I discoverd the function this morning, browsing through results from a literature search. And let me emphasize, it's not obsolete, yet. This function may be totally obvious and trivial to you, but it struck me as noteworthy. After all, I think journals are the dinosaurs of scientific communication and in today's age only group scientific publications by pseudo-tag. The function I noticed of course hasn't anything to do with quality, that's ridiculous. We all know quality doesn't go by journal. It wasn't novelty, either. From all my papers at least, the one with the least amount of novel content was published in Science. What does seem to work at least to some extent, though, is the scope of the paper for readers outside of the field. The 'bigger' journals usually attract articles that cover more ground, even if not everything is novel in the field. For someone working in a related field who's not really an expert, but close enough to be able to understand it, this sort of articles allow him/her to keep up with the field on a superficial level. This description certainly fits our Science paper: we validated and confirmed much of the previous experiments in our field combining different experiments in the same publication. Thus, while the quality and novelty content in individual papers may vary independently of journal name, it seems to me that on average journal hierarchy is a pretty decent indicator of the scope of the articles in that journal.
In a way this is trivial as already the names of the journals give away their scope: Science, Nature or General Physics Relativity Astronomy and Mathematical Physics and Methods. However, in journal rank 'generality' is a derived score: something is supposed to be so groundbreaking that even non-specialists would be interested in it. The fact is that for a non-specialist even a short review over the current literature would be groundbreaking, because a non-specialist doesn't know that literature! Which may explain the higher citation rates for GlamMag papers: more people will be non-specialists and not even realize that much of the results of these papers isn't even new. I wonder how much of the citation advantage of GlamMagz lies in this review-character of many of their papers?
Conversely, when journals have died out, what label do we use to indicate how valuable a paper is for a non-specialist? How do we indicate scope? Is this the sort of rating that would work for scientists? I'm looking at you, PLoS One! In a way, Frontiers in Neuroscience is attempting to develop such a label.
This is all probably obvious to everybody else, but I only realized it this morning.

Maybe I have to take back my 'pseudo-tag' one-liner, at least for now...