W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > January to March 2007

Re: HTTP URI scheme

From: Julian Reschke <julian.reschke@gmx.de>
Date: Wed, 17 Jan 2007 13:20:30 +0100
Message-ID: <45AE148E.5000209@gmx.de>
To: Jay Daley <jay@nominet.org.uk>
CC: w3c-dist-auth@w3.org, werner.donne@re.be

Jay Daley schrieb:
>> I think the important point is that having different protocol handlers 
>> may be the wrong way to approach that problem.
> 
> Here are some concrete examples of why having a separate protocol handler 
> would help:
> 
> 1.  Suppose a new browser comes out that is super-fast and I decide to use 
> it as my standard browser and so my OS is set to use this as the default 
> browser.  However this browser does not support WebDAV.

So, do you actually have a browser that "supports" WebDAV? Otherwise 
that concern seems to be really irrelevant.

> I then receive an email with a link to a collection on a WebDAV enabled 
> server, that has the form http://....  So I click on the link and my 
> browser does what it can but that is no use to me since I need the WebDAV 
> compliance.

The only platform-agnostic way to cause a specific WebDAV compatible 
client to be launched is described in RFC4709, and was written so it can 
be implemented in any user agent that allows to invoke applications 
based on MIME types. As far as I can tell, all common browsers allow 
that (normally relying on the underlying operating system's support for 
that).

> I then have to cut and paste the URL into a WebDAV compliant application 
> and access it that way.
> 
> The alternative is that the link arrives in the form webdav://.... So my 
> OS now has two defaults set up, one for the default browser and one for 
> the default WebDAV application.  I have only changed the former when I 
> installed my new browser so clicking on the link still invokes my WebDAV 
> application and it all works.

That's one alternative. RFC4709 describes another one which doesn't 
require an additional protocol.

> 2.  Suppose I am a fairly technical IT person, who has never heard of 
> WebDAV.  I receive an email with a link to a WebDAV compliant resource 
> that is of the form http://....  Well I am none the wiser.  I don't know 
> this is a WebDAV server, I don't know that I could do so much more than 
> just retrieve the document.  In fact if I receive a "Forbidden to access" 
> error then I might believe the whole thing is broken.

Why would you receive a "forbidden to access" error?

> However, suppose I receive an email with a link of the form webdav://.... 
> Well I realise I don't know what this is so I go and learn about it.  I 
> download a WebDAV compliant app, or I discover that some of my apps are 
> already WebDAV compliant.  I even start to think about how I might use 
> WebDAV inside my company. 
> 
> 
> The insistence on the http:// scheme means that WebDAV is *invisible*. 
> After a decade it is probably less well know than something like ENUM.  I 
> would suggest that the use of the http:// scheme is one of the biggest 
> factors in contributing to this invisibility.

That sort-of implies that "visibility" is an advantage. I don't agree 
here. I prefer seamless integration with the classic Web.



Best regards, Julian
Received on Wednesday, 17 January 2007 12:20:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 2 June 2009 18:44:15 GMT