Two ways to become better

5 June 2017
5 Jun 2017
West Lafayette, IN
5 mins

It’s the movie-version Mark Zuckerberg in The Social Network talking. Just like the real guy, except with 10% the smiles and 10x the sass.

Cursors fly across the screen, scripts run, files download and upload, and all along, through the lens of the film, we stare at Mark’s eyes, fixated on something beyond the glowing pixels on his early-2000’s laptop. We see a solo genius, just a short decade away from a half-trillion dollar valuation with almost two billion users across the globe.

“I need the algorithm,” he says, and types along.

Startup culture has a strange love affair with the idea of working alone. We like to celebrate the cliche of the lone artist, slaving away through the night, a can of Red Bull on her right and a wall of displays on the left. Surely, this is the self-portrait of disruption at work! This is how a technology revolution eats the world.

But on the other side of the Slack chat (so to speak), we love collaboration. Startup culture spearheaded the push into more open office designs and more integrated, more connected teams. We wanted email to be faster and calls to be less intrusive. We replaced cubicle dividers with emoji-studded Slack threads to encourage collaboration, at the cost of cutting isolation. Here, we kill off lone artists in favor of powerhouse teams.

So which is it? Do you believe in the promise of isolation, or the potential of connection?

Ideas grow differently when we work alone than when we collaborate.

Let’s take an example. In machine learning, to solve difficult problems in computer vision, researchers created a type of system called a convolutional neural network. Mimicked after the visual cortex of the brain at a high level, convolutional networks are the best tool we have to make computers see objects today. But it turns out, when we applied the same kind of system to a completely different problem, recognizing natural language, it worked quite well. Convolutional neural networks, designed for vision, are also used widely in natural language processing today, to make computers understand what we mean when we talk conversationally.

There’s two kinds of innovation to unpack in this story. One is the advent of convolutional neural nets that helped us create better computer vision systems. We saw a problem, and tried to find a better solution to the set of related problems. But there’s a second kind of innovation that followed, which took the idea and found a completely different application, multiplying the impact of a single idea into a new area.

Working alone helps you take a deeper look at a problem for better insights or more optimal solutions. A lone genius, working alone, might find a better way to solve a particular problem than anyone else. But collaboration in a community brings a fundamentally different kind of benefit. Rather than taking one problem and looking for better solutions, we can now take a single idea and try to take it in a thousand different directions. This isn’t possible simply by focusing more intensely on a single problem.

So here’s a roundabout answer to the isolation-versus-collaboration question: working alone and working together help us move forward in different ways. We need to recognize the different roles played by individual expertise and collaboration. One is not a substitute for the other.

This is why I love the open-source communities and coworking spaces like The Anvil. They bring in their respective areas the kind of benefit that would not be possible by simply having smarter people in isolated pockets.

The best work arises when we rely on individuals to create, and communities to amplify the impact.

Finding better problem-solvers isn’t a surrogate for creating more diverse groups, because the ways that individual work and collaboration move us forward are different.

Perhaps more importantly, this is why diversity is irreplaceable.

Having the most innovative people can only help so much if the people we seek and the people we welcome have shared stories and ways of thinking. The benefit of collaboration, the fundamental reason we build teams rather than just chase the single best person, is that sharing ideas and working together grows the footprint of a single idea beyond the abilities of a single person. And when we choose to ignore diversity or chase it as a vanity figure, we deprioritize the reason we build teams in the first place.

In a recent interview, Marc Andreessen pointed out a characteristic of Silicon Valley he thought was a core strength, that when there’s an interesting trend or idea, we carry that to the extreme, trying every possible permutation of that single innovation to find the places where it works to solve existing problems, better. For a while, this causes some confusion and rush, but it helps technology have a much greater impact that it would have one at a time.

To me, it seems like there are two ways of becoming better, both in teams and as an industry: finding better people, and finding people who are different. I don’t think we can simply depend on one or the other to push us along.

Purdue has more entrepreneurs

The cost of a polarizing choice

I share new posts on my newsletter. If you liked this one, you should consider joining the list.

Have a comment or response? You can email me.