Okay, after hearing your logic, I think agree with your general ideas
here, but I'd really like #1 and #2 to be explicitly specified
somewhere. It would give us something to direct new players to, and
something to cite in CFJs when the principle comes up. Would you be
opposed to such an explicit provision?
-Aris
On Fri, Jun 21, 2019 at 9:20 PM omd <c.ome...@gmail.com> wrote:
>
> Proposal: Deregulation (AI=3)
>
> Repeal Rule 2125 ("Regulated Actions").
>
> Amend Rule 2152 ("Mother, May I?") by appending after
>
> 5. CAN: Attempts to perform the described action are successful.
>
> the following:
>
> For game-defined actions, the meaning of an "attempt" depends
> on the mechanism the rules define for performing the action.
> If no mechanism is defined, it is not possible to attempt to
> perform the action.
>
> [[[
> Rule 2125 is in an uncomfortable middle ground where it's too
> simplistic to achieve what it wants to achieve, yet explicit enough to
> break things. Potentially lots of things.
>
> Jason Cobb has a proto that tries to make it more explicit, but I
> don't think it does enough to address the deficiencies of the rule –
> such as ambiguity about what it means to "limit", as well as (as far
> as I can tell) the question e emself just raised in a different
> thread, about whether the definition of an "action" includes who is
> performing it or its parameters. (I think it does not, but if we're
> going to flesh out the language, that's the kind of thing that should
> be made explicit.)
>
> I don't want to stomp on eir parade, but personally I'd rather take a
> much different approach. I think we could avoid the difficulty of
> defining a regulated action by splitting the "CAN only be performed"
> clause into two simpler precepts:
>
> 1. The gamestate can only be changed as explicitly specified.
>
> 2. A game-defined action can only be performed as explicitly specified.
>
> #1 implies that any action which inherently involves changing the
> gamestate as part of its definition – like, "the action of making omd
> a player" – can only be performed as specified.
>
> #2 deals with actions like "distributing a proposal" where gamestate
> changes are only a *result* of performing the action. IMO, what most
> distinguishes these actions is not the fact that the Rules allow or
> disallow them, per se, but the fact that they have no inherent
> definition: that they're just Rules-defined platonic "thingies". To
> make this clause fully explicit, it could be expanded in two ways: it
> could use a more expansive definition of "game-defined action", along
> the lines of what I just wrote, and it could clarify that there's a
> presumption in favor of interpreting the rules as defining new actions
> rather than referring to existing ones, even when they use
> ordinary-language terminology ("distribute").
>
> On the other hand, the presumption in favor of 'Rules-defined platonic
> thingy' doesn't just apply to actions but also to things, like coins
> or proposals. We don't have any rule making it explicit in the latter
> case, which suggests that we don't need one for the former either.
>
> In fact, I think #1 and #2 can both be left entirely implicit... which
> is why this proposal doesn't introduce any language to replace the
> "CAN only be performed" clause but just repeals it. :)
>
> The "methods explicitly specified" clause is different. It can't be
> left implicit – it's contrary to precedent established before the
> clause was enacted. And that precedent makes sense: if a rule
> explicitly says that something CAN be done (but doesn't say how), who
> are we to decide, without any rule to the contrary, that it CANNOT?
>
> Hence the new language added to "Mother, May I?". The idea is that by
> saying "it is not possible to attempt to perform the action", it can
> still be (vacuously) true that all attempts succeed. That way there's
> no need to override the rule that says the action CAN be performed.
> ]]]
