I recently solved a problem that for some reason I have no intuition about.
Imagine a box of width $w$, length $l$, and height $h$ filled with a fluid of density $\rho$ on the earth (so under the influence of gravity). Assume that $l > w$, so we can refer to a "long side" of the box (see attached figure). Compare the force on the long side of the box with the weight of the fluid. 
So, the weight of the fluid in the box is given by $W = \rho g V = \rho g lwh$. Meanwhile, we can find that the force on the long side of the box is given by $F=\frac{1}{2}\rho g lh^2$. (This is just the average pressure on the long side of the box times the area of that side.) Then, looking at the ratio of the force on the long side of the box to the weight of the fluid, we find $F/W = h/(2w)$. This result is what I find puzzling. If I make a very narrow (small $w$) but tall (large $h$) box, for a small amount of fluid, the force on the long side can still be very large (much larger than the weight of the fluid). If I make the long side of the box so it can slide (causing $w$ to change), for small enough $w$ and large enough $h$, I won't be able to hold the side of the box in place, even if the volume of fluid (e.g. water) that the box can hold is very small. We can further decrease $w$ and/or increase $h$ to the point that something like a tractor or bulldozer still won't be able to keep the wall of the box from sliding out.
One thing that makes me feel a little better about this is knowing that for very small $w$, if the wall of the box slides out just slightly, the water level in the box will drop drastically and therefore so will the force that the wall applies on me or the bulldozer or whatever. But still, this idea seems strange. Does anyone have a good way to think about this to make it more intuitive?
