​Getting away from the design of book covers to the more general discussion
of automation and machine intelligence.


​I believe that automated design tools of the future will change the way
designers work, allowing designers to concentrate upon their creativity and
overall judgment instead of their skill at mechanical tasks and even at the
traditional craft skills of drawing, and materials, etc.

I've been talking with the folks at Autodesk: consider their Dreamcatcher
(DC) program.

https://autodeskresearch.com/projects/dreamcatcher

Here, the designer states the goals and constraints and DC produces
candidates. The designer critiques them, rejecting some encouraging others,
perhaps modifying the goals and constraints as the designer comes to
understand how they impact the suggested results. DC uses an AI technique
called Genetic Algorithm to develop thousands of alternatives, modifying
its suggestions based upon feedback from the designer.

DC is especially powerful in developing suggestions for items manufactured
by additive techniques, where the materials are unusual or varied, and
where leaving holes and open spaces can reduce cost and weight without
impacting strength or thermal capability (actually, it can enhance thermal
capability). DC can do all sorts of calculations during the design for
strength, rigidity, flexibility, heat transfer, weight, cost ..
simultaneously considering them while doing the design. It is ideal at
creating biomimicry, not by deliberate mimicry but because it is not
constrained by the need to do linear components, straight lines and rigid
angles. It's natural domain is that of smooth curves.

The point I want to make is not that you will all like the resulting
designs, but that this is an early example of a design tools that replaces
much of the detailed handwork of the designer (using, say, a standard CAD
tool) with the higher-level judgment and sensibilities of the designer.
Designers can now think at a higher level.

This has long ago happened in engineering, where instead of spending long
hours with a slide rule, integral tables, and hand calculators, engineers
can now think of the real problem, write down the equations and give them
to the machine to solve and plot -- even complex calculus equations. The
engineers no longer have to do the tedious mechanical part of mathematics
but can concentrate upon the objects they are designing, doing things that
were never possible before these powerful computational tools.

​DC is still immature. I discuss it not to promote it, but rather as an
example of a tool that will drastically modify the job of many designers.
At this point it is quite limited, but it is a good indicator of what the
future will bring.

Will this replace designers? Yes, it will replace those who only have
low-level skills. But for the rest of us, it will empower us to do
ever-more exciting and interesting designs.

Don

Don Norman
Prof. and Director, DesignLab, UC San Diego
[log in to unmask] designlab.ucsd.edu/  www.jnd.org  <http://www.jnd.org/>

Multiple faculty positions in design at UC San Diego
http://d.ucsd.edu/jobs/


-----------------------------------------------------------------
PhD-Design mailing list  <[log in to unmask]>
Discussion of PhD studies and related research in Design
Subscribe or Unsubscribe at https://www.jiscmail.ac.uk/phd-design
-----------------------------------------------------------------