Well that would be one thing, but I can see there's plenty of problems with trying to capture someone in the middle of the battle. There's plenty of enemies that end up getting disabled before they are killed and if you can provide some assistance so they live a bit longer, then you can capture a few after the dust settled. You could also equip soldiers with hammers, as it would decrease mortality.
1)Will the new aquifer changes means it could be viable not to use constructed walls but instead construct a drainage system of grates and underground tunnels that naturally drain the aquifer somewhere else? Or will the water flow be too much for dwarfs to go through?
2)Will water someday seep through constructed walls, albeit at a slower rate the more walls there are between the aquifer/water source and the walls, making drainage systems for mines an actual necessity?
3)Will different kinds of rocks have different rates of permeability, affecting the speed that water seeps through the aquifer, or it will always be a fixed rate?
Trying to capture someone in the middle of a battle is probably a bad idea, yes. Capture requires extra time and resources (tying material, which can be abstracted, disarming that probably would have to be explicit, and prisoner guards that have to be removed from the actual battle to guard the prisoners). Thus, capture would presumably be reserved for mopping up (e.g. those pesky campers that just sit around twiddling their thumbs when the others flee). Still there's no reason for the game to refuse you the option of giving stupid(/brave/creative) orders.
1. It's possible to just let a current aquifer (i.e. near artesian) pour out in a sufficiently large room to evaporate (I've done that by mistake, trying to muddy a large area for a tree farm, only to realize I needed to build walls to concentrate water in one area at a time). The new aquifers will also require basins/cisterns for large water usage projects such as obsidian farms and magma sea obsidianization (when using the aquifer as the source: cavern lakes and surface rivers will probably increase in importance for these purposes).
2. Implementing this eventually will result in a number of issues (not saying it's bad, only that it needs to be balanced):
- Mud (= subterranean plants and obscured engravings) everywhere. At a minimum we need a way to clear away mud.
- Harder to get a fortress going if it gets slowly drowned in water unless you set up pumps (but evaporation should do it, unless air humidity tracking is introduced to block evaporation when the air is saturated).
3. Given that the new implementation provides for different rates, I guess the question is rather whether the rate is tied to the material (i.e. the permeability) of the layer the aquifer is in, randomly determined, or determined based on multiple factors (layer material, rainfall, slope...) with or without a random distribution.