I'm looking forward to the day when we build description sets of identifying characteristics (attributes for each entity, e.g., for person: preferred name as found on publications, dates, etc.) and relationships (to people, works, publishers, etc.) and let systems select which data is needed in each context for displays - sometimes just last names, sometimes inverted names with dates, etc. The authorized access point strings we created for book and card catalogs are still handy for some types of linear displays, but should not be what we are building behind the scenes for context-appropriate displays. VIAF already accepts that there can be many persons with the same name and focuses instead on the significant distinguishing characteristics to build clusters of the authority records/descriptive sets that are for the same entity and preserves the relationships/links to related entities - so we know, for example, that a specific pseudonym/pen name is related to a "real name" given to a person.

The FRBR-family of conceptual models is meant to provide a perspective or point of view of the bibliographic universe, and we now need to take that to the next step of building data models to then build systems based on those models. I see us instead now taking the conceptual model entities and attributes directly as if they were the data model, rather than going the intermediate step to build the library-application data models needed for building library systems of the future. For that intermediate step, I think the RDA element set could give us a jump ahead towards such a data model and resulting schema, with additions from the LCSH and LCC work where RDA now has subject placeholders (awaiting work that would build on FRSAD). Other subject heading and classification schemes should also be added as needed.

- Barbara (my personal views, not necessarily those of LC)

-----Original Message-----
From: Metadata Object Description Schema List [mailto:[log in to unmask]] On Behalf Of Bruce D'Arcus
Sent: Tuesday, January 04, 2011 11:12 AM
To: [log in to unmask]
Subject: Re: [MODS] Mark Twain Rides Again (was Re: [MODS] MADS/RDF for review)

On Mon, Jan 3, 2011 at 6:32 PM, Young,Jeff (OR) <[log in to unmask]> wrote:
> Bruce,
>
> The naming preferences and use cases of different authority agencies
> need to be accounted for.

This sounds a bit circular though :-)

Correct me if I'm wrong, but:

This whole business of authoritative names, it seems to me, is one rather awkward solution to a more fundamental issue of identity: which "Jane Doe" we're referring to. So libraries came up with solutions that disambiguate by string names. So "John Doe" becomes, say, "John Doe, 1783-1841".

With that, you then need to say who created that string identifier, so you don't get conflicts between different ones. E.g. agency X does "John Doe, 1783-1841" and agency Y does "John S. Doe:1783-1841" and you need some way to make sense of that.

So the indirection, then, appears from the outside to look like a solution that creates its own problems (the requirement you note above).

But aren't there other ways to solve this identity-name problem, such that we accept there may be many people with the name "John Doe"?

Bruce

> Take VIAF for example. That's one example of how the indirection of
> mads:MADSAuthority + mads:AuthorityScheme or skos:Concept +
> skos:ConceptScheme come in handy.
>
> I think the MADS focus on classifying names rather than things is
> upside-down, but that's how many people think about authorities. I
> would suggest that the skos:Concept abstraction (via foaf:focus) is
> the best way for us to flip them right-side up. ;-)
>
> Jeff
>
>> -----Original Message-----
>> From: Metadata Object Description Schema List
>> [mailto:[log in to unmask]] On Behalf Of Bruce D'Arcus
>> Sent: Monday, January 03, 2011 12:57 PM
>> To: [log in to unmask]
>> Subject: Re: [MODS] Mark Twain Rides Again (was Re: [MODS] MADS/RDF
> for
>> review)
>>
>> In plain English, isn't Mark Twain simply a pen-name, denoting an
>> alternate persona, for the person Samuel Clemens? Shouldn't the
>> modeling reflect that basic, fairly simple, logic?
>>
>> E.g. the most simple modeling:
>>
>> <http:whatever.org/1> a foaf:Person ;
>>     foaf:name "Samuel Clemens" ;
>>     mads:penName "Mark Twain" .
>>
>> ... or a little more complex, treat the two as discrete foaf:Person
>> subjects:
>>
>> <http:whatever.org/1> a foaf:Person ;
>>     foaf:name "Samuel Clemens" ;
>>     mads:alternatePersona <http:whatever.org/2> .
>>
>> <http:whatever.org/2> a foaf:Person ;
>>     foaf:name "Mark Twain" ;
>>     mads:alternatePersonaOf <http:whatever.org/1> .
>>
>> I don't get the value of indirection: of treating authors as concepts.
>>
>> Bruce
>