I'm also getting vibes of innocent but can't prove it from BSS, but those are easy to fake amd lynching somebody is still better than lynching nobody. Even though I think she's more likely than random to be innocent, I'm keeping my vote on lynching her because she's still a Schelling point and could be faking the innocence, and realistically our choices are still between lynch her and lynch nobody. If a big enough wagon forms for somebody I think has a higher chance of guilt, I'll join it, but I still think that between the two choices that now look realistic I'll take lynching BSS.
I also agree that being out all day is definitely not indicative of guilt or innocence.
I'm also thinking that the standard Mafia terminology could be off; I have a hard time imagining a real-life lynch mob behaving the way we are.
Re: noise,
Every morning I plan to post a summary of stuff I did overnight, even if that summary is “I slept; what else is the night for?”. I'll use the same pubkey for all of them, but if it's something useless like that I probably won't bother to reveal it.
Effulgence Mafia Game Thread
Re: Effulgence Mafia Game Thread
Here is a scenario.Shoal wrote:But if the person making the random noise says that that is what they're doing then we all know that it's just random noise and nothing to pay attention to. Also random noise produced during the daytime can only be random noise because usually new information only happens at night.
Person A accuses person B of being a killer.
B responds by accusing back citing an ability to observe a single person each night.
He is not very credible because he is clearly trying to deflect blame.
But if he can point to a piece of noise he made the previous day that says "I am going to observe person A this night", he is suddenly much more believable and we won't accidentally lynch the Sherlock.
I actually agree that Alphabeta is above average candidate for mobster even if I don't agree with your reason or the level of confidence.
Staying in my pre-committed vote out of principle. If we can't stick to pre-commitments we can't use strategies that need them.
Probably true, but we don't know there aren't any roles that get info during the day. Or that every good person starts with zero intel.While it's true that they get new info at night
Re: Effulgence Mafia Game Thread
Modrony, I agree that we should stick to precommitments. I built into my commitment the ability to be talked out of it, though, because BSS-or-whoever-it-turned-out-to-be could have been able to defend themselves. I also agree Alphabeta is more likely to be guilty than average, while BSS is less likely, but don’t want to split the lynch vote. I precommit to change my vote to Alphabeta unless that opinion changes as soon as I notice that doing so would put Alphabeta and BSS at a tie (or tilted towards lynching Alphabeta).
Re: Effulgence Mafia Game Thread
precommitments can tie your hands when the circumstances change, when you form new opinions based on new data. it is true that the data we receive in this thread isn't 100% reliable -- people can lie, nobody is going to confess that they did it, but every post from a person is still data of some sort. I think it is good to keep to precommitments when you can, but it is better to leave loopholes in the precommitments for changing your mind and being open to new opinions.
For example, to pick on someone at semi-random, I don't know Paradox. I haven't paid enough attention to their posts before this game to have formed any kind of impression of them at all, and in the 26 pages of this game they haven't been talkative enough for me to from any opinions of them in the context of this game either, so I think there is a 5 or 10% chance they are the killer based solely on the idea that I think there is 1 or 2 killers in the game and the decimal expansion of 1/20 and 1/10. We could precommit to lynch them, just like we all precommitted to lynch BSS, and then Paradox could come and talk more and explain why they don't think they should be lynched and even taking into account that everybody doesn't think themselves should be lynched we could come to have an impression of who Paradox is as a person and what they at least want us to think that their goals are in this game, and then after that we might feel it is more like 1 in 40 that Paradox is the killer, or to go the other way maybe it is 1 in 7. To carry on acting as though we still think it is 1 in 10 or 1 in 20 when we don't, i think that ties our hands. That's why I changed my vote away from BSS. I now feel maybe it is only 1 in 40 for BSS, or 1 in 30, or whatever but less of a chance than 1 in 20 anyway, and for Alphabeta I feel it is a greater chance than 1 in 10.
I think the random data is only has some crediblity if all of the random data posted by a person is unlocked by them at the same time. otherwise there is nothing stopping from a person making ten different posts, one claiming to have intel about sherlock, another claiming to have intel about pen, another with intel about the joker, all of them false, and then only selective reveal. But even with revealing the random data it's not hard to be consistent with the random data, decide ahead of time what you plan to pretend to be and make random data posts pretending to be that. it just means you have to think your lies ahead of time instead on the spot, so you have more time to arrange them and make them fit into a narrative that you want them to.
But if everyone's doing it then it will look suspicious not to do it because that's how mobs work so if most people seem to be posting random data then I'll post recipes or something. At the end of this we can unlock them and you can learn wonderful things to do to potatoes.
For example, to pick on someone at semi-random, I don't know Paradox. I haven't paid enough attention to their posts before this game to have formed any kind of impression of them at all, and in the 26 pages of this game they haven't been talkative enough for me to from any opinions of them in the context of this game either, so I think there is a 5 or 10% chance they are the killer based solely on the idea that I think there is 1 or 2 killers in the game and the decimal expansion of 1/20 and 1/10. We could precommit to lynch them, just like we all precommitted to lynch BSS, and then Paradox could come and talk more and explain why they don't think they should be lynched and even taking into account that everybody doesn't think themselves should be lynched we could come to have an impression of who Paradox is as a person and what they at least want us to think that their goals are in this game, and then after that we might feel it is more like 1 in 40 that Paradox is the killer, or to go the other way maybe it is 1 in 7. To carry on acting as though we still think it is 1 in 10 or 1 in 20 when we don't, i think that ties our hands. That's why I changed my vote away from BSS. I now feel maybe it is only 1 in 40 for BSS, or 1 in 30, or whatever but less of a chance than 1 in 20 anyway, and for Alphabeta I feel it is a greater chance than 1 in 10.
I think the random data is only has some crediblity if all of the random data posted by a person is unlocked by them at the same time. otherwise there is nothing stopping from a person making ten different posts, one claiming to have intel about sherlock, another claiming to have intel about pen, another with intel about the joker, all of them false, and then only selective reveal. But even with revealing the random data it's not hard to be consistent with the random data, decide ahead of time what you plan to pretend to be and make random data posts pretending to be that. it just means you have to think your lies ahead of time instead on the spot, so you have more time to arrange them and make them fit into a narrative that you want them to.
But if everyone's doing it then it will look suspicious not to do it because that's how mobs work so if most people seem to be posting random data then I'll post recipes or something. At the end of this we can unlock them and you can learn wonderful things to do to potatoes.
Re: Effulgence Mafia Game Thread
Shoal, I suspect that there's more than two killers here, and I am certain as I can be that there's more than one for sure. I'd estimate it at 3-4, with twenty people. Some games have a 1-4 ratio of mafia to townsfolk, so I'd say the maximum killers present in this town's likely five.
Re: Effulgence Mafia Game Thread
I've been thinking on this issue, and I think I've got a solution: make all random-noise posts with the same private key, use a different key for each actually important post, and then reveal your random-noise key at the same time as the key for whatever actually-important post you're revealing. That way, if someone makes, for instance, 20 random-noise posts and 3 important ones, and wants to reveal one of the important ones, we won't need to be as suspicious of its content, because we'll know that they only actually have 2 posts still hidden, rather than 22.Shoal wrote:I think the random data is only has some crediblity if all of the random data posted by a person is unlocked by them at the same time. otherwise there is nothing stopping from a person making ten different posts, one claiming to have intel about sherlock, another claiming to have intel about pen, another with intel about the joker, all of them false, and then only selective reveal. But even with revealing the random data it's not hard to be consistent with the random data, decide ahead of time what you plan to pretend to be and make random data posts pretending to be that. it just means you have to think your lies ahead of time instead on the spot, so you have more time to arrange them and make them fit into a narrative that you want them to.
Re: Effulgence Mafia Game Thread
I think that's entirely possible, Aestrix. If it's 3 then it's about 1 in 7 chance for anybody in particular, and if it's 4 then it's a 1 in 5 chance so that's pretty good odds we'll get somebody even just choosing blindly.
Tulip that sounds like a good solution, at least at first glance -- I will think about it some more and/or wait to see what other people think about it. I'm not saying I think there is anything wrong with it, but rather that I think I wouldn't be able to tell right now if there was something wrong. But it sounds right so far.
but really if you are a good guy and you have valid information and you're about to be lynched you should say "carinna is good and bella is good and edward is bad" and not hold anything back, and then either we change our minds or we don't and then when you're lynched Alicorn will say "Charlie Swan was lynched" and then we will be able to guess that means you weren't lying and act on that information. the fact that cryptography was used would only convince us to maybe not lynch you beforehand which is good for keeping you alive and not wasting a lynch on an innocent person but bad because you could still have lied in the cryptographically hidden posts and if we do lynch you then we'll know your information was true 100%.
Tulip that sounds like a good solution, at least at first glance -- I will think about it some more and/or wait to see what other people think about it. I'm not saying I think there is anything wrong with it, but rather that I think I wouldn't be able to tell right now if there was something wrong. But it sounds right so far.
but really if you are a good guy and you have valid information and you're about to be lynched you should say "carinna is good and bella is good and edward is bad" and not hold anything back, and then either we change our minds or we don't and then when you're lynched Alicorn will say "Charlie Swan was lynched" and then we will be able to guess that means you weren't lying and act on that information. the fact that cryptography was used would only convince us to maybe not lynch you beforehand which is good for keeping you alive and not wasting a lynch on an innocent person but bad because you could still have lied in the cryptographically hidden posts and if we do lynch you then we'll know your information was true 100%.
-
BlueSkySprite
- Posts: 531
- Joined: Fri Apr 18, 2014 8:40 pm
- Pronouns: she/her/hers
- Contact:
Re: Effulgence Mafia Game Thread
Okay, I think it seems reasonable for everyone to make an encrypted post summarizing their night. I think I agree that it makes sense for people to make random noise with the same private key, although I'd expect it to be possible to tell if different posts were using the same private key because they would have the same public key. (Unless I misunderstand something about how the encryption works, which is definitely possible)
I'm pretty sure the killers can't communicate with each other by PM, because the Gods have explicitly forbidden it to everyone, and in most games of Werewolf, the wolves don't get to communicate about anything except who they are going to kill.
I'd really prefer not to get lynched due to my incompetence at the cryptography website and outside life, but I don't think I have any new information. If you decide to kill me, I'd like enough warning (preferably during my waking hours, which are I think around 04:30 to 20:00 by this website's time)
I'm still not actually sure I get the math for why it is a good idea to randomly lynch someone. From the first post about it, the game requires 8 players, and is better with 16. If we assume 2-4 killers for game design reasons (based on what Aestrix said), and that it is most fun with at least 16 people, I assume that most of those roles are templates that they have cool ideas for. So unless I am messing up the math, there are around 12-14 templates who are not killers, and I would be surprised if most of them were mostly useless.
(I am not actually good at non-homework math, so please correct me if I am wrong about the maths)
EDIT: FIXED SPOILER BOX
I'm pretty sure the killers can't communicate with each other by PM, because the Gods have explicitly forbidden it to everyone, and in most games of Werewolf, the wolves don't get to communicate about anything except who they are going to kill.
I'd really prefer not to get lynched due to my incompetence at the cryptography website and outside life, but I don't think I have any new information. If you decide to kill me, I'd like enough warning (preferably during my waking hours, which are I think around 04:30 to 20:00 by this website's time)
I'm still not actually sure I get the math for why it is a good idea to randomly lynch someone. From the first post about it, the game requires 8 players, and is better with 16. If we assume 2-4 killers for game design reasons (based on what Aestrix said), and that it is most fun with at least 16 people, I assume that most of those roles are templates that they have cool ideas for. So unless I am messing up the math, there are around 12-14 templates who are not killers, and I would be surprised if most of them were mostly useless.
(I am not actually good at non-homework math, so please correct me if I am wrong about the maths)
EDIT: FIXED SPOILER BOX
Maths for killings lynchings
-
BlueSkySprite
- Posts: 531
- Joined: Fri Apr 18, 2014 8:40 pm
- Pronouns: she/her/hers
- Contact:
Re: Effulgence Mafia Game Thread
Oops, sorry for the double post but I did not finish this sentence
I would like enough warning time to post the private key to my powers post (which should work, I think) and also PM a will to the Narrative Gods.BlueSkySprite wrote: I'd really prefer not to get lynched due to my incompetence at the cryptography website and outside life, but I don't think I have any new information. If you decide to kill me, I'd like enough warning (preferably during my waking hours, which are I think around 04:30 to 20:00 by this website's time)
-
ozymandias
- Posts: 7
- Joined: Tue Mar 25, 2014 10:52 am
- Pronouns: They/them/themself
Re: Effulgence Mafia Game Thread
Statement of role:
HOq7edjiLZNPiiUZB1mUudHkT4ZXB4BILQdKjcoaSIWOY69u7X0gVAhcbIJqmMlNeyzNjY0LEm8ncXlTylZGK6ODei84JBrSrz3vM58ZoCC4KYTGWYkM0+f+FxlNCKlabb93KbilTE4E5CYVrtNvKtoyx8f2uGRw92irriIn2V8=
Statement of powers:
LC063DLZSRP1n/a0m2AGJCxImc/uA6A7zy4vf0u2LtdqEH52Ua+kYhy/sL3kbyP4Vpo5DR3HrawX7kZiO+8C1G3ZAzIhv/5/uzD9OJ5UkmRvHfM3tkVQxAjWyHVLgdP1/R715HlWUxNHymlDkZ+ulDo3BxVNtXHLF/FHzl7HcAk=
HOq7edjiLZNPiiUZB1mUudHkT4ZXB4BILQdKjcoaSIWOY69u7X0gVAhcbIJqmMlNeyzNjY0LEm8ncXlTylZGK6ODei84JBrSrz3vM58ZoCC4KYTGWYkM0+f+FxlNCKlabb93KbilTE4E5CYVrtNvKtoyx8f2uGRw92irriIn2V8=
Statement of powers:
LC063DLZSRP1n/a0m2AGJCxImc/uA6A7zy4vf0u2LtdqEH52Ua+kYhy/sL3kbyP4Vpo5DR3HrawX7kZiO+8C1G3ZAzIhv/5/uzD9OJ5UkmRvHfM3tkVQxAjWyHVLgdP1/R715HlWUxNHymlDkZ+ulDo3BxVNtXHLF/FHzl7HcAk=