> An airplane that has constantly filtered air is a hot spot but Costco is relatively benign?
An airplane where you are packed in with other people in a small tube close enough that you are essentially (and often actually) in physical contact on one (or both if you are neither in a window or aisle seat) side is a hot spot, but Costco where you are moving, everyone else is moving, there are wide spaces where you don't have to be particularly close, and high ceilings and wide bay doors so it's essentially outside, is relatively benign.
> Your risk calculus is out of whack.
No, yours is, unless you are on some weird airplanes where everyone is wearing a breathing mask so that the constant filtration happens before your exhalation gets all over your neighbors.
Dispersion matters, as does exposure time. A single inhalation of recently expelled air can be enough. You might waft past someone in Costco, but you could be sitting in the air current of someone shedding virus particles for hours on end, even if the particles are eventually being filtered by A/C.
I have seen airlines discourage that. I wonder whether it makes things better or worse (or whether it is some sort of prisoner's dilemma - making it better for yourself, but worse for others?).
Really? Admittedly I only fly Delta, but they've certainly never discouraged it. Also, I don't really see how it could make things worse for others, since it would also be pushing any viral particles you may be expelling towards the floor where the intake vents reside.
> but you could be sitting in the air current of someone shedding virus particles for hours on end, even if the particles are eventually being filtered by A/C.
Airplane air turnover is _extremely_ high, apparently even higher than many specialized hospital environments. If there's any danger, I would imagine that it would come from the immediate airflow within the cabin, and not from gradual saturation of the cabin's air.
I don't actually have a good model of airplane risks. But I recognize that many of my intuitive assumptions about what's going on would need to be checked against real data.
Your risk calculus is out of whack.