On Tue, 1 Feb 2011, omd wrote: > On Tue, Feb 1, 2011 at 4:57 PM, Kerim Aydin <ke...@u.washington.edu> wrote: > > On Tue, 1 Feb 2011, omd wrote: > >> On Tue, Feb 1, 2011 at 4:41 PM, omd <c.ome...@gmail.com> wrote: > >> >> Past only. future clearly not included. > >> > > >> > The future is certainly "needed to interpret its current and future > >> > behavior". > >> > >> Or, to use your other quote, it must be in the state if the state > >> "describes enough about the system to determine its future behaviour." > > > > Yes, but you don't use that future information until you get to it, at > > which time it's the present. > > I think it's simpler. Does the gamestate (as a tracked value that can > be examined and modified, like your proposal did) describe enough > about the system (Agora) to determine its future behavior? > > If so, it includes future real-live events, and is thus indeterminate. > > If not, that particular definition of state can't apply.
You're extending the scope of the definition beyond what I stated. External (future) forcing of the state (say the future pushing of a pendulum) is not included in the state, but once it's happened, it becomes part of the state. Or, to put it another way (and I missed this); the definition of state as "being predictive of the future state of the system" implies "in the absense of future undetermined forcing" or if you prefer "assuming the system remains closed once the state is set." Once forcing has happened, it is determinate and part of the state. If it hasn't happened yet, it isn't. -G.
