Re: pre-coordination

[Just forwarding this thread to the SKOS list, on which more people may want to react on this]


>  From your example, it sounds like every pre-coordinated "heading" gets its own authority description, as do its individual components. If this were done programmatically with LCSH, wouldn't the result be loads of syntactically "valid" but nonsensical combinations?
>
> Re: MADS RDF, is there anything to read/look at right now, or should I sit on my hands for a week or two?
>
> Cory
>
> On 11/1/10 4:57 PM, Ray Denenberg, Library of Congress wrote:
>> Quote: "Can anyone point me to research or implementations around the
>> question of subject pre-coordination in a linked data context? It seems to
>> me that this is still the elephant in the room when it comes to dealing with
>> LCSH in a meaningful way."
>>
>> The XG use case 'Component Vocabularies',
>> http://www.w3.org/2005/Incubator/lld/wiki/Use_Case_Component_Vocabularies,
>> deals specifically with this; its name (the "component" part) was chosen to
>> emphasize the pre-coordination aspect.
>>
>> For example consider the (hypothetical) LCSH subject heading "Sailboats --
>> Design and construction". A bibliographic description for a book about
>> sailboats might cite this subject heading and link to its authority
>> description in MADS/RDF, which in turn will link to the individual component
>> subject headings, "sailboats" and "Design and Construction".
>>
>> The MADS RDF work, which has been developed over the past year or so,
>> provides the necessary granularity to understand the components of the
>> heading; a MADS/RDF ontology is expected to be available for public review
>> within the next week or two.
> ---
> [This E-mail scanned for viruses by Declude Virus]
>
>
>
>> From the SKOS Primer:
>
> "Rather than commit to a design pattern that has not been proven useful, the Semantic Web Deployment Group decided to postpone the issue of coordination, to allow extension patterns to organically emerge as SKOS is deployed. The hope is that as successful patterns are established, they can be published on the Web as an extension vocabulary to SKOS and documented as a W3C Note or some equivalent."
>
> Have there been any recent developments in this area? Can anyone point me to research or implementations around the question of subject pre-coordination in a linked data context? It seems to me that this is still the elephant in the room when it comes to dealing with LCSH in a meaningful way.
>
> Of particular interest to me is the question of how to express one-to-many or many-to-many matches between SKOS concepts. So, e.g., mapping between pre-coordinated LCSH and post-coordinated AAT & TGN:
>
> LCSH topical "gilt furniture" = AAT Activities "gilt" + AAT Objects "furniture"
> LCSH topical "Leatherwork--Alaska" = LCSH topical "Leatherwork" + NAF "Alaska" = AAT Activities "leatherworking" + TGN "Alaska"
>
> Thanks,
>
> Cory

Received on Tuesday, 2 November 2010 10:43:26 UTC