Hello,

I've always wanted to see a role-playing game wrapped around a pure example of the Prisoner's Dilemma. Jonas Karlsson's One Can Have Her nails it, specifically the iterated version, to the wall.

I think the greatest part is not knowing which scene is going to be final. It really throws the dilemma into high gear, removing long-term strategizing out of the picture even though the characters are focusing almost exclusively on exactly long-term goals. It also reinforces the snap-decisions and desperation inherent to the genre.

However, it is also clear from my notes that this very feature, which is central and indispensable to the crucial decisions of play, also offers the risk of "gamus interruptus." Right when my and others' enjoyment of my character has hit its stride, right when we are ready to rock and see what he'll do next ... wham, the die says the game is over. Um ... hey!!

I really have no idea whether the power of the Prisoner's Dilemma (and the necessary uncertainty over whether this is the last round or not) can overcome the risk of disappointment that "we have to stop now." That is a damn good question, and you know what? It's the first time this question has ever been entered into RPG design in a developed way. For instance, the overriding and often bizarre issue of character death has always been with us ("what? I have to stop playing?") in fucked-up form, but this game in development offers a way to address it more sensibly. Playtesting will tell.

The game also features awesome, efficient description and presentation of noir Color. I wish I'd even begun to approach this level in The Sorcerer's Soul. Combine that with lots of room for theme in the outcomes of scenes and the overall story (yes, even with a slavish devotion to genre and source medium), and it just begs to be played.

The total amount of Coolness/Opposition doesn't diminish, hence unlike Budget/Fanmail in Primetime Adventures, nor does it increase, hence unlike Universalis. That's interesting. You've effectively created a closed economy of "bonus" in the entire play-experience. I wonder whether that will work? I tend to prefer a narrowing-down of economy, personally, as a way to generate rising tension ... but here, you have the "might be over any minute!" at the fourth stage, to generate that effect. Neat! I really want to see that in play.

Now for some mechanics-based comments.

1. I have to say, I'm getting a bit tired of cards used merely as two-sided dice. It seems to me that a standard card deck consists of about eight or ten well-known qualities for comparison or resource-base, and that a given good card game utilizes four or five of them, maximum. When it only utilizes one or two, you get something like War or Spoons. Looking over the mechanics in this game, the only qualities that I see are face-down and red-black, modified by getting more cards. I suggest that this game needs one more card-quality in play ... whether it's a rank thing, a face-card thing, or what, I don't want to suggest (not being the author). I do think that whatever it is, it ought to involve Lula. It might also allow someone else (non-GM) to contribute to the active player's situation.

2. Change the names of Coolness/Opposition objects from "cards" to "points" or "tokens" or something that actually describes them as physical objects. They aren't cards, and calling them "cards" drove me insane while trying to understand the rules. Yes, I understand that having these tokens/points means you get to use more cards, but they themselves, especially as transferred back and forth, are not cards.

3. Scene construction + ratting seems odd. Apparently, you are stating outright that if I get ratted on, I know who did it. How exactly do you (the victim) know who ratted? Furthermore, is it necessary to know who? It seems to me more fun to have that be a mystery, potentially. In other words, let that knowledge play out as people see fit.

4. Should Lula be a bit more centralized or standardized? On the one hand, I like her chameleon-quality, and I really really like the idea that "winning" will actually establish who she really is as a character. On the other, this contradictoriness might make the SIS too vague and irrelevant - which in a card-based, round-based RPG, is an ever-present danger. A playtesting issue for sure.

5. Am I reading correctly, that in the final round, if you lose your conflict, that doesn't have any impact on the final ending for your character? Is the outcome of that conflict merely raw material for the final narrations? It seems to me that conflict success-or-failure needs a lot more teeth in the game, affecting mechanics, throughout play - but especially during that final round.

So finally, how to present and design this thing? The current text needs better explanations and examples, which beefs it up a bit. I'm thinking along the lines of how Paul did the hard-copy (heh) version of Bacchanal, a half-size staple-bound booklet. The more it looks like a pulp zine, the better, with the perfect combination of inexpensive + classy.

Best,
Ron

A quick enthusiastic post: I liked One Can Have Her a lot. It struck me as extremely well-written.

I liked that it genuinely seems to be a noir game, not a noir pastiche. You expect the players to be playing genuinely noir-ish characters, not comedy private detectives. Lula is a perfect noir femme fatale and the idea that all men see what they want to see in her is wonderful.

I'd like to see the text rearranged a bit. The glossary is right near the beginning, but it's too long to read, so you end up skipping over it.

As I said, I liked the way the text is written (e.g. "Regular people are dependable, boring and completely useless for members of the underworld). But I do think it needs editing down. A page and a half of "Theme and Setting" is, I think, too much.

Apart from that, pure enthusiasm.

Hello Ron,

Yes, the prisoner's dilemma is something I've been thinking about for a while, and at least two people besides you have expressed an interest in it. I'm glad to have at least one way of doing it in an RPG now, to see if it'll be fun actually playing. I'm very interested in taking common problems from other fields, masquerade them a bit and use them in play. Besides, great minds have thought a lot about different tactics in the dilemma, so the theoretical side of that part of the game is at least well understood.

Iterated prisoner's dilemma of random length
I used the iterated prisoner's dilemma with a random number of turns since with a fixed number of turns there's an optimal strategy to always choose to rat on the others. In the last turn, if you don't rat you either get a medium reward or none at all. If you do rat you either get a high reward or none at all, so it's always better to choose to rat. By induction you can show that it's better to rat on the second to last turn and so on. But with a random number of turns you can rat on the others without them having a chance to retaliate, if you're lucky.

But yes, the way it is now there's always the risk that you have plans for your character (I don't mean character plans, but rather things you want to express in the game as a player) that are cut short by that end-of-game roll. I wrote it that way because I want the game to feel like a race against time. What your character is trying to do is finish his business before the cops get him, and what you want to do as a player is tell your character's story before the game ends.

One thing I could add to the game is that each of the resolution scenes (not a very good name, but it's all the scenes that might be the last) should be set up as a possible end to the story. The characters should all be in situations that could be the end for the character, even if there should still be things left undone. That way, each extra scene would be seen as a bonus, and a way to make the character grasp for those last straws. The worst thing that could happen, the way I see it, is if the game ends before it has even started to become interesting. If it ends while it's interesting, my hope is that you'll feel a bit sad for the characters that didn't reach their full potential before they were stopped.

Another feature of the iterated dilemma is that it favours cooperation, really. It's an optimal strategy to cooperate until someone defects, then retaliate and then forgive them and keep cooperating. But what I think will happen in play is that someone will rat on the others, they'll all retaliate and the game will end with every character dead or in prison. That would be thematically appropriate. But it's also nice to know that the characters aren't doomed from the start, they can actually all get everything (except the girl) they want.

The use of cards (your point #1) and success/failure mechanics (#5)
You're right that the mechanics lacks teeth; that's actually a very nice way of putting it. Since every player has an equal amount of cards, there's no customization regarding what type of character you play or what types of conflicts you enter. The results of conflicts are purely narrative, it's only a way of finding if the character's story continues this way or that.

I guess you can tell that the prisoner's dilemma was my main idea, and the rest is basically a simple version of PTA ripped out and put here. I wanted something for the players to do during the scenes, and this system provides that. But it doesn't really add anything to the theme of the game, and nothing really changes mechanically depending on whether you win or lose. That's not so good. I agree that you can do a lot of neat stuff with cards, without the game getting overly complicated, and I'll try to think of some nice things that can affect conflicts and fit the theme of the game.

I had some thoughts of using MLwM's type of attributes and let them signal when the game ends. Perhaps a Desperation trait, that increases when you fail or something, and the game can end when someone's desperate enough. A good thing would be that you'' be able to tell your character's story without having the game end by a die roll, but a bad thing would be that it might get a bit to predictable. I thought about a die roll against this Desperation trait to start the endgame, but decided that using the turn number would push people to play aggressively to get things done in time. Also, MLwM traits could be used to show how happy an ending the character gets, but on the other hand I like the simplicity of having it only tied to ratting.

Scene construction + ratting (#3)
I have a couple of reasons why I want all the characters and all the players to know who has ratted, but first I'll have to admit something. I designed it that way because I originally thought about a lot more complex system, where you could rat on a single person and he could rat back on you. The way things are now it doesn't really matter who of the others that have ratted, since you can only rat on everyone or no one. The old system was really hard to build, though, since you get a lot of cases where there's no easy way to say if you should get punished or rewarded, so lets get to the reasons why it's still there.

One reason is because I want the possibility of the situation where a protagonist is caught by a private eye with a gun, working for the police. The character (player) should be able to say "Bernstein (another PC) ratted, didn't he? That's why you're here", because that happens all the time in noir films. People usually figure out who to trust and who not to under the threat of a gun, through great leaps of logic. But on the other hand, that's still possible even if you don't know exactly who's ratted. The fact that the police or someone working for them is in the scene signals that *someone* has ratted, and that should be enough for the character to have a reason to rat back on people.

What's a bit strange about the ratting business is why the character rats on these particular people and how he knows so much about them. Why he's doing it can be answered by them being rivals for the girl, so of course he wants to get rid of them in particular. The second question is a bit harder, but I think I'll just say that the people of the underworld have a way of finding things about each other that might come in handy. The PCs need to have something to tell the police chief, but it felt cumbersome to have them all write down some dark secret that the others know about. Instead I thought that everything the player knows the character knows as well, and can use to sell the others. But perhaps it doesn't matter what he tells the chief's men, and it's the fact that he's ratting that should be emphasized and not what he actually tells them.

I agree that it could be cool to leave it up to the players whether their character finds out who's ratted or not. If they suspect someone, they might ask for a next scene where they go after the other guy to get revenge. Perhaps they also rat on him, to be able to hand him to the police in the scene. All these things could probably be solved on a "meta-level", with the player asking the rest of them who's ratted to see if someone want to answer. If they do, they can all decide which of the other PCs know about it.

Lula Norton, the chameleon femme fatale (#4)
I did a few things to avoid getting Lula to general. I named her, which felt a bit strange, and I made her the police chief's daughter. I also gave her the chameleon like trait to seduce the sucker PC's, but as you say, there's always the risk of her being to undefined. My thought is that she should be created at the same time as the PCs, or directly after. Their view of the dream woman and their future with her should guide the GM when he creates her, and she should definitely be a "real" character and not just some floating ideas of a woman. She should be the in-game personification of the PCs' desires, and she can only be that if she feels like a real person. Perhaps I should look at MLwM's Master creation, and borrow stuff from there. The players and GM could create her together, and then create the PCs. On the other hand, I like it the other way around, building the PCs and then adapt Lula after what they want from her.

Continued design
Yes, examples, definitely. That's one of the things I know that a game needs, but that I didn't find time for. It's really hard to find good examples, if you yourself decide to read them less charitably as you're writing them. I always find that when I write "These are good stakes...", I immediately think "Not if you read that in this way...". I even read the examples of good and bad stakes in PTA that way, trying to find a way to interpret the example in a way it's not supposed to. But that's just stupid, and I have to silence that voice in my head and write some examples. What I would like myself is a longer example of how to play the game with perhaps two players, basically going through a full game (or at least the important steps) and show what people are discussing and what they're thinking. That would show how the card thing works as well, and how to create interesting characters.

Graham, thanks a lot for the enthusiasm, it's very much appreciated.

On theme and setting, I'm actually planning to expand it instead of shortening it. Ever since I submitted the game I've thought about whether it's playable for people who have never seen a noir film. So much depends on people creating the right type of character and putting them in the right situations for it to be noir and not, as you say, noir pastiche or something completely different.

I suppose the best thing would be to have clear decision points in the character creation step, with stuff that the player can write on his sheet for reference during the game. Just selecting a noir archetype is of course enough if you have a clear idea of what that archetype is expected to do in the game. Otherwise it doesn't tell you anything, and is useless as a part of character creation.

What do you think about the part of playing the game in black and white? Is it a silly gimmick, or does it add anything to the game? One reason I like it is just because it feels right, but another is that it forces the players and GM to describe things in more creative ways. Instead of "he has brown hair" they have to say "he has unwashed hair" or "he has well-combed hair", which says more about the guy. What do you think?

Jonas wrote:
Graham, thanks a lot for the enthusiasm, it's very much appreciated.

On theme and setting, I'm actually planning to expand it instead of shortening it. Ever since I submitted the game I've thought about whether it's playable for people who have never seen a noir film. So much depends on people creating the right type of character and putting them in the right situations for it to be noir and not, as you say, noir pastiche or something completely different.

Jonas wrote: What do you think about the part of playing the game in black and white? Is it a silly gimmick, or does it add anything to the game? One reason I like it is just because it feels right, but another is that it forces the players and GM to describe things in more creative ways. Instead of "he has brown hair" they have to say "he has unwashed hair" or "he has well-combed hair", which says more about the guy. What do you think?

Perhaps I should look at MLwM's Master creation, and borrow stuff from there. The players and GM could create her together, and then create the PCs. On the other hand, I like it the other way around, building the PCs and then adapt Lula after what they want from her.