- From: Williams, Stuart <skw@hp.com>
- Date: Mon, 28 Jun 2004 17:49:22 +0100
- To: www-tag@w3.org
I've reviewed sections 1 and 2 and also posted annotated versions in www-archive@w3.org [1]. Below are some notes on the more substantive comments. The rest are mostly editorial/stylistic. As far as publication of a TR page draft I wouldn't make any of these show stoppers. Regards Stuart -- [1] http://lists.w3.org/Archives/Public/www-archive/2004Jun/0037.html [2] http://lists.w3.org/Archives/Public/public-webarch-comments/2004JanMar/1057. html General Comments ---------------- I found myself running up against what I might call a Pat Hayes induced sensitivity to the word identity (see [2]) thoughout. In section 2, which speaks of the assignment of URIs, I've suggested replacing "assignment" with "association" in several places - which 'softens' the notions of ownership and authority. Also, I think specifications 'specify' rather than 'license' and have suggested several substitutions in that vein - I think that makes the language simpler and more direct. Section 1 --------- Section1 Introduction: Right after the first sentence "World Wide Web ( WWW , or simply Web ) is an information space in which the items of interest, referred to as resources , are identified by global identifiers called Uniform Resource Identifiers ( URI )." I'd like us to say the "Anything that can be named or described can be a resource." phrase, or whatver concensus we ultimately achieve over the wording for 2396bis. Section 2 --------- Section 2: I suggest deleting the first sentence. It's fuzzy and likely opens up a philosophical minefield. "In order to communicate internally, a community agrees (to a reasonable extent) on a set of terms and their meanings." Sections 2. and 2.1 speak of URI as an "identifcation mechanism" which grates a little. I'd suggest presenting URI as a "single global system of identifiers" rather than a "single global identification mechanism". Section 2.4 URI Overloading intrinsically accepts that a given URI may be used to refer to many things. However 2.3 opens "As discussed above, a URI identifies one resource." which is contradictory wrt to the concept of overloading and also has a (C)/(D) sense problem wrt to 'identifies'. Section 2.5 URI Ownership. This seems new. If the concept has any validity, I think that we should be able to speak in terms of the rights and obligations associated with ownership. I think that the most significant right associated with 'ownership' of a DNS name and the associated derived URI is the ability to deploy or organise the provision of representations - in the event that a given URI is dereferenced. It is also not clear whether the concepts of authority (which is mentioned) and ownership are synonymous or not. I think this section needs more work to address some of the concerns expressed in response to our LC.
Received on Monday, 28 June 2004 12:50:17 UTC