The Student Room Group

What would be the answer to this integral?

0δx  dx\displaystyle\int_0^{\infty} \delta x\; dx

I can't figure out if it would be zero or infinite? Or maybe 1?
(edited 13 years ago)
Original post by Plato's Trousers
0δx  dx\displaystyle\int_0^{\infty} \delta x\; dx

I can't figure out if it would be zero or infinite? Or maybe 1?


You meant 0δy  dx\displaystyle\int_0^{\infty} \delta y\; dx
Original post by Get me off the £\?%!^@ computer
You meant 0δy  dx\displaystyle\int_0^{\infty} \delta y\; dx


apologies, I did indeed.

What's the answer :wink:
Original post by Plato's Trousers
apologies, I did indeed.

What's the answer :wink:


I don't know. Infinity.
(edited 13 years ago)
Reply 4
Unless I very much miss my guess, you have no sensible definition of what "infinitesimal" means here. So the question has no meaning.
Original post by DFranklin
Unless I very much miss my guess, you have no sensible definition of what "infinitesimal" means here. So the question has no meaning.


it is the usual meaning of δ\delta (as used in working out derivatives from first principles, eg x+δxx+\delta x).
Reply 6
Not from the attachment you've posted it isn't. (In fact a proper proof of derivatives from first principles almost certainly won't use the word infinitesimal except in an illustrative sense).

And if it is, the answer is infinity.
(edited 13 years ago)
Original post by DFranklin
Not from the attachment you've posted it isn't.


What is the difference? The graph shows a line which is an infinitesimal (or if you prefer, arbitrarily small) distance from the y=0 line. Thus the width of the band is δy\delta y

How is that different to the δ\delta used in the derivation of the derivative?
Reply 8
When you derive the derivative, you don't use the word infinitesimal. Therefore, "the infinitesimal distance δy\delta y" is a different object from the δy\delta y you might see in a derivative proof. One of them is infinitesimal (whatever you think that means), and one of them is not.
I was assuming your delta y to have some constant positive value, no matter how small. With this assumption infinity is the answer.
Original post by DFranklin
When you derive the derivative, you don't use the word infinitesimal. Therefore, "the infinitesimal distance δy\delta y" is a different object from the δy\delta y you might see in a derivative proof. One of them is infinitesimal (whatever you think that means), and one of them is not.


I think I see what you mean. When we derive a derivative, we say that δx\delta x is very small and then we let it tend to zero. So I guess it's not infinitesimal.

So are you saying infinitesimal has no meaning? Does it not mean "in the limit δx0\delta x \rightarrow 0
It might do. But doing so doesn't really make sense here.

Formally, you have two main ways of doing that here:

0limδy0+δydx=00dx=0\int_0^\infty \lim_{\delta y \to 0^+} \delta y \, dx = \int_0^\infty 0 \,dx = 0

or

limδy0+0δydx=limδy0+=\lim_{\delta y \to 0^+} \int_0^\infty \delta y \, dx = \lim_{\delta y \to 0^+} \infty = \infty

neither of which seems very useful.

(or you could not worry about ensuring the limit is from above, in which case the 2nd approach gives an undefined integral. Again not useful).
Original post by DFranklin
It might do. But doing so doesn't really make sense here.

Formally, you have two main ways of doing that here:

0limδy0+δydx=00dx=0\int_0^\infty \lim_{\delta y \to 0^+} \delta y \, dx = \int_0^\infty 0 \,dx = 0

or

limδy0+0δydx=limδy0+=\lim_{\delta y \to 0^+} \int_0^\infty \delta y \, dx = \lim_{\delta y \to 0^+} \infty = \infty

neither of which seems very useful.

(or you could not worry about ensuring the limit is from above, in which case the 2nd approach gives an undefined integral. Again not useful).


Ok, thanks for this.

So a bit of a cloth-eared question then :wink:
Well, I suspect there's a fair bit of context you haven't given...
Original post by DFranklin
Well, I suspect there's a fair bit of context you haven't given...


:nope: that was all there was (the attachment). It was billed as being this big, difficult problem that even the finest minds couldn't solve etc :wink:

I think you have shown it for what it is...
Well, I'm guessing they really *do* mean infinitesimal (so not the same as δy\delta y in limδy0\lim_{\delta y \to 0}, where δy\delta y is just a number).
Original post by DFranklin
Well, I'm guessing they really *do* mean infinitesimal (so not the same as δy\delta y in limδy0\lim_{\delta y \to 0}, where δy\delta y is just a number).


but hang on, I thought you were saying "infinitesimal" is fairly meaningless?

Or do you mean that δy\delta y really is infinitely small (ie more akin to dxdx) and so the vale of the integral is zero as you explained?
Original post by Plato's Trousers
but hang on, I thought you were saying "infinitesimal" is fairly meaningless?If you just say the word, then yeah - because what does it actually mean?

But you can do it properly: http://en.wikipedia.org/wiki/Non-standard_analysis

I've not really done any NSA, but from analogy with measure theory I'd guess you can't actually make this integral make sense even under that framework.

On the gripping hand, I suspect all they meant was some vague woolly idea of infinitesimal either. Conceptually the problem is "either it's 0, and the integral's 0, or it isn't 0, and the integral's infinite". Hard to get something inbetween...
Original post by DFranklin
If you just say the word, then yeah - because what does it actually mean?

But you can do it properly: http://en.wikipedia.org/wiki/Non-standard_analysis

I've not really done any NSA, but from analogy with measure theory I'd guess you can't actually make this integral make sense even under that framework.

On the gripping hand, I suspect all they meant was some vague woolly idea of infinitesimal either. Conceptually the problem is "either it's 0, and the integral's 0, or it isn't 0, and the integral's infinite". Hard to get something inbetween...


interesting. Thanks

BTW what does "on the gripping hand" mean?

Quick Reply

Latest