Even brewer's yeast is discarded after a few generations, usually, since it mutates enough to change the flavor of the resulting beer, and that's in an extremely well controlled environment.
Imagine that happening every six weeks in your gastrointestinal tract. Multiply by 6 billion people. I'd guess it's extremely unlikely we're going to thoroughly investigate all of those different variations on the theme.
Is it out of the question for people to be consistently pooping & supplying this data so we have ongoing tracking of the ratios of the various bacteria present in the population? Some clear patterns may arise.
http://academy.asm.org/images/stories/documents/FAQ_Human_Mi...
That represents the consensus view (ie no exotic theories). The really good stuff starts on page 10.
They talk about mother's milk/breast milk being a source of microbes for babies at birth. I have heard this before and I wonder if it could be an alternative to this pill/fecal transplants?
Mother's milk is mostly viruses, so it's unlikely this would work against CDiff. For other users is sounds interesting though!