Representing things RE: RDF and speech acts

Hi Leo,

actually FOAF doesn't use the same mechanism as WordNet.

My understanding of best practice is that a bare URI will often be understood
to refer to the thing that gets returned - i.e. the page. So if you want to
use the wordnet entry for love, it is better to define some "definition"
fragment for all wordnet, such as

and this can clearly identify anything you want, but doesn't mean "the page

FOAF gets around the desire to do this by using a blank Node to identify a
Person -

the person whose email address is

Or in RDF

<rdf:RDF xmlns="I forget right now"
         xmlns:foaf="the foaf namespace">
  <foaf:mbox rdf:resource=""/>

I think for foaf:interest it suggests you use a URI though. The indirection
trick here is neat in that it is relatively easy to apply:

[concept] [described by] [the relevant Wordnet page]

but may be a bit clunkier to use.

(There's a fair bit of discussion around on this topic...)



On Tue, 25 Nov 2003, Leo Sauermann wrote:

>I would be interested in that, too. We are missing some philosophical
>stuff here. If you find anything, please post it here.
>I remember the "URI crisis", that tackles the question of "what does a
>URI identify?". Especially the problem of "how do i represent love?"
>Here is something cut/n/pasted from my diploma thesis about identifying
>general concepts (like "red"):
>A URI can be used to identify an abstract concept. Again we have an
>example, the identification of  the concept of "love". A solution here
>would be to use WordNet  identifiers for the meaning of English words,
>as Dan Brickley suggested in [Brickley2001]. According to his definition
>(and a correction by Libby Miller [Miller2003]), "love" could be
>expressed with this URI.
>Figure 6 The "Love" URI
>Some common concepts can be identified with this method, Dan Brickley
>and Libby Miller used it to identify the concept of a "person" in their
>FOAF, project. [FOAF].

Received on Tuesday, 25 November 2003 12:17:33 UTC