I decided on a whim to look in at the Dilbertblog, where the top post at the moment has Scott Adams calling all atheists that discuss on the net irrational, using a rather neat strawman carbon copy of most discussions of faith between believers (i.e. mostly Christians) and atheists he has seen on the web.
At the end of it comes a question: supposing it could be established that there's a truly infinitesimal (in the physical sense, not the mathematical) chance of a God existing - he quotes "one in a trillion" as a model probability; would it then be fair of an atheist to claim the non-existence to be proven?
Somehow, this illustrates clearly the thesis I occasionally pronounce (when I'm feeling belligerent) that mathematics and theology have a lot more in common than mathematics and physics. For a physicist, the quoted probability of failure is a dream: most physical 'truths' are much looser than that. However, for very many people, the choice between self-identifying as an atheist and as an agnostic falls in the agnostic camp, since Absolute Certainty can never be reached in such a question. For many more, the Lack of Absolute Certainty takes on a handy argumentative baseball bat shape for Believers, who use it to cast doubt on their supposedly wholly rational argumentative opponents.
I'm a mathematician. I acknowledge that things may not be what I perceive them to be. I realize that this may make me into an agnostic - HOWEVER, selflabeling myself as agnostic doesn't change anything. I still view the existence of a God, or several, of descriptions known to man or unknown, as mainly irrelevant.
If you get something out of it - fine, that's great. I, however, don't, and therefore view the hypothesis as irrelevant. I don't believe in a God, and very much not in the (sorry, a) Christian God, since I find the hypothesis preposterous, and of very little impact on my own life.
That would be my answer to Scott Adams - too long to fit in the already too long answer thread.