[
Lists Home |
Date Index |
Thread Index
]
Arjun Ray <aray@nyct.net> writes:
> ari@cogsci.ed.ac.uk (K. Ari Krupnikov) wrote:
> | Arjun Ray <aray@nyct.net> writes:
> |
> | > Now, if INPUT can be differentiated by an attribute, why not TITLE
> | > subelements?
> | >
> | > <author name="James Vleek">
> | > <title>Contracts</title>
> | > <title role="abc">Professor</title>
> | > <title role="xyz">13</title>
> | > </author>
> |
> | Now how is this different form
> |
> | <author name="James Vleek">
> | <title>Contracts</title>
> | <title xmlns="abc">Professor</title>
> | <title xmlns="xyz">13</title>
> | </author>
>
> In this case (in-schema or in-vocabulary differentiation), no difference.
>
> Now tell me, how many colons did you use?
Hm, interesting. Let me see if I get this right.
You'd like to keep xmlns as a reserved attribute that specifies
"roles" (or whatever you might call them). You want to do away with
colonified names and context (elements don't inherit their ancestors'
xmlns attributes) and you want to make xmlns part of the visible
content model again.
So instead of /a:foo/b:bar, you'd write
/foo[@xmlns='a-role']/bar[@xmlns='b-role']
Interesting.
Ari.
|
