versions.txt

emotesV1.gaml
  - implementation of the dilemna prisoner game
  - creation of 2 agents with same guilt aversion
  - the history is a maps : each element is now a pair 
        agent::[nbCoop, totalInter]

emotesV2.gaml
  - modification of the history map : each element is now a pair 
        agent::[nbCoop, totalInter, utilityGained]
  - creation of 20 agents with guiltAversion between 0 and 2 by step of 0.1
  - addition of debug variables
  - save data in csv files
 
emotesV3.gaml
  - introduction of the step number parameter at which the simulation whould be saved
  - introduction of the Rawl's ideality
  - displayParameters
  
emotesV3_ESSA13.gaml    ---- Article ESSA 13 ----
  - ajout pair d'agent to display
  - cycle a partir duquel on observe + le fichier dans lequel on le met

emotesV4.gaml
  - simplification des save
  - reflex global : computeEquilibriumPoint et computeEsymptotPoint   
  - mode batch
  - parametres guiltAversionInitMax, guiltAversionInitMin, guiltAversionStep fonctionnels
  - debut de display sous forme de grille
  
emotesV5.gaml
  - split en mainV5.gaml et peopleV5.gaml 
  - graphical display of the state of the game on a graphic grid.
  
 ================= Introduction of Dynamics ============= 
  
emotesV6.gaml
  - introduction of the replicator dynamics:
    * strategy is constant
    * guiltAversion is now a variable
    * the population now evolves depending on money owned
  - evolve reflex has been implemented:
    * Each stepEvol steps: 
      - we choose 2*nbAgtsEvols
      - compare their payoff, the higher gives his strategy and gA to the lower
      - agents' guilt Aversion is reset
      
emotesV7 :
  - modèle separated in several files (2 kinds of people and the various experiments)
  
emotesV8 : (19/02/14)
  - cleaning of people agents

emotesV8 : (4/04/14)
  - created a species for iterated game of prisonner's dilemma
  - new experiment : sequencial game execution, based on a .csv file containing some configuration (different combinaison of gain), it loads the file and sequentialy simulate each configuration of the PD game a .csv snap shot of agents states is taken at the end of each simulation.
  - model : configFileHandler, handle .csv files containing PD game configurations (load them into one of the sequential experiment variables).
  - new functions that let mechanics of the model easier to understand (reset_game(), create_game(), ...)
  - added color to the grid perspective :
	* blue for betrayors
	* yellow for suckers