UseImpliesConsent

From W3C Wiki
Revision as of 20:48, 2 November 2010 by Dbooth (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

UseImpliesConsent is the general notion that the decision to use a particular URI when authoring an RDF graph is based on the content served at that URI, and so such use of the URI constitutes a kind of consent or approval of that content.

  • Where does this idea come from? I've never seen anything in the RDF or Web specs to suggest it. Thankfully imho! --danbri
  • It comes from the "Gold Standard" argument. Seach for the text "TimBL" in this Oct 2002 e-mail. I argued that if RDF Core didn't go along with this argument, RDF could never be a "viable semantic web language" because the meanings of URIs would always be "floating". I was probably overstating the damage and the lack of alternatives. -- sandro
  • The notion of URI declaration is essentially a crystalization of the ideas behind UseImpliesConsent. -- DBooth

When considering whether and how to FollowLinksForMoreInformation, this principle arises. If each use strongly and completely implies consent, then linked-to content should simply be included as if it were in the original content. But that may raise difficulties, most obviously when then the linked-to content changes after the link is made or when one wants to talk about incorrect statements.

This is an area of practical and philosophical disagreement, particularly in the area of URIs and theories of proper names. See especially SocialMeaning and GoodURIs.

Scenario

Joe maintains an RDF file listing companies he believes have engaged in criminal wrongdoing. He links to their site in detail to help explain who they are and what they are doing wrong. One of the pages he wants to link is an RDF page in which they detail some of their activities. He wants to say that page is wrong. How does it do that?

Scenario

The W3C TR page is maintained via RDF data and cwm rules, including various metadata encoded with the Dublin Core Vocabulary. What if the DCMI decided to include in their served content claims that one of their documents was a W3C TR? Would that result in W3C publishing such a statement? If it did so, would it be incorrect in doing so?

(Probably an inconsistency can be trapped, but then what?)

Older stuff, to be refactored in

Here's the idea:

 Using a URI in RDF implies consent 
 to the theories using that URI
 which are published at that URI
 (following normal web redirections
 if needed).

Could this work? Could it be made to work?

arguments against:

 http://lists.w3.org/Archives/Public/www-rdf-comments/2002OctDec/0050.html
 http://lists.w3.org/Archives/Public/www-rdf-interest/2002Oct/0129.html

How is this different from FollowYourNose? This is much stronger.

This is an inconsistent document (using the semantics of OWL DL and UseImpliesConsent):


@prefix bio: <http://www.w3.org/2000/10/swap/test/demo1/biology#>.
_:foo a bio:Dog, bio:RedWolf.


if the web content at http://www.w3.org/2000/10/swap/test/demo1/biology says:

@prefix bio: <http://www.w3.org/2000/10/swap/test/demo1/biology#>.
@prefix owl: <http://www.w3.org/2002/07/owl#>.

bio:Dog owl:sameClassAs bio:CanisFamiliaris.
bio:RedWolf owl:sameClassAs bio:CanisRufus.
bio:CanisFamiliaris owl:disjointWith bio:CanisRufus.



This is vaguely implemented in http://www.w3.org/2000/10/swap/test/ferrell/myDingo.n3