twelfthrootoftwo

That's bad luck, Bartbug Best wishes. Hope it's not too serious.

My gut feeling is telling me that Koll and Popeye are trustworthy. Koll in particular has made an effort to strategise and plan (even if most of it hasn't been seen), which suggests to me he's not trying to 'fly under the radar' as a new eliminator. Changing my vote to Quitania for the reasons stated above.

As I argued before, there was only one death and that person was a Tineye. Since none of our trusted people had Allomantic tin, they are therefore all still trusted. He'd do exactly that, in the hopes that someone would draw that exact conclusion Though I have reason to believe this isn't the case. I'm actually a bit suspicious of Morgan, since he's posted a lot more this cycle than previously. Though he was challenged by Satrams, and I'm not inclined to discourage frequent posting. In fact I think pushing people to post, trying to catch out the awkward kandra-body interaction, is the next step in finding Joe. Terra and Quitania already have votes on them, so PopEye, what's your status? Quitania is probably a good target for the eventual lynch. Claiming to be vanilla makes her either likely to be a kandra herself, or a good target for the kandra to impersonate. Besides which, from my experience with Quitecontrary as a player, I don't think it's likely she would say she was vanilla if she actually was. Edit: green

Well, there was only one death, which was the person Joe is now puppeting. The alternative way to lose the Tineye power might be to get spiked and the spiker not 'use' the spike, though to my knowledge this was not the Set's plan. And it seems unlikely, unless things have been going on behind the scenes that mean the Set particularly wanted to test the person claiming to be a Tineye. So, I think finding our (former) Tineye will lead us to Joe. Tulir, there were messages explicitly left by a tineye in the writeups: If the people can get behind Alvron’s leadership, we can take our city back from the gangs. This could possibly indicate the Tineye is a Politician. It's likely to be largely RP, since stopping the gangs is Alron's RP motivation. If it's game-motivated, though, it could refer to the author not being a member of one of 'the gangs'. Plus, Alvron started the Politicians PM group, and was the primary force pushing for discussion there. On the other hand, it could be someone from the other two factions bluffing their identity. Something to consider alongside this is that (at least one member of) the Set think Joe moving bodies was partially because he'd figured out all their powers. He was probably planning to move to a different faction, to learn their powers, which means his target was probably a Politician or Rioter. We must follow Aonar’s plan and root out the Kandra! Self-explanatory, and not that helpful, since (to my knowledge) no-one disagreed with Aonar's plan before the end of the corresponding cycle. So: Popeye Araris Known former Feruchemical iron Quitania Terra Alron Jain Known Feruchemical bendalloy Tulir Taliz Claimed Allomantic iron Kamden Morgan Koll Bunnt I'm not sure where to go from here. Edit: blue

For what it's worth, I was contacted by someone claiming to be a Seeker. They correctly identified Aonar as a vanilla player on their night 1 scan, so I'm inclined to believe them. If you wanted a backup for the Soother, we could suggest they Seek the other Set member and then send the results to someone they trust. If we go ahead with this plan, can I put my hand up to be spiked? Not this night cycle, as since you aren't sure I'm actually a Lurcher then it's not a valid way to test the Set spiker. But as it stands, my power isn't much use to us. Either I protect myself, which (since I'm not trusted) gives us next to nothing, or I protect someone else, in which case the kandra already know who's doing the protecting and they just come after me. This way, we can be sure the Lurching power is in the hands of someone trusted, and we can go from there.

Spiking can be used to clear two players, but unless the role in question is a protective role, they won't remain clear for any longer than our other methods. Beyond that, we'll antagonise the Rioters (two of which have already expressed their distrust of the idea). In this game, an alliance between a faction and the eliminators is entirely possible, since the kandra can comfortably make their escape once the Rioters are satisfied with the balance of power and ready to lynch them instead. Further, if we coordinate spikings like that, it means there will be a lot of spikes produced. It will be all to easy for the kandra to come in and collect those spikes (again, with the exception of protective abilities), then slip out with their new powers when the Set realise they can't make spikes themselves. In summary, I really don't think clearing people with spikes is a good idea. --- The next topic is voting, which we all seem to have forgotten about. On the off chance there is actually a Soother or another Rioter in the game, I'll move my vote to someone with no votes on them at the end of the cycle (probably myself, if no-one votes for me). Tulir, you've been checking the politicians PM regularly, and I saw you reading the thread after the day began. You've only posted once each in the PM and in the thread, though, through the entire game. What do you make of the events of last night?

Well, that puts flight to that idea. I'll make an isolated vote next cycle, in the hope that there's a Soother who will be able to confirm me publicly, though since there haven't been any other Soothings, it looks like there might not be. Otherwise, any available Seekers or Feruchemical tin can scan me this cycle, and send the results to Satrams?

Okay, thanks for the clarification. Now I've heard that, I think it's (past) time I put my cards on the table. My vote was never added to the tally, even after I commented on it. I think it's therefore fair to say I was either Soothed or I was filling Feruchemical pewter or Feruchemical gold, removing my vote. Now, this is extraordinarily convenient for Aonar's plan, because I also happen to be a Lurcher. I was debating whether to keep quiet and protect Satrams on alternate nights, or announce myself like this. In the end, I decided to go public because of a comment in one of my PMs - namely, that some players (Riots in particular) might not want to share their power with a member of the Set. (This is basically acknowledging that faction tensions do exist, and has nothing to do with Satrams himself.) Being a Politician and thus only interested in stopping the kandra, hopefully they would be more willing to share that information with me. If there's any doubts, then our (hypothetical) Seeker and Soother can target me over the next night and day. ...What does everyone think?

Joe, taking a body requires the kandra killing the target, which they can't do if the target is protected with a Lurch. They might succeed if the target is instead killed by Feruchemical pewter (which can hit through protection with multiple charges), though this is just speculation on my part. Posting quickly because another (potentially major) problem with Aonar's plan just occurred to me. Are Set kandra allowed to make spikes? If so, then they'll just spike our trusted person, then kill them. They would need to be protected by someone else whose power wasn't known.

I had a vote on Alim; did I get Soothed? ----- I have a few questions about Aonar's methods for identifying the kandra: -Seeker is fairly self-explanatory. Oracle - Allomantic electrum, according to Coppermind - knows if they're going to die, which (unless I'm missing something) won't help us find someone's powers. -Soother checks if the player loses their vote; Allomantic copper is removed as a possibility because Feruchemical bronze/Feruchemical steel can check to see if it was used. -Coinshot checks if the player dies; Feruchemical gold and Allomantic iron can be excluded because Feruchemical bronze/Feruchemical steel can check to see if it was used. -Was Oracle meant to be Feruchemical bronze and Feruchemical steel, which show the commands issued that night twelve hours early? (If it is, what happens to the users of active powers if they put in their orders after that twelve hour deadline?) -It sounds like the hub Lurcher needs to be comparing the results of investigations with a list of claimed metals (to check for loss of powers) or trying to catch the kandra in their original bodies. So the plan requires everyone to claim their power, at least in private? Edit: grammar

Ah, there we go Aonar As to why the kandra didn't steal Lucky's body: -There was no advantage to taking their place: there weren't any networks Lucky had that they particularly wanted/didn't already possess. Chaos and confusion from lots of body-switching is not part of their strategy. -There was a disadvantage to taking their place: Lucky was in contact with people they didn't want to have to deceive. (Having a difficult-to-fake power would also count, if Lucky revealed it.) -There was no advantage to the kandra abandoning their body: there was no suspicion placed on them or trap they had to evade -There was a disadvantage to abandoning the body: the kanda didn't want to fake an identity. Or they didn't want to abandon networks they'd already set up (unlikely, since judging from the thread it doesn't seem there's been enough happening that they'd need to change their plans). Overall a lack of body-switching indicates two things: either the kandra are newer players who want to minimise the amount of deception they use, or (a) more experienced player(s) who prefer a 'transparent' playstyle. LG7 had Alim take exactly that approach. Bartbug, what was the final ruling on whether kandra kills/abandoned bodies would be distinguished in the writeup?

Lucky is an interesting choice for a first kill, unless she's said something significant in private. What's the bet she's our first kandra? Also, I think kandra can't be killed except by lynch - if Meta said anything particularly interesting, it's probably trustworthy. I'm going to vote for Aonar, because the half page of analysis that normally shows up when everyone's quiet hasn't appeared. (And because I need to pick someone likely to respond fully if I want to stir up any debate.) Edit: green

No response from Morgan; I'll move my vote to Tulir. Araris, I don't agree with your logic. Spiking requires knowledge (or a correct guess) of a player's powers, and at this stage it will be very difficult for them to get it right. And killing off the Set won't decrease the rate of Spikes - as a faction, they get only one per night. The only way I can see to keep spikes out of kandra hands through a lynch is to target someone currently holding a spike, and trying to do that essentially starts a gang war.

Thankyou Binnut. To clarify my argument, it's a bit more complex than looking for details in RP to distinguish the kandra. This cycle is our opportunity to establish baselines for things such as posting times, which will inevitably be disrupted if a player is possessed. Depending on the player, RP might be easier to post at this stage than discussion, which is why I mentioned it. Bartbug, can I clarify the obligations of a possessed player? My current understanding is: Depending on the kandra's wishes, they are obliged to: check the game and coordination doc at least once per 24 hour period provide all information about actions and PMs relating to themselves in the period after being possessed act, PM and post according to the kandra's instructions translate content provided by the kandra into their own words ​They are not obliged to: provide information about actions and PMs prior to being possessed contribute game-related content (and if they are, is sabotage allowed?) ​ Also, a vote...let's hear from Morgan Hanks. Edit: I think an example will make my question clearer. Say there's a kandra-body pair using a translation based system. Kandra instructs body to vote for X with Y reasons. Body translates Y reasons into his own language, and posts. But what if kandra just says to vote for player X, and body decides that a vote without reasons is not their style? Are they obliged to come up with a reason, or can they just say "I vote for X"?, If they do need to come up with a reason, does it need to be convincing, or can it be something like "X made a typo in their last post"?

Taliz reviewed the contract. This was the third-party statement; while House Hasting had sponsored their work, it was common for other houses to attempt to buy new research. For a hefty sum, and a guarantee of secrecy, of course. A typical third-party statement from Hasting promised to outbid such offers in exchange for silence. This one was a little different. Hasting was trying for a solid hold over the market, and that meant undermining their opposition. The statement contained a bonus payment, which would be used to develop a version of the design for 'public distribution' - with the implication that it would be designed to fail. Taliz chewed her lip. She was no stranger to these underhanded dealings; bribes were a financial necessity, and she'd happily take money off the richer folk to fund innovation (and her watch collection). Not that it was her decision to make; the team leader had already signed for her. But for once, there was a challenge. How could you make something that was designed to fail? That would be subtle enough to avoid detection and reliable enough to destroy reputations? No, this would be a true challenge. She smiled. There was life in their piston project yet. ...A corrupt engineer was trickier RP than I expected. If I'm reading the game correctly, this day/night cycle is the only time we have where we can be sure every post or PM actually comes from the poster/sender. Even more than most games, volume of information on this first cycle is important. We can't just track a single player's behaviour through the game to identify the kandra; they have the option to move. And identifying the change requires something to compare to. So post some discussion, RP, questions about the rules, what you're doing on the weekend. Half-formed ideas and single sentences are still valuable contributions to the team effort! In that vein, while I will probably switch to a no kill vote later, Binnut gets my vote until he posts.

I've been tossing up whether I'll have the time to play, and it looks like I will. Taliz Duede is an engineer, with a penchant for messy notebooks. Taliz carefully positioned the ruler, and marked in a few more lines. She frowned at the plan. This page was nearly finished, but if she made a mistake now, she'd spend a few more hours drawing it again. Stomach twisting, she looked over at her notes, and flicked through a few sketches to see if she'd ruined her work. No, that was correct. Her sketches hadn't been scaled properly, that was all. This was the worst part of the process. The test models of their new double-flow piston had run beautifully, but the final product required proper drawings. No problem to solve, no challenge to overcome. No promise of discovery just around the corner. Just hours of precise lines and careful angles. Still, it would be nice to see the project she'd been working on for the better part of two years fulfilling its purpose, and that meant preparing for manufacture. The new design promised greatly increased efficiency. Nothing flashy, but then she didn't really want the kind of publicity that, say, an impenetrable armoured car would attract. She shook her head. Now was not the time to start daydreaming. She wanted the plans done for tonight - meaning no mistakes.

Nice GM work, Ren! And well done to all the villagers (in part for not jinxing us by commenting on our good luck). ----- Tulir, I actually want to apologise for doing that analysis. At the time it seemed like a good idea, but as I worked through everything it felt more and more like a dirty trick. In the end, I decided to post it largely because I didn't want the time I spent on it to be wasted, and I regret doing so. Linguistic analysis like that disadvantages multilingual players, those who post greater volumes, and those who otherwise have a distinctive grammatical fingerprint. And while it's not possible to completely remove that aspect from the game, it shouldn't be focused on the way I did. I'm sorry I did that analysis, and I'm sorry you were lynched because of it. In future, I will restrict my analysis to aspects of the game itself.

52197 Thanks for the response, Tahrin, but I'm keeping my vote as it is for now. A brief summary of what Rulit said in their message would be greatly appreciated and not too time-consuming for you. Gart's arguments certainly don't sit right with me, and at the moment I am more suspicious of him than Quismet. If I hear back from Tahrin, that will probably be where my vote ends up. edit: blue

Tahrin, I would very much like to hear some more about your messaging patterns. Were you really messaging Rulit every cycle and still not discussing other players? Or were you also contacting other people that you haven't told us about? I'll add my name to the list of supporters for Gamma's plan. A note for everyone participating: just because we all have keywords doesn't necessarily mean it's alright to make those keywords obvious. If the SB intercept your message and they're all easy to spot, they can identify you because your posts don't have a keyword. I'm not saying it's necessary for everyone to have 'hidden' keywords, just that we will need a couple. (At the other extreme, I doubt Gamma will appreciate having to look for subtle hints in thirteen RPs. Not that I'm putting words in his mouth or anything )

I must admit, the lynch result was a pleasant surprise. Rulit was my #2 guess (after Jain) going in, but after doing the analysis I decided it was less likely than the others I called out. Kudos to Naihar for making the right choice! I'm not so certain the attack on Wurum indicates our SB hasn't been paying attention. I would have taken that as playing protection roulette between him and Gamin. Wurum is protected => better to hit Gamin, who might be protected by squires => WR guesses this and protects Gamin => hit Wurum. And so on. So yeah, don't discount people (including me) just because it looks like the SB made a mistake.

Unfortunately, circumstances change. In this case, I went back to uni between my lynch in the last game and the beginning of this one. It's a lot more convenient for me to make longer posts less often now. As for what I was doing while I was on earlier, it was research: I'm going to try a slightly different approach to this. Mallaw is right; the message Tunnib posted is recognisably not Aladdin. Since the author didn't succeed in imitating Aladdin's language, I think it's reasonable to expect they left their own language instead. So it might be possible to identify where the message came from that way. (Hypothesis: my initial feeling with this was that it sounded like Jain.) I picked out two particular grammatical elements here, which I think are the most distinctive hallmarks: sentences formed by two main clauses without a conjunction, and non-use of contractions. I then went through the posts (starting from signups) of those who haven't had their messages confirmed, looking for these elements. Character: No. instances I noted Quotes Comment Ignoring small sample sizes (ie those who have posted little text), the people who have demonstrated both qualities I was looking for are Jain, Quismet and Tunnib. Jain has demonstrated the fewest. In addition, his example of the structure element doesn't seem like it would appear in a more formal context such as the message. Quismet has examples of using "I'm", in contrast to "I am" from the original. Tunnib, to my knowledge, has not used "didn't" or "don't", both of which appear in the original. ...In summary, I've spent way too much time on a tiny detail. Quismet, I would like to hear what you did with your message.

The [spam] was mine. I kept it short so you weren't wasting your time reading, but in hindsight I should have added some better ID method. I'll post more in a few hours; just wanted to get this out there.

I think it’s safe to assume Meta’s not lying about being a squire, since that would be too easy to call out. Also, some numbers for reference: I get ~45% chance of there being a squire with Full Lashing (FL from here on) (assuming 2 recruits first cycle, and accounting for Jeno and the three other Surgebinders), and 9% for a Henchman with FL (starting with 3 henchmen; it’s about half that for 2). Scenarios for the FL on Meta, sorted by Meta’s role/alignment: Skybreaker: FL, by a WR or Squire, prevents them from using their kill. Here the message did go through, and Meta is lying to make the number of lashings consistent. No powers/TW: WR-WR miscommunication - the FL came from a WR. This could be misreading of a code, or simply one not being on between the change of plans and the cycle end. I expect the WR tried to form a broad plan for the first cycle or two while they had their doc, which means one of the two actions was a substantial deviation (I would lean towards this being the FL). None of Meta’s posts last cycle jump out at me as reason for a change of plan like that (anyone disagree?). WR-Squire miscommunication – the FL was from one of the Squires. No need for a code or change of plan here, just a squire that suspected Meta. Villager: SB concern – a henchman squire with FL thought Meta may have been a Surgebinder. They may have had reason to think that, or they may not have had any suspicions and gone for a player where the power would be most dangerous. Or (unlikely) they were worried about the vote on Aladdin. Setup – This one is contrived, but not as unlikely as, say, the SB Wurum gambit. A henchman, knowing the SB plan to use one RL, receives FL on becoming a squire. They target someone who voted, hoping to set them up as a SB and get them lynched. ----- Something we can do this cycle to figure out what happened is try to find where the lashings went. We should expect a minimum of four and a maximum of six. So far, we have: Division – Wurum Basic – Wurum Full – Metam Reverse – Metam (claimed) Picking Basic and Full lashings on most players won’t be possible, but we may be able to find Reverse lashings if we reveal messages sent/received – which some people have already done. I’ll add to this and say I was one of the many people who messaged Mallaw, and I received a message from Lyce (in response to the message I sent last cycle). If we decide to go this track, it would also be useful to know whether there actually were two squires last cycle, to give us a better idea of how many lashings to expect. ----- I agree that this is strange. It may be that the SB want to leave more targets for the village to lynch, but at the same time they're making it harder for them to find the WR and TW. Collecting a single message is not that likely to identify someone's role. It would make more sense to me to thin out the search field and simultaneously remove known threats before starting to look for unique targets.

(Apologies in advance for any incoherence, it's late and I'm tired.) I'm procrastinating from doing assignments. If I were to post in those visits, it'd be a throwaway line or two, not worth the time it takes to read (or write). If you'd prefer, I can avoid coming on until I'm ready to sit down and decide what to say. ----- Since I don't really have anything to push in the way of suspicions at the moment, I'm going to ask a question. A lot of you are going to think it's really stupid, and I'm probably an idiot for not having figured out the answer. Why do we want to lynch inactive players? I would understand if it was presented as clearing away a hiding spot for eliminators. Or if it was purely a method for getting some discussion happening. Or if the threat of lynching is intended to make them active. But the latter two are purely about information - there's no advantage to going through with the lynch. And the former doesn't seem to come up as often as a straight 'lynch the inactive.' Once the game has started, they're just one more (or one less) player the eliminators need to kill to win. Since they can't contribute to their team, they become (in a sense) a meat shield, which is statistically more likely to be on the village side. There's not much gained or lost by killing them. Lynching them also reveals little information compared to participating players, though it does minimise long-term information loss. The relative values of these two aspects seem to depend on the projected length of the game, so it's most advantageous to do so in long games, not the quick ones where it seems to be brought up more. Clearly I have this completely wrong. What part of the rationale am I missing?

I had the idea of using the number of kills to estimate the number of messages stolen by the Skybreakers, since they need to choose between their Lashings. But after doing some calculations, the chances of there being a Henchman squire with Reverse Lashing is nearly 1/4 after three rounds of perfectly efficient recruiting (for both 2 and 3 henchman cases), so it's not reliable enough to be useful. Revealing messages could certainly be useful, though I'm worried it will allow team evil to identify the Windrunners too easily. Information- and communication-wise, they're in a similar position to team evil, which means we can expect their PM patterns to look similar (or similarly faked). And while the village won't know their alignment for certain, they then become prime targets for Division. Of course, as a mid-range game, the value of immediate information is increased, and powers (arguably) decreased. So the trade-off may be worth it. (If we do decide to reveal, there's little point in using Jerric's security system.)