Earlier quoted context omitted.
This is like saying my contractor put in a low quality roof, but it hasn’t collapsed yet. The problem is while you can’t control randomness, on average those roofs do fail much more. But more broadly it’s amazing how far developers on HN will go to defend things that will mildly inconvenience developers.
No, it's not like that at all. It's like saying the roofer I hired doctored his references, but then proceeded to build a roof that's objectively good enough for me to hire him again to build another roof, on a bigger house for more money. Cheating to bypass a stupid gatekeeping system that's designed to ration 1 job per N perfectly qualified applicants is only a problem if it's a problem. If you claim that people wh…
If an individual cheats but turns out to be competent, it's no loss for the company or anyone else, really. But on average, if cheating is common, more incompetent people will slip through and ultimately cause problems. I can bet my life savings on black and win, but that doesn't make it a good decision.
> Works well enough, but it's a highly non-scientific, non-rational exercise.
If you'd like to know about the research behind interviews, they've found that unstructured interviews aren't better than pulling people out of a hat, regardless of how effective they feel.