I have to disagree - not in principle, but empirically. The Loudness Wars seem to have let up a bit, but there are still way, way too many modern masters of all kinds of music that are mastered too loud.
By "too loud," I don't meant that the mixing and mastering are too compressed - that's an artistic decision, and popular music of all kinds has routinely been compressed ever since the rise of 1960s AM radio rock, if not before.
The issue is that with the advent of look-ahead digital limiters, mastering engineers (usually at the behest of artists and/or labels) are taking those already compressed waveforms and simply turning up the volume on the entire thing, and letting the limiter squish the peaks so it will still "fit" below digital 0.0.
I have no problem with light peak-limiting: If you take a DR10 waveform and turn it up 2dB and let the peak limiter attenuate a small number of the loudest peaks accordingly, so you end up with a DR8 waveform, most people will probably not be able to hear the difference. But if you take, say, a DR11 waveform of music that has tons of loud peaks because of its style, and you turn it up 5dB to end up with a DR6 waveform with 1000s or 10s of thousands of peaks severely limited/attenuated, the result is going to sound like garbage.
And now it's largely pointless anyway, since all the streaming services normalize volume (to avoid jarring volume changes during shuffle play). So your DR6 track is just going to be normalized by the streaming service to 5dB lower volume than your original DR11 mix down would have - your track will play at the same volume it would have, and you'll just have a more boring, monotonous-sounding track because 5dB worth of dynamic peaks will be missing.