Geometry · Eqora guide
Dot product in math: does zero mean a right angle?
Pair the coordinates, preserve their signs and decide what a zero result proves before calculating an angle.
Two arrows can point in different directions even when neither has length zero. Their dot product can nevertheless be zero. The dot product in math combines two vectors into one number, connecting coordinate arithmetic with geometry. The delicate point is deciding whether a conclusion about a right angle is justified. A zero vector gives the same numerical result without having any direction at all.
We will work in real two-dimensional or three-dimensional Euclidean space, using perpendicular coordinate axes with equal unit lengths. These assumptions make the familiar coordinate formula valid. You will compute an example, interpret positive and negative results, recover an angle, and separate a projection length from a projection vector. Paper is enough for the exact calculations; inverse cosine is needed only for a final decimal angle.
Eqora publishes this independent learning guide, not a guarantee of correct answers or grades. Check the task, notation, assumptions and result, and follow your course rules. Eqora is a learning aid, not a substitute for exam work. The sail-loft photographs suggest directions and comparisons; they are not scaled vector diagrams. Do not infer numerical coordinates, angles or lengths from the wooden objects in a photograph.
The dot product in math pairs matching coordinates
For u = (u₁, u₂) and v = (v₁, v₂), calculate u·v = u₁v₁ + u₂v₂. In three dimensions, add u₃v₃. Multiply first coordinates together, second coordinates together, and so on; then add the products. The result is a scalar, meaning a number rather than another arrow. Multiplying a vector by a scalar is a different operation that produces a vector. Distinguish these operations before deciding how many coordinates your answer should have.
Let u = (2, −1, 2) and v = (1, 4, 1). Their product is 2·1 + (−1)·4 + 2·1 = 2 − 4 + 2 = 0. All three terms matter: stopping after two would give −2 and the wrong conclusion. Keep negative coordinates in parentheses until each product is formed. Georgia Tech’s introduction to dot products connects this scalar construction to lengths and orthogonality. Our worked coordinates are independent examples, not measurements taken from the photographs.
Check the coordinate system before applying the formula
The sum of coordinate products represents the usual geometric dot product in an orthonormal basis: its basis directions are perpendicular and each basis vector has length one. Unequal drawing scales can make a graph’s angles look misleading without changing its stated mathematical coordinates. A genuinely oblique basis is different. Its coordinate lists require the appropriate basis inner products; blindly multiplying matching entries need not yield the Euclidean geometric answer. State the coordinate assumption instead of judging it from a sketch.
Points also need translating into vectors. For A = (1, 2), B = (4, 3) and C = (0, 5), the angle at A uses AB = B − A = (3, 1) and AC = C − A = (−1, 3). Their product is −3 + 3 = 0. Using position vectors B and C would test a different angle at the origin. Our guide to reading geometry diagrams helps distinguish given facts from appearances before choosing vectors and operations.
Zero establishes orthogonality, with a special case
For two nonzero vectors, u·v = ‖u‖‖v‖ cos θ, where θ is the angle between their directions from 0° to 180°. Both lengths are positive, so a zero product forces cos θ = 0 and therefore θ = 90°. Conversely, a right angle gives zero cosine and hence a zero product. Our earlier three-dimensional vectors are both nonzero: their zero result establishes a right angle without evaluating inverse cosine or drawing a three-dimensional perspective view.
Now take z = (0, 0, 0). It satisfies z·v = 0 for every v, but has no direction. The angle expression would divide by zero. Algebraically the zero vector is defined to be orthogonal to every vector; geometrically you must not announce a measurable right angle involving it. Record the nonzero assumption when concluding that θ is 90°. A zero dot product also does not require either vector to vanish, unlike the zero-product rule for ordinary real-number multiplication.
Interpret the sign before using inverse cosine
For nonzero vectors, the product has the sign of cos θ. A positive result means θ is less than 90°, a negative result means it exceeds 90°, and zero means it equals 90°. A negative dot product is valid; it is not a negative length. Magnitudes also affect its size: doubling one vector doubles the product without changing its direction when the multiplier is positive. The raw product is therefore not a cosine and cannot be entered directly into arccos.
For p = (1, 1) and q = (2, 0), p·q = 2 while their lengths are √2 and 2. Dividing gives cos θ = 1/√2, so θ = 45°. Replace q by −q = (−2, 0): the product becomes −2 and the angle becomes 135°. For a decimal angle, apply arccos to the normalized quotient and check whether the calculator returns degrees or radians. Preserve exact radicals until this last step instead of rounding the lengths prematurely.
Check the norms and the allowed cosine range
The dot product of a real vector with itself is its squared length: u·u = ‖u‖². For u = (2, −1, 2), this gives 4 + 1 + 4 = 9, so ‖u‖ = 3, not 9. For v = (1, 4, 1), the squared length is 18 and the length is √18. Norms are nonnegative, and only the zero vector has zero norm. A negative square root does not represent the vector’s length, even if some coordinates are negative.
The normalized product must lie between −1 and 1. This is the Cauchy–Schwarz bound |u·v| ≤ ‖u‖‖v‖. A quotient such as 1.2 cannot be passed to real arccos: recheck matching components, squares and denominator. Tiny floating-point overshoots can occur, but do not silently clip a substantially wrong quotient. Parallel nonzero vectors normalize to 1 or −1. A raw product of 1 alone proves neither parallelism nor a particular angle, because the two lengths still matter.
Separate projection length from projection vector
Project a = (4, 2) onto the nonzero vector b = (2, 0). The signed scalar projection is (a·b)/‖b‖ = 8/2 = 4. The vector projection is [(a·b)/(b·b)]b = (8/4)(2, 0) = (4, 0). The multiplier 2, signed length 4 and vector (4, 0) are three distinct objects. Do not report the multiplier as a length or put the scalar projection inside a two-coordinate answer. Write the requested type of quantity beside your calculation.
Subtract the projected vector from a: r = a − (4, 0) = (0, 2). The residual satisfies r·b = 0, independently checking the construction. Reversing b changes the signed scalar projection but not the projected vector on the same line. The target vector cannot be zero because the formulas divide by its length or squared length. These projections are onto the full line through b, not automatically onto a ray or a bounded physical segment drawn on the page.
Find unknown coordinates without adding conditions
Suppose a = (3, 2) and b = (t, −6) must be orthogonal. Set their product to zero: 3t + 2(−6) = 0, so 3t − 12 = 0 and t = 4. Substitution gives 3·4 − 12 = 0. Both vectors are nonzero, so a right angle follows. Orthogonality does not require equal lengths. Adding that requirement without instruction changes the problem and can make you reject a valid solution. Check the actual condition rather than a remembered picture of a square.
In the plane, a perpendicular to (x, y) is (−y, x), because x(−y) + yx = 0. If the original vector is nonzero, the rotated vector is nonzero too. In three dimensions one orthogonality equation generally leaves a plane of possibilities, not one unique vector. For u = (2, −1, 2), every (s, 2s + 2t, t) has zero product with u. This includes the zero vector when s and t are zero; exclude it when the task requires an actual direction.
Try a changed pair and verify independently
Try c = (2, 3) and d = (3, −2) before reading further. Their product is 6 − 6 = 0 and both squared lengths are 13, so the angle is 90°. Also, ‖c − d‖² = (−1)² + 5² = 26 equals ‖c‖² + ‖d‖². This follows from ‖c − d‖² = ‖c‖² + ‖d‖² − 2c·d. It checks the same geometry by a different calculation, not by measuring a drawing that may not be to scale.
Change d to (3, −1). The product becomes 3, the norms are √13 and √10, and cos θ = 3/√130. The angle must now be acute, approximately 74.745°. Our guide to checking answers without an answer key encourages independent structural comparisons. State the coordinate assumptions, nonzero checks and exact quotient before rounding. The final number is useful when you can explain why its sign, range and geometric interpretation all fit the original question, rather than merely repeating a calculator display.
