Jerry Coyne has published this useful piece at The Conversation, briefly setting out his case against the religion-science accommodationist position (i.e. the position that religion and science are, in some sense, straightforwardly compatible). He followed up with a post about this on his Why Evolution is True site.
As a reminder, I sometimes write for The Conversation, via its Australian set-up (my profile is here). In January 2016, I wrote a piece for The Conversation on this exact topic, entitled "Against accommodationism: How science undermines religion."
(My preferred version can be found here, if you have access to Academia.edu. However, there is little difference: I've merely reworded some phrasing about Stephen Jay Gould that caused unnecessary, unintended, and unexpected offence.)
This piece responded to (and acted as a sort of review of) Jerry's excellent book Faith vs Fact: Why Science and Religion are Incompatible. You can, perhaps, detect subtle differences in our approaches, but we're on the same wavelength here. I've long argued that science has a tendency - logically, psychologically, and historically - to undermine religion, which I understand in much the same way as the Canadian philosopher Charles Taylor.
It is also true that the humanities have tended to undermine religion's authority as they've explored the historical provenance of particular religions, not least Christianity.
As for what counts as a religion, I think "religion" is inevitably a fuzzy concept. The boundaries between religions, political ideologies, and various other worldviews are not sharp. Some ideologies show many of the trappings of religion. Indeed, they show many of the worst features of monotheistic religions in particular: their identification of orthodoxies and heresies; their reverence for certain figures and demonization of others; the psychological transformations they can produce, often leading to zealotry; and their intolerance of dissent.
Nonetheless, we can describe the phenomenon of religion in a way that tracks most people's intuitions. See, for example, my discussion of religion on pages 5-10 of Freedom of Religion and the Secular State and the short discussion of religion and ideology on pages 101-103 of The Tyranny of Opinion: Conformity and the Future of Liberalism.
In The Tyranny of Opinion, I describe religions in this way (p. 101): "Religions are (very roughly) systems of belief and practice that include otherworldly or supernatural elements of some kind: they involve belief in something that transcends the observable world or eludes empirical inquiry. Further, religions teach that there is an otherworldly dimension to human lives, and that this has at least something to do with how we ought to act in this world. Religions almost always involve rituals of some sort, and they typically include moral norms that are given a supernatural or otherworldly rationalization."
In recent years, I've become slightly less concerned about the claims of religion - and about religion's threats to our liberties - as I've become more focused on political ideologies and their dangers. Nonetheless, debates about religion and its value, and about its intellectual and moral authority, will not go away any time soon.
To be sure religion itself is slowly fading in Western countries (and some others), as Jerry notes in his post at Why Evolution is True, but my fear is that many people who are turning away from religious faith are turning to other belief systems that offer comfort and the sense of meaning that comes with claims to esoteric knowledge, while showing authoritarian tendencies. That, in fact, seems to be a danger with any belief system except the most open-ended and individualistic. Even if religion eventually goes away, as is clearly happening in Western and Northern Europe especially, there won't be an end to the struggle against authoritarian systems of belief.