Re: Domain of :key

On 3/29/13 1:32 PM, Melvin Carvalho wrote:
>
>     dbPedia:EiffelTower mlv:key key .
>
>     is a sentence that your ontology allows since the domain is
>     anything. So what would it mean
>     for a shoe, the eiffel tower, or an atom to have a key? How would
>     I know it is true or if it is false?
>
>
> there's probably an uncountable number of triples that are ambiguous 
> or nonsensical
>
> what does it mean for a rainbow to have a smile?
>
> isnt it an exercise for the reader?
>
> you can use the sem web to write nonsense, that can never be 
> forbidden, what does it prove?

Melvin,

Let's break this down for clarity.

entityX ends up in a relation with a Public or Private key due to the 
following events:

1. keypair generated
2. certificate generated -- following signed CSR re. usual cert. 
generation flow
3. entityX is the cert. Subject -- the cert. graph is basically a 
collection of claims about entityX.

entityX has to have the capacity to perform 1-3. Thus, entityX is a 
foaf:Agent and thereby defines the domain of the "key generation" 
relation, for it to be effectively usable.

The issue here is (I suspect) a little cross communication. I think you 
want to associate a Certificate (which is a Web Resource) with Keys, and 
if so, see our ontology [2].

Links:

1. 
http://id.myopenlink.net/describe/?url=http%3A%2F%2Fwww.w3.org%2Fns%2Fauth%2Fcert%23key
2. 
http://id.myopenlink.net/describe/?url=http%3A%2F%2Fwww.openlinksw.com%2Fschemas%2Fcert%23 
-- our specific certificate ontology which meshes with the cert. ontology .


-- 

Regards,

Kingsley Idehen	
Founder & CEO
OpenLink Software
Company Web: http://www.openlinksw.com
Personal Weblog: http://www.openlinksw.com/blog/~kidehen
Twitter/Identi.ca handle: @kidehen
Google+ Profile: https://plus.google.com/112399767740508618350/about
LinkedIn Profile: http://www.linkedin.com/in/kidehen