bug / improvement request

Hello Daniel,

Firstly thanks for Kangaroo! I'd like to request some improved behaviour in the ways by which objects are transformed so that it will be easier for users to create models.

I've enclosed a copy of a rhino v5 file as well as two grasshopper .80010 files that demonstrate two different behaviors that could be made more consistent.  Basically - it would be nice for the kangaroo engine to work the same way in both cases to facilitate easier modeling.

Case 1 (the working case):

This is a simple example of a tether for something like a balloon that always rises. There is a constraining anchor point, a kangaroo generated tether point (that tries to float), and two forces, a vector (up) and a spring (two points generating a line in gh) the spring acts to constrain the distance and thus act like a tether. Case 1 works because the initial constraining anchor point coincides with one of the two points that form the constraining spring.  When the simulation starts it always works - you can see that a second point will rise and slowly position itself vertically over the anchor.

Case 2 (the inconsistent case):

This case uses the same elements as case 1 except that I have transfered the anchor point to a different point that does not coincide with the constraining spring. In my mind there is no reason why the anchor should be required to coincide with the spring in order to work.  To further demonstrate the inconsistency if you go back to the working case 1 (after having started that simulation) and then you alter the anchor point to the anchor given in case 2 you'll see that then the simulation works.  It would be nice if the simulation always worked regardless of the anchor point.

Thanks for listening - I hope that you can fix this.

One last suggestion, it would be wonderful if, in a working version of case 2, one could give grasshopper a list of anchor points and then have each anchor generate its own tether while using the single spring and vector forces.