[Home] [By Thread] [By Date] [Recent Entries]



> Well I guess the objection is that XML already has a data model
> for keys and
> values, so building another one on top is a waste.
>
> Eg:
>
> <key name="XXX" value="YYY" />
>
> ...might be written:
>
> <XXX>YYY</XXX>
>
> The name=... and value=... is perhaps pointless when XML already
> has a way of
> associating a name with a value.
>
> It might be compared to implement key/value pairs in Java using a List of
> Maps, where each Map has keys called "NAME" and "VALUE"... as
> opposed to just
> using a single Map.

Sounds reasonable. The main reason I was curious is because I've done (more
or less) a refactoring of java.util.Properties to use RDF. A possibility was
to use a similar king of structure as you describe, with a URI for the name,
but as I only need this locally within an app, opted for a literals there,
giving the simpler:

<rdf:_1
	rdf:type='http://ideagraph.org/xmlns/idea/features#Feature'
	ftrs:name='defaultHtmlTemplate'
	ftrs:value='default.html'
/>

I'm still in two minds over this, so was quietly hoping for a 'third way'...

Cheers,
Danny.





Site Map | Privacy Policy | Terms of Use | Trademarks
Free Stylus Studio XML Training:
W3C Member