Your point on the difference in the evidence for each is valid. You've convinced me that the evidence strongly supports exotic dark matter; I don't think it irrefutable, but no more convincing refutation exists, nor is it likely to emerge, which is a form of supporting evidence in itself. There's no evidence against it, which also helps immensely, and is something most GUTs only wish they could say. As long as they keep predicting proton decay at levels experiment proves impossible, Pati-Salam remains the best by default, and we may be approaching a point where simply warping the theory to allow longer lived protons won't be enough, because if the half-life is so high it CAN'T be experimentally detected (bearing in mind that an experiment that can't be replicated is worthless) even a valid GUT will have left the realm of science. With that in turn in mind the distinction of exotic dark matter and GUTs as equal classes of theories also seems valid, though it also means they both raise the same potential problem that invalidating any or all known theories can't invalidate the class unless it involves invalidating some necessary common attribute.

I assume it goes without saying at this point that eliminating conservation of baryon numbers would put us back at square one on MACHOs. I'd really REALLY like to see a new GUT that doesn't require proton decay (or strong evidence for Pati-Salam in its entirety) because of the issue to which I keep returning on GUTs: Even if protons DO decay, if they do so so rarely that we can't observe a reproducible event we don't have a viable scientific theory. It would really suck to have the right theory about something so pivotal and no way to verify it, but that seems increasingly likely; if we need a neutrino detector the size of the Moon we're screwed. Meanwhile, theories that simply revise the half-lifes lower limit upward indefinitely unnervingly remind me of the particle zoo.

And that is very much as it should be. I like a neat creative theory as much as the next guy, and am all for simplifying the big questions whenever possible, but the data has to tell the story on its own. If a verifiable theory's valid, doing experiments in the area it addresses will sooner or later prove it whether anyone wants to or not. If it's not, the same experiments will prove that, too. That's THE reason why I don't want to rule out anything, however unlikely, until/unless it becomes impossible, and likewise not accept as fact anything, however probable, until/unless it becomes irrefutable. I don't really think reputable scientists need to be TOLD that, per se, but sometimes it's hard to see the forest for the trees, so a periodic reminder couldn't hurt.