Currently only the object geometry pulls through from Speckle into Revit when loading via the Speckle connector. It would be great and very useful to have this pull object metadata through too, similar to how things come through into Rhino.
Rhino’s control on object data is a flat dictionary of key:value pairs
Revit (and Archicad and others) have more tightly controlled logistics around what properties exist, what can be edited or not.
The question is put, should an object hitherto not in any way described as a BIM object with the property foo and value bar - how should that be presented in Revit on receive?
Our prototype for Revit Parameter Updater will only allow the editing of certain pre-existing Revit parameters and only then when received back to the source Revit model from which they were authored.
tbh I wouldn’t even be sure on how to logically implement this.
In Revit, there’s a differentiation between family and project parameters, shared and non-shared parameters, type and instance parameters.
So, when pulling data that previously wasn’t in the model back into Revit presents one with the challenge of deciding, case-by-case, how this data should be handled inside Revit.
Furthermore, like Jonathon said, many parameters in Revit are protected or read-only to the API, so when pulling large amounts of data from many different parameters, you would probably end up with an incoherent mess of successfull and failed parameter write transactions.
I think that - again, like Jonathon suggested - you would need to implement this in a very controlled way, for example on a parameter-by-parameter basis, by declaring (on the Speckle side) very explicitely how the data for that parameter shall be handled inside Revit.