> In the context of assigning Bayesian priors that represent complete
> ignorance, Jonathan Weiss asks:
>
> 1) Someone presents you with a huge deck of cards (not standard playing
> cards -- each card has a spot of a given color on it). Before even
> one card is seen, what is the probability that the first card dealt
> is red?
>
> The problem as stated is ill-posed until we know what set of alternatives we
> are considering.
It seems to me that the problem is as well posed as any other problem
involving degrees of belief. Most texts on probability take for granted
that we have a sample space (or that the sample space is obvious). In
practice, this is frequently not the case. It seems to me that a
nontrivial difficulty in using probability (or pretty much any other
approach to uncertainty that I'm aware of) involves describing the sample
space. The difficulty is compounded when we don't know some potentially
relevant features of the sample space (as in this case). But that doesn't
make the problem of assigning degrees of belief in this case ill-posed.
-- Joe