Let's go back to basics here, and talk about theoretical ideals, and then move to whys and why-nots.
There are two basically accepted notions for engineering a pipe. One is the idea that the airway should be smooth and uniform, with basically no changes of volume (cross-sectional area if you prefer). The idea is that you reduce/minimize condensation by not offering any turbulence inducing zones, no restrictions. Get the smoke in one smooth tube from bowl to button.
This works. And pipes with ragged tenons or obstructive designs generally will gurgle and fuss.
You can up the anti on this idea a bit, and add a nice touch: as a gas moves faster, it exerts less pressure on the side walls of a tube containing it (this is why planes can fly). So you might build a pipe that has an airway that gets smaller and smaller with the idea being that as the smoke cools on its way out of the pipe (and is therefore more likely to condense) it should be moved faster and faster. So a tapering airway seems to work too.
Now, a funneled tenon on one hand is a condensation cause - it's not going to offer as smooth a tube as a perfectly fitting non-funneled unit. But very many pipes do not and cannot have perfect fitting tenon/mortise joints (bent pipes for example) in terms of smoke-hole lining up. So the funnel is a concession in that direction - and it does not seem to do any harm at all.
Personally, I build big airways that taper a lot. But as long as things are at open enough and built well enough, even a fairly small draught hole will work - I have an excellent smoking old Sasieni that is drilled just under 1/8".