W3C home > Mailing lists > Public > w3c-dist-auth@w3.org > January to March 2004

Re: locking clarifications/extensions vs BIND draft vs RFC2518bis

From: Elias Sinderson <elias@cse.ucsc.edu>
Date: Mon, 02 Feb 2004 11:30:50 -0800
Message-ID: <401EA56A.8040202@cse.ucsc.edu>
To: w3c-dist-auth@w3.org

+1, for much the same reasons outlined in Geoff's post below.

Q: 2518 defines two levels of compliance, depending on whether locking 
is supported or not - would it make sense to try and extend this 
approach to cover all of the related specs?

Cheers,
Elias

Geoffrey M Clemm wrote:

>I agree with Julian, and strongly advocate approach #1
>(take locking out, keep bind out, make everything more modular, for the 
>following reasons):
>
>- WebDAV is already a family of specs (3253, ACL, redirect, ordering),
>each of which defines an optional feature-package beyond what is defined
>in the base spec.  It would be more consistent to handle locking
>(which is an optional feature-package) the same way.
>
>- Having a smaller "base WebDAV spec" I believe will make WebDAV more
>accessible to new implementors, since the base spec will be less daunting 
>in
>size.  You don't have to read/understand the locking extensions to
>understand versioning, ACL, redirect, or ordering, but the current 
>packaging of locking in with the base protocol makes it look like you do.
>
>- It allows us to make more rapid progress on getting the locking
>functionality standardized (i.e. it doesn't have to wait until we've
>resolved all the other issues in 2518bis).
>
>Cheers,
>Geoff
>
>
>Julian wrote on 01/17/2004 03:52:29 AM:
>
>  
>
>>In an off-list mail, Jim Whitehead wrote:
>>
>> > I'm tempted to just put BIND right into 2518bis -- worst case
>> > we recycle at
>> > Proposed, which I don't see as being a major adoption
>> > impediment anymore (we
>> > could perhaps call it WebDAV v2, to make it clear that we're making
>> > progress).
>>
>>Well, I think that would be a radical change to our strategy...
>>
>>Seems that opinions vary between
>>
>>1) take locking out, keep BIND out, make everyhing more modular and
>>2) keep locking in, add BIND, publish RFC2518+BIND (staying at the 
>>"proposed level")
>>
>>I'd definifively vote for 1).
>>
>>Julian
>>
>>-- 
>><green/>bytes GmbH -- http://www.greenbytes.de -- tel:+492512807760
>>
>>    
>>
Received on Monday, 2 February 2004 14:31:20 UTC

This archive was generated by hypermail 2.3.1 : Wednesday, 5 February 2014 07:17:51 UTC