A quick tit-for-tat bot

titfortat.erl

@@ -0,0 +1,22 @@
+%%% The rumored-to-be-optimal-ish strategy for prisoners' dilemma
+
+-module(titfortat).
+-export([init/0, play/2, result/2, stop/1]).
+
+init() ->
+    dict:new().
+
+play(Opponent, State) ->
+    case dict:find(Opponent, State) of
+        {ok, OpponentLastMove} ->
+            {OpponentLastMove, State};
+        error ->
+            {cooperate, State}
+    end.
+
+result({Opponent, _OwnChoice, OpponentChoice}, State) ->
+    dict:store(Opponent, OpponentChoice, State).
+
+stop(_State) ->
+    ok.
+