Hi,
I just remebered one last semi-related thing that this thread got me thinking about again.
By training I'm a naval architect & mariine engineer, but I have an interest in all types of engineering, especially in looking at where there could be cross-overs, or in how other related engineering fields handle similar tasks to what I do. Along these lines I've bought a number of books on both car design and airplane design.
With regards to the clay modeling that is done for cars, I kind of wondered if there might be some applicability to ship design. Specifically, in the past a hullform shape (or "ship's lines") would be drawn iteratively, plotting contours out in multiple views and checking that the lines were "fair", and slightly tweaking points if they were not. In doing this it was theorectically possible to also try and take into consideration how the hull would actually be built up by individual plate sections, and try and take into consideration limiting compound curvature in certain places to make it simpler to bend the plate when building the full scale ship.
However, with modern computer tools alot of hullform development is done with 3D computer drafting tools that use fancy mathematical surfaces (like something claaed a Non Uniform Ratioinal B-Spline - or NURBS - surface). The hard part that I have had with these type computer tools is in trying to envision how that computer model can be broken into the individual plating pieces that will be used to build the full-scale ship, and then feeding back into the computer model things like "if you strighten this section out, you can eliminate compound/complex curvature here" which could make the ship easier and less expensive to build.
The bottom line to all this is that I've kind of wondered if you couldn't take the initial output from the computer program and build a clay (or wax) model of a hull. Then you could cut out thin plastic sheets representing plating sections to try and "fit" them to the model, and experiment around with either "shaving the clay (or wax) in some areas" or "building the clay (or wax) up in other areas" to make the model better match the contours that the semi-rigid plastic sheets (and hence, hopfully, the full scale steel plates) could easily be bent into.
I even suspect that now with some of the high-tech 3D scanners available it might then be possible to re-enter the shape of the revised model back into the computer in order to recheck how those changes made in the clay mught affect other aspects of the desigm.
Sorry for going off on a tangent, the whole thought of clay modeling has just gotten me thinking about a lot of different things.
Pat