W3C home > Mailing lists > Public > public-html@w3.org > September 2012

Re: Extension spec for hgroup (Was: Re: Getting HTML5 to Recommendation in 2014)

From: Henri Sivonen <hsivonen@iki.fi>
Date: Tue, 25 Sep 2012 15:44:18 +0300
Message-ID: <CAJQvAud-KHVZyRgHqibGVsq8_y7wh8-i-6CDm+EC+Or1CBZmFw@mail.gmail.com>
To: public-html WG <public-html@w3.org>
On Tue, Sep 25, 2012 at 12:24 PM, Steve Faulkner
<faulkner.steve@gmail.com> wrote:
> Can you be a bit more specific? So I can add the appropriate bits back in.

I mean all changes made to the section "Parsing HTML documents" and
its subsections:
http://www.html5accessibility.com/HTML5extensions/HTML5.html#parsing

On Tue, Sep 25, 2012 at 12:37 PM, Jirka Kosek <jirka@kosek.cz> wrote:
> What are the real practical consequences of instantiating element as
> HTMLUnknownElement instead of HTMLElement?

That's not what I mean. Interfaces are not part of the parsing algorithm.

> How can it break site?

The DOM will be a different on any site that relies on hgroup implying
a paragraph close.

> Are there any sites which will get broken?

I am not willing to spend time to find out when we already have
interop across all the major engines and the problem of finding out
could be avoided by not changing the parsing algorithm.

On Tue, Sep 25, 2012 at 2:37 PM, Sam Ruby <rubys@intertwingly.net> wrote:
> If retaining more of the parsing gets us to Recommendation faster, then that
> may be a win.

Currently, Gecko, WebKit, Presto and Trident all behave the same way
with this example:
http://software.hixie.ch/utilities/js/live-dom-viewer/saved/1796

In other words, on this point, we currently have interoperability
across the board. Changing the spec here would move us further from
interoperability and, therefore, further from exiting the CR.

The same consideration applies to *any* change to the “Parsing HTML
documents” section. If it was allowed to progress on the REC track as
an individual item, we could declare REC on the parsing algorithm
*today* with 4 independent interoperable browser implementations.
Therefore, instead of raising these points repeatedly about various
individual parser changes as side effects of other proposals, I’d like
to ask the Chairs to pre-emptively exclude the parsing section from
removals or additions of elements for the HTML5 snapshot.

--
Henri Sivonen
hsivonen@iki.fi
http://hsivonen.iki.fi/
Received on Tuesday, 25 September 2012 12:47:38 UTC

This archive was generated by hypermail 2.3.1 : Monday, 29 September 2014 09:39:34 UTC