"As you have brewed [the soup], so you must drink [the soup]." by Lisa Rotzinger

August 10, 2016

"As you have brewed [the soup], so you must drink [the soup]." by Lisa Rotzinger

Lisa Rotzinger is one of our newest employees. She is a Designer for Networked Matter. You may recall Lisa from last year when she was an intern and presented at the Autodesk Leadership forum at Autodesk University in Las Vegas. At my urging, Lisa submitted this posting to my blog.

As you have brewed [the soup], so you must drink [the soup].

My name is Lisa Rotzinger, and I have been an official industrial designer since the beginning of this year. I am now a trainee in the Office of the CTO and member of the Corporate Strategy & Engagement (CS&E) team under Jon Pittman, VP of Corporate Strategy.

Why this team, you ask? Long story short: An Autodesk Visiting Fellow — who happened to be part of the CS&E team for a year — was searching for an intern to help him explore the (and I quote) "future of design." I was in my last year of college (University of Wuppertal, Germany), studying abroad at the University of Cincinnati, Ohio, and quite curious to jump right into not just what constituted my desired profession, but the very future of it.

I got the opportunity to spend several fascinating months at Autodesk experiencing the future from the point of view of Mickey McManus, Visiting Fellow, designer and former CEO of Maya Design, as well as my Autodesk peers, customers, and collaborators. At the center of our research — code named Primordial [soup] — lay the convergence of three big trends:

Advanced Manufacturing
Machine Learning
Internet of Things (IoT)

The latter (IoT) might be the most controversial and vague of the three, as it can generate different trains of thought for different people: some think of the IoT as consumer products that automate your home, others think of it as predictive maintenance in huge factories (Industrial Internet of Things, IIoT), and others just think of the mysterious big data issue.

That is why, for the rest of my story, I would like to stop using the term at all. What we basically tried to convey is the following: Given that technology develops exponentially, and therefore complexity increases exponentially, how do we need to change our approach to design to still get to thoughtful and valuable outcomes? (If you want to dive deeper, read the articles published by Mickey on networked matter:

Besides the theory, we also decided to do our first experiment in designing a thing as part of a greater system — Hack Rod — a car chassis created as the result of feeding sensor data back into Autodesk's generative design software.

Although we succeeded in designing the chassis and are hoping to manufacture the first physical prototype in the near future, there is still a long way to go. One of the toughest challenges will probably be the creation of an actual closed feedback loop from real-world data to a design software.

This might require the automation of data capture, storage, and analysis processes and — as a next step — the integration of data that goes beyond stresses, like data about the environment, the race track, and the electrical activity of the driver's brain.

Only recently, I might have stumbled upon a tiny little (but important) detail that could be part of why we got stuck with the data thread of Hack Rod. I happened to be part of a machine learning meet-up at Autodesk, where current experiments and practices were shared, to represent Hack Rod and Primordial. Throughout the conversation something became clear to me: for data analysis and the application of machine learning methods, it is important to have a really good understanding about what questions you want to ask and what's possible in terms of data analysis before you instrument a thing, a person, a space, or entire networks of them.

What sounds trivial seems to be overlooked by a lot of people trying to design connected things or places as part of complex systems. Does increasing complexity make us forget to ask the right questions?