Great principle. I agree with it, and think that most of the people offering objections here do as well. Our question is whether it's it's being applied correctly in this case. For me, the sticking point is whether surface temperature can be used as a proxy for the brightness we care about.
BTW, what's your obsession with Munroe?
I have none. He's a great explainer, probably occasionally gets details wrong, and (so far as I can tell) is a positive force for spreading scientific understanding. My "obsession" seems to be that I have poor tolerance for overly broad smackdowns of critics. If you are going to tell someone else that they are wrong (as opposed to Munroe who is trying to explain what he believes is true) I think you have a higher obligation to get all the details right. I presume I'm sensitive to it because I'm often on the receiving end.
How about you? Why does it bother you so much that some people say that Munroe's argument is logically flawed?
and I agree it's counterintuitive and it's OK for people to come up with ideas, but when you hit the point of "hey, the thermo people have a nice collection of proofs demonstrating this, and it fits very well with the underlying theory, oh, and if you do manage to violate etendue, you could probably build a perpetual motion machine", if you willingly continue to argue and get shot down, it's time to go back and re-read the books.
BTW, what's your obsession with Monroe? What he's referring to is a scientific phenomenon, Monroe is just a science popularizer, and if he got the details wrong- well, the point of xkcds like that is more to inspire people with ideas, than get the exact details right.