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

Re: [Page Visibility] LC issues

From: Boris Zbarsky <bzbarsky@MIT.EDU>
Date: Wed, 29 Feb 2012 12:06:44 -0500
Message-ID: <4F4E5B24.5070606@mit.edu>
To: public-web-perf@w3.org
On 2/28/12 4:58 PM, Jatinder Mann wrote:
>> I'm not quite sure what "if the User Agent is to unload the Document contained by the top level browsing context" means.  Does it mean "is
>> about to unload, with the 'salvageable' state set to false"?   I would assume so.  I'm not sure how to phrase that reasonably, though.  It would
>> be simple to define that one of the "unloading document cleanup steps" this specification defines is to change the state to "unloaded", but
>> those happen after the unload event fires.
>
> I have updated the Processing Model to make the expected behavior here more clear. Please review and let me know if this is clear enough.

I don't think redefining the unload steps here completely is reasoanble. 
  It'll just end up with this spec contradicting the HTML5 spec in the 
end, which is not where we want to be....

The text as written makes the visibility state change happen after 
unload.  Was that the intent?

>> I suggest coordinating with the HTML5 editor to figure out how to define this properly and whether we need new hooks in the
>> document-unloading process to do so...
>
> I would like to avoid making the HTML5 spec take a dependency on this spec if possible. If we do need a hook there, I can follow up.

I agree that having HTML5 depend on this spec is probably wrong.

I think having HTML5 have a hook between pagehide and unload via a hook 
is fine...  I'd vastly prefer that to overriding part of the algorithm 
directly in this spec, and I bet so would the HTML5 editor.

-Boris
Received on Wednesday, 29 February 2012 17:07:23 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Wednesday, 29 February 2012 17:07:24 GMT