« Taking an example of a UI, a window with 5 buttons on it it will likely requires 5 time more testing than one with only one button, even if the later brings a lot more value, which suggests that test coverage is crap. »
Actually, that suggests that that model of test coverage is crap. The idea that one button requires a certain amount of testing, and that five buttons naturally requires five times as much is terrible measurement and worse testing. Are there interdependencies between those buttons? What do those buttons do? Does each do the same thing? Are we talking about an online dating service or a nuclear power plant, here?
