Provide guidance on the use of the occurranceID and dcterms:references terms.

[GoogleCodeExporter]

The iDigBio request for comments on a GUID policy 
https://www.idigbio.org/content/globally-unique-identifiers-guid  raises the 
issue of what the GUIDs in various TDWG DarwinCore terms are intended to 
reference.  We should provide guidance on the use of the occurranceID and 
dcterms:references terms.  

We should probably amplify the DarwinCore statement about occurranceID being an 
identifier of the underlying thing (and clarify if we expect this to be a 
herbarium sheet or a duplicate set) rather than it's digital representation.

I'll suggest we should support and amplify the DarwinCore suggestion of using 
the scheme urn:catalog:[institutionCode]:[collectionCode]:[catalogNumber] for 
occurranceID when a GUID is not available.

We could also suggest use of dcterms:references as a pointer to a canonical 
digital representation of the occurrance.  This appears to be what is done in 
the DarwinCore examples of 
references=http://mvzarctos.berkeley.edu/guid/MVZ:Mamm:165861  and 
occurranceID=urn:catalog:FMNH:Mammal:145732

Original issue reported on code.google.com by mole@morris.net on 2 Feb 2012 at 8:11