16710 2012-04-12 07:44:52 +0000 [Shadow]: Clarify lower-boundary encapsulation existence for shadow insertion point 2012-05-15 03:26:10 +0000 1 1 1 Unclassified WebAppsWG HISTORICAL - Component Model unspecified PC All RESOLVED WONTFIX P2 normal --- 15480 1 hayato dglazkov public-webapps-bugzilla oldest_to_newest 66642 0 hayato 2012-04-12 07:44:52 +0000 Is there any *boundary* concept between <shadow> insertionPoint and its olderShadowRoot (and its children)? In current spec, I cannot read that there is a boundary between them. But I am wondering whether it is okay or not that nodes in olderShadowRoots happen to see any nodes in youngerShadowRoot? For me, it's similar to the situation where a ShadowHost cannot see any noes in its shadowRoot in any cases. This question arises from my mental model that a concept of multiple shadow roots is like nested shadow roots, where the younger shadow root is attached to the older shadowRoot. The OlderShadowRoot acts like a shadowHost for the youngerShadowRoot. 67683 1 dglazkov 2012-05-14 18:53:13 +0000 There is definitely a boundary. Since "shadow insertion point" is just a type of an insertion point, the lower-boundary encapsulation rules apply just the same. Does this seem reasonable? Should I clarify this in the spec? 67701 2 hayato 2012-05-15 03:26:10 +0000 Okay. That makes sense. <shadow> is just an InsertionPoint and there is a lower boundary. We can close this bug as WONTFIX. Thank you!