Moreover, creatures from outside "Town" will be encountered more often in small hamlets. Without a critical mass, field mice from the surrounding area will be more present than city rats.
So you have a nice grayscale when you gain little from hamlets but all of it from cities.
Oh, dear, now we have to track the home of every stray rat NPC.
Do the inhabitants think of it as a town? Then it's a town.
you mean you don't do that as a matter of course?
At least as common. Depending on where and when in Mythic Europe we are.
Some places dispersed farms are common, some places "everybody" lives in villages.
Just as a side note, in England the historical difference between a village and a hamlet was not size, but that a village had a church.
That would work for me.
Good heavens. A debate about a virtue and skills revolving around the technical and practical definitions of a 'town.'
In my game, a settlement large enough to be townlike and have built structures and an ongoing population of more than a handful of people is a town.
If it is abandoned it might also qualify as a Forest (that is, a wooded area; I don't mean a game preserve), or Desert, or whatever, depending on context of ability use.
What about Townsville in Australia. Town or Village?
As the SG controls the world; a town is what the SG says, is the not too helpful answer, however, if there are multiple SGs, I can see why you would want consistency.
I do like the market, and/or church options.
Wouldn't that be between a village and a ham? «Hamlet» is a diminutive, so that's explicitly small, if we are historical enough.
Not by the time the word had reached England and the English language.
It is from the Anglo-Norman hamelet which came from Old French hamelet which was a dimunitive of hamel, which in turn was a dimunitive of ham.
When you observe a market, you collapse the wave function, and either it is (and always has been) a town, or it isn't 
No wonder you're confused about this, scientists have been for decades!
And when you reach the end of Hamlet everyone, including Rosencrantz and Guilderstien are dead so you don't even have a settlement of any type.
Ambassador from England?
This is why you either love or hate Ars Magica.
I for one vastly prefer these kinds of discussions as compared to...another kind of fantasy rpg.
Some of us said regular market, not just any market. It is that mix of residents, regulars, and itinerant traders which makes a medieval town in my mind. Ways of the Town should apply when that ambience makes it into the narrative. The seasonal fair should have a subtly different ambience.
A lot less so than popular media have tried to convince the population.
Mostly it's just math, really.
It depends. Scientists can nail down the math and most of the predictions, its trying to make a model that explains them that drives them buggy, especially since so many of the rules seem to contradict each other. What we really have for QM is less a scientific model and more a record of outcomes from which we can interpolate, which was the state of most modern sciences before the invention of the scientific method. Alchemy was fine for smelting ores and people "understood" it as far as they used it.
And your background for this statement is?
Once you have nailed down the math, you have a scientific model.
Because that is really all a scientific model is - a mathematical description of reality which can be used for analyzing and predicting the behaviour of reality.
Quantum Mechanics is as good a scientific model as any other. It is incomplete and imperfect, which is well known, but it works fine.
What drives scientists buggy is not making a model - we already have that - but trying to explain what it "means" since it involves so many concepts that just aren't part of human experience.
If you define a model to mean only the math you are correct. However when the atomic structure was being argued between "plumb pudding" and solid object models the math in terms of the probability an electron would pass straight through or be deflected was already established. What was lacking was an explanation of those numbers that constituted an actual model.