- From: Marcos Caceres <marcosc@opera.com>
- Date: Wed, 7 Oct 2009 16:11:15 +0200
- To: Arthur Barstow <Art.Barstow@nokia.com>
- Cc: public-webapps <public-webapps@w3.org>
On Wed, Oct 7, 2009 at 3:46 PM, Arthur Barstow <Art.Barstow@nokia.com> wrote: > On Oct 7, 2009, at 9:25 AM, ext Marcos Caceres wrote: > >> (Apologies up front, the following is going to to seem like a rather >> dumb and slightly condescending discussion. I honestly do not mean it >> to be, but its necessary to help me identify where I need to fix the >> specification. Please bear with me.) > > LOL! > >> On Wed, Oct 7, 2009 at 2:24 PM, Arthur Barstow <art.barstow@nokia.com> >> wrote: >>> >>> Since the schema and Authoring guidelines are both non-normative, the P+C >>> spec is not clear if an element's attributes are required or not. >> >> When you say "required" (passive voice), do you mean: > > My expectation is the spec will normatively state whether an element's > attributes (e.g. <widget> element has id, version, etc.) are required or not > in a configuration document. The spec does not set conformance criteria for configuration documents. They are no longer considered a class of product. The specification is exclusively concerned with the behavior of user agents processing zip files and xml files. Those XML files may be in a namespace that identifies them as "configuration documents". If that is that case, apply Step 7. > I think this information is clear in the 23-July-2009 CR but it is not clear > in the TSE where the required information is embedded in non-normative > Authoring Guidelines. This kind of information is now handled by the conformance checker specification. -- Marcos Caceres http://datadriven.com.au
Received on Wednesday, 7 October 2009 14:16:53 UTC