- From: Hayato Ito <hayato@google.com>
- Date: Thu, 28 Mar 2013 11:21:02 +0900
- To: "Tab Atkins Jr." <jackalmage@gmail.com>
- Cc: Dimitri Glazkov <dglazkov@google.com>, "www-style@w3.org" <www-style@w3.org>, Steve Orvell <sorvell@google.com>
On Thu, Mar 28, 2013 at 12:25 AM, Tab Atkins Jr. <jackalmage@gmail.com> wrote: > On Tue, Mar 26, 2013 at 8:55 PM, Hayato Ito <hayato@google.com> wrote: >> I have one concern for the naming of ':root'. >> >> I am afraid that It might be misleading that ':root' matches >> 'top-level elements in the distributed set' rather than 'insertion >> point' itself. >> There are multiple such roots. That's not intuitive for me. ':root' is >> likely to imply 'insertion point' itself because it's *root* of such >> elements. >> >> So instead of reusing ':root' in this context, how about having a more >> intuitive name, like ':child-of-shadow-host' or something? > > The issue with that is that we run into the exact same problem with > ::shadow() - if you want to select only the top-level elements inside > of a shadow root, what selector do you use? Do we invent *another* > pseudoclass that's identical except for the name? That's good point. I agree that we should avoid such a situation, inventing another pseudoclass, if we can reuse existing pseudoclass without any ambiguity of the meaning. As for ':root' pseudo class which is used in the parameter of '::distributed', there is no ambiguity. A '<html>' element never appears in the elements which '::distributed(..)' may match. > I'm personally okay with :root meaning "an element without a parent in > this view of the tree". That is helpful to my mental model of ':root'. Thank you! > > ~TJ -- Hayato
Received on Thursday, 28 March 2013 02:21:50 UTC