Ask your own question, for FREE!
Mathematics 19 Online
OpenStudy (freckles):

\[\int \frac{1}{x} dx \\ =\int \frac{\ln(x)}{\ln(x)} \cdot \frac{1}{x} dx=\int \frac{1}{\ln(x)} \cdot \frac{\ln(x)}{x} dx \\ =\frac{1}{\ln(x)} \cdot \frac{\ln^2(x)}{2}-\int \frac{\frac{-1}{x}}{\ln^2(x)} \cdot \frac{\ln^2(x)}{2} dx \\ =\frac{\ln(x)}{2}+ \frac{1}{2} \int \frac{1}{x} dx \\ \text{ So we have } \\ (\frac{-1}{2}+1) \int \frac{1}{x} dx=\frac{1}{\ln(x)} \cdot \frac{\ln^2(x)}{2} dx \\ \frac{1}{2} \int \frac{1}{x} dx=\frac{\ln^2(x)}{2 \ln(x)} \\ \frac{1}{2} \int \frac{1}{x} dx=\frac{\ln(x)}{2} \\ \\ \int \frac{1}{x} dx=\ln(x) \\ \text{ Now we just need to tact on that } +C \]

OpenStudy (freckles):

for x>0 of course :) @SolomonZelman just wanted to explore that question a little

OpenStudy (solomonzelman):

got disconnected, tnx for mentioning

OpenStudy (freckles):

I used u=1/ln(x) and dv=ln(x)/x dx

OpenStudy (freckles):

but to do this by integration by parts it does require us to know d/dx ln(x) =1/x but i don't see a what to do it by integration by parts without knowing that

OpenStudy (solomonzelman):

\(\large\color{slate}{\displaystyle\int\limits_{~}^{~}1/x~dx}\) \(\large\color{slate}{x(1/x)-\displaystyle\int\limits_{~}^{~}(x)(-1)(1/x^2)~dx}\) \(\large\color{slate}{1+\displaystyle\int\limits_{~}^{~}(1/x)~dx}\) \(\large\color{slate}{\displaystyle\int\limits_{~}^{~}1/x~dx=1+\displaystyle\int\limits_{~}^{~}1/x~dx}\) \(\large\color{slate}{1=0}\)

OpenStudy (solomonzelman):

but there are limits to by parts, no?

OpenStudy (anonymous):

I would like to say\[ \int \frac 1xdx\neq \int \frac 1x dx \]

OpenStudy (solomonzelman):

(Which doesn't make sense as mush as 1=0 )

OpenStudy (solomonzelman):

but there was no rule violation in what I posted, or was there?

OpenStudy (solomonzelman):

I guess that..

OpenStudy (anonymous):

It's a bit like saying \[ \pm x\neq \pm x \]

OpenStudy (anonymous):

Not that they're always unequal, but sometimes they aren't equal

OpenStudy (solomonzelman):

\(\large\color{slate}{\displaystyle\int\limits_{~}^{~}f(x)~dx=\displaystyle\int\limits_{~}^{~}f(x)~dx+C}\)

OpenStudy (solomonzelman):

such as: \(\large\color{slate}{\displaystyle\int\limits_{~}^{~}f(x)~dx=\displaystyle\int\limits_{~}^{~}f(x)~dx+1}\)

OpenStudy (solomonzelman):

and such a function f(x)=1/x

OpenStudy (solomonzelman):

it is actually making sense, ain't it?

OpenStudy (solomonzelman):

it gets absorbed by +C, but I guess the violation comes when you subtract \(\large\color{slate}{\displaystyle\int\limits_{~}^{~}\frac{1}{x}~dx}\) from both sides, as wio, I guess was saying

OpenStudy (solomonzelman):

@wio ty

OpenStudy (solomonzelman):

ty @freckles

OpenStudy (solomonzelman):

I figured. I should have asked it myself, but didn't consider it important... tnx for raising it.... (bye, I guess ?) see you!

OpenStudy (freckles):

i thought "is there a way to integrate 1/x w.r.t x by parts" an interesting question.

OpenStudy (freckles):

Mostly because I never tried

OpenStudy (solomonzelman):

oh, I just say that post that Kainui made (question asked by Marki, to prove that 1=0 using by parts). And I struggled to figure out when is the violation starting. No guideline on by parts, guideline on subtracting. Not that I would for real integrate 1/x by parts. I try to find easiest approaches to any problem be it math or any other subject. And, freckles, don't try... lol

ganeshie8 (ganeshie8):

i dont see anything wrong in by parts for 1/x as you said, if we see that derivative of lnx is 1/x, we're done. and ofcourse we gona miss all the rest of fun stuff ;)

OpenStudy (freckles):

And yes integrating 1/x w.r.t x by parts is probably the most unnecessary thing one can do while knowing d/dx ln(x) =1/x But like you said it takes the fun away :)

ganeshie8 (ganeshie8):

XD im going thru this discussion on fake proof 0=1 http://math.stackexchange.com/questions/806254/using-integration-by-parts-results-in-0-1

ganeshie8 (ganeshie8):

\[\int f(x) dx \equiv \int f(x) \pmod {\text{constant functions}}\]

OpenStudy (anonymous):

Perhaps if we want to say: \[ \int dx = \int dx \]And when we let \(u=1, dv=dx\) we have to remember to say: \[ \int dx = \big(x+C\big)-\int 0~dx \]

OpenStudy (freckles):

lol @SolomonZelman So I guess your question wasn't really to find a way to integrate 1/x w.r.t x. by parts .

OpenStudy (solomonzelman):

no, I was questioning the proof only.

ganeshie8 (ganeshie8):

proving 0=1 using by parts is similar to proving 0 = 4 in mod 4 : \[0 \equiv 4 \mod {4}\] error comes because of treating "congruence" as "identcally equal"

ganeshie8 (ganeshie8):

@freckles i think @SolomonZelman was refering to this q http://openstudy.com/users/Marki#/updates/54aa9adde4b05eecdbea5652

OpenStudy (solomonzelman):

yes, indeed that question. Took time to find it, I would not be that kind if I were you, or probably wouldn't be.... that is very nice of you tnx.

OpenStudy (freckles):

Thanks @ganeshie8 . I was curious.

ganeshie8 (ganeshie8):

yeah searching through answered questions is a pain in the neck google finds the question quicker if you remember any special keywords in the thread : https://www.google.co.in/webhp?sourceid=chrome-instant&ion=1&espv=2&ie=UTF-8#safe=off&q=site%3aopenstudy.com%20%22marki%22%2b%22parts%22%2b%22fake%22 :P

Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!
Can't find your answer? Make a FREE account and ask your own questions, OR help others and earn volunteer hours!

Join our real-time social learning platform and learn together with your friends!