Dunno if Mr M is still monitoring this discussion, but I think the argument from his perspective would be that DOMUS, in great contrast to TNR, is typically dealing with above average to terrific photos. So, even though DOMUS tends to display most of the "image content," or interferes less in relative terms, it's interfering with something disproportionately more worthy of being seen for itself. Still, it's an editorial and aesthetic decision subject to criticism case by case. The photo editor or editors are clearly trying to add aesthetic value or place the overlays according to aesthetic principles.

There was a recent short post by John Judis at TNR that simply dispensed with any photo at all: The image area was just a black space. I still think that it would be preferable to have some kind of default design somehow relating to authors or subjects, along with general encouragement to think about implicit statements.

The weekend profile of Christie - a "big" article - was to me hurt by both of my two TNR pet peeves: Another crystal clear look at a man's forehead and things of little interest, instead of his countenance (even if it's wire service countenance) and a luridly inane article title that is obviously supposed to appeal more to the kids:

http://www.newrepublic.com/article/116601/chris-christies-rise-and-fall

Thanks for your very helpful and thoughtful reply.

First, I just want to note that the Rand Paul example was for me a laugh-out-loud example of the stylization working very well. I've also noted in my follow-up post that the approach also works, if less cheekily, on some other photos - highlighting O's eyes in the example also mentioned above, or filling virtually empty bottom-center space as against interesting framing or background imagery in a recent photo illustrating an article on Edward Snowden.

In the follow-up post's comments we speculated that, if in some arrangements the results seem awkward, it could be more the fault of the photo editors than of the designers, attributable to inattention to detail or concept. I don't know who is responsible for such decisions at TNR - my guess is that responsibility varies greatly from article/post to article/post depending on presumed importance - but it may be that not everyone has been adequately briefed on the concept.

It's hard to do any interesting or unusual design that's also foolproof or inattention-proof. Still, if we agree that the Stelter results above or the Perkins results noted in the other post are sub-optimal, and if such results are in fact likely to recur, since run-of-the-mill illustrative photos are in many or most cases going to center on a face or profile (rather than, say, a crotch), then presumably a designer could have taken the question into account, and have provided fallback options. The run-of-the-mill photo could be cropped in a way that de-centers the subject. It could be reduced and tiled, or it could be enlarged. Or the translucency of the overlay could be adjusted. Or a default illustration or set of default illustrations could be employed instead, perhaps tending to distinguish blog or other "high-metabolism" posts from featured (and more attentively designed) articles. To varying extents, such options could be automated: If the lower level editors or writers or whatever responsible parties can't be trusted, then they should not be put in the position of having to make decisions.

I understand what you are saying about homogeneity and branding. The headline area is or can be a place for the latter, while under a re-affirmed preference for word/idea over image/design, the concerns will soon reverse themselves, in the extended test. "Homogeneity" is also transparency or universality of the shared literary medium. When I'm reading an essay, I don't want to be thinking about the quality of the paper or virtual paper, or the highest quality paper and the best font will be the paper and font that let me forget about paper and font.

I agree with you that the headline area, like the headline itself, is something different, however. It seeks attention or is exposed to a different kind of attention or initial access. As it happens, I'm not a big fan of TNR headline-writing either, since I find it often sensationalizes and distorts serious pieces, putting their arguments in a bad light. (See initial comments here, for example.) It's a common problem and TNR isn't the worst offender, but in a way that further underlines the same point: The click-baitiness of TNR headline-writing somewhat contradicts the branding effort. It makes TNR more like all of the others. Even thinking too hard about branding at all is a somewhat "generic" tendency, paradoxically enough: It says, we want to be different, just like everyone else.

My personal preference, which I think would also be more consistent with TNR's literary aspirations and traditions, would be for a headline style, both as a matter of design and as a matter of verbal expression, that, when in doubt, defaulted to understatement or the statement of no statement or to a less that was more.