Resource Timing CR issues

As of 2013-11-05

when onresourcetimingbufferfull callback is trigged (Pan Deng)

• Re: when onresourcetimingbufferfull callback is trigged. (Tony Gentilcore)
• RE: when onresourcetimingbufferfull callback is trigged. (Jatinder Mann) (accepted)

A resource request without a network connection (James Simonsen)

the goal is to show how much time resource requests were blocked on the network. So anything that doesn't go out on the wire doesn't need to be reported. We may need to beef up our definitions in the spec.

Resource request: failure and success? (Andy Davies)

If a request for an resource is really fast, was it because the resource was retrieved quickly or that the resource doesn't exist and an error response delivered?

byte-size in resource info? (Yoav Weiss)

Is there any reason the resource's byte-size & compressed byte-size were not added to the ResourceTiming API?

• not worth burdening all clients since the information is on the server side already (James Simonsen) rejected

Adding protocol information (Mike McCall)

After some internal discussions, a colleague recently started a thread[1] on the spdy-dev mailing list, asking about having an interface for developers to leverage to determine whether or not a web page resource was fetched via SPDY (or in the future, HTTP 2.0).

• Already available on server side (Arvind Jain) rejected
• spdy and HTTP (Arvind Jain) rejected (too early)
• Wire Protocol (Mark Nottingham)
• Now is the time (Jatinder Mann) agreed, L2

Statements about cross-origin redirect should be more clearly (Pan Deng)

What is the meaning of "Timing-Allow-Origin HTTP response header rules are met"?

• specify a redirection-origin-clean flag in the processing model (Jatinder Mann) agreed

audio and video should be removed (James Simonsen)

we should explicitly say these elements are excluded. It's already on our charter to support them in Resource Timing 2.

• done (Jatinder Mann) fixed

Why does connectEnd exclude the SSL Handshake? (Andy Davies)

I understand why the spec states that connectEnd excludes SOCKS authentication etc., but don't quite understand why it excludes the SSL Handshake

• The spec doesn't exclude it anymore (Jatinder Mann) agreed

Missing example to avoid buffer overflow? (Nic Jansma)

I like Arvind's simpler model as well. By triggering the onresourcetimingbufferfull callback at N instead of N+1, it gives the developer the opportunity to either call clearResourceTimings() or grow the buffer incrementally via setResourceTimingBufferSize() , theoretically, with little chance for 'lost events'.

Maybe an example best practice could be provided so developers see they shouldn't have to setResourceTimingBufferSize(9999)?

Initial “about blank” for iframe example

• fixed the example (Arvind Jain) agreed

304 responses (Arvind Jain)

304 responses should be including in Resource Timing list. We should make sure the implementations store 304 responses.

• Do we have a test for 304 response?

onresourcetimingbufferfull attribute type and event listeners (Christophe Dumez)

The Performance.onresourcetimingbufferfull attribute should be of type 'Function'. Other DOM objects seem to use the 'EventListener' type (and inherit EventTarget) for similar purposes. Is there any reason why onresourcetimingbufferfull is not an EventListener and why the Performance does not inherit EventTarget?

• onresourcebufferfull attribute is now an EventHandler (Jatinder Mann) agreed
• Handling of "resourcetimingbufferfull" events makes no sense (Boris Zbarsky)
• A proposal to fix it (Jatinder Mann)

Web worker initiatorType? (James Simonsen)

We decided to punt Web Worker support to a level 2 spec. We were mainly talking about resource downloaded inside the worker, like XHRs. What about the worker itself? Its script is actually fetched by the main document, so it seems like it should show up in the buffer and have a "worker" initiatorType.

• agreed (Arvind Jain) agreed (L2)

The processing model section is marked as non-normative?!? (Boris Zbarsky)

• oops, fixed (Jatinder Mann) agreed

Need to define "networking layer" (Boris Zbarsky)

The specification uses, without definition, the term "networking layer".

The clock base to use in ResourceTiming is not defined (Boris Zbarsky)

• Fixed (Jatinder Mann) agreed

resourcetimingbufferfull is supposed to clear the buffer (Juan Carlos Estibariz)

• Good point (James Simonsen)
• some proposals (Juan Carlos Estibariz)