The point is really that we don't nearly know enough to answer any of these questions. We can provide "best current thinking," but with only 1 actual sample (Earth) and no experience with GRBs, these are just guessitmates at best.
That's not true. We know the luminosity output of GRBs very well, as well as their spectrum across the electromagnetic spectrum. For a number of exoplanets (and the planets in the solar system) we know their main atmospheric content. For each composition of atmospheres you can predict the effects of a GRB for any chosen distance in terms of photo-dissociation, heating and radiation pressure. I don't see great unknowns there.
- How many planets might have other special circumstances that protect their ozone (such as a lack of N2 in their atmosphere, or an ozone generating biology in their stratosphere, etc.)
Not sure. I think it is possible to come up with such scenarios as you stated, but it has to be shown that they are frequent occurrences to be relevant for changing the survival rate of complex life.
Exactly. To say that GRB==doom means that all of these possibilities must be very infrequent indeed, and I just don't see how we can say that at present. That makes me dubious about the hypothesis.
I think it can be argued that due to the luminosity output of GRBs, that "GRB==doom" holds, within a certain radius and for typical orientations. That serves as a useful starting point. For special orientations, or special atmospheres that one could imagine, this may not hold. But then the burden is on the person dreaming up these scenarios to show that these can more happen frequently than expected due to random orientations and atmospheres representative of the gas make-up of observed stellar and star-forming systems (which is well-studied as well, the technical term is metallicity and (heavy) element abundance). Until then, I think "GRB==doom" is a suitable working hypothesis we can adopt.
I do not. The trouble with GRB==doom is that you have to bring in statistics, and we don't know what they are, but we do know that they have to be pretty extreme. For example, a galaxy might have 100 billion "Earths." Suppose that there 1/1000th that number of "warm Venus's," (a Venus type planet, but far enough out from its star that there never was a run-away greenhouse), and that the chance of advanced life forming on such a body is 1/1000th of the typical "Earth," but none of the Venus's are wiped out due to GRBs. That still leaves 100,000 warm Venus's to form complex life. Now, are those numbers reasonable? Sure. Are they true? You guess is as good as mine. However, even if the Venus situation had a probability of one chance in a billion of occurring it would still leave 100 systems where complex life could arise, and that is a lot more than 0.
I think that there are a lot of these situations - I think the proponents of "GRB==doom" have to show that each one has probability 10^-11 or so, and I don't see how they can do that.