Ray Vickson said:
You start with an equation involving both ##x## and ##y##, so for any given value of ##x## you can solve the equation to get one or more values of ##y##.
Okay, just to make sure I'm understanding:
x + y = 6
y = 6 - x
So if x = 2, 4, 5, that would make y = 4, 2, 1, and so on.
Ray Vickson said:
In your case, when you give ##x## a value such as ##x = 1## or ##x = -7## (or whatever) you will get a quadratic equation in ##y## that has two roots that you can find by the elementary methods you studied years ago.
Hmm, okay. I don't really see it?
Tribo said:
Using the product rule I got:
(2xy + x2) + (y2 + 2y) = 0
The parentheses might be wrong, and I might've taken the derivative of 6 too soon, but I don't see where to go from here regardless. Surely it's not quadratic or going to be if there's no longer anything squared? Also... how do I chain 2xy?
Ray Vickson said:
So, indeed: we are interested in the case where y depends on x---very much so! The only problem is that although y depends on x we do not have an actual formula that tells us exactly what that dependence is. That is, while we know that we will have ##y = g(x)## for some function ##g## (that is, the function ##g(x)## exists), we don't have a formula for ##g(x)##. The question is asking you how you would find the derivative ##dy/dx= dg(x)/dx## without actually knowing the formula for ##g## in the relationship ##y = g(x)##.
What you're saying makes sense but I don't see it in the problem.

We have y = a function but we don't know what the function looks like (it's formula).
On a side note
, dy/dx =
dg(x)/dx means
the derivative of y in respect to the derivative of x =
the derivative of function g(x) in respect to the same derivative of x
...but what does that MEAN?