Wayne Folta sends in this. It seems nuts to me (although I was happy to see that no mention was made of this horrible argument of a related sort). But I know nothing about theoretical physics so I suppose it’s all possible. I certainly have no sense of confidence in anything I’d say about the topic.
It's roughly equivalent to claiming that because young people have fewer landlines, so you have to add young people in to your polling models, that means there are real, virtual 20-somethings popping into existence and then vanishing whenever you take a poll. Because you used a math trick, and that must correspond to reality.
If its the technique of breaking the multiverse into finite samples that causes this end-of-time, aren't there an infinite number of ways of dividing up that multiverse for sampling? So shouldn't many of those destroy our universe at every possible moment? Ouch.
Plus, it strikes me that the whole calculation is incorrect. They use an example of setting an alarm clock with 50% odds and either waking up in one second or one billion years, and how because you cut some universes in the middle, the billion year olds are more likely to get cut off. But that assumes you're only cutting the older half of universes off. You should cut off half of the beginnings of the universes, such that suddenly someone starts existing 500 million years into a nap. This should offset the people who were cutoff, and restore valid values to the model.
I don't know physics, and reading a popularized account distorts things more — though I have usually found New Scientist to be reasonable, which is why the article caught my attention.
As I understand it, the article seems to be making a similar argument to saying that we use MCMC techniques to analyze financial data, therefore financial markets must have underlying quantum mechanisms. (Because MCMC maintains multiple, random chains in parallel. I'm stretching my analogy a bit here.)
I'd sent it to Andrew because it seems to be — as far as I can understand it — a great example of taking your model way too seriously. Maybe even going so far as to assume that the mechanisms underlying your model must underly reality.
Or, perhaps the opposite: assuming that if the same mechanisms do not underly both your model and reality, your model must be "wrong". Reminds me of the quote about all models being wrong, but some being useful.