On 11/7/14 5:32 PM, Robert Sanderson wrote:
> To provide information about the resource in its own context, rather 
> than relying on the property referencing it from another resource to 
> imply additional information about it.
> For example:
> <w> a Work ;
>   bf:issn <i> .
> <i> a Identifier ;
>   rdf:value "1234567890" .
> If I dereference <i>, I have no idea that the resource is an ISSN.
> Thus we need to include that information associated with the 
> resource.  That leads to identifierScheme ... which is a URI.
> So now we have:
> <i> a Identifier ;
>   scheme <http://.../issn/> ;
>   value "1234567890" .
> But an ISSN is a *type* of Identifier, and redundant with the class 
> bf:Identifier.  There can be no resource which has a scheme of <issn> 
> which is not an Issn Identifier.
> Thus we simplify the model to:
> <i> a IssnIdentifier ;
>   value "1234567890" .
> QED?

Thanks, Rob. I do see what you are intending here. From this, I assume 
that a reason for using subclassing rather than domains is that the 
effort to infer that Issnidentifier is a sub-class of bf:Identifier is 
less than (or different from) the effort to infer that same information 
based on a property with a domain. Is that the case?

>     Do we anticipate particular searches that make use of them? 
> How about, off the top of my head:
> * Associate a prefLabel with each of the classes for display.
> * Provide additional linking or information based on the class
> * Provide styling or ordering based on the class, due to local preferences
> * Do validation on the value to ensure that it's a legal instance of a 
> <class> value
> * etc.

It seems, however, that any of the above could also apply to a 
property/sub-property relationship. Say in the case of wanting to 
display the label for IssnIdentifier as "Identifier" or "ISSN" - how 
does this differ from using the rdf:label (or skos:preflabel) for a 
property in the ontology? or all sub-properties of a property?

BTW, ordering based on the class and ordering of properties are among 
the requirements for the validation work being done at DC, as is value 
checking. The SPIN and ICV validations are "class-based" in that the 
class provides the link between the validation rules and the instance 
data. Other solutions, like resource shapes and shape expressions, are 
not fixed on types but on graphs. It's two different ways to look at the 
problem, and the work is in its early stages so a comparison of 
capabilities has not yet been done.

I'm not sure what you mean by "a legal instance of a class value" - how 
do classes have values? I'm only aware of value definitions related to 
the property range. If you have a class definition that shows this, such 
an example would really help.
> Disagree. For example:

Ah. I was talking about the BIBFRAME process that this list is 
supposedly a discussion list for. I like those use cases and it would be 
great if you could get them added to the BIBFRAME use cases.

>     Without answers to these questions, I don't see how we can
>     evaluate BIBFRAME as it exists today. If we don't know what needs
>     it is responding to, how can we know if it meets any needs at all?
>     This is system development 101, folks. I'm not asking anything out
>     of the norm.
> Agreed... how about you start a thread to gather use cases, starting 
> with some of your own? :)

Rob, if you mean use cases for BIBFRAME... if LC both wants them and is 
willing to consider them, that would be a great idea. Let's see if this 
prompts that.

However, a different site that isn't limited to BIBFRAME would be a 
really great idea. I started something like that about a decade ago, but 
I think it was too soon. Is there a way that your project could open up 
a section to gathering of use cases from the community? We could also 
link from the validation use cases that DC and W3C are gathering, and 
see where we have overlaps. Or is there some third-party neutral site 
where we could do this?


Karen Coyle
[log in to unmask]
m: +1-510-435-8234
skype: kcoylenet/+1-510-984-3600