On Tue, 8 Sep 2009, ais523 wrote:
> On Mon, 2009-09-07 at 15:56 -0500, Pavitra wrote:
>> Either the rule is broken platonically (UNDETERMINED is appropriate), or
>> it's broken pragmatically (it's possible to exploit it as I described).
>> In no case is it working properly.
>
> I see what you mean... I'm inclined to think it's broken platonically,
> and if I understand your point correctly, it's that platonic random
> numbers are impossible to determine; this seems to be correct, but a
> mess.
Interesting. Making random choices is one of those things that used to
be covered by the Rules, but it was repealed because a precedent was
straightforward. All of R1079's (a-e) below are reasonable precedents
where the rules are silent (especially as (e) defers to the courts) with
the exception of (b) which is the key here. At the time of the repeal,
all random choices were associated with the person who makes the choice
so this wasn't an issue. A stretch that the recordkeepor could coalesce
unknown randomness into known randomness by making an after-the-
fact random selection; there's an indirect precedent that it's the
process of making the randomness known, not rolling the dice, that
resolves the action (CFJ 1435). Still even that was associated with an
act (making a random choice) and here it's pretty clear that the rules
say "something random happens platonically upon event X".
Rule 1079/4 (Power=1)
Definition of "Random"
(a) When a Rule requires a random choice to be made, then the
choice shall be made using whatever probability distribution
among the possible outcomes the Rules provide for making
that choice. If the Rules do not specify a probability
distribution, then a uniform probability distribution shall
be used.
(b) Where the Rules do not indicate who is required to make a
particular random choice, it shall be made by the Speaker.
(c) When making a random choice as required by the Rules, a
Player may rely on any physical or computational process
whose probability distribution among the possible outcomes
is reasonably close to that required by the Rules.
(d) For the purposes of this Rule, tossing a platonic solid that
is not specially weighted has a probability distribution
among the possible outcomes that is reasonably close to
uniform.
(e) For other methods, the Courts are the final arbiter of
whether a method's probability distribution among the
possible outcomes is reasonably close to that required by
the Rules.
