It's been I think an unmitigated success in physics. Yeah you get the occasional ridiculous paper. But since you are writing to people in the field usually people figure that out pretty quickly. What's really nice is that you get a range of papers that I don't think normally would get published in journals. I'm here thinking of some of Lee Smolin's papers that go well beyond the typical science paper. I think it also aids in both promoting mathematical "trials" of ideas that may go no where but are helpful as well as more negative results. (The latter being quite important in my mind) It's kind of surprising that after 15 years since LANL started doing the physics repository (now arXiv.org) that some disciplines are still wondering about following suite. I think this has more to do with ingrained power brokers than science.

To second Kurt's point Rodney Stark made a similar claim about the growth of Mormonism into a major world religion. Yet, at least in the US, the overall proportion of self-identified Mormons remains close to 1%. (i.e. despite proselytizing and a relatively high birth rate relative to the average many Mormons leave the fold at about a rate enough to end up with stable growth on par with overall American growth) So I think Stark's claim is pretty hard to buy. John, I never read the book but I read several intereviews with him. I think his point was more that even if Republicans have more kids they end up moving to areas where they structurally will tend towards becoming Democrats. (i.e. out of more rural areas into cities) Where I think his prediction may prove wrong is that I'm not sure the growth of big cities will continue. It's hard to say, His point about how strongly the type of environment you live in affects your vote seems pretty dead on though. I sometimes wonder that if we broke up high population density areas into individual states if people would continue to vote the same way.