Applied Mathematics, 2015, 6, 1069-1075

Published Online June 2015 in SciRes. http://www.scirp.org/journal/am

http://dx.doi.org/10.4236/am.2015.66097

More Properties of Semi-Linear Uniform

Spaces
S. A. Alhihi
Department of Mathematics, Al-Balqa’ Applied University, Alsalt, Jordan
Email: suad.hihi@bau.edu.jo, suadalhihi@yahoo.com

Received 22 April 2015; accepted 7 June 2015; published 10 June 2015

Copyright © 2015 by author and Scientific Research Publishing Inc.

This work is licensed under the Creative Commons Attribution International License (CC BY).
http://creativecommons.org/licenses/by/4.0/

Abstract
In this paper we shall generalize the definition given in [1] for Lipschitz condition and contrac-
tions for functions on a non-metrizable space, besides we shall give more properties of semi-linear
uniform spaces.

Keywords
Lipschitz Condition, Contractions, Not Metrizable Spaces, Semi-Linear Spaces, Uniform Spaces,
Fixed Point

1. Introduction
The notion of uniformity has been investigated by several mathematician as Weil [2]-[4], Cohen [5] [6], and
Graves [7].
The theory of uniform spaces was given by Burbaki in [8]. Also Wiels in his booklet [4] defined uniformly
continuous mapping. For more information about Uniform spaces one my refer to [9].
In 2009, Tallafha, A. and Khalil, R. [10], defined a new type of uniform spaces, namely semi-linear uniform
spaces and they gave example of semi-linear space which was not metrizable. Also they defined a set valued
map ρ on X × X , by which they studied some cases of best approximation in such spaces. More precisely,
they gave the following.
Let ( X , Γ ) be a semi-linear uniform space; E ⊆ X is proximinal if for any x ∈ X , there exists e ∈ E
such that ρ ( x, E ) = ρ ( x, e ) . They asked that “must every compact is proximinal”, they gave the answer for the
cases—i) E is finiate; ii) If xn converges to x, then { x, x1 , x2 ,} is proximinal.
In [11], Tallafha, A. defined another set valued map δ on X × X , and gave some properties of semi-linear
uniform spaces using the maps ρ and δ . Also in [1] [12], Tallafha defined Lipschitz condition and con-
tractions for functions on semi-linear uniform spaces, which enabled us to study fixed point for such functions.

How to cite this paper: Alhihi, S.A. (2015) More Properties of Semi-Linear Uniform Spaces. Applied Mathematics, 6, 1069-
1075. http://dx.doi.org/10.4236/am.2015.66097
S. A. Alhihi

Lipschitz condition, and contractions are usually discussed in metric and normed spaces and never been studied
in other weaker spaces. We believe that the structure of semi-linear uniform spaces is very rich, and all the
known results on fixed point theory can be generalized.
The object of this paper is to generalize the definition of Lipschitz condition, and contraction mapping on
semi-linear uniform spaces given by Tallafha [12]. Also we shall give a new topopological properties and more
properties of semi-linear uniform spaces.

2. Semi-Linear Uniform Space

Let X be a set and DX be a collection of subsets of X × X , such that each element V of DX contains the
=
diagonal ∆ {( x, x ) : x ∈ X } , and = −1
V V= {( y, x ) : ( x, y ) ∈ V } for all V ∈ DX (symmetric). DX is called
the family of all entourages of the diagonal.
Definition 1 [10]. Let Γ be a sub collection of DX , the pair ( X , Γ ) is called a semi-linear uniform space
if,
i) Γ is a chain.
ii) For every V ∈ Γ , there exists U ∈ Γ such that U  U ⊆ V .
iii)  V = ∆ .
V ∈Γ

iv)  V= X×X .
V ∈Γ
Definition 2 [10]. Let ( X , Γ ) be a semi-linear uniform space, for ( x, y ) ∈ X × X , let
) {V ∈ Γ : ( x, y ) ∈ V } . Then, the set valued map ρ on X × X is defined by ρ ( x, y ) =
Γ( x , y=  V.
V ∈Γ( x , y )

Clearly for all ( x, y ) ∈ X × X , we have ρ ( x, y ) = ρ ( y, x ) and ∆ ⊆ ρ ( x, y ) . Let

Γ \ Γ( x , y ) = ( Γ( x , y ) ) = {V ∈ Γ : ( x, y ) ∉ V } , from now on, we shall denote Γ \ Γ( x , y ) by Γ( x , y ) .
c c

Definition 3 [11]. Let ( X , Γ ) be a semi-linear uniform space. Then, the set valued map δ on X × X is
  V if x ≠ y,
 c
defined by, δ ( x, y ) = V ∈Γ( x , y )
φ
 if x = y.
The following results are given in [12].
Proposition 1. Let ( X , Γ ) be a semi-linear uniform space, and Λ is a sub collection of Γ , then V ≠ ∆ ,
V ∈Λ

if an only if there exist U ∈ Γ such that U  V .

V ∈Λ

Corollary 1. Let ( X , Γ ) be a semi-linear uniform space. If ρ ( x, y ) ≠ ∆ , then,

1) There exist U ∈ Γ such that U  ρ ( x, y ) .
2) U ⊆ δ ( x, y ) .
Let V ∈ DX the family of all entourages of the diagonal, then for all n ∈  , by nV, we mean V  V  V
n-times and 0V = ∆ , so for all V ∈ DX , V ⊆ nV and nV ∈ DX for all n ∈  .
Proposition 2. Let ( X , Γ ) be a semi-linear uniform space. If x ≠ y , then nδ ( x, y ) =  nV .
V ∈Γc
( x,y)

Proposition 3. Let ( X , Γ ) be a semi-linear uniform space. If Λ is a sub collection of Γ , then

 nV = n  V .
V ∈Λ V ∈Λ

Question. Does  nV ⊆ n ρ ( x, y ) ?
V ∈Γ( x , y )

3. Topological Properties of Semi-Linear Uniform Spaces

Definition 4 [13]. For x ∈ X and V ∈ Γ . The open ball of center x and radius V is defined by
B ( x, V ) { y : ( x, y ) ∈ V } , equivalently=
= B ( x, V ) { y : ρ ( x, y ) ⊆ V } .

1070
 S. A. Alhihi

Clearly if y ∈ B ( x, V ) , then there is a W ∈ Γ such that B ( y, W ) ⊆ B ( x, V ) . So { B ( x, V ) , x ∈ X , V ∈ Γ}

is a base for some topology on X. This topology is denoted by τ Γ .
More presicly τ= Γ {G ⊆ X : for all x ∈ G, there exist V ∈ Γ,such that B ( x,V ) ⊆ G} .
In [10] it is shown that τ Γ is Hausdorf, so if X is finite then we have the discreet topology, therefore interest-
ing examples are when X is infinite. Also, if X is infinite then Γ should be infinite, other wise ∆ ∈ Γ , which
implies also that the topology is the discrete topology.
Proposition 4. Let ( X , Γ ) be a semi-linear uniform space and Γ′ a subcollection of Γ satisfies,
i) For every U ∈ Γ , there is a V ∈ Γ′ such that V ⊆ U .
ii)  V= X × X .
V ∈Γ ′

iii) V = ∆.
V ∈Γ ′
Then ( X , Γ′ ) is a semi-linear uniform space and τ Γ = τ Γ′ .
Proof. Since Γ′ is a subcollection of Γ , then Γ′ is a subcollection of DX and i), iii), iv) in definition
(1.1) are satisfied. Now for U ′, V ′ ∈ Γ′, ther exist U ∈ Γ such that U ⊆ (U ′  V ′ ) , so there is a W ∈ Γ′
such that W ⊆ U ⊆ (U ′  V ′ ) . So ( X , Γ′ ) is a semi-linear uniform space. Now τ Γ′ ⊆ τ Γ is clear. Let O be a
nonempty open set in τ Γ , then if x ∈ O ther exist U ∈ Γ such that B ( x, U ) ⊆ O. Let V ∈ Γ′ such that
V ⊆ U , so B ( x, V ) ⊆ O, hence τ Γ ⊆ τ Γ′ .
Theorem 1. Let ( X , Γ ) be a semi-linear uniform space and τ Γ the topology on X indused by Γ , then.
i) Γ can be consider as asubset of τ Γ × τ Γ .
ii) Fore all x, y ∈ X , δ ( x, y ) ∈ τ Γ × τ Γ .
=
Proof. i). Let { }
Γ′ U  : U ∈ Γ , where U  is the interior of U with respect to the topology τ Γ × τ Γ on
X × X . By Proposition 2.2, ( X , Γ′ ) is a semi-linear uniform space indusing the topology τ Γ on X × X .
Sine we deal with the toplogy τ Γ , we may replace Γ with Γ′ , so for all V ∈ Γ′, V = U  for some U ∈ Γ,
hence V ∈ τ Γ × τ Γ .
ii) Is clear by definition of δ ( x, y ) .
In [11], Tallafha gave some important properties of semi-linear uniform spaces, using the set valued map ρ
and δ .
Now we shall give more properties of semi-linear uniform spaces.

4. More Properties of Semi-Linear Uniform Spaces

Let ( X , Γ ) be a semi-linear uniform space, then Γ is a chain so there exist a well order set ( R, ) , such
that= Γ {U r ,r ∈ R} . For x, y ∈ X , x ≠ y , then, ther exist U ∈ Γ , such that ( x, y ) ∈ U . That is, there exist
r0 ∈ R such that ( x, y ) ∈ U r0 . This implies, if ( x, y ) ∉ U r then U r ⊆ U r0 . Hence S ={r ∈ R : ( x , y ) ∉ U r }
is bounded above by r0 . By Zorn’s Lemma S has a maximal element r( x , y ) . So δ ( x, y ) ⊆ U r( x , y ) ⊆ ρ ( x, y ) .
This copleet the proof of the following lemma.
Lemma 1. Let ( X , Γ ) be a semi-linear uniform space, then= Γ {U r ,r ∈ R} , where R is a well order
set, then for x, y ∈ X , x ≠ y , ther exist r( x , y ) ∈ R, such that δ ( x, y ) ⊆ U r( x , y ) ⊆ ρ ( x, y ) .
Remember that for all V ∈ DX the family of all entourages of the diagonal, V satisfies the following nice
properties, V ⊆ nV and nV ∈ DX for all n ∈  . Let Λ , δ and ρ be a subcollections of DX , defined by
=δ {δ ( x, y ) : x, y ∈ X , x ≠ =
y} , ρ { ρ ( x, y ) : x, y ∈ X , x ≠ y} and
= Λ δ  ρ  Γ. For all A ∈ Λ , by Co-
1
rollary 1.6, there exist U ∈ Γ such that U ⊆ A . So we can define A , for an element A ∈ Λ by.
n
1
Definition 5. For n ∈  and A ∈ Λ . Define A by,
n
1
= A  {U : nU ⊆ A}.
n U ∈Γ

1 1 1
Clearly ∆ = ∆ and A ∈ DX for all A ∈ Λ . But A need not be an element in Γ , evin if A ∈ Γ .
n n n

1071
S. A. Alhihi

But we have.
Lemma 2. Let ( X , Γ)
be a semi-linear uniform space,= Γ {U r ,r ∈ R} , R is a well order set, and A ∈ Λ .
1
Then there exist rA ∈ R , such that A ⊆ U rA and if r ∈ R satisfied rA  r , then nU r \ A ≠ φ .
n
Proof. For A ∈ Λ , there exist r0 ∈ R such that A ⊆ U r0 , so for every U ∈ Γ satisfies nU ⊆ A , we have
U ⊆ U r0 . Hence T = {r ∈ R : nU r ⊆ A} is bounded above by r0 . Hence rA be the maximal element of T.
Theorem 2. Let A ∈ Λ , and σ a subcolection of Λ . For n ∈  , we have
1 
i) n  A  ⊆ A .
n 
1
ii) If B ∈ Γ satisfies nB ⊆ A , then B ⊆ A.
n
1 1
iii) A ⊆ A.
n +1 n
1
iv) A⊆ A.
n
1 1
v)  A=
n A∈σ
 n A.
A∈σ

1 1
vi)  n A⊆ n  A.
A∈σ A∈σ

( s0 , sn ) ∈ n 
1  1 
Proof. i). Let A  , then there exist s1 , , sn −1 such that ( s j , s j +1 ) ∈  =
A  , j 0,1, , n − 1.
n  n 
( )
So there exist U1 , U 2 , , U n in Γ such that s j −1 , s j ∈ U j and nU j ⊆ A, j = 1, , n. Since Γ is a chain,
there exist j0 ∈ {1, , n} such that U j ⊆ .U j0 for all j = 1, , n. So ( s0 , sn ) ∈ nU j0 ⊆ A.
1
ii) and iv), are clear by definition of A.
n

A  {U : ( n + 1)U ⊆ A} ⊆  {U : nU =
1 1 1 1
iii) = ⊆ A} V . v)  A ⊆ A , for all A ∈ σ , so
n +1 U ∈Γ U ∈Γ n n A∈σ n
1 1 1
 A ⊆  n A . Conversly, Let ( s0 , sn ) ∈  n A, then, for all A ∈ σ , there exist s1 , , sn −1 such that
n A∈σ A∈σ A∈σ

( )
( s j , s j +1 ) ∈ n A, j= 0,1,, n − 1. So there exist U1 ,U 2 , ,U n in Γ such that s j −1 , s j ∈ U j and nU j ⊆ A,
1

j = 1, , n. Since Γ is a chain, there exist j0 ∈ {1, , n} such that U j ⊆ .U j0 for all j = 1, , n So
( s0 , sn ) ∈ nU j0 ⊆ A.
Let n ∈  and Aα ∈ DX , α ∈ Λ, replacing A by  Aα , or by  Aα , we have.
α ∈Λ α ∈Λ
Corollary 2. For x, y ∈ X where ( X , Γ ) is a semi-linear uniform spaces, we have,
 1
 c n
V if x ≠ y
1
i) δ ( x, y ) = V ∈Γ( x , y ) .
n 
φ if x = y
1 1 1
ii) =ρ ( x, y )  V ⊆  nV .
n n V ∈Γ( x , y ) V ∈Γ( x , y )

Corollary 3. Let n ∈  and Aα ∈ DX , α ∈ Λ, then

1 
i) n   Aα  ⊆  Aα .
 α ∈Λ  α ∈Λ
n

1072
 S. A. Alhihi

ii) If B ∈ V satisfies nB ⊆ (  { Aα : α ∈ Λ} ) , then B ⊆

1
n
 { Aα : α ∈ Λ} .
1 1
iii) 
n + 1 α ∈Λ
Aα ⊆  Aα .
n α ∈Λ
1 
iv) n   Aα  ⊆  Aα .
 n α ∈Λ  α ∈Λ
1
v) If V ∈ DX satisfies nB ⊆  Aα , then B ⊆  Aα .
n α ∈Λ
α ∈Λ

1   1 
vi)   Aα  ⊆   Aα  .
n + 1  α ∈Λ  n  α ∈Λ 
Also we have the follwing corollary.
1 
Corollary 4. For x, y ∈ X where is a semi-linear uniform spaces, then n  ρ ( x, y )  ⊆ ρ ( x, y ) .
( X , Γ)
n 
1 1 1 
Corollary 5. For n ∈  , V ∈ Γ, if V ∈ Γ, then V is the largest element in Γ satisfies n  V  ⊆ V .
n n n 
Also by Definition 3.6, for x ⋅ y ∈ X where ( X , Γ ) is a semi-linear uniform spaces, we have,
1 1 
i) = ρ ( x, y )   V : V ∈ Γ ( x , y )  .
n n 
 1 
  V : V ∈ Γ( x , y )  if x ≠ y,
c
1
ii) δ ( x, y ) =   n 
n φ
 if x = y.
Proposition 5. Let x ⋅ y be any distinct points in semi-linear uniform spaces ( X , Γ ) . Then,
1 
n  δ ( x, y )  ⊆ δ ( x, y ) ⊆ ( nδ ( x, y ) ) .
1
n  n
 1  1 
Proof. Let ( s0 , sn ) ∈ n  δ ( x, y )  , then ( s0 , sn ) ∈ n  V : V ∈ Γ(cx , y )  , then there exist s1 , , sn −1 such
n  n 
1 
that ( s j , s j +1 ) ∈   V : V ∈ Γ(cx , =
y) 
, j 0, , n − 1. So there exist U1 , U 2 , , U n ∈ Γ(cx , y ) such that
 n 

( s j −1 , s j ) ∈ n U j , but Γ is a chain implies the existence of j0 ∈ {1,, n} such that U j ⊆ .U j0 for all
1

1 
j = 1, , n. So ( s0 , sn ) ∈ n  U j0  ⊆ U j0 ⊆ .δ ( x, y ) . On the other hand by proposition 1.7,
n 
nδ ( x, y ) =  nV . So for all V ∈ Γ(cx , y ) , nV ⊆ nδ ( x, y ) , which implies V ⊆ ( nδ ( x, y ) ) .
1
V ∈Γc n
( x,y)

Definition 6. For n ∈  , A, Aα ⊆ X × X , α ∈ Λ. and r ∈ + , define

1  n
i) rA = n  A= , r , n, m ∈  and the greatest common devisor of n, m is 1.
m  m
ii) r  { Aα=
: α ∈ Λ}  {rAα : α ∈ Λ} .
iii) r  { Aα=
: α ∈ Λ}  {rAα : α ∈ Λ} .
Definition 7. Let x ⋅ y be any points in semi-linear uniform spaces ( X , Γ ) . For r ∈ + ,
1  n
rδ ( x, y ) = n  δ ( x, y )  , where
= r , n, m ∈  and the greatest common devisor of n, m is 1.
m  m
Using the a bove definition and Proposition 1.7, we have.
Proposition 6. Let V ∈ Γ. and r1 , r2 ∈ + , r1 < r2 , then rV
1 ⊆ r2V .

1073
S. A. Alhihi

Proposition 7. Let x ⋅ y be any points in semi-linear uniform spaces ( X , Γ ) . If r1 , r2 ∈ + , r1 ≤ r2 , then

r1δ ( x, y ) ⊆ r2δ ( x, y ) .
Definition 8. Let rn be an increasing sequence of positive rationals. If rn → r ∈ + , then
∞
lim rnδ ( x, y ) is defined by
n →∞
 rnδ ( x, y ).
n =1

Proposition 8. Let x, y ∈ ( X , Γ ) and rn an increasin sequence of positive rationals. If rn → r ∈ + , then

lim rnδ ( x, y ) ⊆ rδ ( x, y ) .
n →∞

Proof. It is an immediate consequence of Proposition (7) and Definition (8).

5. Contractions
In [10] the definitions of converges and Cauchy are given. Now we shall discuss some topological properties of
semi-linear uniform spaces. Since the semi-linear uniform space is a topological space then the continuity of a
function is as in topology. The concept of uniform continuity is given by Wiels [4], so we have:
Definition 9 [4]. Let f : ( X , Γ X ) → (Y , ΓY ) , then f is uniformly continuous if ∀U ∈ ΓY , ∃V ∈ Γ X , such
that if ( x, y ) ∈ V , then ( f ( x ) , f ( y ) ) ∈ U .
Clearly using our notation we have:
Proposition 9. Let f : ( X , Γ X ) → (Y , ΓY ) . Then f is uniformly continuous, if and only if ∀U ∈ ΓY ,
∃V ∈ Γ X such that, for all x, y ∈ X , if ρ X ( x, y ) ⊆ V , then ρY ( f ( x ) , f ( y ) ) ⊆ U .
The following proposition shows that we may replace ρ by δ in Proposition 2.2.
Proposition 10 [12]. Let f : ( X , Γ X ) → (Y , ΓY ) . Then f is uniformly continuous, if and only if ∀U ∈ ΓY ,
∃V ∈ Γ X , such that for all x, y ∈ X , if δ X ( x, y ) ⊆ V , then δ Y ( f ( x ) , f ( y ) ) ⊆ U .
In [10], Tallafha gave an example of a space which was the semi-linear uniform space, but not metrizable. Till
now, to define a function f that satisfies Lipschitz condition, or to be a contraction, it should be defined on a
metric space to another metric space. The main idea of this paper is to define such concepts without metric
spaces, and we just need a semi-linear uniform space, which is weaker as we mentioned before.
Definition 10 [1]. Let f : ( X , Γ ) → ( X , Γ ) , then f satisfied Lipschitz condition if there exist m, n ∈  such
that mδ ( f ( x ) , f ( y ) ) ⊆ nδ ( x, y ) . Moreover if m > n , then we call f a contraction.
Now we shall give a new definition of Lipschitz condition and contraction called r-Lipschitz condition and
r-contraction.
Definition 11. Let f : ( X , Γ ) → ( X , Γ ) , then f satisfied r-Lipschitz condition if there exists r ∈ + such
that δ ( f ( x ) , f ( y ) ) ⊆ rδ ( x, y ) . Moreover if r < 1 then we call f a r-contraction.
Question. Let ( X , Γ ) be semi-linear uniform spaces and f : ( X , Γ ) → ( X , Γ ) , what is the relation be-
tween Lipschitz condition and r-Lipschitz condition, contraction, and r-contraction.
Remark 1 [12]. Let ( X , Γ ) be semi-linear uniform spaces, then if ∆ ∈ Γ, then the topology indused by
( , Γ ) is the descrete topology which is metrizable. Therefore we can asumme that ∆ ∉ Γ.
X
Question [1] [12]. Let ( X , Γ ) be a complete semi-linear uniform space. And f : ( X , Γ ) → ( X , Γ ) be a
contraction. Does f has a unique fixed point?
Question. Let ( X , Γ ) be a complete semi-linear uniform space. And f : ( X , Γ ) → ( X , Γ ) be r-contraction.
Does f has a unique fixed point?

References
[1] Tallafha, A. (2013) Open Problems in Semi-Linear Uniform Spaces. Journal of Applied Functional Analysis, 8, 223-
228.
[2] Weil, A. (1936) Les recouvrements des espaces topologiques: Espaces complete, espaces bicompact. Comptes Rendus
de l’Académie des Sciences, 202, 1002-1005.
[3] Weil, A. (1937) Sur les espaces a structure uniforme et sur la topologie generale. Actuarial Science India, 551, Paris.
[4] Weil, A. (1938) Sur les espaces à structure uniforme et sur la topologic générale. Hermann, Paris.
[5] Cohen, L.W. (1937) Uniformity Properties in a Topological Space Satisfying the First Denumerability Postulate. Duke
Mathematical Journal, 3, 610-615.

1074
 S. A. Alhihi

[6] Cohen, L.W. (1939) On Imbedding a Space in a Complete Space. Duke Mathematical Journal, 5, 174-183.
[7] Graves, L.M. (1937) On the Completing of a Housdroff Space. Annals of Mathematics, 38, 61-64.
[8] Bourbaki, N. (1940) Topologie Générale (General Topology). Paris.
[9] James, I.M. (1987) Topological and Uniform Spaces. Undergraduate Texts in Mathematics. Springer-Verlag, Berlin.
[10] Tallafha, A. and Khalil, R. (2009) Best Approximation in Uniformity Type Spaces. European Journal of Pure and Ap-
plied Mathematics, 2, 231-238.
[11] Tallafha, A. (2011) Some Properties of Semi-Linear Uniform Spaces. Boletin da Sociedade Paranaense de Matematica,
29, 9-14.
[12] Tallafha, A. (2015) Fixed Point in a Non-Metrizable Spaces. To Appear.
[13] Engelking, R. (1968) Outline of General Topology. North-Holand, Amsterdam.

1075