Re: Summary: RFC

On Sat, 20 Nov 1999, Stefan Decker wrote:
> sure.
> Entity-relationship modelling is capturing the relations between
> generic objects (classes).
> RDF however, unifies the instance and the class level.
> 
> So i don't directly see, how ER-modelling is applicable with RDF.
> And i don't see that it is conceptually simpler than RDF.

One perspective is that RDF's terminology and specification style
obscures the similiary between RDF and the ER model. 

RDF says that there are only entities (resources) and their relationships
(properties). That relationships/properties are a special type of
resource, as are these classes themselves. When we talk about reification
and resources and predicate/subject/object and properties and statements and
rdf descriptions, we risk obscuring the essential simplicity: the RDF
information model is just the Web information model. On the Web,
everything is a resource and nameable with simple text identifiers in an
agreed format (URIs); in RDF we simply call out that resource types 
themselves, and their typed inter-relationships are also interesting
classes of resource and namable with URIs. After that, the {resource,
relationship, resource} information model of the Web itself is easily
recycled for data representation and interchange.

So, while classical E-R may or may not be 'simpler', we can certainly make
life simpler for developers by making more explicit the similarities (and
differences) between RDF and (amongst other things) E-R.

Dan

--
danbri@w3.org