Notes on Topology

David K. Zhang
Last Modified 2023-11-20

Definition: Pseudometric, Symmetry, Triangle Inequality

Let XX be a set, and let FF be an ordered field. An FF-valued pseudometric on XX is a function d:X×XFd: X \times X \to F that satisfies the following requirements:

  • d(x,x)=0Fd(x, x) = 0_F for all xXx \in X.
  • Symmetry: d(x,y)=d(y,x)d(x, y) = d(y, x) for all x,yXx, y \in X.
  • Triangle inequality: d(x,z)d(x,y)+d(y,z)d(x, z) \le d(x, y) + d(y, z) for all x,y,zXx, y, z \in X.

When the ordered field FF is not explicitly specified, it is usually assumed to be the ordered field R\R of real numbers.

Pseudometrics are Nonnegative

Theorem: Let XX be a set, let FF be an ordered field, and let d:X×XFd: X \times X \to F be an FF-valued pseudometric on XX. For all x,yXx, y \in X, we have d(x,y)0Fd(x, y) \ge 0_F.


Proof: Let x,yXx, y \in X be given. By the defining properties of a pseudometric, we have

0F=d(x,x)d(x,y)+d(y,x)=2Fd(x,y) 0_F = d(x, x) \le d(x, y) + d(y, x) = 2_F d(x, y)

which implies that d(x,y)0Fd(x, y) \ge 0_F. (Note that 2F0F2_F \ne 0_F in all ordered fields FF.)

Definition: Metric, Distance Function, Positivity

Let XX be a set, and let FF be an ordered field. An FF-valued metric on XX, also known as an FF-valued distance function on XX, is an FF-valued pseudometric d:X×XFd: X \times X \to F that satisfies the following additional requirement:

  • Positivity: For all x,yXx, y \in X, if d(x,y)=0Fd(x, y) = 0_F, then x=yx = y.

In other words, the only difference between a pseudometric and a metric is that a pseudometric is allowed to assign an distance of zero to two distinct points, while a metric must assign positive distances between distinct points.

Definition: Generalized Pseudometric Space, Underlying Set, Distance Field, Point

A generalized pseudometric space is an ordered triple (X,F,d)(X, F, d) consisting of a set XX, called the underlying set, an ordered field FF, called the distance field, and an FF-valued pseudometric dd on XX. The elements of XX are called points.

Definition: Open Ball, Br(x)B_r(x), Closed Ball, Br[x]B_r[x]

Let (X,F,d)(X, F, d) be a generalized pseudometric space, xXx \in X, and rFr \in F. The open ball of radius rr centered at xx, denoted by Br(x)B_r(x), is the subset of XX defined by

Br(x){yX:d(x,y)<r}. B_r(x) \coloneqq \{ y \in X : d(x, y) < r \}.

Similarly, the closed ball of radius rr centered at xx, denoted by Br[x]B_r[x], is the subset of XX defined by

Br[x]{yX:d(x,y)r}. B_r[x] \coloneqq \{ y \in X : d(x, y) \le r \}.

Definition: Continuous at a Point, Everywhere Continuous, Continuous

Let (X,FX,dX)(X, F_X, d_X) and (Y,FY,dY)(Y, F_Y, d_Y) be generalized pseudometric spaces. A function f:XYf: X \to Y is continuous at a point x0Xx_0 \in X if, for every ε>0FY\varepsilon > 0_{F_Y}, there exists δ>0FX\delta > 0_{F_X} such that f[Bδ(x0)]Bε(f(x0))f[B_\delta(x_0)] \subseteq B_\varepsilon(f(x_0)).

We say that ff is everywhere continuous, or simply continuous, if ff is continuous at every point x0Xx_0 \in X.

Definition: Interior Point, Interior, AA^\circ, Open Set

Let (X,F,d)(X, F, d) be a generalized pseudometric space, and let AXA \subseteq X. A point xXx \in X is an interior point of AA if there exists r>0Fr > 0_F such that Br(x)AB_r(x) \subseteq A. The set of all interior points of AA, denoted by AA^\circ, is called the interior of AA.

A{xX:r>0F:Br(x)A} A^\circ \coloneqq \{ x \in X : \exists r > 0_F : B_r(x) \subseteq A \}

We say that AA is open if every point in AA is an interior point of AA (i.e., AAA \subseteq A^\circ).

Definition: Close, Touches, Closure, AA^\circ, Closed Set

Let (X,F,d)(X, F, d) be a generalized pseudometric space, and let AXA \subseteq X. A point xXx \in X is close to AA, or touches AA, if for every r>0Fr > 0_F, there exists aAa \in A such that d(x,a)<rd(x, a) < r. The set of all points close to AA, denoted by A\overline{A}, is called the closure of AA.

A{xX:r>0F,aA:d(x,a)<r} \overline{A} \coloneqq \{ x \in X : \forall r > 0_F,\, \exists a \in A : d(x, a) < r \}

We say that AA is closed if AA contains every point that is close to AA (i.e., AA\overline{A} \subseteq A).


Definition: Topology

Let XX be a set. A topology on XX is a collection TP(X)T \subseteq \powerset{X} of subsets of XX that satisfies the following requirements:

  • Empty set is open: T\varnothing \in T.
  • Whole set is open: XTX \in T.
  • Closed under arbitrary unions: If STS \subseteq T, then ST\bigcup S \in T.
  • Closed under finite intersections: If STS \subseteq T and S\abs{S} is finite, then ST\bigcap S \in T.

Definition: Topological Space, Point, Open Set

A topological space is an ordered pair (X,T)(X, T) consisting of a set XX, called the underlying set, and a topology TT on XX. The elements of XX are called the points of the topological space (X,T)(X, T), and the elements of TT are called its open sets.

Definition: Closed Set

Let (X,T)(X, T) be a topological space. A subset FXF \subseteq X is closed if XFTX \setminus F \in T.

In other words, a set is closed if and only if its complement is open. Because taking complements is an involution (i.e., X(XA)=AX \setminus (X \setminus A) = A for all AXA \subseteq X), the converse is also true: a set is open if and only if its complement is closed.

Note that, in topology, “open” and “closed” are not opposites! It is possible for a set to be both open and closed, and it is also possible for a set to be neither. In fact, there is a special name for sets that are both open and closed.

Definition: Clopen Set

Let (X,T)(X, T) be a topological space. A subset of XX is clopen if it is both open and closed.

Every topological space (X,T)(X, T) has at least two clopen sets: the empty set \varnothing and the set XX itself.