On Sat, 4 Apr 2009, João Eiras wrote: > > On , Jeremy Orlow <jor...@google.com> wrote: > > > I think this also applies: "NOTE: The lifetime of a browsing context > > can be unrelated to the lifetime of the actual user agent process > > itself, as the user agent may support resuming sessions after a > > restart." > > Should that restore sessionStorage data ? Aren't you making > sessionStorage much more complicated while the same use cases are > covered by localStorage ? sessionStorage could be optimized to be just a > volatile amount of data in memory, but these requirements require > sessionStorage to implement the same disk IO heuristics, and a complex > heuristic to decide when to erase sessionStorage completly. > > I vote for the data to be present just while a page is open or is > restored from history or by going back.
User agents aren't required to do the above; but they are allowed to if they so desire. So the complexity is entirely optional (and I don't expect most UAs to avail themselves of this option). -- Ian Hickson U+1047E )\._.,--....,'``. fL http://ln.hixie.ch/ U+263A /, _.. \ _\ ;`._ ,. Things that are impossible just take longer. `._.-(,_..'--(,_..'`-.;.'