An Intelligent Constraint

Think for a moment of the architecture of the Walt Disney Concert Hall. The undulating brushed stainless steel exterior is a multi-planed work of adventurous curves, and it’s a stunning visual fixture in its Los Angeles neighborhood, like nothing else in the surrounding physical landscape. The interior too, is strange and striking, a monument to flawless acoustics, the centerpiece organ a spray of over 6,000 pipes reaching towards a ceiling of swooping planes of Douglas fir.

Now imagine Frank Gehry, the architect, in fits of creative genius as he created his vision for the space. It's easy to think of him as a Romantic visionary, unfettered by the rules that hold mortal men and women back from achieving such works of greatness.

It’s easy to imagine such a thing, sure -- but in fact, that’s not the case at all. The myth of the unbound visionary, untethered by rules and constraints, is just that -- a myth.

In fact, it turns out that genius or vision -- whichever you call it -- functions best, and most readily achieves greatness, when constrained rather than unbound. As the Harvard Business Review explains it:

An intelligent obstacle or constraint is one laden with creative tension, whether stated in the form of a well-defined problem (“How might we simultaneously decrease both inventory and back orders?”) or a challenging goal (NASA’s 1990s mission to land a rover on Mars in half the time and a tenth the budget of the previous mission). An intelligent constraint informs creative action by outlining the 'sandbox' within which people can play and guides that action not just by pointing out what to pursue but perhaps more importantly what to ignore.

In Gehry’s case, this “sandbox” consisted of very specific requirements and rules governing what he could and could not build in order to achieve perfect acoustics. These internal restrictions then naturally led to the distinctive multi-planed exterior of the Concert Hall. Far from being limiting, these constraints served as a puzzle within which Gehry was able to apply specific design principles and best practices, and within that, let his creativity run amok.

In fact, Gehry has said his worst architectural challenge was when he was asked to design a house with zero constraints. “I had a horrible time with it,” he told Forbes. “I had to look in the mirror a lot. Who am I? Why am I doing this? What is this all about?” It’s better to have some problem to work on,” Gehry said. “I think we turn those constraints into action.”

And when it comes to the relationship between vision and constraint, the world of product development is no different. Traditionally, the human mind, left to ramble freely with no rules, is terrible at predicting the precise product needs of the future. Hoverboards and flying cars are examples from fiction, but in the real world, without the right kind of constraints, products like the Newton Message Pad and Google Glass get released into the world .. to a resounding thud.

Our Prediction Problem

Indeed, history is rife with examples of predictions gone horribly amiss. Over 100 years ago, a popular Victorian-era scientist, John Henry Pepper, advised people to sell their gas shares because, he predicted, “the electric light has no future.” Around the same time, an internal Western Union memo declared that the telephone had “no future as a means of communication.”

In our own century, in 1943, then-IBM chairman Thomas Watson speculated as to the future world market for computers. His conclusion? There would only ever be a need for a total of five. Tech predictions had grown no more accurate 50 years later, as this 1996 prediction from Time magazine makes clear:”Apple [is] a chaotic mess without a strategic vision and certainly no future." Even the usually prescient Steve Jobs wasn’t immune to an epic failure of prediction, as in 2001 he expressed his firm belief that the Segway (then known by its prototype name, “Ginger,”) would be bigger than the PC.

Flash forward to the present: Pundits and pollsters alike were blindsided by the Brexit vote, and, most recently, the electoral win by supposed underdog Donald J. Trump. But this likely came as no shock to University of Pennsylvania professor Philip Tetlock (co-author of Superforecasting), who conducted a 20 year study and discovered that, over time, the accuracy of political predictions amounts to little more than chance.

It’s tempting to think that the fault lies in the methodology or skills used to predict these events, but the truth is, predicting the future is more often than not a total crapshoot. And people kind of suck at it.

The question then becomes, with so much data at our disposal, why are we still so bad at predicting the future -- and perhaps more curiously, why do we even keep trying?

One reason may be that it is, quite literally, in our very nature. Humans evolved to see patterns as a matter of species survival -- detecting patterns, after all, can help us predict impending danger or threats. The trouble is, as Nate Silver points out in The Signal and the Noise, our brains can’t seem to turn this survival technique off, and we end up seeing patterns everywhere, even where none exist, and then we go on to use these (non-existent) patterns to make predictions about our future.

Humans also tend to think that we as individuals are exceptional. According to Duncan Watts, author of Everything is Obvious Once You Know the Answer, “Around 90 percent of Americans believe they are better-than-average drivers, and a similarly impossible number of people claim that they are happier, more popular, or more likely to succeed than the average person.” It’s a very small leap to conclude that we also think we are better at predicting the future than everyone else who has tried, and failed, to do so.

We as a species are also great at constructing narratives to explain past events, as Philip Tetlock and Dan Gardner discusses in Superforecasting. How easy is it to understand, in hindsight, how the self immolation of one young man sparked the Arab Spring, or how a series of decisions and actions led to the horrible events of 9/11? It all seems so obvious once we construct a narrative to explain our world, that we make a leap of faith and believe that if only we could have known enough, or been discerning enough, we could have predicted this narrative before the fact.

Alas, no one has managed to master that level of prognostication with any degree of sustained success. For one thing, there’s simply too much information to sift through. Nate Silver writes that “The human brain is quite remarkable; it can store perhaps three terabytes of information. And yet that is only about one-millionth of the information IBM says is produced in the world each day.” Given that level of information overload, our puny brains have no choice but to be very selective. Add to this the myriad of personal biases with which we are all burdened, and the many historical failures of prediction became less laughable, and more understandable.

Creation Within Constraint

Think back to the minds behind the massive flops like the Newton Thinkpad and Google Glass. What were the visionaries behind these massive flops lacking? For one thing, the right kind of constraints, in the form of design principles and best practices. When it comes to the field of product development, constraint can help act as a foil to our fallibility, by intelligently selecting boundaries within which to create and innovate.

In fact, these constraints are inseparable from product vision. No one person can create a product; it takes a team, and that team must be challenged by limitations in order to be able to creatively solve problems in the right way. Otherwise, all work will lack direction, focus, and purpose -- and it will be impossible to translate vision into a concrete product.

Just as Frank Gehry achieved greatness through the limitations of acoustics and space, so too are game-changing products like the Palm Pilot able to hit the market. The device was so successful in no small part because creator and visionary Jeff Hawkins placed an inviolable design constraint on his team: the device must be no larger than a small block of wood he carried in his pocket. All solutions and functionality must, he insisted, work within that specific limitation. Sure, the company eventually went under due to complicated buyouts, but not before enjoying incredible success and laying the groundwork for the future of handheld gadgets.

The relationship between vision and restriction may seem paradoxical, but as design innovation expert Marty Neumeier explains, “Too much freedom can lead to mediocrity. Why? Because without boundaries there’s no incentive to break through them.”

‍