W3C home > Mailing lists > Public > www-webdav-dasl@w3.org > October to December 2003

Re: SEARCH by last path segment, Was: SEARCH for displayname

From: Julian Reschke <julian.reschke@gmx.de>
Date: Wed, 19 Nov 2003 17:33:56 +0100
Message-ID: <3FBB9B74.5070705@gmx.de>
To: Lisa Dusseault <lisa@xythos.com>
Cc: "'Wallmer, Martin'" <Martin.Wallmer@softwareag.com>, 'Kevin Wiggen' <kwiggen@xythos.com>, www-webdav-dasl@w3.org

Lisa Dusseault wrote:

> Case 1.  If "hiddenness" is a property of a resource, then 
>  a) you need write permission on the resource to change the is-hidden
> property
>  b) the property value stays with the resource when it is moved or renamed
>  c) the resource is hidden in every collection to which it is bound, because
> bindings merely expose the properties of their resources.
> Case 2.  If "hiddenness" is a property of a collection's membership data,
> then 
>  a) you need write permission on the *collection* in order to change the
> value.  
>  b) When you MOVE or rename a resource it would no longer be hidden because
> that was a property of its source membership.  
>  c) one binding of a resource may be visible, and another hidden.

I don't agree with b). A server would be free to not only create a new 
binding and remove the old, but also to re-create as much information as 

> Case 3.  If "hiddenness" is a property of a binding, then
>  a) you need write permission on ?? to change it
>  b) When you MOVE/rename a binding, its hidden value would go along with it
>  c) One binding of a resource may be visible and another hidden
> For MOVE and rename functionality, and the flexibility of one binding being
> hidden and another not, I prefer Case 3.  However I agree the ACL issue is
> problematic and unless we come up with a general way to deal with it we
> might prefer Case 1 despite the fact that it has less flexibility.
> I do not prefer case 2 because I wouldn't expect a rename of a resource, or
> a MOVE to another collection, to "unhide" it.  Yet that's the natural
> implementation if hiddenness is a property of the collection's membership
> list.
> If hiddenness is a property of the collections membership list but it *does*
> go along in a MOVE, then we have somewhat of a hybrid model, IMO.  Or if
> moving to another collection behaves differently than a rename, we may have
> a problem with MOVE.

How would it? There is no distinction between moving and renaming in WebDAV.

Anyway, this really has little to do with DASL, so if we want to pursue 
this further, I'd suggest doing that on the generic WebDAV mailing list.


<green/>bytes GmbH -- http://www.greenbytes.de -- tel:+492512807760
Received on Wednesday, 19 November 2003 11:35:12 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 20:22:43 UTC