Re: namespaceDocument-8 background

Norman Walsh scripsit:

> The technical hurdle we face is that RDDL relationships are naturally
> quads:
> 
>   (namespace name, nature, purpose, resource)

I don't think they are, really.  The nature(s) of a resource are
independent of the purpose(s) they serve with respect to some
namespace(s).  So this naive model is not in second normal form.

> The current finding proposes a graph like the following:
> 
> 
>  +------------+  some-purpose  +----------+  nature  +--------+
>  | namespace  |--------------->| resource |--------->| format |
>  +------------+                +----------+          +--------+
> 
> That is, a namespace is related by some purpose to a resource which
> has a RDDL nature. 

I think this model is entirely correct, except that it should allow
multiple "nature" arcs from a resource.  Purposes are RDF verbs, natures
are RDF objects.

> I believe very strongly that the graph has to be small and simple. If
> we wind up with a huge, hairy model, we'll never be able to make it
> fly.

+1

> There is something asymmetric about the current model with a variety
> of "some-purpose" arcs and exactly one "nature" arc.

There's no reason why there shouldn't be multiple natures.  A document may
have the HTML nature, the HTML 4 nature, and the HTML 4 Strict nature.
Likewise, it might have the (hypothetical) plain-text nature as well as
the RFC nature.

> I can think of two other models that seem plausible to me:
> 
> 
>  +------------+  ancil-resource  +----------+  nature  +-------------+
>  | namespace  |----------------->| resource |--------->| some nature |
>  +------------+                  +--------+-+          +-------------+
>                                           |
>                                           |   purpose  +--------------+
>                                           +----------->| some purpose |
> 				 	               +--------------+

This model fails to associate the purpose with the namespace correctly;
a resource (a document whose nature is "RELAX NG Schema", say) might
serve the purpose "validation" for one namespace, and another purpose
such as "non-normative reference" for another.  Your second model is more
complicated but doesn't solve this underlying problem.  By using the
particular purpose to link the namespace to each resource as above,
the problem goes away.

-- 
Your worships will perhaps be thinking          John Cowan
that it is an easy thing to blow up a dog?      http://www.ccil.org/~cowan
[Or] to write a book?
    --Don Quixote, Introduction                 cowan@ccil.org

Received on Saturday, 10 March 2007 19:34:46 UTC