A lot of users will not need fully constrained elements. Given the typical size of AEC models, MCAD-style fully constrained component or part models constrained together using mates (Mstn 3d Constraints) are going to be comparatively rare. The vast majority of elements would be under or un-constrained.
What I think will be really useful / popular is a simple way to 'trammel' lines to an orthogonal (or polar/spherical) coordinate system without the hassle of setting up fix points + helper geometry lines etc etc. Most MCAD sketchers have this built in. I think that the solving would also be more scalable as any mods can be solved without reference to another element, only the host coordinate system.
The user would just use the HVN Constraint tool on a selection and then rely on the constraint to keep everything orthogonal, level or vertical... which is what is needed 99% of the time.
The Properties Panel would display which 'host' plane the element is assigned to: Model, ACS or even a Face to Solid or a Cell?
It would be good to extend this to planar elements (Shapes, Smart lines, Circles) and Solids like Slabs, Linear Solids etc
In the meantime, the Model's coord system would be good. The 3D Angle Constraint would be a good entry point. Constrain a Face on a Slab. User gets menu for H,V,N... if user selects a Face on screen then the current workflow applies.
When the user creates a Cell with the Constraints, the Constraints uses the Cell's coordinate system's H and V. N would refer to the associated Face (always included even when not selected when creating the Cell?)
Apologies for the late reply:
ACS: Yes, this would definitively be good to have. But, Mstn would need to provide an ACS object that is availabe to the Constraints solver? As you know, the way 2d Constraints work, its very difficult to see what plane is being used. An ACS would definitely help here.
Also, when the Constraints and geometry are packaged up as Cell, there needs to be a way to include the ACS with the Cell.
Hi Dominic, would H & V be relative to the active ACS?