RE: [ResourceTiming] initiator types

That will align more closely with our previously spec'd behavior. I will special case XHRs.



Thanks,

Jatinder



________________________________
From: ojan@google.com [ojan@google.com] on behalf of Ojan Vafai [ojan@chromium.org]
Sent: Thursday, June 28, 2012 11:13 AM
To: James Simonsen
Cc: Jatinder Mann; public-web-perf@w3.org
Subject: Re: [ResourceTiming] initiator types

On Thu, Jun 28, 2012 at 10:32 AM, James Simonsen <simonjam@chromium.org<mailto:simonjam@chromium.org>> wrote:
On Wed, Jun 27, 2012 at 10:21 PM, Ojan Vafai <ojan@chromium.org<mailto:ojan@chromium.org>> wrote:
On Wed, Jun 27, 2012 at 10:14 PM, Ojan Vafai <ojan@chromium.org<mailto:ojan@chromium.org>> wrote:
Right. That's all I had in mind there. I suppose, by this definite it would also apply to Image(), which also seems fine to me.

Actually, Image() creates an img element, so perhaps it already falls under the localName clause.

On second thought, the current wording you have in the spec sounds good to me. I don't see a problem with using "script" as the initiatorType for XMLHttpRequests. No need to add any rules regarding JS constructor names.

I don't like that. I think XHRs should show up as XHRs. It's going to be useful for a website to collect timing info about XHRs and scripts separately from each other by using getEntriesByInitiatorType().

Oh, good point. Should we just special-case XHR?