I have a conceptual question. Everybody knows the difficulties in order to
model/implement Slowly changing Dimensions (type 2) that change or depend of
the time dimension.
I have read a lot on this subject and still don't know which is the best
way/practice to implement this kind of changes in a DataMart/Datawarehouse
model. I think that the idea is to avoid rewrite the history in summarize
processes ... no? In that case, could I incorporate the field id_date in
the SCD dimension table?... so the relation between fact table and dimension
table is cover with the id_key_dimension and id_date?
Thanks
Rodrigo