W3C home > Mailing lists > Public > public-web-perf@w3.org > October 2010

Re: [Navigation Timing] Clarification needed on DOM timers and document.open()

From: Zhiheng Wang <zhihengw@google.com>
Date: Mon, 4 Oct 2010 10:16:10 -0700
Message-ID: <AANLkTinvrEaoH_XWJxn5DEqvUPQ+-Uved_CmefRKx3gk@mail.gmail.com>
To: Tony Gentilcore <tonyg@google.com>
Cc: Anderson Quach <aquach@microsoft.com>, James Simonsen <simonjam@chromium.org>, "public-web-perf@w3.org" <public-web-perf@w3.org>
   +1 to taking the first instance, which IMO is more consistent with other
variables in terms of
finding the critical path.

   I am updating the draft with some recent feedback including this one. It
should be up later today.

thanks,
Zhiheng

On Mon, Oct 4, 2010 at 9:51 AM, Tony Gentilcore <tonyg@google.com> wrote:

> Reporting the first time in the case of repeated dom* events makes
> sense to me. Zhiheng, are you okay with adding something to the spec
> to mention this corner case?
>
> On Fri, Oct 1, 2010 at 8:15 PM, Anderson Quach <aquach@microsoft.com>
> wrote:
> > Great feedback! The expectation here is that the  domLoading,
> domInteractive
> > and domComplete markers should be set upon the first time the
> > readystatechange as described in the HTML5 spec [1] and [2]. You can
> observe
> > this behavior today in the implementation available in the IE9 beta and
> the
> > IE9 platform preview available on the Test Drive site [3].
> >
> >
> >
> > [1] http://dev.w3.org/html5/spec/dom.html#resource-metadata-management
> >
> > [2] http://dev.w3.org/html5/spec/the-end.html#the-end
> >
> > [3] http://ietestdrive.com
> >
> >
> >
> > Best Regards,
> >
> > Anderson Quach
> >
> > IE Program Manager
> >
> >
> >
> > From: public-web-perf-request@w3.org [mailto:
> public-web-perf-request@w3.org]
> > On Behalf Of James Simonsen
> > Sent: Thursday, September 30, 2010 5:34 PM
> > To: public-web-perf@w3.org
> > Subject: [Navigation Timing] Clarification needed on DOM timers and
> > document.open()
> >
> >
> >
> > Hello,
> >
> >
> >
> > The Navigation Timing spec says that the domLoading, domInteractive, and
> > domComplete timers should be set to the time immediately before the
> > document's readiness changes. What is the expected behavior when the
> > document reaches the same state multiple times?
> >
> >
> >
> > For example, if document.open() is called, the document returns to the
> > loading state. Should the domLoading timer reflect the first time we
> reached
> > the loading state? Or the most recent?
> >
> >
> >
> > The domContentLoaded timer has the same issue, since its associated event
> > can fire multiple times as well.
> >
> >
> >
> > Thanks,
> >
> > James
>
Received on Monday, 4 October 2010 17:16:42 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 21 December 2010 18:13:55 GMT