Listen

Description

This story is also a podcast. Press play on the audio player to listen, or keep reading. Same story either way.

Forty years ago, we used computers to give animators perspectives they couldn’t easily create by hand. Today, I’m using artificial intelligence to discover perspectives I might not have considered.

The technology has changed. The curiosity hasn’t.

I’ve been thinking about that lately, because I spend my days with people trying to figure out what AI might mean for their organizations, their work, and their lives. And it keeps taking me back to a room full of Disney animators who wanted to fly a camera through Big Ben. Straight through it, which at the time couldn’t be done. Someone asked, “What if we could?” and that question opened a door none of us fully appreciated.

I’ve spent a good part of my life in rooms where I wasn’t the smartest person, and sometimes wasn’t sure I understood the conversation. Those have often been the most interesting rooms. I’ve sat with animators at Disney, visualization experts at Lockheed, artists at Industrial Light & Magic, and engineers who explained mathematics that made my head spin. I’ve also spent years in kitchens, listening to people describe a meal they wanted without having the slightest idea what it was.

The most useful thing I brought into every one of those conversations was what I wanted to understand.

Squash and stretch

At Wavefront Technologies, the computer animation software company I co-founded, we worked with Disney as they explored what computers might add to traditional animation. I remember animators explaining something they needed our software to do, using a car on a mountain road as the example.

A General Motors truck takes that curve and obeys the laws of physics. A Disney car takes it fast. It stretches, skews sideways, and deforms in ways that would give an automotive engineer a headache. That exaggeration communicates speed, energy, and personality, and the principle behind it is called squash and stretch. The animators were telling a story, and physics could take a back seat.

I understood the picture. I’m a visual thinker and I’ve always painted pictures with words, so I carried what I’d seen back to our engineers. One of them said, “Oh, what they want is nonlinear transforms.” I remember thinking, What the hell is that? Then he explained it. You put a three-dimensional box around an object, bend the box mathematically, and the geometry inside bends right along with it. Suddenly I could see it.

The engineers knew how to build something I didn’t know existed. The animators knew what they wanted but not how to describe it mathematically. I sat in the middle. We’d build something, take it to Disney, and hear either “Yes, that’s it” or “Almost. What if it could do this?” Then around we’d go again.

I never needed to know how to write the software. I needed to understand the artists well enough to explain them to someone who could, and then I needed to listen again.

The trouble with knowing

Being good at figuring out what people want carries a danger. You start to believe you already know. The moment I project my own answer onto someone else’s problem, I stop listening, and I’m steering them toward what I’ve already decided makes sense.

So I reflect back what I hear. I’ll say, “Let me see if I’ve got this right,” or ask for an example, a habit I picked up from listening to Ira Glass interview people on This American Life. An example turns an abstract idea into something that happened, something we can both picture. And when someone says, “No, that’s not quite what I meant,” I’m delighted. The correction shows me the distance between what they intended and what I understood. That distance is the work.

OMV and a sandwich

For years I’ve used a simple framework called OMV: Objective, Measure, Value. What are you trying to accomplish? How will you know you did? Why does it matter?

Take a hungry customer who wants something crunchy, a little salty, maybe a little sweet. I suggest a BLT, and they say, “That’s exactly what I wanted.” The objective was a satisfying lunch. The measure was food that looked good, tasted good, and filled them up. The value was leaving happy and feeling that someone understood them.

The Disney animators answered the same three questions. They wanted more expressive animation, tools that let them experiment, and better stories at the end of it. Different rooms, same questions.

Twelve departments

Fast forward four decades. I’m sitting with people from the Santa Barbara Museum of Natural History, talking about how AI might help their organization. Someone mentions that the museum has twelve to fifteen departments, and that they all speak different languages. Paleontology, astronomy, and every other discipline has its own vocabulary, methods, and priorities. They share an institution without sharing a language.

That sounded familiar. At Wavefront, I sat between artists and engineers. Today, through Coastal Intelligence, the Santa Barbara AI consultancy I co-founded, I sit with organizations trying to understand what AI can do for their work. They know their customers, their missions, and their problems. What they may not know is how to say it in a language an AI system can act on.

Why should a paleontologist have to become an expert in prompting? Why should a nonprofit executive need to study context engineering to produce a meaningful impact report? Questions like those shaped our framework, DEVA: Data, Envision, Validate, Apply. We listen to what someone has to work with, understand what they’re trying to accomplish, create something, show it to them, ask whether we got it right, and put it to work.

The technology has changed dramatically. The human part feels remarkably familiar.

Coastal OS

Recently my partners and I started building Coastal OS, our own set of tools for the work of running a consultancy: presentations, surveys, workshops, reports, proposals. We wanted tools shaped around how we actually think and serve our clients, tools that look beautiful, reflect our brand, and are a delight to use.

Forty years ago I helped create tools for animators, and we needed teams of software engineers to turn ideas into working systems. Today AI lets me take part in building those systems directly. I still find that extraordinary. It brings back the kitchen, when someone described a meal they couldn’t imagine and I could see how to make it. It brings back the animator who said, “Wouldn’t it be cooler if we could do this?” That joy of making something out of nothing hasn’t left me. If anything, I’m more curious now.

I’m even developing this story by talking to an AI. I talk, I remember, I wander. The AI asks questions, sometimes it misunderstands me and I correct it, and occasionally I discover a connection between experiences I’ve carried around for decades. I already knew the Disney story, squash and stretch, the kitchens, the AI work. What was new was how they connected.

Forty years ago, computers calculated visual perspectives that were brutal to produce by hand. Today, AI helps me find intellectual perspectives I hadn’t considered. Understanding takes context, experience, questions, a willingness to be corrected, and curiosity about what someone else sees that you don’t.

The bridge

For years I described my work as translating between people who speak different professional languages: artists and engineers, customers and developers, business leaders and technologists. Translation loses something along the way. A bridge is a better word, because a bridge lets each side stay itself. The animators stayed animators and the engineers stayed engineers, and they gained a way to connect their expertise.

I’m always looking for gaps. The distance between what someone wants and what they can currently do. The difference between what one person says and another hears. The space between an idea and the tools to make it real. Some gaps are worth keeping, because the differences between people are often what mak…



Get full access to Through Another Lens at marksylvester.substack.com/subscribe