1,535 karma · joined January 31, 2020
You see this everyday right now during the pandemic. Policy making is very different from say, civil engineering, where failure scenarios can be quantified within tightly defined bounds. At the nation state level every decision may affect everybody. Nobody wants to be told their specific circumstances are the "statistical outliers". But science rarely gives good answers. The only time science has been reliable is when a large amount of resources is dedicated to figuring out answers in a single, specific problem. Unfortunately the life sciences don't have the luxury of physics and the formal sciences. There is no concerted level of effort to cure cancer that is comparable to the Apollo program. The military prefers to buy ten more hellfires than to redo life science ethics. If you quantify the ultimate utilitarian value, the damage ten missiles can do to humanity may be the same as a FDA that focus more on progress than covering their own ass. Yet the former is more acceptable than the latter. Science has failed because it no longer gives useful answers. When somebody's grandmother is lying sick in the hospital, they want a cure that is immediately accessible, not yet another feel-good post on Reddit or Facebook about "Scientists claim experimental cure for cancer found! Appearing in your friendly neighborhood pharmacy in ten years".
If you want an exact replica of the cow mammalian milk production process that is molecule-to-molecule compatible then you need at least two dozen tissue engineers. This is slightly beyond what a "fungus" to do.
This argument is circular logic because everyone has written bad code at some point in their life. Assuming that the author has not suffered any form of death between birth and now, and assuming that when he was born he had no knowledge of programming, by intermediate value theorem of course he has written bad code.