: I wondered what those numbers meant, I assumed it's % change in length/area?
: The two fairly curvy ones I built last year required 11 and 14 iterations
: but had errors on the order of +/- .001 or less

Always pays to read the manual! :-) I was just reading this:

After the plates have been unfolded to 2D,FREE!ship compares the length
of the unfolded interior edges to the length of these edges in 3D. If this length is smaller then the edges were compressed. If the unfolded edges are longer then these edges were stretched.The min. error shown at the top is the largest compression error that occurred (in world space dimensions,so meters or feet). The max. error is the largest amount of stress of an edge.

Compressed or stretched edges may be visualized by turning both the visibility of interior edges and highlighting of compressed edges on.

.....the displayed edges the number of iterations it took to unfold the selected panel is shown. FREE!ship makes up to 25 developments of each panel and uses the one with the smallest over all error as the final one. Generally surfaces that are truly developable are unfolded in 1 iteration,and have min.and max.errors of 0.0.

Surfaces that are not exactly developable can in most cases still be unfolded but might have significant errors due to the fact that the surface is curved in two directions. Think of it as the top half of a sphere,you can not press this surface down to a flat surface without stretching or compressing certain areas,unless you make some cuts of course. So it's very important to check these errors when you actually want to use the unfolded plates for construction purposes!