CAD Working Group Strategy

a. Build examples of X3D CAD models to provide use cases and use case and best practices.

Preserve product structure in CAD node name fields and metadata.

b. Resist the urge to refactor CAD product nodes, unless a clear improvement in CAD consistency is shown. Preferred is good practices rather than restrictions that are difficult to enforce.

c. Goals - Primary goal. CAD definitions converted to X3D geometry, preferring parametric representations for brevity and precision. - Secondary goal. Option to save CAD product structure in the use of named X3D CAD product structure nodes, with addition of embedded typed metadata nodes. - Third goal. Define, implement, demonstrate and distribute the workflow for converting CAD models to X3D.

d. Next steps for activity. - Compare vocabularies for ISO STEP, Macro Parametric Approach, B-REPS (general definitions) and B-REPS (proposed X3D nodes). - For each table row, define corresponding mappings to existing parametric X3D nodes (e.g. NURBS, extrusion, primitives). - Consider implementing these BREPS nodes as NURBS-based X3D prototypes - These prototypes are an independent implementation that make it easy for either authors or browser to use such BREPS nodes - Can also update the support for these nodes in Xj3D open source - List all pros/cons of BREPS nodes as first-class X3D v3.4 nodes - At this point it is much easier to decide whether to add BREPS to X3D

e. Constructive Solid Geometry (CSG) is best considered part of model definition, something that helps to produce BREPS and surfaces, and not something that needs to be supported in browsers and content. CSG is computationally heavyweight, and does not belong in a lightweight portrayal format like X3D. If browsers are heavy, no one uses them. "Lighter, stronger, simpler" is what Hyo hears from his customers.