Have you programmers ever tried to fix a problem with your program by trying random things? Did it work? I am so tired of hearing about some random thing that

Have you programmers ever tried to fix a problem with your program by trying random things? Did it work? I am so tired of hearing about some random thing that supposedly helps with mental illness as if trial and error without understanding was a reasonable method. What mental illness is, is an ordinary illness that occupies a social niche where it's exempt from normal scientific or medical study. It evades our social immune system the way retroviruses evade our physical immune system. The sane, rational, reasonable approach would be to put everything into developing tools and measurement techniques to establish what mental illnesses are. Depression, mania, and psychosis need, above all else, tools like blood sugar tests and A1C provide for diabetes. 99% of solving any problem is describing it. Administering random treatments to suffering people and lab animals is not progress, it is a tragic waste, cargo cultism, part of the disease that transcends the actual victims. The only thing I know of that's really seemed like progress in ages is opto-genetics. Just...develop tools because without them there is no evaluating treatments. Stop using surveys. I would like to think that researching mental illness would lead to ways of enhancing normal cognition that would help everyone. I would expect that people would think of this possibility, and therefore there would be plenty of funding available. But it doesn't seem to work that way.