I've been giving a lot of thought lately to parametric design: that oh-so-fashionable method of design which turns architects into code-writers and computers into designers. I don’t mean to bear down on this method too much, because I do see many purposes for its implementation as a design tool. I can see how parametric design can numerically arrive at what could potentially be the most efficient layout of a space, or best use of a material, or efficient structure. I also see how parametric design can take raw data about wind velocity and temperature, humidity, orientation to the sun, average rainfall, etc, and optimize the layout of a structure such that it provides the maximum benefit of its environment with minimal impact. The utilitarian use of parametric design fascinates me.
What worries me more is using parametrics as a design driver. Putting my finger on just what makes me worry about it has been more difficult, but I think I’m on the right track: With parametric design an architect can, with a bit of computational magic, create a set of mathematics which results in a perfect sculpture. (Note: the use of “perfect” here does not necessarily signify any aesthetic or cultural exactitude, but instead a faithful application of some theoretical input) But while parametric designer applies mathematics to create sculpture, an artist will create sculpture that inherently uses mathematics. This inverse relationship is not equal. What gets lost in translation is not the ability to create a perfect sculpture – A computer can scan, digitize, and recreate Michelangelo’s “David” down to the slightest knick in the Carrara marble. What the computer cannot recreate is an ephemeral “warmth” or soul to the piece which can only be wrought with the hands of an artist.
[via ArchDaily]
In a similar way I’ve always preferred the drawn line to the plotted line. Something in the inconsistency, the variability, the flow and the changing pressure of graphite on vellum, always seemed to being more life to a drawing than the cold, unwavering plot line.