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

RE: [Resource Timing] Opt in vs Opt Out and initiatorType attribute

From: Alois Reitbauer <alois.reitbauer@dynatrace.com>
Date: Thu, 8 Mar 2012 09:40:03 +0000
To: Andy Davies <dajdavies@gmail.com>, "public-web-perf@w3.org" <public-web-perf@w3.org>
Message-ID: <A8638EB8DF9DE24D8C1E12D326D97AC39ABC5B@ATEX01.dynatrace.local>
Andy,

I like the idea of having a github project. I was working on a very simplistic bookmarklet as well. The ideas that you brought up are exactly what I wanted to implement as well. I think that this could be a great reference implementation. This could also be one of the deliverables of the working group. I know that there is also a Performance Interest Group and I was thinking about having something like a Community Group - like coremob for example - where we work on a reference implementation.

What do you think?

Alois 

-----Original Message-----
From: Andy Davies [mailto:dajdavies@gmail.com] 
Sent: Saturday, January 07, 2012 10:26 AM
To: public-web-perf@w3.org
Subject: [Resource Timing] Opt in vs Opt Out and initiatorType attribute

Hi All,

I've got some ideas on how I want to use the Resource Timing API but also got a couple of questions...

I currently want to use the RT API in two ways:

1. Create a bookmarklet that creates a visualisation of the waterfall and offers the ability to save it as a HAR file e.g. on jdrop etc.

2. Create a 'beacon' that can be embedded in a page and reports back the waterfall to a host - might not use it for all visitors e.g. could specify that it's only invoked for certain UAs or when page load time is slow.

What's the status of the thinking on Opt out vs Opt in default?

It isn't a problem for #1 as the bookmarklet user is that one in control of the choice but opt in makes #2 very difficult (presume the whole issue of opt out/in is around privacy?)


In the latest editor's draft of the spec The initiatorType attribute has changed from an integer to a string, can anyone explain the thinking behind this (it's more for my curiosity, but I see the attribute in PerformanceNavigation spec are integers)


I'm aware that actual implementations of ResourceTiming are some way off (checked both the Chrome and Firefox source trees) so I've started creating a set of mock objects to work with (their values are initiated from a HAR file) - if it's of interest to anyone will put it on GitHub in the next few weeks.

Thanks for your work on this...

Cheers

Andy
Received on Thursday, 8 March 2012 09:40:41 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Thursday, 8 March 2012 09:40:41 GMT