Bizer & Carroll
Trust + TriX
Semantic Web Interest Group
1-2 March 2004
What is a context?
Minimalist answer:
The context of a statement is its surrounding graph
log:implies) 
  A uriref or a blank node is interpreted as a graph.
Syntax
 G1 (Monica ex:name  Monica Murphy . 
      Monica rdf:type ex:Person . 
      Monica ex:homepage  )
 _:G3 (G1 dc:author Chris. 
      _:G3 dc:author Chris )
  
 Also TriX (Carroll-Stickler), N3
SELECT ?x, ?y 
WHERE 
    ?a (?b rdf:type ex:Person . 
        ?b rdf:hasName ?x . 
        ?b ex:homepage ?y) 
    (?a dc:date ?c) 
AND ?c > "1/1/2003"
G1 (Monica ex:name  Monica Murphy . 
    Monica rdf:type ex:Person) 
G2 (G1 dc:author Chris. 
    G1 dc:date  2/10/2004 ) 
G3 (Monica ex:hasSkill ex:Programming) 
G4 (G3 dc:author Peter. 
    G3 dc:date  2/3/2004 )
SELECT ?x ?y ?z 
WHERE ?a (?x ?y ?z) 
     (?a dc:author Chris)
G1 (Monica ex:hasStatus Admin. 
    Monica rdf:type ex:Person. 
    G1 ex:author Andy. 
    G1 dc:date  2/10/2004 ) 
G2 (G1 ex:hasSignature  xd2shfl22k4jdsre& . 
    G1 ex:Signer Andy) 
G3 (Andy ex:publicKeyURL http://bla.bla.bla)
SELECT ?a ?b ?c ?d WHERE ?d (Monica ex:hasStatus Admin) (?d dc:signer ?a) (?a ex:publicKeyURL ?b) (?d ex:hasSignature ?c)
Unconvincing?
 A consistsOf Metal . 
B consistsOf Water . 
{ Metal denserThan Water } log:implies { A sinksIn B } .
{ Mass a FixedValue } in NewtonianMechanics . 
{ Mass a Variable } in RelativityTheory . 
{ RelativityTheory approximates NewtonianMechanics } 
     when { RelativeVelocities lessThan halfC } .
{ <http://example.org/liar> 
     log:implies { 
       <http://example.org/noone> a owl:Nothing .
     } . 
} owl:sameAs <http://example.org/liar> .
<http://example.org/liar> a log:Truth ?
Workaround for de dicto "ex:thisURI"^^xsd:anyURI
G1 (Monica ex:hasStatus Admin. 
    Monica rdf:type ex:Person. 
    G1 requiresAccessRight Admin)
   
  G1 (Monica ex:name  Monica Murphy . 
    Monica rdf:type ex:Person. 
    Monica ex:hasSkill ex:Programming. 
    G1 dc:author Chris) 
G2 (Monica ex:hasSkill ex:Programming) 
G3 (G2 ex:truthValue ex:false) 
G4 (G2 dc:author Peter. 
    G3 dc:author Peter)
  
  SW statements are claims not facts.
We evaluate claims pragmatically, based on who said them, in what context, etc.
Reputational info, and digital sigs helpful but neither necessary nor sufficient.
"Distrust everything a vendor says about its competitor."
"Believe information which has been stated by at least 5 independent sources."
Do not use document URL to identify either a graph or a graph set.
Use explicit names for graphs
Use RDF collections and containers for graphsets
N3 - yes
TriX - no
Yes, but not blank predicates?
File