I found the following article by Stereophile quite interesting: http://www.stereophile.com/content/winning-loudness-wars

Of note to me was that music had 16dB of PLR during the 1982-1984 period vs. 9dB in the 2006-2009 period... Maybe the '80s were not that bad (relatively speaking.)

This is what I got from the article: TV (and possibly radio) are implementing Loudness Normalization through EBU R 128 and ITU-R BS.1770 (vs. Peak Normalization) across the board which may discourage it "during mastering". My best guess is that the article's assumption is that once the standards are in place, further compression (post-normalization) would sound awful and inconsistent (not 100% sure if that was the point though.) Another highlight seem to be that album normalization is preferred as it preserves the soft quality of certain songs.

I didn't quite understand why 11dB of PLR was great, or why -16, -23, or -14 LUFS/LKFS was the way to go in Loudness Normalization settings, but those numbers where thrown quite a bit.

My concern though is that TV and radio seem to use Peak Normalization according to the article, and I fear that the move to Loudness Normalization may not necessarily translate in recording studios moving to less aggressive compression of dynamic range.

I'm not very versed in this, so I'm very interested on more qualified opinions, corrections, and comments.