W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > October to December 2005

Re: ETags, next call, was: Notes on call from today ...

From: Julian Reschke <julian.reschke@gmx.de>
Date: Thu, 24 Nov 2005 21:51:41 +0100
Message-ID: <438627DD.9010706@gmx.de>
To: Cullen Jennings <fluffy@cisco.com>
CC: Geoffrey M Clemm <geoffrey.clemm@us.ibm.com>, WebDav <w3c-dist-auth@w3.org>, w3c-dist-auth-request@w3.org

Cullen Jennings wrote:
> 
> So the usual IETF thought pattern around profiles is say we have might 
> have profile A and B. Are their clients that want both? Are there 
> reasons a server would only support one? How will this negation and if a 
> server only does A and a client wants B, what will the interoperation be 
> like?

Let's call that profile "filestore".

This profile would be a true subset of WebDAV. A server such as Slide 
(to take a real-world example) might want to advertise "filestore" on 
some resources (such as in a FS backend), but not others (such as a 
Tamino XML database).

A client that would absolutely need the features defined as part of the 
"filestore" would simply refuse to interop with some of the resources on 
the server.

IMHO that would still be far better than to forbid servers to implement 
useful WebDAV stuff on resources like these (because that's what a MUST 
or even a SHOULD requirement would essentially mean).

We also should keep in mind that we're in HTTP's area here, because 
we're talking about PUT. We simply can't overrule HTTP here. We *can* 
define a way for servers to advertise that their PUT support fulfills 
certain requirement, so that clients can find out.

Best regards, Julian
Received on Thursday, 24 November 2005 20:53:02 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:44:11 GMT