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

Re: [Web Timing] Interface names

From: Tony Gentilcore <tonyg@chromium.org>
Date: Thu, 21 Oct 2010 12:44:44 -0700
Message-ID: <AANLkTikBezoKBiEC9oViDdEG9efwxHSDke-AGzUP6jqC@mail.gmail.com>
To: Zhiheng Wang <zhihengw@google.com>
Cc: Anderson Quach <aquach@microsoft.com>, "public-web-perf@w3.org" <public-web-perf@w3.org>
On Thu, Oct 21, 2010 at 11:52 AM, Zhiheng Wang <zhihengw@google.com> wrote:

>
>
> On Thu, Oct 21, 2010 at 11:26 AM, Tony Gentilcore <tonyg@chromium.org>wrote:
>
>> Sorry I was unable to join the call yesterday. Did this get resolved?
>> Should be a detail that is easy to close the loop on.
>>
>> On Mon, Oct 18, 2010 at 12:07 PM, Anderson Quach <aquach@microsoft.com>wrote:
>>
>>> Thanks for catching this. I agree we should be consistent with the
>>> interface names, especially moving towards an implementation where the we
>>> can drop the vendor prefix.
>>>
>>>
>>>
>>> The open question is that do we want to move NavigationInfo Ó Navigation
>>> or keep NavigationInfo.
>>>
>>>
>>>
>>> Proposed interface names:
>>>
>>> Performance
>>>
>>> NavigationInfo ľor- Navigation
>>>
>>> NavigationTiming
>>>
>>
>> I like the idea of using Performance to scope the sub interfaces like you
>> did in the IE implementation.
>>
>> What about Performance (window.performance), PerformanceNavigation
>> (window.performance.navigation) and PerformanceTiming
>> (window.performance.timing)?
>>
>
>    sounds good to me.
>
>    Since we are on this, we should settle that for ResourceTiming and
> UserTiming as well. How about:
>    ResourceTiming: window.performance.createResourceTimingCollector()
>

Hard to say for resource timing since I don't think we are sure how it will
look. If we go with this scheme and the interface looks like that, I'd
expect the interface name for the collector to be
PerformanceResourceTimingCollector.


>  UserTiming: window.performance.timing.mark() - or - <something else>?
>

If we go with this scheme and the interface looks like
http://dvcs.w3.org/hg/webperf/raw-file/tip/specs/UserTiming/Overview.html#DOMPerformanceTimingList,
then I'd expect that what is called UserTiming in that spec is not a new
interface at all but a supplement to the Performance interface. It would add
new PerformanceMark and PerformanceMarkList interfaces.


>
>
> cheers,
> Zhiheng
>
>
>>
>>
>>>
>>>
>>> As for the name of the interface prototype object, on IE we believe that
>>> the WebIDL spec [1] is crucial to get right. The naming conventions of the
>>> interface prototype object should be left up to that working group.
>>>
>>>
>>>
>>> [1] http://www.w3.org/TR/WebIDL/
>>>
>>>
>>>
>>>
>>>
>>> *From:* public-web-perf-request@w3.org [mailto:
>>> public-web-perf-request@w3.org] *On Behalf Of *Tony Gentilcore
>>> *Sent:* Thursday, October 14, 2010 8:35 PM
>>> *To:* public-web-perf@w3.org
>>> *Subject:* [Web Timing] Interface names
>>>
>>>
>>>
>>> Since user agents expose the interface names to the DOM, I'd like to
>>> reconsider the interface names as a whole to get a consistent scheme.
>>>
>>> Spec              IE                                WebKit
>>> Performance       MSPerformancePrototype            Performance
>>> NavigationInfo    MSPerformanceNavigationPrototype  Navigation
>>> NavigationTiming  MSPerformanceTimingPrototype      Timing
>>>
>>> I recommend we standardize on Performance, PerformanceNavigation, and
>>> PerformanceTiming.
>>>
>>>
>>>
>>> For every other interface I checked, IE uses a Prototype suffix, but
>>> other UAs don't use it and it isn't in any specs that I've found. So I
>>> assume that we should standardize on something without Prototype and IE will
>>> might add a Prototype suffix for internal consistency.
>>>
>>>
>>>
>>> Thoughts?
>>>
>>> -Tony
>>>
>>
>>
>
Received on Thursday, 21 October 2010 19:53:31 GMT

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