Why doesn't $\omega ^ {\omega}$ have cardinality equal to real numbers? One set of order type $2\omega$ is the set {1,2} $\times  N$
One set of order type $3\omega $ is {1,2,3} $\times N$
Similarly, order type $\omega ^2$ would be the set $N \times N$
$\omega ^3$ would be $N\times N \times N$
.
.
.
$\omega ^{\omega}$ would be the never-ending cartesian product $N\times N\times N\times N\times.........$
However, this infinite cartesian product simply corresponds to the set of all infinite sequences of natural numbers, aka the real numbers. Then, doesn't this mean that a set of order type $\omega ^{\omega}$ would have cardinality equal to reals?
Moreover, you could apply the diagonal argument to the set $N\times N\times N\times......$ to directly prove that it's uncountable.
 A: In the case of cardinal arithmetic, where $\omega^\omega$ is the set of all functions $f\colon\omega\to\omega$, then yes, you are correct. That set has the same cardinality as the real numbers. And indeed, it corresponds to the infinite product of copies of $\omega$ itself.
However, in the case of ordinal arithmetic, where $\omega^\omega$ is the supremum of $\omega^n$ for $n<\omega$, there $\omega^\omega$ is countable. To see that, note that what it actually corresponds to is the functions in the infinitary products above which are eventually $0$. So we can identify those with the finite sequences of natural numbers, which is a countable set.
What we see here is an example of continuity of ordinal arithmetic, which is defined to be continuous: limit steps are literally the limit of their previous steps; and cardinal arithmetic, which is defined by taking cardinality of certain sets, and it is very discontinuous, especially when infinite exponentiation is involved.
A: In ordinal exponentiation, $x^\omega$ = sup($x^n$) for natural number n. As such, it is different form cardinal exponentiation.
In The Book of Numbers, by John Conway and Richard Kenneth Guy, there is a bijection from the natural numbers to a set of order type $\omega^\omega$. The bijection is specified by these two rules:

*

*If the greatest prime factor of m is greater than the greatest prime factor of n, then m comes later than n.

*If m/x comes later than n/x, then m comes later than n.

One can verify that this well-order has order type $\omega^\omega$.
