in half. (See Haeckel6.jpg in Full brief) This seems mechanically repetitious, but "processes" never get tired. Two, 4, 8, 16, 32 -- nothing comes more easily to a generative process.

11. Now we've got a scattered galaxy of graphic elements that no human being could have planned or foreseen.
Using human judgment, pick out three processed elements as the "art."
(See element1.jpg, element2.jpg, element3.jpg in Full brief)

END.

Element1, element2, element3. (See element1.jpg, element2.jpg, element3.jpg in Full brief)

These are your three graphic elements for the Generative Art workshop.
You have these three elements, you have no more, and no less. You are required to use all three of these elements and you can't use anything else. These are your constraints.

Your assignment is to DESIGN A PROCESS that uses these three elements to create a COMPLETE AND FINISHED WORK that you will show at the end of the workshop.

You might do wallpaper, an animation, a music video, fabric, or a poster. These are all typical applications for generative art.

You might even print out the elements, put them on cardboard, and use them as 3-d elements to generate architecture or furniture. These are rarer, but generative art is getting into generative building and generative manufacturing.

You can take these elements and copy them, multiply them, repeat them, invert them, rotate them, mirror them, change their scales, assemble them in flocks and patterns, overlap them, even colorize them or cut
them in fragments. But: these elements are all the graphics you have. Don't add anything else.

Generative art has two unique challenges. First, it never directly designs any artwork. It *generates* art through process. You are designing a process. Your process has to be explicit and rigorous. It's a set of instructions.
Your process has to begin with the word BEGIN and end with the word END.

This process has to be so clearly phrased, so simple and so accurate that ANYONE can execute it anywhere and get the same results (or very similar results). You'll find that this is quite hard to do. Pressing the button and running the process is easy. Backing off to think deeply about process in a clear and intelligible way is the hard craft-work of generative art. That's where the action is.

Furthermore, you're not going to execute this process yourself (even though you can practice doing it so as to refine your process). You're going to be handing your process to SOMEONE ELSE to "execute." If they can't understand what you mean and follow the instructions correctly, you've failed!

The second unique problem is the "Sorcerer's Apprentice" problem, or, getting the process to stop in some complete and finished position. Since processes aren't human and have no aesthetic judgment, they tend to grind on endlessly. As a generative artist, it's up to you to choose where to stop your process and how to frame the results as your completed work of art.

This is also hard to do; but if you fail to design a process that produces a coherent artwork, you haven't made any art; at best, you've made a tool. Almost anything can be a generative tool -- a ruler, scissors, a repeating drip of paint.
Making tools is not the point here. When the workshop is over you must have something complete and coherent to show.

In conclusion, a few general words of advice that seem to have helped other people struggling to create good generative processes.

Start simple. Let the complexity grow from the process itself. Repeat it till you get a good intuitive feel for it. Get comfortable with its "processuality."

Be fast, clear, small, elegant and bug-free; avoid unnecessary stunts that will create random confusion.

The rules come out of the parts, while the parts come out of the rules. The details reflect the logic of the whole creation.

If it's too tightly controlled, it will look boring and mechanical; if it's too loose, it will look sloppy and chaotic. Try to find a sweet spot where the process seems to "live."

Good generative art is for human beings; it combines regularities with surprises in a humanly pleasing way.

References (see Full brief)

Generative Nature
Aesthetics, repetitiveness, selection and adaptation
Marco Mancuso
Milano 2008

In his De Rerum Natura, Lucretius denies any kind of creation, providence and original bliss and maintains that people freed themselves from their condition of need thanks to the production of techniques, which are transpositions of nature. A god and some gods exist, but they did not create the universe, nor do they deal with people's actions. Lucretius maintains that the rational knowledge of nature shows us an infinite universe that is made up of complex forms and constituted by atoms; it follows natural laws, it is indifferent to people's needs and can be explained without gods.

When an artist uses a conceptual form of art, it means that all of the planning and decisions are made beforehand and the execution is a perfunctory affair. The idea becomes a machine that makes the art. – Sol Le Witt

Modern ecology began with Charles Darwin's studies. In his "theory of evolution" published in 1859 in On the Origin of Species, he underlined the adaptation of the different organisms to the various kinds of environment, which are subjected to the age-long examination of natural selection. However, the word was coined by Ernst Heinrich Haeckel in 1869 and comes from the Greek óikos meaning "house" and logos meaning "discourse". It is therefore a biological science that studies environment and the relationships that the different living organisms establish between each other and with the environment itself. For some time, Haeckel was a strong supporter and popularizer of Darwin's theories, but he soon became one of his most bitter enemies; he firmly refuted the process of natural selection as the basis of the evolutionary mechanism, in favor of a thought that was more focused on the environment as a direct agent on natural organisms, which is able to produce new species and generate diversity.

Ernst Heinrich Haeckel's thought and work represent the starting point of this critical reflection. First of all, because it was the theoretical and practical cue suggested by Bruce Sterling during his workshop for Fabrica, to which the text refers. Secondly; because it allows me a philosophical and critical reflection, aspiring to find a possible point of contact between nature, theories of evolution and programmatic and generative art. Is it impossible? Well, I would say no, on the contrary. Above all if we try to compare and amalgamate, like the colors on a canvas, the German biologist's research on one side with some works of conceptual and minimalist artist Sol Le Witt and the possible relationship between mathematics and nature on the other, and what is known today as art and generative design.

Nature as an art
"Kunstformen der Natur" literally means "artistic forms of nature": this is the title of biologist Ernst Haeckel's 1898 most important text, his most complex and fascinating research. Moreover, this is the text from which Bruce Sterling took, for those participating in his workshop, some primary images that could be the graphic material and starting point for an aesthetic and methodological reflection on the practices of generative repetitiveness. By watching the richly decorated plates in Haeckel's text, it is undeniable that nature is able not only to create spontaneously real "art forms", but also to produce a direct correspondence between a certain generative aesthetics, starting from a fundamental unit/nucleus to come to a complex entity, and a consequent adaptive and evolutionary practice.

In other words, if the stages of the embryological development of a species actually trace the evolution phases that led it to its position in the natural order, the survival of each species basically depends on its interaction with the environment. According to Haeckel, the mechanism thanks to which new species and a new diversity have origin is that of a gradual addition of a certain development trajectory starting from an initial unit, which is determined by imposed external (environmental) parameters, which are able to influence the gradual direction of the trajectory itself.

At this point, a first important reference to the theoretic and methodological bases of Generative Art seems evident, as one of the pioneers of this discipline, Italian architect Celestino Soddu, suggests: "Generative Art is the idea realized as genetic code of artificial events, as construction of dynamic complex systems able to generate endless variations. Each Generative Project is a concept-software that works producing unique and non-repeatable events, as possible and manifold expressions of the generating idea strongly recognizable as a vision belonging to an artist/designer/musician/architect/mathematician. This generative Idea/human-creative-act makes an unpredictable, amazing and endless expansion of human creativity. Computers are simply the tools for its storage in memory and execution. This approach opens a new era in Art, Design and Communication: the challenge of a new naturalness of the artificial event as a mirror of Nature. Once more man emulates Nature, as in the act of making Art […]."

Although, over the centuries, biologists and morphologists have widely denied a so close correspondence between ontogenesis and phylogeny, and so between unity and complexity, the germ of thought is interesting and I think it is worth continuing to nourish it…

Forms, colors, lines and instructions
As everybody knows, US conceptual and minimalist artist Sol Le Witt, who died not long ago, is one of the spiritual fathers of modern artists and generative designers. By reducing art to a series of instructions thanks to which everybody is able to draw forms, colors and lines in the two-dimensional and three-dimensional space, creating geometric elements that are repeated and modulated according to standard space proportions, Le Witt loved reminding that "all the people are able to participate in the creative process, to become artists themselves". It is well-known that the artist tended to separate the planning stage from the realization of the work; he devoted himself to the former, whereas his assistants devoted themselves to the latter: if the artistic process thus lies in the conceptual planning of the work, the (basic, elementary and geometric) execution can be carried out by everybody, thanks to a series of detailed instructions that are suggested by a thinking unit with a procedural approach. He also claimed: "There are several ways of constructing a work of art. One is by making decisions at each step, another by making a system to make decisions."

In this kind of approach the work of the last years of some of the most important generative artists and designers in the world (Casey Reas, Ben Fry, Jared Tarbell, Theodore Watson, Lia, Toxi, Andreas Schlegel, Marius Watz, Robert Hodgin, to mention only some of them) is reflected: if the human being identifies himself/herself with the author of a series of mathematical instructions that can be suggested to a computer, the resulting work of art will be the sum of the operations that the computer has carried out autonomously. Therefore, as for Sol Le Witt the emotional elements of the authors, their joy at a moment, their frustration, their apathy were constituent elements of a free interpretation of the instructions that had been suggested to them and so of the resulting work of art, in the universe of digital software as well (from Processing to VVVV to Open Frameworks, to mention the most widespread) we can hazard the thought that the instructions given by the artist/designer can be freely interpreted by a kind of "emotiveness" of the "thinking" computer.

Le Witt's conceptual indifference to any kind of aesthetic judgement, the aversion to prearranged aesthetic conventions that are assimilated by the public, a general indifference to any kind of distinction between old and new are perfectly reflected in the words of one of the most important generative artists in Italy, Fabio Franchino: "In the evening I give some instructions to the computer, which processes data and autonomously generates lines, forms and colours during the night; in the morning, when I wake up, I judge the results. If I like the product I will keep it, if it is not satisfying I will throw it."

Well, I do not know what these things suggest to you: I think that also in this case we can make a comparison with the natural universe. If we assimilate the environment, nature in its widest meaning, as the entity that is able to cause a series of changes, evolutions and dynamics, then the organisms living in contact with it (again, the concept of "ecology") are able to interpret these vital codes, to assimilate them, in order to react to them and autonomously generate a series of forms, colors and systems that can be seen as the result of their evolutionary process, which comes to a complex final system from a starting unit. The difference maybe lies in the "spontaneity" with which this process begins: if an artist/designer decides, in advance, a series of instructions that will be given to the computer, it is difficult not to think that nature operates by following only its evolutionary spontaneity. At the same time, it is fascinating even to think that as the artist/designer does not know the final effects of the instructions, giving the computer the freedom to interpret them, similarly nature does not care about the effects it produces on the organisms living in it, giving them evolutionary freedom of forms and elements that we, human beings, only afterwards could maybe consider as "works of art".

Numbers in evolution
Today, one of the most fascinating mathematical theories is undoubtedly that of fractals: according to the definition of their discoverer, Polish mathematician Benoît Mandelbrot, they are geometric shapes, characterized by the endless repetition of the same pattern at ever smaller scales. This is the most intuitive definition that can be given to shapes that exist in nature in an impressive number but do not still have a precise mathematical definition. The natural universe is rich in forms that are very similar to fractals, forms that do not follow the norms of the Euclidean geometry: a stretch of coast, the branches or the roots of a tree, a cloud, the snowflakes, the ramifications of a lightning and the dentation of a leaf are example of fractal forms originating spontaneously in nature. Among these, the fractal form par excellence is the spiral, the constituent element of the shell of many annelids and conches, which is one of the main objects of study of Ernst Heinrich Haeckel's theories and one of the most beautiful and fascinating geometric forms.

If we shift the field of analysis to mathematics, to numbers, to equations and algorithms, the level of intersection between science, technology, art and nature does not change. And if the procedural and generative method is what we have chosen as the guiding element of this treatise, it is not surprising to think that the construction of fractals follows a reiterated process, that is, the repetition of a starting element for a theoretically infinite number of times until, after a while, the human eye cannot distinguish the changes in the starting element any longer. We must not forget the fact that, as it is acknowledged, fractals are influenced by certain controlled casualness. There is thus again the element of casualness, of spontaneity, as the distinctive (or unifying) element between computer and nature, according to which evolutionary mechanisms cannot be predicted from their constituent elements and it is often impossible to reconstruct them, starting from their visible manifestations.

At this point of the text, the procedural, generative, iterative and evolutionary element may be considered as the pillar of the thought underpinning a modern "computational ecology": between Turing's revolutionary theories on "morphogenesis" (every living organism is able to develop complex bodies, starting from extremely simple elements and basing on processes of self-assembly, without the aid of a guide following a prearranged plan) and the most recent studies that have been carried out on "genetic algorithms" (a particular class of evolutionary algorithms using techniques of mutation, selection and recombination, so that a certain population of abstract representations of possible solutions to an optimization problem evolves into better solutions) almost 50 years of studies, analyses and research passed; they aimed at underlining the nearly computational properties of Mother Nature on one side, and the ability of digital machines to simulate and repeat complex natural phenomena. Frankly, I do not wonder any longer what is the most fascinating form of art or the most difficult process…

Moreover, I think that the most interesting answers to these themes can be found in the studies and theories of Karl Sims, the famous artist and researcher from Mit Media Lab; in particular, we can find them in his 1993 work, Genetic Images, which drew inspiration from his paper Artificial Evolution of Computer Graphics, where he described the application of the "genetic algorithms" for the generation of 2D abstract images, starting from complex mathematical formulas. Therefore according to Sims, Darwin's evolutionary theories can be simulated by means of a generative software or appropriate mathematical algorithms; in this way, "populations of virtual entities specified by coded descriptions in the computer can be evolved by applying these same natural rules of variation and selection. The definition of fitness can even be altered as the programmer desires." I think that what is interesting in Genetic Images is the fact that this work was presented as an interactive installation: in other words, it was the public who could choose and select the most interesting images and forms from an aesthetic point of view, among those generated by a computer simulating a process of artificial evolution. The selected images were then recombined by the computer to create new ones, basing on alteration and mutation methods, similar to those of natural species during their evolutionary process. Karl Sims thus wonders whether these interactive evolutions can be considered a creative process. If yes, is it the public who develop an independent creative attitude or is the presence of a designer making the computer follow precise creative paths necessary? Or is it maybe the computer that develops autonomous creative tendencies?

In his treatise, Sims duly mentions biologist Richard Dawkins who, in his book The Blind Watchmaker, talks about the ability of natural evolutionary processes to create complex forms without the external presence of any designer or programmer: "It is thus possible that these generative techniques challenge an important aspect of our anthropocentric tendencies, according to which it is difficult for us to believe that we are planned not by a God but by casualness showing through the codes of a natural evolution", concludes Sims. Maybe true art lies exactly in all these things.

http://caliban.mpiz-koeln.mpg.de/~stueber/haeckel/kunstformen/natur.html
http://www.flickr.com/photos/origomi /sets/72157601323433758/
http://www.celestinosoddu.com/
http://it.wikipedia.org/wiki/Benoît_Mandelbrot
http://www.karlsims.com/genetic-images.html

For further information:
Fabrica Press Office
fabrica@fabrica.it
tel. +39 0422 516349
fax. +39 0422 516347
Fabrica, Via Ferrarezza
31020 Catena di Villorba, Treviso