Cosmic dimensions

Martin's Axiom for σ-centered partial orders implies that there is a cosmic space with non-coinciding dimensions.     

Some remarks concerning C-spaces

We comment on the definition of C-spaces in [D.F. Addis, J.H. Gresham, A class of infinite-dimensional spaces. Part I: Dimension theory and Alexandroff's Problem, Fund. Math. 101 (1978) 195-205] and [W.E. Haver, A covering property for metric spaces, in: Topology Conference at Virginia Polytechnic Institute 1973, in: Lecture Notes in Math., vol. 375, 1974, pp. 108-113]. Furthermore we introduce two types of ‘finite’ C-spaces one of which gives an internal characterization of all spaces having a metrizable compactification satisfying property C. We also introduce a transfinite dimension function for those finite C-spaces. Several questions arise that are related to Alexandrov's problem.     

Dimensions of strong n-point sets

n-point sets (plane sets which hit each line in n points) and strong n-point sets (in addition hit each circle in n-points) exist (for n?2, n?3 respectively) by transfinite induction, but their properties otherwise are difficult to establish. Recently for n-point sets the question of their possible dimensions has been settled: 2- and 3-point sets are always zero-dimensional, while for n?4, one-dimensional n-point sets exist. We settle the same question for strong n-point sets: strong 4- and 5-point sets are always zero-dimensional, while for n?6, both zero-dimensional and one-dimensional strong n-point sets exist.     

Lelek maps and n-dimensional maps from compacta to polyhedra

For a natural number m?0, a map from a compactum X to a metric space Y is an m-dimensional Lelek map if the union of all non-trivial continua contained in the fibers of f is of dimension ?m. In [M. Levin, Certain finite-dimensional maps and their application to hyperspaces, Israel J. Math. 105 (1998) 257-262], Levin proved that in the space C(X,I) of all maps of an n-dimensional compactum X to the unit interval I=[0,1], almost all maps are (n−1)-dimensional Lelek maps. Moreover, he showed that in the space C(X,Ik) of all maps of an n-dimensional compactum X to the k-dimensional cube Ik (k?1), almost all maps are (n−k)-dimensional Lelek maps. In this paper, we generalize Levin's result. For any (separable) metric space Y, we define the piecewise embedding dimension ped(Y) of Y and we prove that in the space C(X,Y) of all maps of an n-dimensional compactum X to a complete metric ANR Y, almost all maps are (n−k)-dimensional Lelek maps, where k=ped(Y). As a corollary, we prove that in the space C(X,Y) of all maps of an n-dimensional compactum X to a Peano curve Y, almost all maps are (n−1)-dimensional Lelek maps and in the space C(X,M) of all maps of an n-dimensional compactum X to a k-dimensional Menger manifold M, almost all maps are (n−k)-dimensional Lelek maps. It is known that k-dimensional Lelek maps are k-dimensional maps for k?0.     

Proper hereditary shape equivalences preserve property C

In a recent paper [6], van Mill and Mogilski prove that a proper hereditary shape equivalence preserves property C, if its domain is σ-compact. In this note, the same result is established without the hypothesis of σ-compactness.     

A note on the Brouwer dimension of chainable spaces

A general method produces from a compact Hausdorff space S a compact Hausdorff space T with IndT=IndS+1. We show that if S is chainable, then T is also chainable while DgT<IndT, where Dg denotes dimensionsgrad, the dimension in the original sense of Brouwer. This leads to a chainable, first countable, separable space Xn with DgXn<IndXn=n for each integer n>1.     

Two-dimension-like functions defined on the class of all Tychonoff spaces

Two-dimension-like functions are constructed on the class of all Tychonoff spaces. Several of their properties, analogous to those of the classical dimension functions, are established.     

Some characterizations of spaces that have transfinite dimension

We characterize separable metrizable spaces that have small transfinite dimension and metrizable spaces that have large transfinite dimension modifying two classical characterizations of countable-dimensional spaces and applying the notion of a strongly point-finite family.     

On M-structures

We introduce the notion of M-structures and consider the class $\text{M}$ of stratifiable spaces with M-structures. Especially we study the relation between the class $\text{M}$ and that of M₁-spaces.     

On the relationship between cmp and def for separable metrizable spaces

For each pair of positive integers k and m with k?m there exists a separable metrizable space X(k,m) such that cmpX(k,m)=k and defX(k,m)=m. This solves Problem 6 from [J.M. Aarts, T. Nishiura, Dimension and Extensions, North-Holland, Amsterdam, 1993, p. 71].     

On nagata's problem for paracompact σ-metric spaces

Let (A) be the characterization of dimension as follows: Ind X?n if and only if X has a σ-closure-preserving base $\text{W}$ such that Ind B(W)?n?1 for every W?$\text{W}$. The validity of (A) is proved for spaces X such that(i) X is a paracompact σ-metric space with a scale {X_i} such that each X_i has a uniformly approaching anti-cover, or(ii) X is a subspace of the product ΠX_i of countably many L-spaces X_i, the notion of which is due to K. Nagami.(i) and (ii) are the partial answers to Nagata's problem wheter (A) holds or not for every M₁-space X.     

Rigid continua and transfinite inductive dimension

We introduce a general method of resolving first countable, compact spaces that allows accurate estimate of inductive dimensions. We apply this method to construct, inter alia, for each ordinal number α>1 of cardinality ?c, a rigid, first countable, non-metrizable continuum Sα with . Sα is the increment in some compactification of [0,1) and admits a fully closed, ring-like map onto a metric continuum. Moreover, every subcontinuum of Sα is separable. Additionally, Sα can be constructed so as to be: (1) a hereditarily indecomposable Anderson-Choquet continuum with covering dimension a given natural number n, provided α>n, (2) a hereditarily decomposable and chainable weak Cook continuum, (3) a hereditarily decomposable and chainable Cook continuum, provided α is countable, (4) a hereditarily indecomposable Cook continuum with covering dimension one, or (5) a Cook continuum with covering dimension two, provided α>2.We also produce a chainable and hereditarily decomposable space Sω(c⁺) with , , trind₀Sω(c⁺) and trInd₀Sω(c⁺) all equal to ω(c⁺), the first ordinal of cardinality c⁺.     

A space with maximal discrepancy between its Katêtov and covering dimension

It is shown, using a non-measurable partition of the real line, that the covering dimension of a modified Niemytzki space is infinite while its Katêtov dimension is zero.     

Spaces N ∪ R and their dimensions

About spaces N∪$\text{R}$ (see [2, Exercise 5I]), the following are proved: (1) dim $\text{N∪}\text{R}\text{= dim β(N∪}\text{R}\text{)?N∪}\text{R}\text{,(2)if|β(N∪}\text{R}\text{)?N∪}\text{R}\text{|<2}{}^{\mathrm{?}}{}^{\mathrm{o}}$, then no real-valued continuous fu ction on N∪$\text{R}$ is onto (and hence, dim $\text{N∪}\text{R}\text{=0}$), (3) any compact metric space without isolated points is homeomorphic to some $\text{β(N∪}\text{R}\text{)?N∪}\text{R}$ and (4)there are spaces X,X₁ and X₂ of the form N∪$\text{R}$ such that X=X₁∪X₂,X₂andX₂ are zero sets of X, and dim X=n, dimX₁=dimX₂=0, where n=1,2,… or ∞.     

Certain classes of weakly infinite-dimensional spaces and topological games

In [V.V. Fedorchuk, Questions on weakly infinite-dimensional spaces, in: E.M. Pearl (Ed.), Open Problems in Topology II, Elsevier, Amsterdam, 2007, pp. 637-645; V.V. Fedorchuk, Weakly infinite-dimensional spaces, Russian Math. Surveys 42 (2) (2007) 1-52] classes w-m-C of weakly infinite-dimensional spaces, 2?m?∞, were introduced. We prove that all of them coincide with the class wid of all weakly infinite-dimensional spaces in the Alexandroff sense. We show also that transfinite dimensions dimwm, introduced in [V.V. Fedorchuk, Questions on weakly infinite-dimensional spaces, in: E.M. Pearl (Ed.), Open Problems in Topology II, Elsevier, Amsterdam, 2007, pp. 637-645; V.V. Fedorchuk, Weakly infinite-dimensional spaces, Russian Math. Surveys 42 (2) (2007) 1-52], coincide with dimension dimw2=dim, where dim is the transfinite dimension invented by Borst [P. Borst, Classification of weakly infinite-dimensional spaces. I. A transfinite extension of the covering dimension, Fund. Math. 130 (1) (1988) 1-25]. Some topological games which are related to countable-dimensional spaces, to C-spaces, and some other subclasses of weakly infinite-dimensional spaces are discussed.     

More remarks on Souslin properties and tree topologies

We investigate separation properties of ω₁-trees. We show that the property γ of Devlin and Shelah is equivalent to hereditary collectionwise normality. We show that monotone normality and divisibility are both equivalent to orderability. Finally we show that Souslin trees are examples of trees with property γ which are not retractable.     

Multistage n-dimensional universal spaces and extensions

Let Iτ be the Tychonoff cube of weight τ?ω with a fixed point, στ and Στ be the correspondent σ- and Σ-products in Iτ and στ⊂(Σστ=ω(στ))⊂Στ. Then for any n∈{0,1,2,…}, there exists a compactum Unτ⊂Iτ of dimension n such that for any Z⊂Iτ of dimension?n, there exists a topological embedding of Z in Unτ that maps the intersections of Z with στ, Σστ and Στ to the intersections , and of Unτ with στ, Σστ and Στ, respectively; , and are n-dimensional and is σ-compact, is a Lindelöf Σ-space and is a sequentially compact normal Fréchet-Urysohn space. This theorem (on multistage universal spaces of given dimension and weight) implies multistage extension theorems (in particular, theorems on Corson and Eberlein compactifications) for Tychonoff spaces.     

There is no compactification theorem for the small inductive dimension

We give an example of a perfectly normal first countable space X¹ with ind X¹ = 1 such that if Z is a Lindelöf space containing X¹. then ind Z=dim Z=∞. Under CH, there is a perfectly normal, hereditarily separable and first countable such space.