On Sun, 1 Oct 2017 at 03:28 Aris Merchant < thoughtsoflifeandligh...@gmail.com> wrote:
> Create a power 3.0 rule entitled "Conditionals and Determinacy", with the > following text: > > A conditional is any textual structure that attempts to make a statement > affecting any part or aspect of the gamestate (the substrate), or the > permissibility or possibility of any action affecting such a part or > aspect, > dependent on the truth value or other state of a textual structure > (the condition). The condition is said to be "affixed" to the substrate > (inverse "to be conditional upon"). > Isn't "to be conditional upon" the same as affixation? The condition is affixed to the substrate; the condition is conditional upon the substrate. So they are synonyms. > A condition is inextricable if it is unclear, ambiguous, circular, > inconsistent, paradoxical, depends on information that is impossible or > unreasonably difficult to determine, or otherwise requires an > unreasonable > effort resolve; otherwise it is extricable. A conditional is > inextricable if > its condition is inextricable; otherwise it is extricable. A player > SHOULD NOT > use an inextricable conditional for any purpose. > > If a restricted value, or the value of a conditional, or a value > otherwise > required to determine the outcome of a restricted action, CANNOT be > reasonably > determined (without circularity or paradox) from information reasonably > available, or if it alternates instantaneously and indefinitely between > values, then the value is considered to be indeterminate, otherwise it is > determinate. > If an action would, as part of its effect, make a restricted value > indeterminate, it is void and without effect unless it is explicitly > permitted > to do so by a rule; this restriction should be interpreted in accordance > with existing precedent, and this rule defers to judicial discretion and > game custom. > I'm worried that if the game state gets into an ambiguous situation, this rule just becomes actually harmful as it nullifies all the actions that made the gamestate ambiguous, preventing us from picking a resolution and forcing us to treat things as if the ambiguity never happened. It gets worse if the ambiguity is introduced platonically sometime after the action in question: does this retroactively invalidate the original action? > > Create a power 3.0 rule entitled "Conditional Announcements", with the > following > text: > > A player SHALL NOT deliberately make an action taken by announcement > conditional on an inextricable condition, and any such conditional is > INVALID, and its substrate void and without effect; these restrictions > should > be interpreted in accordance with existing precedent, and this rule > defers to > judicial discretion and game custom. > > Extricable conditionals do not necessarily fail; however, they must be > reasonably resolvable given complete knowledge of the gamestate at the > time > the message takes effect. This knowledge CAN require interpretation of > data > in non-trivial ways (e.g. interpretation requiring CFJs), but such > interpretation must be achievable without absurd effort. No > by-announcement > conditional may ever be conditional upon information which cannot be > deduced > from the knowable gamestate at the time the message takes effect, nor can > such conditionals ever change the time the message takes effect. > > Loops are generally viable, subject to the above restrictions. However, > long loops used abusively may fail, at the discretion of a judge; the > presumption is in favor of the loop being sucessful. > I strongly oppose letting judges discretionarily and retroactively cause actions to fail, especially without a stronger procedure in place to define what that means (e.g. return to binding CFJs. I also generally dislike the amount of platonic reliance on things like "unreasonable" that we have here. I've heard some arguments defending its use in AIAN in order to reduce the scam potential (as that *really* isn't a rule that should be scammable), but I don't think it adds value here as it adds uncertainty to the gamestate especially, as mentioned, in the absence of binding CFJs. > > This rule is intended as a codification and clarification of existing > precedent. It does not attempt to overrule existing precedents, but only > to > make explicit the principles by which its subject matter is to be > understood. > > Create a power 3.0 rule entitled "Rational Action", with the following > text: > > An irrational action is one that is either deliberately hidden from view > inside a larger message (e.g. a report) or contains excessive repetitions > or complex loops that make the message unreasonably hard to comprehend or > respond to. All other actions are rational. > > A player SHALL not take an irrational by announcement action, and any > such > actions fail. > Again an example of platonically trying to capture a very subjective definition; everyone's view of what is excessive or complex or even deliberate is going to be different. > > Amend Rule 1023, "Common Definitions", by removing the third item of the > top level list, and renumbering appropriately. > In general, I'm not a fan. I don't think this actually works significantly to clarify the territory and introduces a huge amount of subjective platonism. I don't think there's a huge amount of value, and I don't understand why this needs to be in place for binding contracts to work. Perhaps you could elaborate on that? -Alexis
