Let T denote a completion of ZF. We are interested in the number μ(T) of isomorphism types of countable well-founded models of T. Given any countable order type τ, we are also interested in the number μ(T,τ) of isomorphism types of countable models of T whose ordinals have order type τ. We prove: (1) Suppose ZFC has an uncountable well-founded model and $\kappa \in \omega \cup \aleph ₀,\aleph ₁,2^{\aleph ₀}$. There is some completion T of ZF such that μ(T) = κ. (2) If α <ω₁ and μ(T,α) > ℵ₀, then $\mu (T,\alpha )=2^{\aleph ₀}$. (3) If α < ω₁ and T ⊢ V ≠ OD, then $\mu (T,\alpha )\in 0,2^{\aleph ₀}$. (4)...

We continue the work of Shelah and Casanovas on the cardinality of families of pairwise inconsistent types in simple theories. We prove that, in a simple theory, there are at most ${\lambda }^{<\kappa \left(T\right)}+2^{\mu +\left|T\right|}$ pairwise inconsistent types of size μ over a set of size λ. This bound improves the previous bounds and clarifies the role of κ(T). We also compute exactly the maximal cardinality of such families for countable, simple theories. The main tool is the fact that, in simple theories, the collection of nonforking extensions...

We study the distribution of rational points on a certain exponential-algebraic surface and we prove, for this surface, a conjecture of A. J. Wilkie.

Gruenhage asked if it was possible to cover the real line by less than continuum many translates of a compact nullset. Under the Continuum Hypothesis the answer is obviously negative. Elekes and Stepr mans gave an affirmative answer by showing that if $C_{EK}$ is the well known compact nullset considered first by Erdős and Kakutani then ℝ can be covered by cof() many translates of $C_{EK}$. As this set has no analogue in more general groups, it was asked by Elekes and Stepr mans whether such a result holds for...

We prove that it is consistent that the covering number of the ideal of measure zero sets has countable cofinality.

We formulate a Covering Property Axiom $CP{A}_{cube}$, which holds in the iterated perfect set model, and show that it implies easily the following facts. (a) For every S ⊂ ℝ of cardinality continuum there exists a uniformly continuous function g: ℝ → ℝ with g[S] = [0,1]. (b) If S ⊂ ℝ is either perfectly meager or universally null then S has cardinality less than . (c) cof() = ω₁ < , i.e., the cofinality of the measure ideal is ω₁. (d) For every uniformly bounded sequence $⟨fₙ\in {ℝ}^{ℝ}{⟩}_{n<\omega }$ of Borel functions there are sequences:...

For any three noncollinear points c₀,c₁,c₂ ∈ ℝ², there are sprays S₀,S₁,S₂ centered at c₀,c₁,c₂ that cover ℝ². This improves the result of de la Vega in which c₀,c₁,c₂ were required to be the vertices of an equilateral triangle.

We construct a compact set C of Hausdorff dimension zero such that cof(𝒩) many translates of C cover the real line. Hence it is consistent with ZFC that less than continuum many translates of a zero-dimensional compact set can cover the real line. This answers a question of Dan Mauldin.

We develop the theory of topological Hurewicz test pairs: a concept which allows us to distinguish the classes of the Borel hierarchy by Baire category in a suitable topology. As an application we show that for every ${\Pi }_{\xi }^0$ and not ${\Sigma }_{\xi }^0$ subset $P$ of a Polish space $X$ there is a $\sigma$-ideal $ℐ\subseteq 2^X$ such that $P\notin ℐ$ but for every ${\Sigma }_{\xi }^0$ set $B\subseteq P$ there is a ${\Pi }_{\xi }^0$ set $B^{\text{'}}\subseteq P$ satisfying $B\subseteq B^{\text{'}}\in ℐ$. We also discuss several other results and problems related to ideal generation and Hurewicz test pairs.

This paper deals with questions of how many compact subsets of certain kinds it takes to cover the space ${}^{\omega }\omega$ of irrationals, or certain of its subspaces. In particular, given $f\in {}^{\omega }(\omega \setminus \left\{0\right\})$, we consider compact sets of the form ${\prod }_{i\in \omega }{B}_i$, where $|B_i|=f\left(i\right)$ for all, or for infinitely many, $i$. We also consider “$n$-splitting” compact sets, i.e., compact sets $K$ such that for any $f\in K$ and $i\in \omega$, $\left|\right\{g\left(i\right):g\in K,g\upharpoonright i=f\upharpoonright i\left\}\right|=n$.

Answering a question of Miklós Abért, we prove that an infinite profinite group cannot be the union of less than continuum many translates of a compact subset of box dimension less than 1. Furthermore, we show that it is consistent with the axioms of set theory that in any infinite profinite group there exists a compact subset of Hausdorff dimension 0 such that one can cover the group by less than continuum many translates of it.

Si espongono alcuni risultati, provati dall’Autore negli articoli citati nella bibliografia, a proposito della complessità del teorema d’interpolazione di Craig: con ciò si intende la relazione tra la lunghezza (cioè il numero di simboli) della formula $\chi$ e la lunghezza di $\phi$ e $\psi$, ove $\phi \to \psi$ è un’implicazione valida, e $\chi$ è un interpolante, come esibito dal teorema di interpolazione stesso. Si intende altresì sottolineare la rilevanza dello studio della complessità dell’interpolazione per far luce su alcuni...