Review of current CUBA keys

[kemattil]

Few comments/remarks related to the current CUBA keys:

1. MATERIAL_TYPE:
   definition: Material dimension and type
   shape: [1]
   type: integer

   Is this needed anymore as materials will be defined using the Material class (from metadata)?

   If it is still needed, the definition should be more precise (a single integer
   expressing both the dimension and type of a material?) ...

2. EQUATION_OF_STATE_COEFFICIENT:
   definition: Equation of state for multiphase fluids
   shape: [1]
   type: double

This is probably not needed anymore, because the equation of state will be
defined using materials relations (i.e. metadata) ...

If it is still needed, the type (scalar double) seems odd for defining
an equation of state ...

3. HEAT_CONDUCTIVITY:
   definition: Heat conductivity
   shape: [1]
   type: double

Should this be "thermal conductivity" or, e.g., "heat transfer coefficient"?
I mean, is "heat conductivity" a standard physical variable in some field?

4. Many of the descriptions only repeat the key, do not convey much information,
   or are overly restrictive. See e.g. the VELOCITY below: why only velocity of a
   point or node? VELOCITY can be used to define velocity of anything ...

   VELOCITY:
   definition: Velocity of a point or node
   shape: [3]
   type: double

[ahashibon]

Agree with the above, these are good suggestions:

1. MATERIAL_TYPE is made obsolete and surpassed in context by CUBA.MATERIAL
2. EQUATION_OF_STATE_COEFFICIENT: is made obsolete and surpassed by CUBA.EQUATION_OF_STATE
3. HEAT_CONDUCTIVITY: should be thermal conductivity as you suggest. An ontology of thermal models is needed in general, but for now we can change it as you suggested.
4. this is a good comment, velocity (and other basic attributes) should be defined without reference to the computational method details (i.e., as pure physics variables).