Karl,
Here it is with a little extra frosting.
The real wish is a performance standard:
To be able to have elements set to the correct layers by view setting and type.
For example:
When I switch to the Electrical Plan view: I want the Line, Arc, Polyline, Spline, Label and Text tools set to the Electrical Notes layer (or E-ANNO, or +E-NOTE, or whatever). I want the ceiling light fixture Objects set to the Electrical Ceiling Fixtures layer (E-LITE-CLNG). I want the wall sconce Objects set to the Electrical Wall Fixtures layer (E-LITE-WALL). AND I want the Symbols for receptacles, switches, phone jacks, CATV, etc. to be on their appropriate layer or layers. (Of course, the particulars depend on the specific office standards and project requirements.)
This would be accomplished by two separate but related means:
1. The View selected in the Navigator can define default settings for some or all of the tools. The default would be "current settings". This would allow the feature to be implemented with no change the the current behavior of the program, and to be enabled only as needed for advanced users or template makers. (This wish is primarily about layers but it is only natural to expand this to other settings.)
2. An additional option in the above procedure would be "by Subtype" for Objects or "by Fill or Composite" for walls, slabs and roofs. In order for this to work the user (or author of the template) would have to assign layers to each subtype and the variety of subtypes would need to be sufficient to cover most needs.
This could be expanded to include "by IFC Class" for even more detailed (insanely compulsive?) levels of control. This would require a look-up table for mapping all the IFC classes to the currently available layers.
BTW I have some concerns about this in the long run. While I really like the idea of the auto-layering for immediate productivity, I think the "layer" analogy is problematic for the Virtual Building. If this wish could be tailored to lead us away from a dependence on "layering" for organizing our models and toward a more multi-dimensional organization of assemblies, subassembies, superassemblies and components then I am all the more enthusiastic about it. If, on the other hand, it is organized in such a way that the "layer" system becomes more entrenched and ossified, then we might be better off to approach the whole issue from a different angle.
