W3C home > Mailing lists > Public > public-rdf-comments@w3.org > June 2013

Re: bNodes as graph identifiers

From: Sven R. Kunze <sven.kunze@informatik.tu-chemnitz.de>
Date: Wed, 05 Jun 2013 16:02:25 +0200
Message-ID: <20130605160225.Horde.ZejVB3ybQqB8aUrAJiAjoA1@mail.tu-chemnitz.de>
To: public-rdf-comments@w3.org
Zitat von William Waites <wwaites@tardis.ed.ac.uk>:

> On Wed, 05 Jun 2013 09:36:23 -0400, Kingsley Idehen  
> <kidehen@openlinksw.com> said:
>
>     <http://tu-chemnitz.de/sven> { <http://tu-chemnitz.de/sven#i>
>
> Yes, one could do that, but it only works neatly with the # convention
> for the http-range-permathread issue, and my point was that many
> people, perfectly reasonably, don't do this.
>
> -w

Thank you, William; I see your point.
However, I must admit that this argument of disjoint sets also would  
apply in "plain" RDF.
Yet, we can conceive such statements like
:s a :Pet; a :Car; a :Table.
They are perfectly valid and if someone use RDF in such way (well I  
wouldn't go so far and say mis-use, because we aren't so great as to  
foresee the future and declare that there won't be any car that can be  
a table and a pet simultaneously).

When there is such thing as:
[] a rdf:Graph, a foaf:Person.
so be it and we inherit all the power we already have in RDF tools and  
RDF vocabs.

Furthermore, it would, well, not force but push people to properly  
design their domain data. Yet, their are free to choose whether a  
Person is the same entity as the graph describing it. Thus, I do not  
see a problem with that use case.
Received on Friday, 7 June 2013 06:41:03 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:29:57 UTC