W3C home > Mailing lists > Public > uri@w3.org > September 2002

Re: Possible new issue: Things with and without identity?

From: Stefan Eissing <stefan.eissing@greenbytes.de>
Date: Tue, 10 Sep 2002 18:31:13 +0200
Cc: Miles Sabin <miles@milessabin.com>, uri@w3.org
To: "Roy T. Fielding" <fielding@apache.org>
Message-Id: <B899A8A1-C4DA-11D6-9F78-00039384827E@greenbytes.de>


Am Dienstag den, 10. September 2002, um 14:46, schrieb Roy T. Fielding:

>> WebDAV working group decided to call a URL without a resource
>> (e.g. 404 on GET) an "unmapped URL".
>
> WebDAV got that wrong.  The filesystem object at the end of a request
> is not the resource -- it is only one representation of the resource.

I agree that the filesystem object must not be the resource. It can
be the resource in certain implementation. And even then, it must
not be the representation as I could imagine a translation service
on a text resource which hands out representation for different
content-languages.

Yet, there are URLs (at least on our server) which are not mapped
to anything (at a certain point in time). That's what WebDAV wants
to call a (currently) "unmapped URL". I do not see the mistake
in that.

> The mapping is the resource and the resource always exists, which is
> is why 404 Not Found does not imply that the resource does not exist,
> but rather says that no current representation of the resource exists
> at the time of the request.

While it is attractive that every character sequence following the
RFC 2396 syntax identfies a resource from sometime around 1992
to eternity, I find it problematic to derive any useful properties
from that definition. It certainly avoids the messy dependency on
time...

> WebDAV talks about it that way because, for the purpose of authoring,
> it doesn't matter whether there is a distinction between the resource
> and the current representation of that resource -- they become 
> effectively
> the same thing because WebDAV actions exist at an instant in time and
> only work on resources that are directly authorable (one-to-one
> relationship between representation and resource).

Is it impossible to model Dan's Toyota as a WebDAV resource? Maybe
one has to disallow certain operations on it...

More serious: variant handling in WebDAV is most certainly a
very interesting topic and currently not defined.

//Stefan
Received on Tuesday, 10 September 2002 12:32:08 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:25:04 UTC