W3C home > Mailing lists > Public > public-web-perf@w3.org > May 2012

RE: [ResourceTiming] Clarifying a few things...

From: Jatinder Mann <jmann@microsoft.com>
Date: Fri, 4 May 2012 18:34:03 +0000
To: James Simonsen <simonjam@chromium.org>, Andy Davies <dajdavies@gmail.com>
CC: "public-web-perf@w3.org" <public-web-perf@w3.org>
Message-ID: <AE5FFD9402CD4F4785E812F2C9929D6504EEB905@SN2PRD0310MB383.namprd03.prod.outlook.com>
>> 2. Should secureConnectionStart be included the list of attributes
>> (Section 4.5) that are set to zero when the Timing-Allow-Origin
>> response header isn't present?
>
> Yeah, sounds like we should update the spec.

I have updated the spec to also set secureConnectionStart, if supported by the user agent, to zero when the Timing-Allow-Origin response header isn't present. 

In regards to Timing-Allow-Origin rules, I also clarified the startTime definition to state startTime must have the same value as redirectStart when there are redirects from the same origin or if the Timing-Allow-Origin response header rules are met.

>> 4. setResourceTimingBufferSize - trying get a clear understanding of
>> this sentence:
>> 
>> "If this method is not called, the default maximum number of
>> PerformanceResourceTiming resources stored must be 150, unless
>> otherwise specified by the user agent."
>>
>> Is this saying "the buffer must be a minimum of 150, but a UA can
>> specify it to be larger", or does it leave open the possibility that a
>> UA can specify it to be smaller than 150?
>
> The intent was that 150 is a minimum. We figured something like 90+% of pages would be fine with 150.
>
 >We should probably clarify that.
> 

I updated the spec text to state the user agent should store at least 150 PerformanceResourceTiming resources in the buffer, unless otherwise specified by the user agent.

Thanks,
Jatinder
Received on Friday, 4 May 2012 19:07:05 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:04:32 UTC