> Date: Tue, 3 Jun 2003 12:36:13 +0100
> From: "Oldroyd, Bill" <[log in to unmask]>
> Say I have 20 different systems serving digital objects.
> There is a single server indexing these 20 systems, say by crawling
> or harvesting. The server provides a search interface.
> The user performs a search against the search server. The records
> are retrieved from the digial object servers.

Yes, this is a sensible scenario.  But I still don't see why this
needs any protocol changes.  All you'd do is have your harvester build
metadata records representing the resources from the twenty systems,
and roll those records into the SRW server's database.  Those metadata
records would include a "URI from which to retrieve the actual
resource" field.

What am I missing?

 _/|_    _______________________________________________________________
/o ) \/  Mike Taylor  <[log in to unmask]>
)_v__/\  "By the time you discover that you need a two-inch-thick
         book to figure out how to format chapter headings, your
         check has already been deposited in the Bank of Redmond" --
         Jakob Nielsen, "Designing Web Usability"

Listen to my wife's new CD of kids' music, _Child's Play_, at