- From: Yaron Goland <yarong@Exchange.Microsoft.com>
- Date: Thu, 10 Feb 2000 23:50:21 -0800
- To: "'w3c-dist-auth@w3.org'" <w3c-dist-auth@w3.org>
- Message-ID: <7DE119D3D0E15543874F7561EECBDBED0261A16A@BEG.platinum.corp.microsoft.com>
The first sentence of the first paragraph of section 7 reads: "A URI of a redirect reference resource can be an internal member URI of a collection just as the URI of a non-reference resource can. " Given the previous comments it would seem that this sentence needs to be rewritten to specify that all the resources involved must be WebDAV compliant. It is not enough that the parent of a redirect resource is a WebDAV compliant collection. The redirect resource itself needs to be WebDAV compliant. However, specifying all this would only reproduce the namespace requirements stated in RFC 2518, requirements that apply to all WebDAV compliant resources regardless of their type. Hence a reasonable reader is likely to draw the conclusion, based on this sentence, that RFC 2518 is wrong and that there do potentially exist WebDAV compliant resources that are the children of WebDAV compliant collections that may not be internal members of that collection. After all, were this not so, it would not be necessary to explicitly state that redirect resources can be the child of a WebDAV collection. As this sentence seems to clarify something that does not require clarifying and as it does it in a manner likely to produce confusion. I move that it be deleted. The last sentence of the first paragraph of section 7 reads: " The methods that can accept a Depth header are PROPFIND, COPY, MOVE, DELETE, and LOCK." As a general rule of thumb it is a bad idea to give lists in standards for features that can and will change. A reader seeing this sentence would come to the conclusion that the Depth header can only be used with the listed methods. This is, of course, false. The Depth header may be used with any number of methods that have yet to be defined. While I suppose we could qualify this sentence to say that this is the list of methods that use Depth known at the time the standard was written I can find no utility for such a statement. As such I propose that this sentence be removed from the spec.
Received on Friday, 11 February 2000 02:50:41 UTC