[OAI-implementers] Question about sets definition

Simeon Warner simeon at cs.cornell.edu
Mon Jul 30 16:46:13 EDT 2007


I agree with Michael, it would be bad to use sets as a hook to obtain 
different record from one items and such use leads to undefined behavior 
as noted.

This stems from the conception of sets as a mechanism for selective 
harvesting
(http://www.openarchives.org/OAI/2.0/openarchivesprotocol.htm#Set):

"A set is an optional construct for grouping items for the purpose of 
selective harvesting. ..."

One should think of set membership as an item level property. It would 
be legitimate (though not necessarily useful) to have sets which group 
items that have certain metadata formats available; but not to use the 
set parameter to affect the content of records that are returned by 
ListRecords. The metadata records themselves are obtained from items and 
  are (http://www.openarchives.org/OAI/2.0/openarchivesprotocol.htm#Record):

"... identified unambigiously by the combination of the unique 
identifier of the item from which the record is available, the 
metadataPrefix identifying the metadata format of the record, and the 
datestamp of the record."

Cheers,
Simeon

Michael Nelson wrote:
> On Mon, 30 Jul 2007, Tsach Moshkovits wrote:
> 
>> Hi all,
>>
>> I have a question regarding implementing "sets" and "md-format" :
>>
>> If I understood correctly, the record-id together with the md-format
>> identifies the data uniquely?
>>
>> Meaning, looking at record "x" of format "oai_dc", must I see the same
>> data whether the scope is set "A" or set "B"? or is it valid to see
>> different data for the same record and format in different scope
>> (something like, set "A" will contain thin oai_dc and set "B" will
>> contain richer oai_dc)?
> 
> Tsach,
> 
> No, you can not legally use sets to influence the contents of the record.
> 
> One thing to remember is that GetRecord does not take the "set" argument,
> so record X would have to belong to sets A & B (which is fine), but when
> you do:
> 
> ?verb=GetRecord&metadataPrefix=oai_dc&identifier=X
> 
> there is no way to specify if you want "thin" or "thick" oai_dc version of
> X.
> 
> go ahead and create separate metadata formats to handle this situation.
> 
> regards,
> 
> Michael
> 
>>
>>
>> TIA,
>>
>> Tsach Moshkovits
>>
>>
>>
>>
> 
> ----
> Michael L. Nelson mln at cs.odu.edu http://www.cs.odu.edu/~mln/
> Dept of Computer Science, Old Dominion University, Norfolk VA 23529
> +1 757 683 6393 +1 757 683 4900 (f)
> 
> _______________________________________________
> OAI-implementers mailing list
> List information, archives, preferences and to unsubscribe:
> http://www.openarchives.org/mailman/listinfo/oai-implementers




More information about the OAI-implementers mailing list