Given that you can read the opponent’s source code, self-recognition is trivial to implement anyway (trivial but annoying, since you need to do quining).
I am aware of the potential strategic implication you bring up. However, as the organizer of this game, I believe it is my place neither to confirm nor deny beforehand any strategic implications of these rules.
Explaining my rationale necessarily involves discussing the strategic implications. I will explain my rationale for this choice after the game.
If I had to guess, the reasoning behind it is to nudge the game closer to a ‘true’ prisoner’s dillemma (trying to work out if your opponent is willing to cooperate, rather taking focus away from it towards the shallower problem of trying to work out if your opponent is a copy of you)
I agree, and this design avoids that problem, but seems to introduce a much larger one, assuming the intent also includes measuring bots on their ability to survive in progressively more “difficult” bot mixes, which “Darwin” seems to imply.
This choice also nudges me from “has noodled around the idea of hosting a similar competition many times and probably won’t” to “same, but slightly more likely to actually do it”. :D
My suggestion for a future Darwin Game tournament is to get rid of the 0, 1, 4, and 5 moves, and leave 2 and 3 as the only legal moves. Serious bots generally don’t need or use the other moves, so they mostly just make you add annoying special case logic to get a few more points out of GoofBots in the early game.
Update: There are two primary reasons for free auto-recognition. Firstly, I promised to write bots for non-programmers and I didn’t want to promise to implement an arbitrary number of bad crypto systems written by non-specialists. Secondly, I believe is is good game design to reward early aggression.
Reading the other’s source code is a bigger change (actually, a superset of auto-recognizing self)
It actually makes coordinating trivial and enforceable, since you can just check for the correct behavior & password before cooperating.
And if you can run your opponent’s source code...
If i had to guess on the motives, the last time a similar game was played (non publically) the meta developed to be about self recognizing, this is likely a rule to avoid this. Winning strategy was some key string to identify yourself, cooperate with yourselt, play 3 otherwise. (Granted number of iterations was low, so people might not have moved to be non cooperating strategies enough (something like grim trigger))
Free self recognition seems like it makes game less interesting, And that anyone who gets a big lead early just wins?
Given that you can read the opponent’s source code, self-recognition is trivial to implement anyway (trivial but annoying, since you need to do quining).
Since quining is “trivial but annoying”, I am willing to provide a quining function in the
extrapackage if anyone requests it.I am aware of the potential strategic implication you bring up. However, as the organizer of this game, I believe it is my place neither to confirm nor deny beforehand any strategic implications of these rules.
Explaining my rationale necessarily involves discussing the strategic implications. I will explain my rationale for this choice after the game.
Eh, okay, but (prediction) this choice nudges me from “probably participate” to “probably ignore”.
If I had to guess, the reasoning behind it is to nudge the game closer to a ‘true’ prisoner’s dillemma (trying to work out if your opponent is willing to cooperate, rather taking focus away from it towards the shallower problem of trying to work out if your opponent is a copy of you)
I agree, and this design avoids that problem, but seems to introduce a much larger one, assuming the intent also includes measuring bots on their ability to survive in progressively more “difficult” bot mixes, which “Darwin” seems to imply.
This choice also nudges me from “has noodled around the idea of hosting a similar competition many times and probably won’t” to “same, but slightly more likely to actually do it”. :D
My suggestion for a future Darwin Game tournament is to get rid of the 0, 1, 4, and 5 moves, and leave 2 and 3 as the only legal moves. Serious bots generally don’t need or use the other moves, so they mostly just make you add annoying special case logic to get a few more points out of GoofBots in the early game.
It’s a good point but in the original Darwin Game story, the opening sequence 2, 0, 2 was key to the plot.
Update: There are two primary reasons for free auto-recognition. Firstly, I promised to write bots for non-programmers and I didn’t want to promise to implement an arbitrary number of bad crypto systems written by non-specialists. Secondly, I believe is is good game design to reward early aggression.
Reading the other’s source code is a bigger change (actually, a superset of auto-recognizing self) It actually makes coordinating trivial and enforceable, since you can just check for the correct behavior & password before cooperating. And if you can run your opponent’s source code...
If i had to guess on the motives, the last time a similar game was played (non publically) the meta developed to be about self recognizing, this is likely a rule to avoid this.
Winning strategy was some key string to identify yourself, cooperate with yourselt, play 3 otherwise. (Granted number of iterations was low, so people might not have moved to be non cooperating strategies enough (something like grim trigger))