W3C home > Mailing lists > Public > www-rdf-interest@w3.org > January 2005

Re: xml without rdf, but with an ontology [0]

From: Henry Story <henry.story@bblfish.net>
Date: Fri, 14 Jan 2005 18:01:42 +0100
Message-Id: <F72520FC-664D-11D9-9C7C-000A95D9FA7A@bblfish.net>
Cc: www-rdf-interest@w3.org, atom-owl@googlegroups.com, jsled@asynchronous.org
To: users@bloged.dev.java.net


On 14 Jan 2005, at 17:07, Ian Davis wrote:

> On 14/01/2005 15:40, Henry Story wrote:
>> so we can find the definition of foaf:homepage at
>> http://xmlns.com/foaf/0.1/#homepage
>
> There's no requirement from the web that this should be true. What if 
> the extension namespace were urn:foo:1234567:foaf ?

Then clearly it will be more difficult to work with that xml. But that
will be true in any case.
If I write an xml schema and place it on my laptop and don't place it
in an obvious location, and ask people to fill in lots of forms before
I send it to them, then if it is anything complicated, it is likely 
that few people will use it or bother to understand it.

>> [snip]
>>> So, there need to be some sort of structural rules that define what  
>>> types of triples should be produced. These rules should be 
>>> consistent  and work from the instance document. For the specific 
>>> case of Atom the  following suggested equivilences between markup 
>>> and NTriples might  serve:
>>>
>> Before I go into these examples, can you tell me how my proposal does
>> NOT deal with them?
>
> 1. It assumes that every extension will have an OWL schema at the 
> namespace URI. I don't believe that extensions will have OWL schemas 
> in general, or that they will be always accessible at that URI.

No. It just makes it a lot easier to use that xml. If there is no 
ontology
available (and we don't want to write one out) then we can still get the
minimal graph full of blank nodes I showed in my example. Not very 
useful graph,
but neither will it be any more than what we are allowed.

> 2. It also handles different instances of the same extension in 
> different ways. I'm sure you've already done this but can you list the 
> production rules for triples in the scenarios I listed. Assume that 
> <extension> is typed as a property in an OWL schema.

((that is what I would always do ;-))

> Imagine then that the extension contains contact information for an 
> author but your software can't know this up front so it has to work 
> out how to display it. What query do you write against the triple 
> store to get this information?

Can I take the foaf ontology, or any other well known ontology, as an 
example? And assume that I have some atom xml that uses those 
ontologies? And that I have a BlogEditor that does not understand 
ontologies but knows how to get the ontology?


Henry


> Ian
>
> ---------------------------------------------------------------------
> To unsubscribe, e-mail: users-unsubscribe@bloged.dev.java.net
> For additional commands, e-mail: users-help@bloged.dev.java.net
>
Received on Friday, 14 January 2005 17:01:45 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:52:12 GMT