>>10924562Have you looked at the actual formalism of CH? There is a step where you verify that the individual histories are consistent with each other, and if they are, great now you can use classical logic with this particular family of histories. Decoherence is great too because it makes it easier for the family to be consistent.
But the problem is that we can make more than one family. Each family is consistent with itself, but they are inconsistent with each other. This is what Lubos means by "splitting the pie" in the quote above. Which family do you pick? This is also a very concrete definition of the idea of complementarity. Both families are equally valid but they are consistent with each other.
This idea is core to the CH interpretation.
Lubos is saying that in verifying the predictions of QM by performing an experiment we are selecting which complementary family to use. In that sense the observer is important.
>>10924571Yes they seem the same to me to, but you hear claims that CH is observer free, which is a big deal
