Counterpoint: https://xkcd.com/1425/ https://xkcd.com/1425/ Sometimes, in software, things are hard because they're still research problems. Solving those prob

Counterpoint: https://xkcd.com/1425/ https://xkcd.com/1425/ Sometimes, in software, things are hard because they're still research problems. Solving those problems means paying Ph.D.s to try to make progress on them, with no guarantee of results. If you had guarantees of results, it wouldn't be research, would it? Of course not. Since most businesses aren't going to allocate money for grants, and since firing someone for negative results still doesn't get you your money back or a return on your investment, trying to negotiate with a client whose problem involves a fundamental breakthrough in computer vision, or speech recognition, or attempting to do what Theranos promised isn't going to be fruitful, because success in that field isn't a matter of money. It's a matter of, more than anything else, luck, in getting the right minds in the right context to make the right advances, assuming the advances are even possible.