I'm very much in favor of what you wrote there. I've been thinking to start a separate thread about this some time. Though feel free to beat me to it, I won't be ready to do so very soon anyway. But here's a stab at what I'm thinking.

This is from the welcome thread:

A note for theists: you will find LW overtly atheist. We are happy to have you participating, but please be aware that other commenters are likely to treat religion as an open-and-shut case. This isn't groupthink; we really, truly have given full consideration to theistic claims and found them to be false.

This is fair. I could, in principle, sit down and discuss rationality with a group having such a disclaimer, except in favor of religion, assuming they got promoted to my attention for some unrelated good reason (like I've been linked to an article and read that one and two more and I found them all impressive). Not going to happen in practice, probably, but you get my drift.

Except that's not the vibe of what Less Wrong is actually like, IMO, that we're "happy to have" these people. Atheism strikes me as a belief that's necessary for acceptance to the tribe. This is not a Good Thing, for many reasons, the simplest of which is that atheism is not rationality. Reversed stupidity is not intelligence; people can be atheists for stupid reasons, too. So seeing that atheism seems to be necessary here in order to follow our arguments and see our point, people will be suspicious of those arguments and points. If you can't make your case about something that in principle isn't about religion, without using religion in the reasoning, it's probably not a good case.

What I'd advocate would be not using religion as examples of obvious inanity, in support of some other point, like in this, otherwise great, post:

http://lesswrong.com/lw/1j7/the_amanda_knox_test_how_an_hour_on_the_internet/

Now I'm not in favor of censoring religion out and pretending we're not 99% atheists here or whatever the figure is. If the topic of some article is tied to religion, then sure, anything goes - you'll need good arguments anyway or you won't have a post and people will call you on using applause lights instead of argumentation.

But, more subtly: if the topic is some bias or rationality tool, and religion is a good example of how that bias operates/tool fails to be applied, then go ahead and show that example after the bias/tool has already been convincingly established in more neutral terms. It's one of the reasons why we explain Bayes' theorem in terms of mammographies, not religion.

Feedback would be welcome.