[topicmapmail] Re: Logicians do not rule the world (fortunately)

[Lars Marius Garshol]

* jsinger at meaningmachines.com
>
> Is there no definition of "symmetry" in the context of Topic Maps  
> agreed upon by the Topic Maps community?

There's no piece of ISO or other scripture that defines this term in  
the context of Topic Maps, no.

> Does each Topic Maps query language have a unique treatment of an  
> association with indistinguishable roles, i.e. whether
>
> 	borders-with(norway : neighbour, sweden : neighbour)
>
> implies
>
> 	borders-with(sweden : neighbour, norway : neighbour)?

Well, the data model says that

   (a) the roles within the association have no defined order, and

   (b) these two associations are equal (same type, same set of roles)
       and will therefore merge the two. So you'll be left with a single
       association.

Also, a query language that given the query

   borders-with(norway : neighbour, $N : neighbour)?

failed to provide "sweden" as one of the values for $N (regardless of  
which of your two cases was the input) would have to be severely  
brain-damaged.

It's conceivable that you could create a query language that given  
the query

   borders-with($N1 : neighbour, $N2 : neighbour)?

did not respond with both combination for any of those two inputs,  
but I think it would be clear to everyone that this would not be a  
very good query language.

--
Lars Marius Garshol, Ontopian               http://www.ontopia.net
+47 98 21 55 50                             http://www.garshol.priv.no