Yea, but the problem with that, contagions don't work linearly. It doesn't take many missteps to screw up the whole thing. That, and the whole mask thing is starting to smell more like the crap from the 1950s, "If you see a nuclear flash, jump underneath your desk for protection."... yea... false hope is good and important. But only for a little while.

And don't get me wrong, I technically want to be wrong. But science and human nature butt heads. People are human, perfect lab conditions don't imitate human actions that well. Relying too much on lab condition results and ignoring the idea of, "How will Joe Schmoe fuck this up?" means you wasted lab time.

Hell, how many UI folks here? Or any programmers that have to deal with user testing? In your perfect locked away testing, everything is "perfect". Admit it. You think you solved all the ways users will screw up. Then it goes out into the wild, only to find out how incredibly stupid humans are. How many times have you said, "That's not the way you're supposed to use that!" regarding software and a random person?

Why is the HN community so willing to think this behavior doesn't apply outside computers?

A lot of that stuff from the 50s was actually pretty good advice. It wouldn't help if you were a mile from the blast, but if you were say 5-10 miles away, you have at least 20 seconds between the flash and when the blast wave hits. There may be glass and debris flying everywhere, and you have a much better chance of survival if you're under a desk with your head covered.