On Saturday, February 5, 2005, at 10:00 AM, Bill de h�ra wrote:
Graham wrote:
2. Changing the atom:id of the entry with each revision would severly reduce the duplicate detection value of atom:id.I don't think anyone's proposed this. -1
Wrong: http://www.imc.org/atom-syntax/mail-archive/msg13161.html
-1 also. I think we'd do better to focus on the specification text for id - that will be much more effective than adding qualifying text. I think it falls under editorial work and requires no pace.I have no idea what you mean by this comment. Part of the purpose of the Pace does seek to improve on the specification text for id (see the abstract, and compare the proposed text to the current draft) for the benefit of those who don't think it's perfectly clear if you don't refuse to keep reading after the first sentence. Do you think the proposed change doesn't help? If not, how about a counter suggestion or at least something more concrete than "...that will be much more effective than adding qualifying text. I think it falls under editorial work and requires no pace."
Do you not think that we need to specify whether or not multiple instances of a resource can appear in a single document (with the same atom:id)? Language specifying that they may is hardly mere editorial work--it requires a decision by the working group to decide whether we do want to allow it or not.
Warning publishers that consumers might not display each version of a resource if the do choose to include multiple revisions in a single document might be considered editorial work, but there's certainly no reason not to include such language in a proposal whose primary purpose is to clarify whether putting multiple revisions in the document is allowed.
