[Freeciv-Dev] Re: patch: make ai understand peace and alliances 2 (PR#1277)

[Gregory Berkolaiko <gberkolaiko@xxxxxxxxxxx>]

 --- Raahul Kumar <raahul_da_man@xxxxxxxxx> wrote: 
> 
> --- Gregory Berkolaiko <gberkolaiko@xxxxxxxxxxx> wrote:
> >  --- Raimar Falke <hawk@xxxxxxxxxxxxxxxxxxxxxxx> wrote: 
> > > On Mon, Feb 25, 2002 at 02:07:00AM -0800, Per I. Mathisen wrote:
> > > 
> > > So we have now active diplomacy and passive diplomacy. The question
> > > is: is it worth implementing passive diplomacy because you have to
> > > reaudit the code when active diplomacy is implemented?
> > 
> 
> That is a big if. Unless you know someone else is working on a diplomacy
> patch, better the current patches go through.

quite right

> > 
> > I am fully supportive of what Per is doing.  My reasons:
> > 1. Passive diplomacy is useful, fx for scenarios
> > 2. "Reaudit" is a strong word.  To add active diplomacy on top of passive,
> > you wouldn't need to touch most of the existing code.  It would probably 
> > amount to
> > profit estimation and hysteretic decision making.  
> 
> Hysteretic decision making. Sounds cool, but conveys no meaning to me.

Suppose you have one function dip_attitude(player1, player2), which takes into
account everything (historical love/hatred, military fear/sense of
superiority).  When dip_attitude is <0 you would probably like to have a war. 
When it is >0, you want peace.  Then you'd set up two boundaries, say -20 and
20 and say: when dip_attitude drops below -20 we switch state to WAR, when it
rises above 20, we switch state to PEACE.  In science this effect is usually
referred to as hysteresis.  In particular it prevents you from switching
WAR-PEACE if your attitude oscillates around zero.

G.

__________________________________________________
Do You Yahoo!?
Everything you'll ever need on one web page
from News and Sport to Email and Music Charts
http://uk.my.yahoo.com