There is a core capability requirement, that one or more statements defining a statistical process are defined with respect to another aspect of the data set.
The common use case is that a statistical process is defined with respect to a coordinate.
This issue forces us to investigate multiplicity, I believe. Specifically:
Will a data set contain more than one statistical process definition?
if so, will these be an ordered list?
Will a statistical process definition reference more than one coordinate?
An implementation which assumes which assumes that 1 & 2 will always be 'no' is a simpler implementation, but may be a significant limitation to our progress.
There is a core capability requirement, that one or more statements defining a statistical process are defined with respect to another aspect of the data set.
The common use case is that a statistical process is defined with respect to a coordinate.
This issue forces us to investigate multiplicity, I believe. Specifically:
An implementation which assumes which assumes that 1 & 2 will always be 'no' is a simpler implementation, but may be a significant limitation to our progress.