Who is best qualified to assess architecture? A peer review committee at the American Institute of Architects? A museum curator? A design critic with a graduate degree in art history? Here’s a (radical) idea: if we’re designing for the public, how about also asking them?

And if you’re an architect, and want to hear some real perspectives on architecture, you might think about asking a cab driver. Every day they pass thousands of houses, hospitals, churches and schools, so it’s no wonder many of them have strong opinions about the kinds of buildings they like. They may in fact be one of the built environment’s most authentic focus groups.

The first time I caught a cab to the airport after building my house in an historic New Orleans neighborhood, the driver said, “That’s an interesting house…”

“Oh, really?” I said, “What do you like about it?”

“I like that it’s modern, but it fits in with the neighborhood,” he said. “I like the big wrap around porch, the tall windows and sloping roof with the solar collectors. The ceiling fans on the porch remind me of the house I grew up in.”

After that, I began to routinely ask cabbies what they thought about some of the buildings they drove by every day. When I was in San Francisco recently, I broke the silence of my Uber ride with: “What’s the most beautiful building in the city?”

“The Transamerica Building,” he answered. “I like the way it looks from the Bay Bridge. How it stands out on the skyline. But they must have only designed it to be seen from a distance, because I drive around it a lot, and on the street it’s pretty overpowering. It doesn’t seem to fit with its neighbors very well.”

One of my most memorable exchanges came from a taxi driver in Bilbao, Spain.  I asked what I thought would be an obvious question (with an obvious answer). “How do you like the Guggenheim Museum?” I said, expecting to hear pride of ownership in the international icon.

“We like the money it brings to town,” he responded.

“No, I mean, what it looks like,” I said.

“Well, I like the outside better than the inside.”

“How do you think most people in Bilbao feel about it?”

“I think they mostly like buildings with columns.”

So maybe he was tired of answering the same question from visiting architects and designers. And perhaps, if I had taken another cab ride, at another time, I might have found a driver as enthusiastic about Frank Gehry’s masterpiece as I was.

As an architect, I know the thought of ceding artistic control is like taunting the ghost of Howard Roark; it contradicts everything we’ve been taught. We’re the experts after all: design is what we’re paid to do.

But the driver’s response raised a more interesting question: what if architects and designers included the public in the design process as partners, not in the obligatory phase that every architect labels “research,” not to “educate” or pretend to listen, but to authentically engage and collaborate with them? As an architect, I know the thought of ceding artistic control is like taunting the ghost of Howard Roark; it contradicts everything that we’ve been taught. We’re the experts, after all: design is what we’re paid to do.

But the idea of co-design, of creators and users collaborating, is happening elsewhere. In a recent paper entitled “Designing with the Beneficiary,” Joanna Levitt Cea and Jess Rimington, researchers at Stanford University’s Global Project Center, argue that designers can learn more by listening to the people they’re designing for than from experts. Again, this is counterintuitive to what we have been taught, but there is sound scientific support for the idea. They point out that many of the most innovative companies today have moved away from the old model, where products and services are created through closed-door, top-down, expert-based processes, toward more open, crowd-sourced strategies, where the end-user takes a prominent role in shaping and creating the final product.

Why couldn’t this work for architecture? Would it mean abdication of our design responsibilities, or a more humble and authentic approach to place making? Would it compromise our control, or would it provide new insights into how we might define success? Would we (heaven forbid) have to share the credit? There are obviously decisions that by law can’t be shared. But the truth is, like it or not, architects have already ceded a lot of authority and control to consultants and building contractors anyway. Why not, when appropriate, broaden that already large circle to include the people who’ll actually use the building? They understand what works much better than we give them credit for.

Once when I was on my way to the airport in Washington, DC, I asked my cab driver what she thought of Eero Saarinen’s Dulles terminal. “I love it,” she said. “It has this uplifting feeling—like it’s flying.”  She was clearly a sharp and perceptive observer, and one of the many voices worth listening to. Otherwise, how will we know when we’ve got it right?