Not sure if it's the exact research that the author is referring to, but it could be this one: https://www.computer.org/csdl/magazine/so/2023/04/10176199/1...

I don't think anybody talking about "latency" without a qualifier is thinking about build latency.

But it's a nice article. The idea that giving-up on waiting for a delay has a simple exponential distribution is something that I never thought. (And now I'm fixed on understanding why... Something must have biased me against it.)