# Dot product, cross product

i need help with the following:
note that the big dots represents the dot product

1. suppose that $$a \bullet b = c \bullet b$$ for all vectors $$\overrightarrow{b}$$. show that $$\overrightarrow{a} = \overrightarrow{c}$$.

i suppose i can't simply divide out the b, right? anyway, i tried writing out the components of each vector - for example a=(a1, a2, a3), b=(b1, b2, b3), c=(c1, c2, c3). i got as far as:

(b1, b2, b3) $$\bullet$$(a1-c1, a2-c2, a3-c3) = 0

but i don't know how to simplify this further to eliminate the b. any help is appreciated.

## Answers and Replies

Related Introductory Physics Homework Help News on Phys.org
cross product

my second question is:

2. suppose that $$\overrightarrow{a} X \overrightarrow{b} = \overrightarrow{c} X \overrightarrow{b}$$ for all vectors $$\overrightarrow{b}.$$ how are $$\overrightarrow{a}$$ and $$\overrightarrow{c}$$ related?.

i wrote the vectors in terms of its components, took the cross product, and i got as far as:

(b3(a2-c1) + b2(a3-c3), b1(a3-c3) + b3(a1-c1), b2(a1-c1) + b1(a2-c2)) = 0

again, i'm not sure how to simplify this further or what conclusion to draw about the relationship between a and c. any help is greatly appreciated.

For the first, did you know that $\vec{a} \bullet \vec{b} = |a||b|cos(\theta)$? That should be all you need.

There is a similar identity for the cross product.

i have considered that, but the angles between a and b, and c and b, are different (presumably). so i end up with

|a| cos(t) = |c| cos (s)

...which doesn't seem to help very much.

Curious3141
Homework Helper
For the dot product, here's a hint. Consider the cases where $$\vec{b}$$ is a unit vector first in the direction of $$\vec{a}$$ then in the direction of $$\vec{c}$$. Draw up some equations and see what you can conclude about the angle between vectors $$\vec{a}$$ and $$\vec{c}$$, then what that implies about the relationship between their magnitudes.

Curious3141
Homework Helper
Here's the solution for the dot product :

Let the angle between vectors $$\vec{a}$$ and $$\vec{c}$$ be $$\theta$$.

Let $$\vec{b}$$ be a unit vector in the same direction as $$\vec{a}$$

Then $$\vec{a} \bullet \vec{b} = |\vec{a}|$$

$$\vec{c} \bullet \vec{b} = |\vec{c}|\cos\theta$$

These two are equal, so we have :

$$|\vec{a}| = |\vec{c}|\cos\theta$$ --eqn(1)

Now let $$\vec{b}$$ be a unit vector in the direction of $$\vec{c}$$

Working through it in the same way gives :

$$|\vec{c}| = |\vec{a}|\cos\theta$$ --eqn(2)

Solving those we get :

$$|\vec{c}|\cos^2{\theta} = |\vec{c}|$$

Either $$|\vec{c}| = |\vec{a}| = 0$$, meaning both of them are null vectors (in which case they're equal in any case),

or $$\cos^2{\theta} = 1$$

giving $$\cos{\theta} = 1$$ (the negative root is inadmissible because if you look at equations 1 and 2, the magnitudes are positive, so the cosine term can't be negative).

So we have $$\theta = 0$$, and putting that back into either equation 1 or 2, we get $$|\vec{a}| = |\vec{c}|$$.

So the two vectors $$\vec{a}$$ and $$\vec{c}$$ are of the same magnitude and oriented in the same direction. By definition, therefore, $$\vec{a} = \vec{c}$$ (QED)

Galileo
Science Advisor
Homework Helper
Write the equation in a different form:

$$(\vec a -\vec c)\times \vec b=\vec 0$$
for any vector $\vec b$.

(You can do the same thing with the dot product also).
What can you infer about $\vec a-\vec c$ from this?

dextercioby
Science Advisor
Homework Helper
The first exercise is incorrect.

$$\vec{a}\cdot\vec{b}=\vec{c}\cdot\vec{b}$$

Using the distributivity,one gets

$$\left(\vec{a}-\vec{c}\right)\cdot\vec{b}=0$$

Which has the infinite set of solutions:

$$\vec{a}-\vec{c}=\left\{\vec{d}\times\vec{b}|,\vec{d}\in V,\vec{b}\in V \right\}$$.

The "V" is a linear space and the $\vec{d}$ and $\vec{b}$ are arbitrary vectors.

So,please,think deeper before giving erroneous advice.

Daniel.

Galileo
Science Advisor
Homework Helper
The exercise is to show that $\vec a \cdot \vec b=\vec c\cdot \vec b\forall \vec b \Rightarrow \vec a = \vec c$.
This is clearly true, you just have to pick a clever value for b, say b=a-c

Last edited:
dextercioby
Science Advisor
Homework Helper
That's not true.That difference is not zero,but an arbitrary vector.Zero vector is not arbitrary.

Daniel.

Galileo
Science Advisor
Homework Helper
Are you sure you have the same understanding of the problem as I do?
$\vec b$ can be any vector.

dextercioby
Science Advisor
Homework Helper
Yes.

HINT:$$\vec{c}\cdot\vec{b}=\vec{c}\cdot\left(\vec{b}+\vec{c}\times \vec{d}\right)$$

,where \vec{d} is an arbitrary vector...

Daniel.

here is how I would handle the two dimentional case:

a * b = (ax + ay)(bx + by) = ax*bx + ay*by

c * b = (cx + cy)(bx + by) = cx*bx + cy*by

Two vectors are equal only if their corresponding components are equal. So if a * b = c *b then we have:

ax*bx = cx*bx

ay*by=cy*by

Which shows that a and c are equal.

arildno
Science Advisor
Homework Helper
Gold Member
Dearly Missed
You are wrong, Daniel.
$$\vec{a},\vec{c}$$ are fixed vectors, and the $$\vec{a}\cdot\vec{b}=\vec{b}\cdot\vec{c}$$ is supposed to hold for ALL $$\vec{b}$$
This is equivalent to:
$$(\vec{a}-\vec{c})\cdot\vec{b}=0$$
In particular, this must hold for the choice $$\vec{b}=\vec{a}-\vec{c}$$ from which follows $$\vec{a}=\vec{c}$$

Last edited:
dextercioby
Science Advisor
Homework Helper
That's wrong

$$\vec{a}\cdot\vec{b}$$ is a scalar

and hence

$$a_{x}b_{x}+a_{y}b_{y}=c_{x}b_{x}+c_{y}b_{y} \nRightarrow \left\{\begin{array}{c}a_{x}b_{x}=c_{x}b_{x}\\a_{y}b_{y}=c_{y}b_{y} \end{array}\right$$

Okay??

Daniel.

dextercioby
Science Advisor
Homework Helper
arildno said:
You are wrong, Daniel.
$$\vec{a},\vec{c}$$ are fixed vectors, and the $$\vec{a}\cdot\vec{b}=\vec{b}\cdot\vec{c}$$ is supposed to hold for ALL $$\vec{b}$$
This is equivalent to:
$$(\vec{a}-\vec{c})\cdot\vec{b}=0$$
In particular, this must hold for the particular choice $$\vec{b}=\vec{a}-\vec{c}$$ from which follows $$\vec{a}=\vec{c}$$
I'm not wrong,my solution does not exclude the 0 vector.That "d" is arbitrary,just as "b",so it can be the 0 vector.

Daniel.

arildno
Science Advisor
Homework Helper
Gold Member
Dearly Missed
dextercioby said:
The first exercise is incorrect.

$$\vec{a}\cdot\vec{b}=\vec{c}\cdot\vec{b}$$

Using the distributivity,one gets

$$\left(\vec{a}-\vec{c}\right)\cdot\vec{b}=0$$

Which has the infinite set of solutions:

$$\vec{a}-\vec{c}=\left\{\vec{d}\times\vec{b}|,\vec{d}\in V,\vec{b}\in V \right\}$$.

The "V" is a linear space and the $\vec{d}$ and $\vec{b}$ are arbitrary vectors.

So,please,think deeper before giving erroneous advice.

Daniel.
Whatever are you talking about???
$$\vec{a},\vec{c}$$ are FIXED vectors; their difference do not change if $$\vec{b}$$ changes.

dextercioby
Science Advisor
Homework Helper
Who said "a" and "c" are fixed?

Daniel.

Last edited:
arildno
Science Advisor
Homework Helper
Gold Member
Dearly Missed
jeanf said:
i need help with the following:
note that the big dots represents the dot product

1. suppose that $$a \bullet b = c \bullet b$$ for all vectors $$\overrightarrow{b}$$. show that $$\overrightarrow{a} = \overrightarrow{c}$$.
It says for all b here.

dextercioby
Science Advisor
Homework Helper
Yes,and my solution exploits the equality of the 2 scalar products.

Daniel.

arildno
Science Advisor
Homework Helper
Gold Member
Dearly Missed
No, it doesn't do any such thing.
For the particular choice $$\vec{b}=\vec{a}-\vec{c}$$
we get
$$(\vec{a}-\vec{c})^{2}=0$$
which holds if and only if $$\vec{a}=\vec{c}$$

dextercioby said:
The first exercise is incorrect.

$$\vec{a}\cdot\vec{b}=\vec{c}\cdot\vec{b}$$

Using the distributivity,one gets

$$\left(\vec{a}-\vec{c}\right)\cdot\vec{b}=0$$

Which has the infinite set of solutions:

$$\vec{a}-\vec{c}=\left\{\vec{d}\times\vec{b}|,\vec{d}\in V,\vec{b}\in V \right\}$$.

The "V" is a linear space and the $\vec{d}$ and $\vec{b}$ are arbitrary vectors.

So,please,think deeper before giving erroneous advice.

Daniel.
To be in a state of complete misunderstanding, and say something like that. And then continue to argue about it. It takes balls. I respect that.

You did not give a "solution". The problem was to give a proof that a = c , and that's what Galileo and arildno did. You just didn't follow it.

dextercioby
Science Advisor
Homework Helper
Let's look at it more abstractly and u'll see that my solution includes yours.

My solution asserts that

If $\vec{b}\in \mathcal{P}_{1}\subseteq \mathcal{P}$ with $\bar{\mathcal{P}_{1}}=\mathcal{P}$,then

$$\left(\vec{a}-\vec{c}\right) \in \mathcal{P}_{1}^{\perp}$$

and $$\mathcal{P}_{1}\oplus\mathcal{P}_{1}^{\perp}=\mathcal{P}$$

You assumed that

$$\vec{b}\in \mathcal{P}$$.Since any improper preHilbert subspace is closed and its orthogonal complement is made up from the zero vector,your solution follows from mine.

Daniel.

P.S.Arildno

dextercioby
Science Advisor
Homework Helper
HackaB said:
To be in a state of complete misunderstanding, and say something like that. And then continue to argue about it. It takes balls. I respect that.

You did not give a "solution". The problem was to give a proof that a = c , and that's what Galileo and arildno did. You just didn't follow it.

Maybe there was a missunderstanding,the text of the problem was vague,specifically that "for all b".I got balls... :surprised :tongue2:
My solution was more general.It had their solution as a particular case.

Daniel.

No, what you posted was an attempt to confuse the issue, and then show off some of your new found functional analysis jargon. You aren't helping anyone. You are just being a jackass.