Konqueror WebDAV integration question

[Julian Reschke]

Bert Bos wrote:
> On Wednesday 07 September 2005 15:23, Julian Reschke wrote:
>> Thiago Macieira wrote:
> 
>>> If that is so, this proposal of yours will allow the servers to
>>> tell clients, but it doesn't allow one other benefit of the
>>> "webdav" scheme we use in KDE: allow the user to decide.
> [...]
>> 1) Well, I agree it may be useful; and I didn't propose to take that
>> away. On the other hand, this is an unregistered URI scheme, and it
>> only exists one one particular platform.
>>
>> 2) The whole point of this format is that you don't have to go
>> through these steps manually; instead just click on a link and the
>> WebDAV view of things will open, no matter what platform you are on.
> 
> I agree with Julian that there should be a standard solution, a way to 
> link to a document on a WebDAV server that does not depend on the 
> browser you are using.
> 
> But Julian's solution (a davmount file with a single HTTP URL in it) 
> seems rather less practical than KDE's (a "webdav:" URI scheme). 
> Julian's solution requires you to make an extra file, just to be able 
> to attach a second MIME type to a document that already has one.

No, it doesn't require a second file. It requires a second *resource*, 
which a WebDAV server in general can generate dynamically.

> It also requires an extra round-trip to the server: first to get the 
> small file, then to get the actual document.

So? I don't think that one additional round trip is a problem when the 
subsequently activated client is going to submit lots of other requests.

> And WebDAV is a protocol, so a URI scheme definitely feels more natural 
> than a MIME type.

WebDAV is an HTTP extension, not a new protocol, so it really makes no 
sense to define a new URI scheme. See 
<http://www.w3.org/TR/2004/REC-webarch-20041215/#URI-scheme>.

> Julian mentions W3C's Web Architecture document[1] in his ID[1]. W3C 
> indeed says that you should not unnecessarily make new URI schemes 
> (because it is costly, all Web software will have to be updated). But:
> 
>   1) A new MIME type is just as costly.
 >
>   2) Existing URI schemes and MIME types do not provide what's needed.

 From the above mentioned document:

"When designing a new data format, the preferred mechanism to promote 
its deployment on the Web is the Internet media type (see Representation 
Types and Internet Media Types (§3.2)). Media types also provide a means 
for building new information applications, as described in future 
directions for data formats (§4.6)."

In particular, I disagreee that a new MIME type is as costly. User 
agents are designed to handle new MIME types, not new protocols.

> So in my view, it is acceptable to create a new URI scheme in this case.
> 
> (One caveat: I know very little about WebDAV, even though W3C has 
> implemented a server and I'm using it every once in a while.)
> 
> 
> As W3C's document also says: don't use URI schemes without 
> registering[3] them with IANA! So the ideal solution, in my mind, would 
> be if Julian and the inventors of "webdav:" got together and submitted 
> a specification for a URI scheme to the IETF.

Well, I personally feel that this would be a Very Bad Idea, so I'm 
certainly not going to do that...

> [1] 
> http://www.ietf.org/internet-drafts/draft-reschke-webdav-mount-00.txt
> [2] http://www.w3.org/TR/2004/REC-webarch-20041215/
> [3] http://www.iana.org/assignments/uri-schemes

Best regards, Julian