In a place I can't link to, I encountered the somewhat boggling statement that "Nature leans more in the direction of Popular Science than Critically Peer Reviewed [Journal]." Thus, a quick poll:
Context is for the weak.
Nature and Science are hybrids. The front section of each includes news, views, and commentary on some of the articles/reports in the journal (and/or competing journals). That is the "Popular Science" part of the journal. Then there are the peer-reviewed research reports (and occasional review article) toward the back (this is the part that distinguishes these two journals from publications like Physics Today. At the very end you will usually find some kind of classified ad section.
Perceived trendiness is a major factor in determining which submitted manuscripts get sent out for review instead of triaged in the editor's office. So although I think the quote goes a bit too far, it does have some basis in reality.
I'm quite sure that all of the researchers who consider their Nature publication to be the centerpiece of their career will be pleased to learn that they acheived nothing more impressive than getting their work into Omni.
I have to go with, "like all around you, dude. Whoa." Want to know about Cell?
A physicist I once knew described papers published in Science and Nature as "extremely important, if the results turn out to hold up." That is, the articles are cutting edge, and probably have a higher rate of being overturned by new data at some point than papers published in top-level field-specific journals, which tend to involve more theoretical synthesis.
I call Nature and Science the "yellow rags"- more along the lines of sensationalist journalism touting "extremely important advances" in science with little or no substantial details.
I'll echo what Harlan said at #4.
If you see it in Nature, you know it's important, and probably wrong.
Not the purpose of this post, I know, but isn't Nature (the TV show) still running? I'm positive I watched it last week...
I'd go with #5's definition.
What I was originally looking for was "a pain in the butt". Nature papers are much harder to get your hands on electronically if you aren't at an institution with a subscription than many other journals. That's pain in the butt #1. Pain in the butt #2 is that if you are foolish enough to submit there, they sit on it for a long time trying to decide if it's worthy of them. Pain in the butt #3 is that they then want to rewrite your abstract, so that it says something wrong and is poorly written to boot. Pain in the butt #4 is when they leave off the names of one of the co-authors (causing internal strife), and then, when they issue preprints with him added back in, misspell his name.
Pain in the butt #5 is when you're trying to figure out what an astronomer did in a Nature paper, you're sort of out of luck. The papers aren't long enough to hold all the necessary details, and if the authors didn't go on to publish a longer and more detailed paper in a real journal, then you're left guessing what really happened.
I will never in my life submit to Nature. Was on a collaboration that did once, watched a grad student go through it once, and have been frustrated too many times at not being able to get papers. A lot of the publishing companies that control our journals are twisting and tweaking the process of science communication in ways that are anathema to science, but Nature really takes the cake on that. Just say no....