The notion of faking a feature until you figure out how to make it for real, has real world consequences when it becomes just part of the background approach to

The notion of faking a feature until you figure out how to make it for real, has real world consequences when it becomes just part of the background approach to product development. So what that the problem you're aiming to solve is impossible, that all the techniques you can think of to try don't actually do what you need them to? Implement them anyway, maybe some combination of them will eventually stick. If you don't, you'll lose, well, fail to gain market share!