(It looks like you've been downvoted a lot. For what it's worth, it wasn't me.) I didn't claim that CRUD apps don't have complexity in them. I claimed that CRU

(It looks like you've been downvoted a lot. For what it's worth, it wasn't me.) I didn't claim that CRUD apps don't have complexity in them. I claimed that CRUD apps don't typically have the sort of thing in them for which this kind of small-scale semi-formal stuff is useful in them. I am happy to stand by that claim; do you actually disagree with it? I get the impression that you think I was being sniffy about CRUD apps. I wasn't. In particular, (1) I don't think CRUD apps are low-value. There are plenty of CRUD apps that add much more value to the world than almost any piece of code of the sort that invariants are super-helpful for. And (2) I don't think that CRUD apps are easy. Well, doubtless some are, but as you say they commonly have huge amounts of complication in them, of a sort that isn't amenable to the kind of formalized reasoning I was talking about. To answer the implied question in your first sentence: I wasn't trying to make a point, I was giving an answer to a question you asked, namely why mathematical induction might be important for programmers. Briefer and more explicit version of that answer: For people writing certain kinds of code, understanding mathematical induction makes it easier to reason about that code via techniques such as loop invariants, which makes it easier to make that code bullet-proof. There are many other kinds of code for which nothing closely related to mathematical induction is of any value at all. (And, to reiterate lest I set the bee in your bonnet buzzing again: Whether a kind of code benefits from this sort of technique has basically nothing to do with how important it is, or how difficult it is to write overall, or what anyone should think of the people writing it.)