From: "Ed Summers" <[log in to unmask]>
> Who is it who needs this functionality
This feature has been discussed a number of times but most recently raised
as a requirment by Martin Morrey of Intallect.
> why can't they simply define a new recordSchema like OCLC
> does in Johan's example?
>
>
http://www.worldcat.org/webservices/registry/search/Institutions?version=1.1&operation=searchRetrieve&recordSchema=info:rfa/rfaRegistry/schemaInfos/adminData&maximumRecords=10&startRecord=1&resultSetTTL=300&recordPacking=xml&query=local.oclcAccountName+all+"Library+of+Congress"
As for Joe's example, I don't understand it yet.
But strictly speaking this extension isn't needed because you can always
retrieve a record according to both a "real" schema and a metadata schema,
but that means two separate requests, which some see as a burden, so the
extension could be seen as a convenience.
> Also what is the relation between "x-info-99-metadata" and the use of
> accept extensions? [1]
> [1] http://www.loc.gov/standards/sru/accept.html
We briefly discussed trying to shoehorn this metadata functionality into the
accept extension, but decided it wasn't a good idea. (So where you see
'recordMetadata', 'responseMetadata', and 'metadata' as values, that was a
relic of that discussion, and these should probably be removed.)
--Ray
|