21887 2013-05-01 03:49:11 +0000 [Custom]: Add a note about using TYPE and NAMESPACE to identify definitions despite NAMESPACE being redundant 2013-05-21 22:19:07 +0000 1 1 1 Unclassified WebAppsWG HISTORICAL - Component Model unspecified PC Linux RESOLVED FIXED P2 normal --- 18720 1 dominicc dglazkov public-webapps-bugzilla oldest_to_newest 87060 0 dominicc 2013-05-01 03:49:11 +0000 The steps for register say "If there already exists a definition with the same TYPE, throw a NotSupportedError and stop." This means that TYPE uniquely specifies a definition. However when referring to definitions, the spec usually (always?) refers to them by TYPE and NAMESPACE. For example, this from parsing: "Let ELEMENT be the result of creating a node implementing the interface appropriate for the element type corresponding to TYPE and NAMESPACE" I agree that this "TYPE and NAMESPACE" is a convenient way to sneak in a namespace check without complicating steps with explicit exits for mismatched namespaces. However it would be nice if there was non-normative language pointing out that the spec does this and there is this implicit namespace checking going on all over the place, which when it fails, is the same as a definition not being available. 88029 1 dglazkov 2013-05-21 22:19:07 +0000 I think I'd rather just make this checks explicit: https://dvcs.w3.org/hg/webcomponents/rev/e96be051a09a