- From: Antoine Zimmermann <antoine.zimmermann@insa-lyon.fr>
- Date: Thu, 03 Mar 2011 15:49:59 +0100
- To: Pat Hayes <phayes@ihmc.us>
- CC: Jiří Procházka <ojirio@gmail.com>, Reto Bachmann-Gmür <reto.bachmann@trialox.org>, Richard Cyganiak <richard@cyganiak.de>, William Waites <ww@styx.org>, nathan@webr3.org, Ivan Shmakov <oneingray@gmail.com>, semantic-web@w3.org
Le 02/03/2011 20:52, Pat Hayes a écrit : > > [...] Bnodes have local scope, a syntactic issue rather than a semantic one. Strange. I would have said the opposite. Syntactically, scope has no importance and even has no meaning, unless maybe if you forbid a name to appear in two contexts (but this is obviously not the case in RDF). An RDF graph is a set of triples which is composed of URIs (elements of an infinite set), literals (an infinite set disjoint from URIs) and bnodes (an infinite set disjoint from URIs and Literals). Syntax-wise, this is all we have (in 2004 standard). The scope of URIs and bnodes could even be swapped (make URIs local and bnodes global) and the syntax would stay identical. Nothing in the syntax specifies the scope of these elements. The semantics does. Note that this is true in other languages like programming languages, but sometimes, programming languages have additional syntactical constraints such as the obligation to declare a variable within a context, which makes scope more explicit. But think of less constrained languages like javascript, visual basic, PHP... > [...] -- Antoine Zimmermann Researcher at: Laboratoire d'InfoRmatique en Image et Systèmes d'information Database Group 7 Avenue Jean Capelle 69621 Villeurbanne Cedex France Lecturer at: Institut National des Sciences Appliquées de Lyon 20 Avenue Albert Einstein 69621 Villeurbanne Cedex France antoine.zimmermann@insa-lyon.fr http://zimmer.aprilfoolsreview.com/
Received on Thursday, 3 March 2011 14:50:35 UTC