W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > April to June 2008

Re: Thoughts on relation to WebDAV

From: Werner Baumann <werner.baumann@onlinehome.de>
Date: Sun, 25 May 2008 11:14:38 +0200
Message-ID: <48392DFE.6080506@onlinehome.de>
CC: w3c-dist-auth@w3.org


Petr Tomasek wrote:
> On Sat, May 24, 2008 at 06:13:54PM +0200, Werner Baumann wrote:
>> What the hell is a "read-only WebDAV implementation"? WebD*A*V is about 
>> *Authoring*.
> 
> Sorry, but this is silly! According to such an interpretation one
> couldn't use e.g. the HTTP protocol for anything else, than a hypertext!
> 
> Rather, the WebDAV is most commonly used as a network filesystem
> implemented over HTTP and as such it makes much sense to have
> a read-only filesystem.
>
WebDAV is not a network file-system and never was intended to be. Please 
read the documents, e.g.
ftp://ftp.rfc-editor.org/in-notes/rfc2291.txt and
ftp://ftp.rfc-editor.org/in-notes/rfc4918.txt

I am maintaining a WebDAV-file-system (davfs2, which is not a network 
file-system but an intermediate solution for authoring tools without 
built-in WebDAV-support). My experience with this work tells me:
It is impossible to cleanly map WebDAV-resources into a 
Unix-file-system. Though I am not really happy with RFC 4918, this point 
is not the fault of the specification, because it was never intended for 
this use.

It has become common practice, to reuse existing protocols for new 
applications. That's OK. But when it turns out, that the chosen protocol 
does not really meet the requirements of this application, there is the 
bad habit of tweaking the original protocol, up to the point where it 
gets unusable for its original intention. This must stop.

"preaching to the choir" again
Werner
Received on Sunday, 25 May 2008 09:15:29 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:44:16 GMT