Le 16 mars 09 à 18:27, kidehen a écrit :
>
> Alex,
>
> On Mar 16, 4:12 am, Alexandre Passant <
alexandre.pass...@deri.org>
> wrote:
>> Hi Kingsley,
>>
>> Could you please add the feature in the wiki using the new process
>> [1], thanks.
>> One issue I see with your proposal is that any item contained in a
>> wishlist will be instanciated as a sioc:Item (due to the
>> sioc:container_of relationship), and I don't really like things like
>> DVDs, records, cars, etc. being instances of such classes.
>
> Why isn't a DVD a sioc:Item ? Which is an owl:Thing? All you need to
> do is qualify the sioc:Item with an rdf:type link to an ontology that
> defines a class for describing consumer goods for instance.
>
To my mind, sioc:Item are on-line things.
A flash video file can be a sioc:Item, a webpage describing a DVD as
well, but the DVD, that you can buy in a store, is imho no more an
online object, i.e. a sioc:Item.
If everything becomes a sioc:Item, what's, at the end, its difference
with owl:Thing ?
> I am just modelling the fact that a "WishList" is a collection of
> "Things".
I agree, but currently container_of is a collection of Items
>
>
> That said, what's you alternative modelling suggestion? The more eyes
> the better as this is an important tweak to SIOC.
Well, I agree on the container, not sure yet how to represent the
inclusion of the item.
But as John suggested (if I understood well), an option is indeed to
have:
:x a sioc:Container ; a sioct:WishList ;
container_of :coolDVDPage .
:coolDVDPage a sioc:Item ;
sioc:topic :coolDVD .
:coolDVD a foo:DVD .
There's one more path in the graph, but that sounds more coherent to me.
Alex.