On Wed, Jun 19, 2013 at 6:23 PM, Fool <fool1...@gmail.com> wrote: > Now, usually to win it is necessary to stop others from winning. Around here > the two things are often unrelated. Win by paradox seems like a perfect > example, it looks like it basically does nothing, so this doesn't affect > anyone else's chances of winning in the slightest. To what extent do people > try to stop others from winning?
In my case, while I am usually happy to see other players pull off scams that award wins, I feel obliged to make a reasonable effort to stop them from succeeding to make sure that they are actually being rewarded for their cleverness rather than the apathy of other players. I consider wins by paradox to fall into this category, but since they tend to come out of nowhere and be unstoppable (typically the undefined behavior is achieved and CFJed on in a single message), there is no real way to stop them. To some extent, the idea of stopping them also applies to judging - as clever as their construction may be, if it's based on an unreasonable interpretation of a rule, it does need to be thrown out - but if it's correct, then in these cases I don't think we usually begrudge players their wins. The same somewhat applies to dictatorship scams, but despite the long tradition of players relinquishing the dictatorship after a time and not actually abusing it, the idea of "some player has absolute power over me" retains the tendency to spark drama and emotion, and accompanying deregistrations; counter-scams tend to be a much higher priority, and judges tend to be much more adversarial to and skeptical of the would-be dictators, which, while arguably legal due to Rule 217, I think is somewhat unfair. The reason I think wins by paradox somewhat cheapen the concept of winning is that while I do believe paradoxes deserve a reward (even if we could continue on perfectly well by simply disregarding the theory of propagating undefinedness, rewarding people for spotting potential issues is an inherently fun mechanic), they usually don't compare with the difficulty of other sorts of wins. This includes scam wins, since they tend to be easier than other scams: they depend on a type of bug that is easier to make, more common, and less interesting (especially as we are really only considering them vulnerabilities by choice) than those that could award dictatorships/wins/mousetraps. Then again, the sample size is fairly low, and I guess I've been burned by a slate of liar's paradox type trivial paradoxes lately (including one achieved only because the rule meant to prevent it was repealed as cruft), which should be fixed now!
