W3C home > Mailing lists > Public > www-rdf-interest@w3.org > August 2004

Re: query!

From: Richard Newman <r.newman@reading.ac.uk>
Date: Tue, 31 Aug 2004 14:03:04 +0100
Message-Id: <18A150AA-FB4E-11D8-BBE5-000A957D9564@reading.ac.uk>
Cc: www-rdf-interest@w3.org
To: go4chaitu_eng@yahoo.co.in

You would be lying. When I write an RDF file using the FOAF vocabulary, 
my use of it is essentially a contract: I subscribe to that ontology, 
which means that when I say foaf:firstName (which expands to 
http://xmlns.com/foaf/0.1/firstName), I mean the person's first name.

If you define your own ontology in that namespace, you're trying to 
speak for someone else --- like claiming your CV is at 
http://microsoft.com/ChaitanyasCV.doc. Furthermore, it won't work: a 
crawler that goes to look up the ontology/schema by hitting 
http://xmlns.com/foaf/0.1/ won't find your terms.

FOAF defines firstName. If you defined favouriteFruit as

<http://xmlns.com/foaf/0.1/favouriteFruit> a owl:ObjectProperty ;
rdfs:range <http://example.com/fruit/Fruit> ;
rdfs:domain <http://xmlns.com/foaf/0.1/Person> .

Then anyone picking up your RDF file would now have a new FOAF 
property. That's fine. However, other people who haven't got your file 
loaded would have no idea about foaf:favouriteFruit, and they couldn't 
find out by accessing the standard FOAF ontology (which is a bad thing 
for you!). If you redefined FOAF (rather than extending it), using the 
same namespace, it's the same as saying you speak English but changing 
the meaning of the words. Interoperability relies on people using 
shared terms, and using your own namespace is the chosen way of 
controlling vocabulary.

I think what you're wanting to do is mix terms. The right way to do the 
above is to put favouriteFruit into your own namespace. You can then 
reference that ontology in your FOAF files, and describe your favourite 
fruit. You don't have to only use one namespace. The following is the 
way to do it:

@prefix ex: <http://example.com/example/>
@prefix foaf: <http://xmlns.com/foaf/0.1/>

ex:favouriteFruit a owl:ObjectProperty ;
rdfs:range <http://example.com/fruit/Fruit> ;
rdfs:domain <http://xmlns.com/foaf/0.1/Person> .

:Me a foaf:Person ;
foaf:name "Richard" ;
ex:favouriteFruit ex:Apple .

There is no incentive to overloading an existing namespace that belongs 
to someone else; define your own. There is an incentive to using the 
same terms as everyone else (hence FOAF's popularity), which is how you 
get interoperability.

There's also no point in restricting namespaces. The Semantic Web is an 
enormous graph; your application should just take the slices out of it 
that it needs. If you only want to look at people, use FOAF; there's no 
reason why people can't also add other triples into the same file --- 
you can safely ignore them.

Apologies for the long, rambling reply, but I hope that covers what 
you're trying to ask.

Richard Newman

School of Systems Engineering
University of Reading

On 31 Aug 2004, at 13:05, Chaitanya Saragadam wrote:

> Hi all,
>   I sent a query few days back and didn't get a good
> answer for the same.I think i have'nt expressed it
> correctly.I will try to explain it again.
>   Take an example of FOAF Vocabulary,
>     where a <foaf:person> has some properties like
>        <foaf:mbox>,<foaf:img> etc.
>  And he gave his namespace as,
>            http://xmlns.com/foaf/0.1/
> ok,so all the rdf files with foaf tags will be used in
> searching applications of some subject like person.
> Now my question is ,
>     i will use the same namespace (ie).,
>        http://xmlns.com/foaf/0.1/
> but i will define my own properties for a person.I can
> do it right!
> And ur search engine will search my file by some
> crawling mechamism and searches for a specific
> property
> of a person which wouldnt be avaliable as it is
> defined by me.
> So, what is the purpose of writing searching
> applications if i cannt restrict the namespaces to be
> used.Is there any body governing that?
> If not no one sticks to a spec as they define their
> terms according to their applications and so their
> needs will be satisfied but not a solution for global
> applications.
> Please clarify this , even now if u cant understand, i
> will explain it still more clearly.
> But i hope u people got my point.
> Thanx for helping,
> bye.
> __________________________________
> Do you Yahoo!?
> New and Improved Yahoo! Mail - Send 10MB messages!
> http://promotions.yahoo.com/new_mail
Received on Tuesday, 31 August 2004 13:04:00 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 7 January 2015 15:07:52 UTC