Notes on q-normed Spaces in Constructive Analysis

Douglas S. Bridges · 2026

Read Notes on q-normed Spaces in Constructive Analysis by Douglas S. Bridges (2026) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2606-25236.txt.

Introduction

Constructive mathematics insists that existence claims come with a way to compute the object claimed. Douglas Bridges writes in 2026 about q-normed spaces, a generalisation of normed spaces built for cases where not every element has a usable norm. Instead of one norm, such a space carries a family of seminorms whose values stay bounded for each element. Spaces like the essentially bounded functions fit this pattern even though a full norm may not be constructively available.

Bridges proves constructive versions of several standard theorems from the theory of normed linear spaces, restated for the q-normed setting. Two other natural looking analogues turn out to be essentially nonconstructive, and he gives examples showing why. The results clarify which parts of classical functional analysis survive when every step must be computable.

Catalog record

CREATORDouglas S. Bridges
YEAR2026
COLLECTIONPAPERS
FORMATTXT
FILE SIZE32 KB
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2606.25236

Browse the full library · About the Tower of Babel

Full text

The opening of the work, from the library's hosted text:

Notes on q-normed Spaces in Constructive Analysis Douglas S. Bridges Abstract What we call q-normed (linear) spaces were introduced into constructive analysis, under the name pseudonormed spaces, in [ 10 ] , as a means of handling spaces, such as L ∞ L_{\infty} , in which not all elements are constructively normable. We prove a number of q-normed-space analogues/generalisations of standard theorems in the constructive analysis of normed linear spaces, and give examples showing that two natural analogues are essentially nonconstructive. Introduction Working within the framework of Errett Bishop’s constructive analysis ( BISH ), 1 1 1 See, for example, Bishop’s original work [ 1 ] , the later reference [ 4 ] , or Ishihara’s excellent exposition in Chapters 8 and 9 of [ 7 ] . in this article we present some constructive work on q-normed spaces, 2 2 2 Johns [ 10 ] used the name pseudonormed space; in both [ 2 , Chap. 7, Sect. 5] and [ 3 ] we used quasinormed space . However, pseudonorm and quasinorm are applied classically to notions different from ours, so I have decided to adopt an entirely new, safer nomenclature. a constructive generalisation of normed linear spaces that was introduced by Johns in his PhD. thesis [ 10 ] . For the reader’s benefit we begin by gathering together some fundamental definitions. We define a q-normed space to be a pair \left(X,\left(\left\|\ \right\|_{i}\right)_{i\in I}\right) comprising a real 3 3 3 For simplicity, we consider only real linear spaces here. linear space X equipped with a q-norm —that is, a family (\left\|\ \right\|_{i})_{i\in I} of seminorms such that (*) for each x\in X the set \left\{\left\|x\right\|_{i}:i\in I\right\} is bounded in \mathbf{R} . The inequality and equality on this q-normed space are defined by x\neq y if and only if there exists i\in I such that \left\|x-y\right\|_{i}>0 , and x=y if and only if \left\|x-y\right\|_{i}=0 for all i\in I . Thus a q-normed space is a special type of locally convex space, 4 4 4 For more on locally convex spaces in BISH, see [ 1 , Ch. 9, Sect. 5] or, for a more up-to-date exposition, [ 3 ] . one in which (*) holds. An element x of X is normable with respect to the q-norm (\left\|\ \right\|_{i})_{i\in I} if the norm \left\|x\right\|\equiv\sup\left\{\left\|x\right\|_{i}:i\in I\right\} (1) of X exists. If every element of X is normable, then we can regard X as a normed (linear) space relative to the norm defined at ( 1 ). A simple example of a q-normed space is a normed space X , where we take I=\left\{1\right\} and \left\|\ \right\|_{1} equal to the given norm on X . Our various definitions for properties and elements of q-normed spaces are chosen so that for a normed space they coincide with their counterparts in the standard theory of normed spaces. A more interesting example of a q-normed space is given by the Lebesgue space L_{\infty}\ of essentially bounded subsets of \mathbf{R} with respect to Lebesgue measure, in which example the classically valid statement Every element of L_{\infty} is normable is not constructively provable; for details see [ 2 , pages 346–350] . For each element a\ of the q-normed space \left(X,\left(\left\|\ \right\|_{i}\right)_{i\in I}\right), and each r>0 , the open and closed balls with centre a and radius r are, respectively, \displaystyle B(a,r) \displaystyle\equiv\left\{x\in X:\exists_{r^{\prime}}(0<r^{\prime}<r\wedge\forall_{i\in I}(\left\|x-a\right\|_{i}<r)\right\}, \displaystyle\overline{B}(a,r) \displaystyle\equiv\left\{x\in X:\left\|x-a\right\|_{i}\leq r\text{ for all }i\in I\right\}. The unit ball of X is B_{X}\equiv\overline{B}(0,1) . These definitions of open and closed balls reduce to the standard ones when X is a normed space. The q-norm topology \tau_{qX}\ on X consists of all subsets S of X such that for each x\in S there exists r>0 such that B(x,r)\subset S . The elements of \tau_{qX} are said to be \tau_{qX} -open (or, if the context is clear, just open). The \tau_{qX} -closure of a subset S of X is the set S^{c} consisting of all x\in X such that B(x,r) intersects S for each r>0 . We say that S is \tau_{qX} -closed (or, again if the context is clear, just closed) in X if S=S^{c} . We say that a subset T of X is dense in S if T^{c}=S , and that S is separable if it has a countable sense subset. Let Y be a q-normed space, with q-norm (\left\|\ \right\|_{j})_{j\in J} . A mapping f:S\rightarrow Y , where S\subset X , is • continuous at the point a\in S\ if for each \varepsilon>0 there exists \delta>0 such that if x\in S\ and \left\|f(x)-f(a)\right\|_{j}>\varepsilon for some j\in J , then \left\|x-a\right\|_{i}>\delta for some i\in I ; • uniformly continuous on S if for each \varepsilon>0 there exists \delta>0 such that if x,x^{\prime}\in S and \left\|f(x)-f(x^{\prime})\right\|_{j}>\varepsilon for some j\in J , then \left\|x-x^{\prime}\right\|_{i}>\delta for some i\in I . We then call \delta a continuity associate for \varepsilon and f ; • bi-uniformly continuous if it is injective and uniformly continuous, and f^{-1} is uniformly continuous on f(S) . Proposition 1 Let \left(X,\left(\left\|\ \right\|_{i}\right)_{i\in I}\right) be a q-normed space and Y a normed space. Let a\in S\subset X and let f:S\rightarrow Y be a mapping that is continuous at a with respect to the locally convex structures given by the q-norms. Then f is continuous at a in with respect to the q-norms. Proof. According to [ 3 ] , the continuity assumption for f means that for each each \varepsilon>0 there exist a finitely enumerable subset F\equiv\left\{i_{1},\ldots,i_{N}\right\} of I and \delta_{0}>0 such that if x\in S and \sum_{i\in F}\left\|x-a\right\|_{i}<\delta_{0} , then \left\|f(x)-f(a)\right\|<\varepsilon . Let \delta=\delta_{0}/2N>0 . If x\in S and \left\|f(x)-f(a)\right\|>\varepsilon , then \sum_{i\in F}(\left\|x-a\right\|_{i}-\delta)=\sum_{i\in F}\left\|x-a\right\|_{i}-N\delta>\sum_{i\in F}\left\|x-a\right\|_{i}-\delta_{0}\geq 0, so there exists \iota\in F such that \left\|x-a\right\|_{\iota}-\delta>0 and therefore \left\|x-a\right\|_{\iota}>\delta . Proposition 2 Let \left(X,\left(\left\|\ \right\|_{i}\right)_{i\in I}\right) be a q-normed space and Y a normed space. Let S be a subset of X , and f:S\rightarrow Y a mapping that is uniformly continuous on S with respect to the locally convex structures given by the q-norms.. Then f is uniformly continuous on S\ with respect to the q-norms. Proof. Similar to that of the previous proposition. A sequence \left(x_{n}\right)_{n\geq 1} of elements of the q-normed space X • converges to the (perforce unique) limit x\in X if for each \varepsilon>0 there exists a positive integer N such that \left\|x-x_{n}\right\|_{i}<\varepsilon for all n\geq N and all i\in I ; • is a Cauchy sequence if for each \varepsilon>0 there exists a positive integer N such that \left\|x_{m}-x_{n}\right\|_{i}<\varepsilon for all m,n\geq N and all i\in I . We say that X is complete if every Cauchy sequence in X converges to a limit in X . Let X and Y be q-normed spaces. Leaving the straightforward proofs to the reader, we state some facts about a mapping f of a subset S of X into Y . – If f is continuous at a\in S and \left(x_{n}\right)_{n\geq 1} is a sequence in S converging to a , then \left(f(x_{n})\right)_{n\geq 1} converges to f(a) . – If f is uniformly continuous on S and \left(x_{n}\right)_{n\geq 1} is a Cauchy sequence in X , then \left(f(x_{n})\right)_{n\geq 1} is a Cauchy sequence in f(S) . – If f\ is a bi-uniformly continuous injection of S

Read the complete text