(d-di(x))+di=s

Ok I spent about seven hours today trying to find some good information on how to do a stretchy-limb setup for Dans character. Unfortunately I totally failed. There just doesnt seem to be any tutorials that work for me, and they all involve either ridiculous amounts of scripting or tons of stuff I just plain cant understand.

The first one I tried was this:
http://www.tutorialized.com/view/tutorial/Stretchy-Arm-Skeleton/26944

This one suggests using a duplicate skeleton with more bones in for smoother deformation, which turns out to be massively complex with tons of intricate constraints and nodes I don't understand just to keep the second bloody skeleton following the movements of the first. I couldnt even get this step right, let alone making any joints stretch. Here's what happened when I followed the tutorial and tried to move the arm:


I got similar results on both attempts of this. Maybe the tutorial works, who knows, but with my limited understanding of what I'm doing when I'm following it I don't have a chance of getting it right; there are just too many variables I don't know about that might be effecting it.

The second tutorial I tried was this:
http://www.creativecrash.com/maya/tutorials/character/c/stretchy-hand--2

This makes it look a bit simpler with no extra skeletons or anything, but has a massive section of scripting and expressions, but only for one side of the body. Even if I tried this it would be a waste of time because I don't know anything about scripting so I wouldnt have a clue how to get the other side of the body done.

The third tutorial I tried was this:
http://www.keegan3d.com/tutorial_view.php?id=7

This tutorial brings cluelessness to a whole new level. I literally have no idea whatsoever what any of it means but nevertheless I had a go at following it. This is what happened somewhere in the middle:



Unaware of exactly why my arm shrunk, I read further on to see if there were any solutions. Nope, it just relentlessly fills my head with nonsense and numbers, at one point proposing that the key formulae is (d-di(x))+di=s, but that we'd need to convert it to expressions.

Im going to give up for the day and come back to it tomorrow.