I will take this one:
 
Greg, you have hit the problem of the overloaded dataset, and I have some
good news and some bad news.
 
The bad news first.  If your split will define columns which are different
data types from the original table, you are out of luck and skip the next
paragraph, ie goto epilog;
 
The good news is if you have an original column which is something like X200.
 Well, create a set a views which substring the column into various, well,
views.  This will work fine.  No need to do multiple splits.
 
epilog:
At present, there is no capability to do multiple mapping for the same table.
 AT IPROF, I suggested that HP look at doing something to resolve this and
put forth an idea on how this my work.  Both the request and the suggestion
went directly to the proper person.  Now we wait for a reply.
 
 
Kind regards,
 
Denys. . .