At the very least I'd say the author should clarify the droplet sizes. Saying droplets are large and small means different things for different people. A person studying nanodroplets will consider 1nm sized droplets small and 1um droplets large. But a person studying spray systems may thing 1um is small and 100um is large.
Generally it is fine to talk about general physical concepts without reference but, for example, when you start talking about droplets going 200mph you probably should add references to that.
The author needs to make it clearer in some parts that the droplets won't all be travelling at the same velocity. They have done in some parts, but not in others. Even if the droplets are all the same size (this seems unlikely to me), same sized droplets may move at different speeds. If the droplets have different sizes they can move with different velocities. This is discussed [
here] although the focus of the paper is collisions and not dispersion.
One comment didn't make much sense to me:
A single breath releases 50 - 5000 droplets. Most of these droplets are low velocity and fall to the ground quickly.
You can't really state any claim like this without knowing a lot more information like the size of the droplets. Even so, "Droplets with a lower velocity" may settle more slowly, not more quickly. This is because a) they may have a smaller velocity component in the direction of gravity and b) I believe they will lose more quickly any of that initial momentum they had from an individual breathing. I would also never qualitatively characterise droplet dispersion by just the droplet velocity. If you want to characterise (qualitatively) the dispersion, you use non-dimensional numbers. Stokes number and terminal velocity ratio (settling parameter) are two that I have come across. Some authors use non-dimensional numbers which don't have a specific name.
I don't believe the example about the restaurant is explained very well. For example, downwind and upwind mean specific things. Actually I think people use "upstream" and "downstream" more frequently. If you have airflow over an object, upstream means any flow ahead (or prior) to that object. Downstream means any flow behind that object e.g. its wake. At least that is my understanding of the terms. Maybe that is not correct but it has not failed me in the past. So I am not sure what the author means by upwind here. If you also look at that room, it is 6m across so I would guess that close physical contact would also be an issue here.
A passing comment - I don't like the author saying droplets can spread across entire rooms or "within seconds" etc. It can happen sure, but there will be instances where it may not happen and really you need to assess each area individually rather than make sweeping comments like this. This may be a situation where simulations can help - modelling rooms like boxes and tracking how far one droplet travels with various configurations. But high quality simulations like this may not be computationally feasible.