The crazy thing that brings hidden complexity is the waterfall model. “We have that thing that get fed to this one”. Ok, how do you report errors, in the upstre

The crazy thing that brings hidden complexity is the waterfall model. “We have that thing that get fed to this one”. Ok, how do you report errors, in the upstream system? The downstream one? Do you have parameters with the downstream process? Now you need the user to understand how they influence the output. My point being that when you put an arrow on a white board you absolutely and immediately have to be skeptical and assume that some data will have to move against the arrow, and that will drag some hidden complexity. Stuff like very far downstream passes in compilers will need the source file line numbers and some inversion system because they might need to report errors or emit debug symbols, and that can be complicated because the code might be unrecognizable at that point.