Hi Eric
I would always use CLUSTER rather than ITEM for the data and other features
in other classes. The alternative is to have far more versions of archetypes
as if you allow element at this point you have to version when cluster is
necessary (which you could argue it always will be at some time in the
future).
Cheers, Sam
Hi Zam! 
I was merely trying to keep most of the same semantic power in the
change suggestion as when the abstract ITEM_STRUCTURE (that subsumed
ITEM_SINGLE, ITEM_TREE etc) was used rather than ITEM_TREE in various
places in the RM model. But you might be completely correct that it
would be better to point to CLUSTER rather than it's abstract
superclass ITEM in some or perhaps even all places in the model where
ITEM_STRUCTURE is used today. I guess other people on the list will
have additional good ideas about this.
Did you have any more info (or link) regarding the "pivot"
semantics/requirements by the way?
Best regards,
Erik(!) Sundvall
erik.sundvall@liu.se http://www.imt.liu.se/~erisu/ Tel: +46-13-286733