Let $\omega^\omega$ denote the collection of all functions $f:\omega\to\omega$. For $f,g\in\omega$ we define

  • $f\leq g$ if $f(n)\leq g(n)$ for all $n\in\omega$;
  • $f\leq^* g$ if there is $N\in\omega$ such that $f(n)\leq g(n)$ for all $n\in \omega$ with $n\geq N$.

The cardinals ${\frak b}, {\frak d}$ are defined as follows:

  • ${\frak b} = \min\{|S|: S \subseteq \omega^\omega \text{ and } \forall f\in\omega^\omega \exists s\in S(s \not \leq^* f)\}$,
  • ${\frak d} = \min\{|S|: S \subseteq \omega^\omega \text{ and } \forall f\in\omega^\omega \exists s\in S(f \leq^* s)\}$.

Let us define ${\frak b}', {\frak d}'$ using $\leq$ instead of $\leq^*$. Can it be proved that ${\frak b}' = {\frak c}$ (which would imply ${\frak d}' = {\frak c}$)? If not: which of ${\frak b}' < {\frak b}, {\frak b}' > {\frak b}, {\frak d}' < {\frak d}, {\frak d}' > {\frak d}$ is consistent?


Using "$\le$" instead of "$\le^*$," I claim $\mathfrak{b}'=\omega$ and $\mathfrak{d}'=\mathfrak{d}$.

To see $\mathfrak{b}'=\omega$, let $f_n$ be the constant function $x\mapsto n$; then setting $S=\{f_n: n\in\omega\}$, there is clearly no function which is $\ge$ every element of $S$.

To see $\mathfrak{d}'=\mathfrak{d}$, suppose $S$ is a set of functions with $$(*)\quad\forall f\in\omega^\omega\exists s\in S(f\le^*s);$$ I'll build an $S'$, with $\vert S'\vert=\vert S\vert$, such that $$(**)\quad\forall f\in\omega^\omega\exists s\in S'(f\le s).$$

To do this we proceed as follows. For $\sigma\in\omega^{<\omega}$ and $f\in\omega^\omega$, let $f^\sigma$ be the function gotten by replacing the length-$\vert\sigma\vert$ initial segment of $f$ by $\sigma$; that is, $f^\sigma(n)=\sigma(n)$ for $n<\vert\sigma\vert$, and $f^\sigma(n)=f(n)$ otherwise.

Now, given an $S$ with the property $(*)$, let $$S'=\{f^\sigma: f\in S, \sigma\in\omega^{<\omega}\}.$$ Clearly $\vert S'\vert=\omega\times\vert S\vert=\vert S\vert$; I claim $S'$ has property $(**)$.

To see this, fix $g\in \omega^\omega$. Since $S$ has property $(*)$, we may find an $f\in S$ with $g\le^*f$. Then by definition of $\le^*$ there is a $\sigma\in\omega^\omega$ such that $g\le f^\sigma$. But this $f^\sigma$ is in $S'$.


${\frak b}'=\aleph_0$ since you can take $S$ to be the collection of constant functions.

${\frak d}' = {\frak d} + \aleph_0$ since you can take any family $S$ realizing $\frak d$ and close it under finite differences, i.e., let $S'$ consist of all functions that are eventually equal to some function in $S$.

  • 1
    $\begingroup$ You don't need the "$+\aleph_0$," since $\mathfrak{d}$ is infinite. $\endgroup$ – Noah Schweber May 18 '15 at 7:38
  • 1
    $\begingroup$ @NoahS true, I was in a hurry there. The argument I gave shows $\frak d\le\frak d'\le \frak d+\aleph_0$ and then an even easier one gives $\frak d\ge\aleph_0$. $\endgroup$ – Bjørn Kjos-Hanssen May 18 '15 at 7:50

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.