appropriate use of aggregation identity

[Tim DiLauro]

Dear OAI-ORE group members,

I'm working on a approach to packaging and I wonder how frowned upon the following approach would be.  I would like to create an aggregation (inside a larger package aggregation) that represents aggregated resources (AR-*) to be ore:aggregatedBy an existing aggregation in a repository.  Sorry, I know that's a lot of aggregating.  As a shorthand, I would like to use the identity of the repository aggregation (A-1) as the identity of the Deposit Package aggregation (the resource maps would be specific to each environment, though).  My main concern is that two aggregations are obviously not the same, but I want the simplest way to match the objects for the deposit.

The idea is illustrated in the diagram below.  I would welcome thoughts on appropriate/better ways to do this.  All constructive criticism welcome.

Cheers,
~Tim