Well, I love abstract generalizations probably more than most, but frankly I don't see much use for non-conventional definitions of mark-up in regards to biological artifacts. For that matter, I don't see much point in trying to map semantic constructs from the programming world to the internal workings of biological processes. The fact that RNA has start and stop codons does let us draw analogies to Turing machines but that's about as far as I'd want to go as to finding shared abstractions.
Now, the day may come when we can interface to synthetic genes via Web Services, but that won't be because of any emergent markup phenomenon that was always inherent in the core of biological processes. Rather, that would be the result of some standardized interface that would likely send Simon off to live as a hermit in some far off land, weaving his own clothes and building furniture lest such an abomination disturb his renaissance man existence....