Pletch,
You can pick my brains at any time (although I must warn you that the pickin's are often very slim).
Couple of thoughts: First, Acer is right about a radiused hole lasting longer. Any sharp edge will heat and erode much faster than a rounded edge. Second, as Dan pointed out, and I tried to indicate by saying that this was a pyrotechnic issue, radiant heat and hot particulate from the pan flash are driving the speed with which the main charge goes off. It is not the flow through an orifice, of whatever shape, as, initially, there is no pressure driving the flow. The pan side combustion is occurring at one atmosphere and the pressure inside the barrel is one atmosphere. While there is some very minor local pressure increase as the pan goes off, it is very, very small and cannot result in much gas flow. Acre is also right in that flow through a counter sunk / coned liner is flow through a venturi, but that comes into play from the inside to the outside as the main charge goes off and there is about 10,000 psi inside the barrel driving gas flow out of the vent. The throat (i.e smallest cross section of the vent) will go sonic (the gas velocity will be the local speed of sound in that gas at that density and temperature). If the vent is coned on the outside, the gas will increase in velocity to super sonic. It is this very high velocity, very hot gas that causes gas cutting and local erosion at the vent.