Don't forget that DRMs aren't here to save Amazon from piracy. Today and for the next couple of years, what they really care about is securing a dominant position. So the DRM security only needs to be good enough to reassure clueless edition managers, who believe that DRM will help their business rather than hurt it. So they have no incentive to get out of their way to bug DRM-stripping customers (in light of which, I'd guess some of the morons who handled Linn Nygaard's case will be fired, and deservedly so).
Same as iTunes DRM actually, which Apple ended up discarding once they were in a position to tell the music editor to just shut up.
Yeah that's the issue...From what I can tell, Amazon differentiates between both and I imagine that would affect how its central database of highlights/bookmarkings keeps track of each individual book. I suppose this would affect you only if you were highlighting a DRM book and then stripped the DRM...which I guess is mostly an edge case.
You need to have a unique ID set in the header. You can do this in calibre.
Specifically, a unique 4 byte character sequence at the 32 byte offset in the header (MOBI header). See: http://wiki.mobileread.com/wiki/MOBI
If that's set (easy in calibre), you will sync across your devices.
So, yes, the main purpose of DRM stripping -- at least for me -- is as part of backing the books up into my Calibre collection. I don't bother to copy the stripped copies back onto my Kindle, there's just not any compelling reason to.