Graphics Reference
In-Depth Information
Imagine that you were assigned a design project, a new perfume bottle.
You don't know how to use design software and you would prefer not to base
your new design on an optical scan of an existing bottle. Instead, you would
sit down at the Endless Forms website and make your irst design choice by
clicking on a crude shape from a collection of several basic variations on the
website. In response, Endless Forms would note your irst design choice,
“think” about it, and rapidly make a few calculations. It would then generate
several new shapes that were somehow similar to the irst shape you chose.
Next, you would choose again from this new set of shapes. Endless Forms
would repeat the process, noting your design choices and generating a new
batch of related shapes for your perusal. Eventually, after several rapid itera-
tions and rounds of shape selection, you would start to see a design emerge.
When you felt that inally, the design offered to you was what you wanted,
you could save the inal design and your design project would be completed.
When Jeff and his students set up the Endless Forms website, they real-
ized they could watch peoples' design process, irst hand. When users are
given a nearly blank slate and allowed to make a series of rapid selections,
interesting designs emerge. In just a few months after the site went live, users
generated over three million oddly shaped objects from lampshades shaped
like mushrooms, fertility goddess igurines, to cubes featuring spooky faces
jutting out of them.
When people use conventional design tools, the interface and software fea-
tures seem to shape their thinking, maybe by making people self-conscious.
In contrast, on Endless Forms, users “bred” designs by impulsively clicking
on a shape they liked and doing that again and again. The process of rapidly
choosing and then sitting back to wait, combined with the fact that the com-
puter did most of the work for them, seemed to bring out a new expression of
people's subconscious.
Endless Forms may not be the best design tool to create machine parts
for the next Mars Rover. When users create a design by impulsively pointing
at appealing shapes, the result will likely not be an object that's honed and
mathematically optimized for maximal performance. Instead, as an early pro-
totype of a new type of human/computer interaction, perhaps the value of the
Endless Forms project was that we proved that it is possible for a computer
and a human to design “collaboratively.”
