I agree the specific example he used is not a great one. I think you nailed it. It seems his main problem was spending a lot of time and resources on a problem

I agree the specific example he used is not a great one. I think you nailed it. It seems his main problem was spending a lot of time and resources on a problem that most likely wouldn't exist in Waze the startup. It's not the kind of problems he wants to spend time on, which is valid. In this particular example he even acknowledged that it was important. But it's not always the case and then I can understand the frustration. As you said "question the rationale for doing something".