[
Lists Home |
Date Index |
Thread Index
]
Simon St.Laurent wrote:
> At 10:14 PM 8/2/2002 -0400, you wrote:
> >ok i'll bite:
> >
> >the URI reference identifies the namespace,
> >in other words
> >the namespace is being named
>
> And that buys us what if the URI is most easily resolved to a porn site?
>
> (Deliberately perverse, yes.)
That would be a deliberately confusing choice of a URI. I'd say: use another
namespace name.
>
> Naming a namespace is _nearly_ as exciting as identifying a resource which
> is a thing that can be identified.
>
> If the URI can be resolved to a RDDL document which purports to be a
> representation of the namespace, I'm happy to buy your argument. If it
> doesn't or doesn't have to, I think I'm off the train completely at this
point.
>
There are two different points here.
The first is using URIs to identify namespaces. URIs have the property of
allowing anyone with a DNS entry to create and manage their own namespace
names. There are other mechanisms of allocating unique URIs (and hence
namespace names) such as urn:uuid's.
Such URIs as namespace names are used, purely as specified by XML
Namespaces, as a syntactic device to disambiguate XML elements, i.e. as a
perfectly good two part naming system.
The second point is that some classes of URIs may be dereferenced e.g.
"http" URIs. Such URIs may be given representations. Now for the general
case, for example with a "urn:uuid:xxx" URI used as a namespace name, it is
just confusing to say that this is a string that happends to have the syntax
of a URI. For the second case, namely "http" URIs, then Yes, the expectation
is that the URI will resolve to a representation of the namespace.
So if your namespace name resolved to a porn site, as in your example, I'd
say that this is a _contradiction_. The user of the namespace, i.e. the user
and/or creator of the XML document is asserting that the namespace URI
identifies a namespace. The owner of the URI is asserting that the URI
represents a porn site. I assert that namespace names and porn sites are
disjoint, hence the contradiction.Contradictions don't give us a logical
problem, they are just wrong. '3 = 2' is wrong that's all.
Jonathan
|
