Overloaded URIs (was Re: XLink: behavior must go!)

[James Tauber]

>One possible solution would be to define a new URL protocol that
>builds on HTTP URLs -- not as fancy as URNs, but they would build on
>something that exists and is well understood:
>
>  hname://www.megginson.com/ns/


I suggested this to TimBL at WWW8 but he didn't buy it. My argument went
something like this:

1. Namespace URIs don't have to point to anything retrievable, they are just
unique identifiers
2. Some URI schemes might not be web-retrievable (eg "isbn:1-23456-789-0")
3. Some URI schemes *are* web-retrievable (eg http://www.megginson.com/ns/)
4. So namespaces URIs *can* point to something, though, such as a
human-readable description of the namespace, if they use a URI scheme that
is web-retrievable.
5. If this is done, how does one distinguish referencing the namespace from
referencing the file describing the namespace
6. My suggestion: come up with a new URI scheme that has the same
scheme-specific syntax as an http URI but that is only intended as a unique
identifier.

TimBL suggested there wasn't a distinction to be made (as did Rohit Khare
who cited Roy Fielding's PhD thesis). I'm still not convinced. Comments?

James


xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:majordomo at ic.ac.uk the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)