And with the growing of farming labors and capitals invested (tools, farm fields, fertilizers), the efficiency can slowly increased to current level or beyond. THAT's the definition of scalability.
I'm aware of the definition, thank you.
Only specialized agriculture society will be able to make a profit out of it, and the level of sustaining a community itself will require a large portion of the populations/industry capability. Those only choose small farmer populations will have to import foods. (Or completely dependent on importing food or hunting)
In case you haven't tried this: download Dwarf Fortress, create a world, start fortress mode, and build a farm on a patch of muddy ground. You will make a profit. It's almost impossible not to. So how can this be changed so that "farming capital" and specialization actually matter? Right now there's no place where "scalability" can change anything--each additional square of farm plot is exactly as productive as the first one, and the labor input is small enough to be irrelevant.
Are you being completely ignorant?
Each tile produces X food. This food requires Y effort to take care of it, with untrained labourers.
From what I understand, he wants the time taken by labourers to pick and plant the crops to be much increased. This is reduced by having highly skilled labourers, tools to aid in the process, and fertiliser. A fort that isn't dedicated to keeping these in good condition will be unable to efficiently produce food via farming.
We can then introduce the efficiency of having larger fields by having a efficiency increase in crops produced in a larger field, while countering the exploitability by forcing a set amount of labour that, in a small amount of food production, would favour the smaller fields.