Re: [NavigationTiming] Unclear note about "maintain the DOM structure of the document in memory"

Just a quick reply from my phone - I believe that part of the spec is meant
to refer to bfcache
On Jun 23, 2012 10:52 PM, "Boris Zbarsky" <bzbarsky@mit.edu> wrote:

> The specification currently says (section 5.1):
>
>  Some user agents maintain the DOM structure of the document in
>  memory during navigation operations such as forward and backward.
>  In those cases, the window.performance.timing and
>  window.performance.navigation objects must not be altered during
>  the navigation.
>
> I don't quite understand what this means, and in particular I can't tell
> whether what Gecko actually does is conformant.  On the face of it, this
> says that if I have page A loaded and then I navigate to page B while
> saving the DOM of page A, then the window.performance.timing after the
> navigation must be the same as before the navigation.  But if I _don't_
> save the DOM of page A, then it can be different.  I have a hard time
> believing this is what was meant.
>
> Is the point simply that if I have a page and I save the
> window.performance.timing for that page in a variable in another Window,
> navigate away from the page, and then go back in history that the object
> identity of the resulting window.performance.timing matches what I had
> before I navigated away?  That seems to be a natural consequence in an
> implementation that saves the Window when it "saves the DOM", but I guess
> one could do something where the "DOM" is preserved but the Window is not.
>  In that case, "maintain the DOM structure of the document in memory"
> probably needs to be defined by this specification for this requirement to
> mean anything.  In particular, I doubt that it actually matters that the
> saving is done "in memory" as opposed to "on disk" or "in the cloud"...
>
> -Boris
>
>