The best defense is to build a technology that isn't cheap to reproduce. There is no better moat than killer IP.
Actually that's the single worst defense. What can possibly be so expensive to produce that it can't be cloned, yet cheap enough to be possible to sell with profit? The fact of the matter is that most software is quite simple, and when it solves a particular problem in an innovative way, there is no way to capitalize on that because there's no way to enforce exclusive use (give or take a few counterexamples left or right, like a super special secret server-side recommendation algo or something like that, but those are outliers).
It's excruciatingly hard to get funding if you do come up with that sort of idea, because the experts all say its impossible. ;)
Don't forget the third part, which is what the author described, i.e. "not very smart people say they can easily do it themselves". They do not have to be able to do it, they just have to be able to convince the CEO to spend $1 million in salary on the internal delivery time to try to do it.