P2P Foundation

Researching, documenting and promoting peer to peer practices


Subscribe

Translate

Eric Hunting on Open Source Parametric Design

photo of Michel Bauwens

Michel Bauwens
10th January 2013


Eric Hunting, On Parametric Design and the Emergence of a Design Science:

“On the blog complexitys.com, devoted to the subject of parametric architecture, an article recently appeared on the related subject of associative design featuring an intriguing video illustration of a design system for urban development planning.

Though the article was in French with no corresponding version on the English side of the site, the video featured is narrated in English and well makes the point of the article, demonstrating the very organic, dynamic, architectures procedurally/generatively synthesized with a specific associative design system based on a fairly simple geometric scheme and a modest number of design program parameters with, ultimately, very complex topological dynamics. It well demonstrates the kind of surprising ‘naturalistic’ complexity of structure that arises in relatively simple parametric systems–structure difficult for the human being to pre-conceive and visualize without the aid of contemporary computers. And here we get a tantalizing taste of the possibilities in the emerging field of parametric design.

What is parametric design? Even the authors of this interesting blog site find it a difficult concept to nail down and so, through their articles and examples, try to present a cloud of ideas that encompass and tentatively converge upon it. But perhaps it may not be so challenging an idea that we must sneak up on it obliquely, thanks to the growing influence of open hardware and desktop fabrication, and 3D printing. Among routine users of desktop fabrication tools we often hears the term ‘parametric model/design file’. This term refers to a data file for an object that is designed to variable in certain parameters while remaining largely the same in basic form and function and some key details. (like sizes of screw holes and mounting points for certain parts) Most often the changeable part is scale, the object freely changeable in one or more dimensions yet still functioning the same. A mundane example of this is the common urn shape that has remained similar since antiquity and yet is found in an endless variety of sizes and variations. Another is the stereotypical pitched roof house typified by the shape of house pieces in the game Monopoly. We see this form replicated in the built habitat in innumerable variations of size and relative dimensions and yet it remains recognizably the same throughout, sometimes reduced to mere symbol or icon.

With some types of objects and 3D file systems a shape may not be a fixed set of form data but rather represented as an algorithm that generates the resulting shape, with desired parameter changes incorporated on-demand. This is a process referred to as ‘procedural modeling/rendering’ and can incorporate complex systems of parameters. The term ‘generative form’ is closely related to this, usually referring more specifically to an abstract mathematical model of a form that produces repeating or fractal-like self-similar patterns, whereas the procedural form may be based on parameters of many kinds characterized less by a ‘formulaic’ expression and more a ‘rule based’ or ‘syntactic’ expression. These parameters need not be strictly physical or topological in nature–in other words, not necessarily relating to matters of space and dimension or engineering constraints. In some cases the parameter influencing a form may relate to functional constraints, environmental and energy factors, or other types of factors, such as aesthetic schemes, social (the number of people in a space and how they use it) or cultural (sizes, shapes, and orientation of space relating to cultural conventions or religious practices) aspects. When many procedural elements are linked dynamically in a network of mutual influence this is called an ‘associative’ system or network because the procedures generating the individual elements are linked by shared parameters or relationships between parameters–changes to any one element influencing/inducing changes in others. Thus an associative design becomes a complex procedural system–often rather organismic or ‘naturalistic’ in character.

Procedural modeling is an area of great promise that has only begun to break through to general design as personal computers have caught up to the needed computational power. Key among its potential virtues is the automation of details in computer based modeling so as to make the process of on-screen design more intuitive. One of the most well known demonstrations of this was with the procedural modeling and rendering system of the computer game Spore which enabled players to craft their own creatures and objects with far greater ease and speed than is typical of 3D modeling tools.

But the real power in this is that the development of parametric design systems, in a more practical sense, is an integration of real world knowledge–physics, engineering, environment, economics, regulations, social systems, behavioral psychology, etc.–into computer aided design tools that allow an interactive exploration of possibilities at a depth, level of complexity, multidimensionality, impossible for the designer of the past to even imagine, pushing the idea of ‘elegance’ beyond the Golden Ratio and the merely visual and pushing function beyond the simply ergonomic. Additionally, we are now incorporating the built/made structure as a feedback mechanism, aiding the gathering of this knowledge. Design is no longer just ‘output’ with the only feedback market share as a measure of fitness. The artifacts it produces now talk back to us by a variety of means and that data feeds back into subsequent design–become concrescent knowledge in parametric systems and thus a kind of genome inherent to the built habitat. Right now our explorations of generative and associative parametric systems are like the X-ray crystallography that first exposed the structure of DNA. And this is transforming the nature of design itself, moving it away from a more ego-centric aesthetics-dominated pursuit it has been in modern times, relying on talent and intuition, to a more social and global design science that uses form and structure to seek knowledge about the world–which, in fact, is what it was before it became ‘professionalized’. Design evolution as scientific method. Through parametric design we are discovering a theory of ‘design science’.

Two common themes to discussions/presentations of parametric design–which we see well represented on the noted website–are structures of naturalistic complexity and the notion of an organic evolution to design. This has become summed-up in the idea of New Nature; the form of nature that has arisen from technology as it becomes more ‘autonomic’ in function and more self-organizing. The prime example is the Internet which has transcended the notion of a planned technological construct to become a self-organizing entity with very organismic characteristics. In contemporary design theory we are seeing the emergence of the idea that designs are no longer merely products of an individual imagination but rather evolutionary entities that iteratively change in form with real world experience. It is this author’s contention that this has, in fact, always been the case with design, particularly in pre-industrial culture when design was a more peer-oriented process. Design has always been a process seeking, and encoding, physical knowledge. (which relates to my assertion that architecture is the definition of place through the encoding of knowledge and information into structure) It was only with Industrial Age task specialization that design and engineering were taken out of a larger community’s hands and relegated to professionals operating in an increasingly isolated professional sub-culture and, furthermore, separated from each other with design becoming mis-construed as a practice largely aesthetic/stylistic in focus, separate and often indifferent to utility and function. I think this can be clearly confirmed and historically illustrated by any anthropologist as so much of their work involves the deductive analysis of the iterative changes in historic/archeological artifact design as a reflection of a culture’s history and evolution. One of the most clear unique hallmarks of contemporary design has been the way aesthetic characteristics of todays structures and artifacts have been disconnected from physical function and material characteristics of things to the point of deliberate subterfuge–to the application of faux and veneer materials/textures to things as a means of ‘applying styles’, disguising or concealing underlying function, and creating illusory impressions of quality and value while accommodating the situation/constraints of mass production techniques.

Thanks to the way computers now augment our perceptions, letting us look through the games, as it were, design is rediscovering its roots as one of the most basic of human activities toward learning about the world–at least for those designers sophisticated and comfortable enough to work with this technology. Through parametric design we are discovering the mechanics of the human species’ epigenetic evolution as carried though a genome of its artifact design. But here is where the proposition of open source design comes in. Conventional professional design can be argued to be an evolutionary dead end because intellectual property constrains the communication necessary for iterative concretion of knowledge. And so it is like the organism trapped on a remote island, becoming increasingly adapted to a very small set of parameters (today mostly ‘market’ oriented) outside of which it is hopelessly dysfunctional. Specialization is evolutionary death. And this is not only a dead-end for a particular design but also for technology because design and technological innovation are linked. By exploring the boundaries of engineering possibility design challenges the limitations of technology, compelling its evolution. Companies that obsess about the control of product design IP as a means to control market share paint themselves, and their industries, into corners–as so well illustrated by the automobile industry and its long-chronic technological stagnation.

At present parametric design remains ‘open’ by default because there are, as yet, no definitive software platforms for it and few to no mechanisms by which the now largely experimental technology can be monetized. Even game and graphics companies have failed to recognize the potential market value of procedural modeling technology–but then they generally have a poor practical comprehension of their end users/markets and how their own products are and might be used. But there is a potential hazard of key technology being locked up as it is developed because of the competitive professional character of the design industry and the companies that commonly produce its tools, impeding the emergence of this new design science. Thankfully, this is rather far off on the horizon. What is important to the progress of this field today is that method and technique be as freely shared as the visual results, and that will require this new community of designers to recognize that they are, indeed, dealing with a new design science and have the same sort of responsibility as other scientific fields to document and cultivate organized venues of communication. For this the open source software culture has created very suitable systems and methods, but whether today’s designers can learn about and use them is another matter. At present parametric design is a concern of a very small community of designers with a rather exceptional fluency in both the cultures of computing and a broader spectrum of the sciences. Can this translate to the mainstream?”

FacebookTwitterGoogle+RedditShare

2 Responses to “Eric Hunting on Open Source Parametric Design”

  1. Francesco Says:

    Hi, I am the curatore of the mentioned blog complexitys.com and I really want to thank you for your interest in our works.

    Some links you may be interested in:

    A conference about Parametric Architecture and Performance Drive Design:
    http://issuu.com/immaginoteca/docs/parametrix_-_intro/6#embed

    A conference about Parametric Design and Inclusive participation process:
    http://www.scribd.com/doc/87038823/Conference

    More references can be found on my personal blog (but the post is in French, we are working on a related english post):
    http://www.immaginoteca.com/collaborative-parametric/

    Thanks a lot,
    Francesco

  2. Michel Bauwens Says:

    Thanks for the extra information.

    The category “P2P Architecture” here in our blog has much related material, including by the same author, Eric Hunting.

    In our wiki, we keep a section on the intersection between p2p and urbanism, http://p2pfoundation.net/Category:Urbanism, as well as design http://p2pfoundation.net/Category:Design.

    Michel

Leave a Reply

XHTML: You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>