Article:		Something Useful This Way Comes
Subject:		"data model"?
Date:		2004-06-09 21:29:01
From:		autarch
I wasn't aware that there was a formal RDF data model, as in: 1. A collection of data structures. In the relational data model this is relations & tuples (also called tables & rows, bleah). 2. A collection of operators which can be applied to those structures in order to retrieve data from those structures, in other words a query language. 3. A collection of integrity constraints which can be used to define valid states for the data. The above definition is a paraphrased version of that given by Fabian Pascal in an article here: http://www.tdan.com/sms_issue28.htm What would these 3 things be for RDF?

Previous Message

Next Message

• "data model"?
  2004-06-10 05:27:34 Kendall Clark [Reply]
  
  I quote the RDF Concepts and Abstract Syntax standard:

  
  --
  3.1 Graph Data Model

  
  The underlying structure of any expression in RDF is a collection of triples, each consisting of a subject, a predicate and an object. A set of such triples is called an RDF graph (defined more formally in section 6). This can be illustrated by a node and directed-arc diagram, in which each triple is represented as a node-arc-node link (hence the term "graph").
  --

  
  However, I will say that I sorta ran together in my piece the RDF data model, which is a graph composed of triples, and the formal semantics of RDF, which is expressed in terms of model theory, which is a kind of logic of sets.

  
  But I think that running these together doesn't hurt my central claim, which is that, whatever one thinks of RDF's canoncial XML serialization, it has a formal semantics behind it, and that's one very big difference between it and XML plus namespaces.
  
  

• "data model"?
  2004-06-09 23:59:54 giovannitummarello [Reply]
  
  assuming that's the right definitio of "data model" (according to me it isnt) you anyway get what you want (if that makes you happy)

  
  a) data structures are triples, the datamodel is a graph. let me remind you tree is simply a specific case so XML expressibity is just a subset of RDF expressivity

  
  b) Operators? Download JENA or any other tool kit and operate as you want, standardization (See query languages) is well on the way

  
  c) ever heard of OWL?

  
  There might be no intereting semantic web based applications NOW but that doesnt mean anything.. YOU try to make a simpler way to express relations between resources in a standard way.

  
  Final word of advice , people that even consider saying "but the syntax is ugly" are making clear that they're so far away from competent the should just plain remain silent.

  
  The rdf datamodel is relatively simplex, there are many syntaxes you can chose from, the stuff its the same anyway once you load it in a computational space that has operators for it (say Jena).

  
  Finally, i have the feeling that XML people saying rdf is complicated clearly havent read their own stuff.. ( http://www.w3.org/TR/xmlschema-1/ http://www.w3.org/TR/xmlschema-2/ )
  
  

  • "data model"?
    2004-06-14 07:28:42 bryan rasmussen [Reply]
    
    "a) data structures are triples, the datamodel is a graph. let me remind you tree is simply a specific case so XML expressibity is just a subset of RDF expressivity"

    
    I really love the whole "a graph is more expressive than a tree" thing, especially as regards rdf which is generally serialized as xml.

    
    So you see, you can express any tree structure in this graph structure which we've expressed in a tree structure.