On Wed, Apr 20, 2016 at 2:44 AM, Tantek Çelik <tan...@cs.stanford.edu>
wrote:
> On Tue, Apr 19, 2016 at 8:41 AM, Ting-Yu Lin <t...@mozilla.com> wrote:
> > To summarize the feedback so far, I'd still like to ship <details> and
> > <summary> without |summary::marker| because
> >
> > 1) No other browsers support summary::marker yet.
>
> This is a reason to ship *with* it IMO, per showing standards
> leadership, something Firefox is known for, leading the open (rather
> than prefixed) web and all that.
>
I agree with you that we should really support ::maker in general to set a
standard.
>
> > 2) From the webcompat point of view, even if we support summary::marker,
> > our usage to customize the triangle will still be different from the
> > |summary::-webkit-details-marker| usage on Blink/Webkit. So implementing
> > ::marker does not solve the webcompat issue.
>
> This is a normal transition phase in the path towards developers using
> standards-based properties (border-radius etc. history has plenty of
> examples). Not a reason not to ship.
>
>
> > 3) The marker (triangle) is still stylable without |summary::marker|.
> > Author could change the triangle via |summary { list-style-type: xxx}| or
> > use pseudo element |summary::-moz-list-bullet| to add more css rules.
>
> It's good that we have a prefixed alternative (that does not depend on
> someone else's prefix).
>
summary::-moz-list-bullet is a consequence of <summary> being a list-item
in our implementation. Please don't advocate its usage.
>
>
> The key question to consider about delaying |summary::marker| support
> is whether or not we (e.g. you) think the spec for ::marker is
> "stable" and "good" enough to ship.
>
> That's a judgment call, and I'd like to know your opinion of that.
>
That's a good point. The ::marker are defined in both Pseudo-Elements
Module Level 4 <https://drafts.csswg.org/css-pseudo-4/#marker-pseudo> and Lists
and Counters Module Level 3
<https://drafts.csswg.org/css-lists-3/#marker-pseudo-element>.
And the one defined in pseudo element spec is easier to implement since the
set of CSS ruless that can be used on ::marker is restricted.
I would say we should implement it and give feedback to spec if we found
anything making no sense.
I don't mind if we finally decided to block shipping details and summary
due to lacking of ::marker support.
The point is that we need more feedback about the real use cases for
details and summary on Firefox.
I had enabled the pref on non-release channel on 48, but I don't see any
bug report about cosmetic issue or site breakage.
We might get more feedback once Aurora became 48 though.
_______________________________________________
dev-platform mailing list
dev-platform@lists.mozilla.org
https://lists.mozilla.org/listinfo/dev-platform
