Quoting Ross Singer <[log in to unmask]>:


> I realize that's an unsatisfying statement, especially since there
> there seem to be no real established Film-based RDF vocabularies, but
> it's important realize that it's not a failure (or a responsibility)
> of BIBO that it's lacking here.

You are right. BIBO gets to be exactly what BIBO wants to be. But that  
doesn't mean that I should feel compelled to use it. My annoyance is  
with folks who argue that libraries should be using [insert favorite  
RDF bibliographic ontology here] and not 're-inventing the wheel.' For  
all that "re-use don't re-invent" is a great idea, you should only  
re-use properties and vocabularies that strictly meet your definitions  
and uses. Since we are in a moment when the word "compromise" is being  
heavily used, I will ask here: "How much should you compromise, and  
what are the consequences?"

I think it's hard to predict those consequences, but I'm wary of the  
compromises. What I think is a better solution is to define your  
ontology to meet your needs, re-using only when the semantic  
equivalence is absolutely clear, and to define relationships between  
your ontology and the ontologies of communities with whom you wish to  
link data. This to me is the "win-win" of linked data. Of course, what  
this does is to put off to the future things like trying to reconcile  
the 6-levels of serials captions and enumeration from MARC Holdings  
with the "volume" and "number" data elements of just about every other  
bibliographic metadata schema. In the end, I think we have to accept  
that some things will be lost in translation.

kc



>
> -Ross.
>



-- 
Karen Coyle
[log in to unmask] http://kcoyle.net
ph: 1-510-540-7596
m: 1-510-435-8234
skype: kcoylenet