Crosscut

I have big ideas that I want to see realized in Caterpillar. But I have to spend my limited resources wisely. And I'm absolutely convinced that iterating on something basic that works, will lead to a better result than building up to some grand vision, without tangible benefits on the way there.

This means I have to define milestones that provide those tangible benefits. And it means I have to decide what won't make it into a milestone. Which is going to be most of what I eventually want to achieve. As a result, these milestones will look very different from the desired end result.

And sometimes, I might need a placeholder. For example, if I decide we need better types, but a static type system is too much work for now, I'll go for a dynamic one. Which is a totally different direction from where I want to go eventually. This might look like wasted work, but I don't think it is. Not having a working product to incrementally improve on, would be worse.