Which is why a "set on terrain" option for individual objects would be nice. This feature is done with populations, I'm not sure why it couldn't be implemented for individual objects.
I've started to write an "object manipulation" program in C#. Right now, it's mainly in the "thinking stage" though I did make a form with some basic data readouts, reference dropdown, tgd reader and parser, and a few buttons. I'm thinking a list of objects in the scene; chose a reference object original position; input the reference object new position; apply the offset to the rest of the selected objects. Of course there would be the problem with the y axis which we have to play with now when we move objects. Probably have to save the tgd and load it since I don't think it could be interactive, but maybe it can.
Currently though, when building models, I typically pick one object as my "reference" and then build and position the rest in my modeling program (Lightwave.) Then, when I import the reference model, I can copy the coordinates and paste that to the rest of the objects that I had arranged in Lightwave, so all the objects will "connect" properly in TG2.
So if I had two objects, a cube and a sphere and wanted the sphere to be 20m to the right of the cube, , I would first build the cube at 0, 0, 0 in Lightwave. Then I would build the sphere at 20,0,0. I can either save both or save each individually. Run them through Poseray and import the cube into TG2. Copy the coordinates and import the sphere and paste the coordinates. This works great for larges networks of objects like roads, bridges, etc.
Anyway, any thoughts on this would be helpful. Don't know when I'll get the chance to code it, but I think it would be useful (at least for me.)