Re: Namespaces -- Out of scope of OWL?

minor correction
Turner, David wrote:

> If I use a namespace 'ns', in some places in the XML the entity '&ns;'
> has to be used; 

I may want to use an entity '&ns;' - there is a style of writing 
RDF/XML, supported by some tools (e.g. Jena XML output module), in which 
all the namespace prefixes are also defined as entities.

> if the namespace is called 'amp' then you cannot use
> '&' because that entity is already defined. I'm led to believe that
> corner-cases to do with namespaces are a recurring support issue in
> Jena, which suggests that getting them correctly specified would be a
> nightmare, and that it's safer to leave it up to implementors to worry
> about them.

In this particular case, jena currently simply does not use that 
particular entity. A different implementation may opt to rename the 
prefix from amp to ampx, so that this idiom works.

There are many many more examples of where the level mismatch between 
namespaces and ontologies would cause problems, if implementations were 
not free to discard or rename namespace prefixes.
In practice, users don't like this, so implementations are under 
significant pressure to keep this to a minimum.

Jeremy

> 

-- 
Hewlett-Packard Limited
registered Office: Cain Road, Bracknell, Berks RG12 1HN
Registered No: 690597 England

Received on Friday, 20 April 2007 09:47:07 UTC