Re: Web Performance Potential Future Work

Hey Paul,

Thank you for your feedback. Let me elaborate on some points in more detail

Ad Perf1:  I agree that there should not be that much difference in timing in certain cases. However there are definitely use cases where this makes sense. First of all I can measure First Impression Time which indicates when the page is visible for the first time. The same is true for above the fold time and other rendering related timings. Currently there is no really reliable way to do so. The second issue is that we need to understand how much layout calculation occurs on whether this is happening within a JavaScript execution.  Very often bad JS code forces relayouting and people are not aware of it.

Ad Perf6: Good idea to work with the groups directly. Maybe the Web Perf WG can provide guidance on how to best expose information. Ideally they will have an easy way to extend the Performance Timeline.

Ad Perf8: It would be great to get more input on your actual requirements here. Did you find the memory profiling support of Chrome useful here?

Ad Perf10: The problem is that if you want to track all user events and time them you need to register event handler for every button etc.  Then we know when a user clicks something that then causes an XHR request for example. Theoretically we can do this today, however it has serious impact on the page as we need to walk the DOM to register event handlers as we also want to get informed about events that get cancelled after some handler triggered. So just registering at the document level does not do the trick.

// Alois


Alois Reitbauer, Technology Strategist    M (US) +1 617 515 1956 | M (EU)  +43 664 8536534 | E alois.reitbauer@dynatrace.com<mailto:alois.reitbauer@dynatrace.com>  | @AloisReitbauer
www.dynatrace.com<http://www.dynatrace.com/> | blog.dynatrace.com<http://blog.dynatrace.com/> | ajax.dynatrace.com<http://blog.dynatrace.com/>

[Description: Description: Description: Description: DT_CPWR_Elec Signature.png]
The contents of this e-mail are intended for the named addressee only. It contains information that may be confidential. Unless you are the named addressee or an authorized designee, you may not copy or use it, or disclose it to anyone else. If you received it in error please notify us immediately and then destroy it.

Received on Wednesday, 30 November 2011 13:32:19 UTC