Coordinate Systems and Linear Combinations: Bases, Spans, and Independence

Short read

This video explains how we describe vectors numerically by using a basis of two special vectors I hat and J hat in the plane. It demonstrates how each coordinate acts as a scalar that scales a basis vector, so a vector is the sum of two scaled arrows. The talk introduces the span, describes what happens when the basis vectors align or when a third vector adds a new direction, and distinguishes between thinking of vectors as arrows versus thinking of collections of vectors as points. It also previews the idea of a basis as a minimal, spanning, and independent set and teases the transition to matrices in future content.

• Vectors as linear combinations of basis vectors
• Span and dimensional reach in 2D and 3D
• Linearity, independence, and the basis concept
• Basis changes and future space transformations

Coordinate systems and the role of basis vectors

The video opens by revisiting how a pair of numbers, such as 3 and -2, can describe a 2D vector when we view each coordinate as a scalar that stretches or compresses a corresponding basis vector. In the familiar XY plane, the unit vectors I hat and J hat point to the right and up, each with length 1. The X coordinate scales I hat, and the Y coordinate scales J hat, so the vector is the sum of two scaled directions. This frames coordinates as a linear combination of basis vectors, which is a central idea in linear algebra. The speaker stresses that the basis vectors are not unique; choosing different pair of vectors yields another valid coordinate system, though it changes how coordinates correspond to vectors.

The video then introduces the term span: the set of all vectors you can reach by taking linear combinations of a given set of vectors. In two dimensions, most pairs of vectors span the entire plane, while two vectors that lie on the same line span only that line, and the trivial case where both are zero lands you at the origin. This is illustrated by thinking of vectors as arrows when considered individually and as points when considered as a collection, with the tip of the vector representing the point.

Beyond two dimensions, the intuition extends to three dimensions. The span of two nonparallel vectors is a flat plane through the origin, and adding a third vector typically expands the span to fill all of 3D space. If the third vector lies in the span of the first two, nothing new is added and the span remains the same plane. If it does extend in a new direction, the span becomes richer, mirroring the increase in dimensionality that the scalars can access. The speaker calls this interplay the power of linear combinations and uses it to motivate what it means for vectors to be independent or dependent.

From span to basis and the idea of independence

The talk introduces terminology that will be foundational in linear algebra. A set of vectors is linearly independent if none of them can be written as a linear combination of the others; equivalently, each vector genuinely adds a new direction to the span. If one vector is expressible in terms of the others, the vectors are linearly dependent and one is redundant for spanning the space. The basis of a space is defined technically as a set of vectors that is both linearly independent and spanning the space. In 2D, the standard basis I hat and J hat is a concrete example that makes the abstract idea tangible: any 2D vector can be reached by choosing appropriate scalars to scale I hat and J hat and then summing. The video teases that a complete treatment of bases and spans will be developed further, including the precise relationship between coordinate systems and different bases.

Visualizing span in higher dimensions and the move toward matrices

With three dimensions, the same logic applies: two vectors generally define a plane, while three non-coplanar vectors can span all of space. When a new vector is not in the existing span, it unlocks access to more dimensions of space by providing an additional degree of freedom for the linear combination. Conversely, if a new vector is already in the span of the existing set, it does not expand what can be reached. This leads to the concept of linear dependence, where one vector is redundant because it lies in the span of others. The speaker invites the viewer to ponder the exact notion of a basis and to anticipate how matrices will come into play in transforming space and coordinates in future lessons.

Takeaway and look ahead

The core message is that describing vectors numerically depends on a chosen basis, and a basis is a special set of vectors that both spans the target space and is linearly independent. The span captures all vectors you can obtain through two fundamental operations, vector addition and scalar multiplication. The video closes by signaling that the next discussion will move from geometric intuition to matrices and linear transformations, showing how these ideas underpin much of linear algebra and its applications.