You need to be very careful to distinguish between lossy DNGs and full-blooded ones. If not, it would be very easy to create lossy DNGs for some reason like taking on the road or saving disc space by compressing frames shot for timelapses, and then confuse them with the real negatives, even overwrite non-lossy DNGs.
A couple of features in Lightroom 4 are helpful. Library's filter panel's File Type now distinguishes between the different types of DNG, and you can examine individual images in the Metadata panel's DNG view.
But outside Lightroom, you may not be able to see them at all, and in Explorer / Finder the only indications that you've a group of lossy DNGs may be file size and that the thumbnail doesn't resolve (at least until an OS update).
I have no real reason to create lossy DNGs, so I'm not doing so. If I did so, then I would be taking extra steps to prevent confusion - probably adding a filename suffix to every lossy DNG. Whatever app I am using, I want to distinguish between the full blooded DNG "120210_0123 Blea Tarn.dng" and its lossy counterpart "120210_0123 Blea Tarn LSS.dng".
I'm not saying there isn't some progress, and I welcome fast load (even if I can't see any speed difference) and I like that the small number of JPEGs I shoot with my iPad can now be saved as a sensible-size DNG. But I do feel we shouldn't dive blindly into these waters. We've got the new capabilities, they do present their risks to the integrity of our picture collections, and have we got sufficient workflow benefits? For me, no, not yet. You?