RE: Generated RDF conformant with good practise?

On Thu, 16 Sep 2004, Howard Katz wrote:

>> From: www-rdf-interest-request@w3.org
>> [mailto:www-rdf-interest-request@w3.org]On Behalf Of Libby Miller
>> Sent: Thursday, September 16, 2004 2:47 AM
>> To: Howard Katz
>> Cc: www-rdf-interest@w3.org
>> Subject: Re: Generated RDF conformant with good practise?
>
> 	[snip ...]
>
> So if I understand, you've inverted the relationship: instead of a bib
> containing books, you now have book as the primary object (in a non-RDF

ah, I misunderstood what a bib was, sorry, serves me right for rushing 
it. It might clarify things to have typed Bib and Book objects perhaps.

Libby

> sense), and its url predicate now points at its bib container. Is that the
> intention? if so, url seems rather generic for such a usage. Wouldn't
> something like bib or owner or container be more meaningful??
>
> Howard
>
>> hm, looking at it again, if it was me, I'd do something like this:
>>
>> <rdf:RDF xmlns:bibterm="http://www.book-stuff.com/terms/"
>>           xmlns:dc="http://purl.org/dc/elements/1.0/"
>>           xmlns:foaf="http://xmlns.com/foaf/0.1/"
>>           xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#">
>>
>>          <bibterm:Book><!--now a class -->
>>                   <bibterm:url
>> rdf:resource="http://www.book-stuff.com/bib"/>
>> <!-- I'm not sure what this url refers to -->
>>                  <bibterm:year>1994</bibterm:year>
>>                  <dc:title>TCP/IP Illustrated</dc:title>
>>                  <bibterm:author>
>>                      <foaf:Person><!--just a suggestion :) -->
>>                          <bibterm:last>Stevens</bibterm:last>
>>                          <bibterm:first>W.</bibterm:first>
>>                      </foaf:Person>
>>                  </bibterm:author>
>>          </bibterm:Book>
>>   </rdf:RDF>
>>
>> so now I've assumed the book itself is not the same as its url, but
>> that it has an associated url (e.g like a foaf:homepage) - the url is
>> not necessary by the way. I've made the book a typed node (which fits
>> better with your xml I think), and there's no need for the 'book'
>> predicate indirection from your first example.
>>
>> I hope that makes sense!
>>
>> Libby
>>
>>
>>> Here's an N-Triples view of the same information:
>>>
>>> <http://www.book-stuff.com/bib> <http://www.book-stuff.com/terms/book>
>>> _:jARP438894 .
>>> _:jARP438894 <http://www.book-stuff.com/terms/year> "1994" .
>>> _:jARP438894 <http://purl.org/dc/elements/1.0/title> "TCP/IP
>> Illustrated" .
>>> _:jARP438894 <http://www.book-stuff.com/terms/author>  _:jARP438895 .
>>> _:jARP438895 <http://www.book-stuff.com/terms/last> "Stevens" .
>>> _:jARP438895 <http://www.book-stuff.com/terms/first> "W." .
>>>
>>> My basic question, given that I'm still fairly new to RDF, is: does this
>>> look like reasonably valid RDF to people, valid in the sense of
>> not seeming
>>> too odd or unusual in some way, as well as being true to the
>> intent of the
>>> original data? I basically want to make sure I'm not producing
>> RDF that's at
>>> odds with what's considered good practise.
>>>
>>> TIA for any thoughts,
>>> Howard
>>>
>>>
>>>
>>
>
>

Received on Thursday, 16 September 2004 18:02:20 UTC