Re: [webcomponents] Style attributes in Shadow DOM Cascade Order Proposal (#316)

Following up the previous @hayatoito 's comment:

*Proposal1* and *Proposal2* are at the bottom of @hayatoito 's [proposal doc](https://github.com/w3c/webcomponents/blob/gh-pages/proposals/Shadow-DOM-Cascade-Order-in-v1.md), which says essentially 2 things if I sum up:

* Rules coming from other shadow trees are compared in tree-of-trees order (as proposed by @rune-opera 's original proposal).
* Regarding how style attribute is compared against other style rules defined in stylesheet (which is the original topic of this thread), Proposal1 follows Option1 in this thread , and Proposal2 follows Option2.

Regarding the former, "Outer wins Inner" is a rule of thumb and inaccurate.  Comparison in terms of tree-of-trees order is more comprehensive and precise definition, as @rune-opera already mentioned.  I think all of us in this thread agree on this.  Both Hayato's *Proposal1* and *Proposal2* follow this comparison order.

For the latter, which is still unresolved.
On the previous @hayatoito 's comment's second bullet, "... likes Option2, which doesn't conflict *Proposal 1*" but actually it **conflicts**.  If nothing changed, Apple still wants *Proposal2* behavior?

To be consistent with how style attribute is handled in scoped style, Option 1 (or Hayato's *Proposal 1*) looks a reasonable choice to me.

At this moment I don't think the feasibility to keep the backward compatibility for Blink implementation is irrelevant to which option we take.

---
Reply to this email directly or view it on GitHub:
https://github.com/w3c/webcomponents/issues/316#issuecomment-161548684