They are. But limiting the number of people you’ve been potentially exposed to to just the people directly next to you is obviously safer than being exposed to every person between your table and the bathroom/exit.
That being said, the virus isn’t magically going to fill the entire space equally from one infected person. There is a distance factor involved, and the closer you are unmasked the higher your risk. So only unmasking at your table cuts your risk pool down to the tables immediately proximate to you, rather than everyone you walked by, and anyone you shared the bathroom with. It’s not as safe as staying home, but it’s better than being unmasked the entire time.
Oh, and the risk isn’t going to hang around hours after the fact, unless if you’re in a completely unventilated space. Commercial restaurants change their air over a recommended 8 times an hour.
2. Research shows takes less time to diffuse than you'd think, and lingers longer than you think. Since spring 2020 research has shown to wait till tomorrow if someone infected was in the space today.
3. Your own mask mitigates this, as does the amount of time you're breathing the infected space, as does the amount of human turnover (how many rolls of the dice of the space hosting an infected breather, and how long are they exhaling in the space).
2. I would like to see that cross referenced against ventilation capacity. Again, restaurants are supposed to be changing their air over with outside air a lot, the idea that a properly ventilated space would have represent an infection risk days later is in the “doubt” category for me. Basically, please provide a source.
Heck, last time I was in Vegas I could barely smell people smoking a few feet away from me, the ventilation was that good. If cigarette smoke is the metaphor we’re going with for viral particles and dividers, that strongly implies that better ventilation yields a lower infection risk.
3. Exactly. This is an argument for masking when you’re not eating. It’s not 0 risk, but it’s lower. How much lower is debatable, and obviously depends on more circumstances than just masking policy alone.
Personally I’d rather just get takeout when local case loads are high, like right now.
Do you have a citation for this? I would be interested in reading this study.