In Scientific American, a discussion between Richard Dawkins and Lawrence M. Krauss:
Krauss: Both you and I have devoted a substantial fraction of our time to trying to get people excited about science, while also attempting to explain the bases of our current respective scientific understandings of the universe. So it seems appropriate to ask what the primary goals of a scientist should be when talking or writing about religion. I wonder which is more important: using the contrast between science and religion to teach about science or trying to put religion in its place? I suspect that I want to concentrate more on the first issue, and you want to concentrate more on the second…
Dawkins: The fact that I think religion is bad science, whereas you think it is ancillary to science, is bound to bias us in at least slightly different directions. I agree with you that teaching is seduction, and it could well be bad strategy to alienate your audience before you even start. Maybe I could improve my seduction technique. But nobody admires a dishonest seducer, and I wonder how far you are prepared to go in “reaching out.” Presumably you wouldn’t reach out to a Flat Earther. Nor, perhaps, to a Young Earth Creationist who thinks the entire universe began after the Middle Stone Age. But perhaps you would reach out to an Old Earth Creationist who thinks God started the whole thing off and then intervened from time to time to help evolution over the difficult jumps. The difference between us is quantitative, only. You are prepared to reach out a little further than I am, but I suspect not all that much further.