[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]
Re: [xml-dev] Trailing slash on a namespace that is a url
- From: Philippe Poulard <philippe.poulard@sophia.inria.fr>
- To: Andrew Welch <andrew.j.welch@gmail.com>
- Date: Tue, 24 Mar 2009 17:06:55 +0100
Andrew Welch a écrit :
> 2009/3/24 Chris Burdess <dog@bluezoo.org>:
>> Jirka Kosek wrote:
>>>> (They do have something at the end of them, and if you visit any of
>>>> them in a browser, the trailing slash gets added in the address bar...
>>>> not a big deal, but has potential for confusion)
>>> Note that the slash is added by web server not browsers -- for example
>>> in XHTML case the namespace document is placed in /1999/xhtml directory
>>> on the web server which means that web server issues redirect to
>>> /1999/xhtml/ to get directory index. Namespace document has to be placed
>>> to /1999/ directory and named xhtml or some rewrite rules has to be
>>> applied on the web server.
>>
>> There is no requirement that namespace URLs resolve to a valid resource,
>> whether via HTTP or otherwise. Namespace URLs are not intended to be
>> resolved, they are opaque identifiers.
>
>
> Sure, but sometimes resources related to the markup are placed at the
> end of the url, and that's quite nice for everyone new to it.
For years, I have suggested to use URNs, but now I find very convenient
to put something pointed by an URL.
We have tried that at INRIA with a "namespace server" which purpose was
to supply informations for humans or programs... well, it was the basic
idea but so far, nothing really interesting was done, the big question
is what would be useful for programs ? Thus, we only have web pages for
humans, and some users also have put related resources. The next step
would be serving informations in several formats, maybe with RDF+XSLT,
GRDDL or microformats, that would describe the actual namespace and
related resources (specifications, schemas, stylesheets, etc). A kind of
catalog. It could also be used to identify and describe our applications
and to specify some entry points, but isn't that what UDDI is supposed
to do ?
> If that technique is used (such as the w3c namespaces) then I wonder
> if it's good idea that the address bar and namespace don't match...
Because of URL rewriting by the server, the referred document *always*
indicates the actual namespace to avoid confusion with what is presented
by the browser in the adress bar. This is what is suggested in our
"namespace server" with other naming conventions.
--
Cordialement,
///
(. .)
--------ooO--(_)--Ooo--------
| Philippe Poulard |
-----------------------------
http://reflex.gforge.inria.fr/
Have the RefleX !
[Date Prev]
| [Thread Prev]
| [Thread Next]
| [Date Next]
--
[Date Index]
| [Thread Index]