[discoursism]

I think that caveat that you cite is operative. Although I take it in a slightly different direction than the author.

The point at which integration tests become important is not really a question of bullet-proof-ness. It's a question of complexity. When the underlying system has sufficiently many moving parts/degrees of freedom, at some point, you need to see what it does when your system interacts with it in a certain way.

Generally this is not necessary for integers. In most domains, you're not going to come upon any boundary conditions in the way they are transformed or compared. Transactions are also a pretty simple concept with little in the way of moving parts. If your system doesn't rely on their behavior near boundary conditions, maybe they are transparent enough that you don't have to test your system's interaction with the transaction concept. But there are a lot of underlying systems that are not like that.

[murftown]

I think that's right, that other than bullet-proof-ness, the boundary conditions and interfaces between nontrivial subsystems matter a lot and need to be integration-tested. Glad you pointed that out.