I Killed Our Own Technology

At Diamond, my team had been building a hardware abstraction layer so developers could target our products more consistently. It made sense. Developers needed a cleaner way to work across hardware, and we were trying to remove that friction.

Then Microsoft DirectX emerged. Once I looked at what Microsoft was doing, I came to an uncomfortable conclusion: they were solving the layer more effectively than we could. We could protect our internal work because it was ours, or we could change direction because the problem mattered more than the solution.

We changed direction. I focused on getting Diamond as close to the DirectX effort as possible and hired Steve Ouimette to become an embedded, all-things-Diamond resource around Microsoft's work in Washington.

Steve was a musician with a music degree, which might not have looked like the obvious résumé for that assignment. He was also talented, curious, and exactly the kind of person who could learn what the outcome required. He went on to build a remarkable career in games and music, including extensive work on Guitar Hero and other franchises.

I have never regretted stopping our technology. The job was not to keep our project alive. The job was to make life better for developers and help Diamond win. Once a better route appeared, our responsibility was to take it.