*Coughs politely. * Karma is a switch, and according to Rule 2162: "If an action or set of actions would cause the value of an instance of a switch to become indeterminate, the instance instead takes on its last determinate and possible value, if any, otherwise it takes on its default value."
-Aris On Sun, Apr 28, 2019 at 8:18 PM James Cook <jc...@cs.berkeley.edu> wrote: > > CFJ: A player deputising as the Herald can legally publish a Herald's > weekly report stating the value of G.'s karma switch is 2. > > I think this is PARADOXICAL. > > Evidence: > > 2019-04-07 Herald weekly report: > +1 G. > > Message published by Falsifian 2019-04-10: > Notice of Honour > +1 G. For getting the ball rolling again, and for the cute "(d) > with Notice." > -1 twg for disappearing without a trace (I hope e's okay, though). > > Message published by Aris 2019-04-22: > ... Actually, I'll make this a Notice of Honor: > > -1 G. for not voting on Proposal 8172 despite saying that e would, and > even after many players had endorsed em > +1 G. for restarting the game by publishing the Referee report and > proposal resolutions > > From Rule 2510: > Karma is a person switch tracked by the Herald in eir Weekly > Report. > ... > When a valid Notice of Honour is published, the entity specified > to gain karma has eir karma increased by one, and the entity > specified to lose karma has eir karma decreased by one. > > From Rule 2143: > A person SHALL NOT publish information that is inaccurate or > misleading while performing an official duty, or within a document > purporting to be part of any person or office's weekly or monthly > report. > > My argument: > > Before Aris sent their 2019-04-22 message, the value of G.'s karma switch was > 2. > > Rule 2510 describes the effect of Aris's action of publishing a Notice > of Honour, by describing two changes linked by the word "and". There > are two reasonable interpretations of this: > > Interpretation A: > > The two changes happened in sequence. > > Interpretation B: > > Generally, rules describing the effect of a single action describe how > the game state immediately after the action is performed relates to > the game state immediately before the action is performed. Let N0 be > the value of G.'s karma switch before Aris's action and N1 be the > value after. Rule 2510 says the following two things about N0 and N1: > 1. "the entity specified to gain karma has eir karma increased by > one": so, N1=N0+1 > 2. "the entity specified to lose karma has eir karma decreased by > one": so, N1=N0-1 > The two parts are joined by an "and", which means both are true. > > Under Interpretation A, the current value of G.'s karma switch is 2, > and under Iterpretation B, it's undecidable. > > The rules give us little guidance about which interpretation is > correct. By Rule 217, we cannot favour A just because B leads to a > contradiction: "...an absurdity that can be concluded from the > assumption that a statement about rule-defined concepts is false does > not constitute proof that it is true.". > > If anything, I would argue that Interpretation B (that is: > interpreting "and" as a logical connective, so Rule 2510 is saying two > different things, which happen to be contradictory in this case) is > more sensible, because Interpretation A would require a notion of > splitting actions into finer-grained temporal steps, which is not > described anywhere in the rules. (Closely related: the last paragraph > of Rule 478 states that separate actions within a message happen in > sequence, but Rule 478 does not discuss sub-dividing actions.) > > To summarize where we are so far: the value of G.'s karma switch is > not determined, because we can't rule out Interpretation B. > > Now, the Herald's office is vacant and no Herald's weekly report has > been published this week, so any player CAN deputise as the Herald to > publish eir weekly report. However, Rule 2143 states that for em to do > so legally, the report must not include any inaccurate information, > i.e. information that is false. Since it is logically undetermined > whether the value of G.'s karma switch is 2, it is undetermined > whether they can do so legally. > > (It looks like Cuddlebeam tried to do something similar with CFJ 3638, > but it got judged based on a detail they didn't account for.)
