==============================  CFJ 1975  ==============================

    ais523 could, at the time this CFJ was called, leave The Gnarly
    Contract by announcement

========================================================================

Caller:                                 ais523

Judge:                                  Pavitra
Judgement:                              TRUE

Appeal:                                 1975a
Decision:                               OVERRULE/FALSE

========================================================================

History:

Called by ais523:                       26 May 2008 16:50:26 GMT
Assigned to Pavitra:                    27 May 2008 01:45:43 GMT
Judged TRUE by Pavitra:                 27 May 2008 02:15:27 GMT
Appealed by omd:                        27 May 2008 13:29:03 GMT
Appealed by ais523:                     27 May 2008 14:10:23 GMT
Appealed by Wooble:                     27 May 2008 14:53:34 GMT
Appeal 1975a:                           27 May 2008 19:20:05 GMT
OVERRULED to FALSE on Appeal:           13 Jun 2008 15:48:35 GMT

========================================================================

Caller's Arguments:

Whether I can leave this contract by announcement depends on my gnarliness,
which
I don't know. Some time has elapsed since I flipped my gnarliness to twisty;
in
the meantime, it has been flipped by the contract automatically. (Some
evidence
that this is possible: rule 1504 reads, in part, "If an exiled entity is ever
a
player, e is deregistered.", which implies that the rules (and so presumably
contracts) can cause things to happen automatically based on some trigger.)

Either a ruling of TRUE or FALSE is inappropriate, because there is no way of
knowing what my gnarliness is right now (and extra information would not help,
because nobody tracks gnarliness, and so there is no way to determine its true
value at the time the CFJ was called; ). IRRELEVANT is inappropriate, because
it
is certainly relevant to know whether or not I can leave the contract. So this
leaves UNDETERMINED and UNDECIDABLE. The statement of the CFJ is not
nonsensical,
and is not vague (especially given the evidence section). It is probably
possible
to make a case that the evidence available to the judge is insufficient to
tell
between TRUE, FALSE, and UNDECIDABLE; however, there is no additional
information
that would help, other than information on how to apply the rules to trigger
loops
such as the one in the contract. (The fact that no trigger loops previously
existed does not prevent their potential to exist; imagine, for instance, if
there
was a rule that said "If an exiled entity is ever not a player, e is
registered.")
Therefore, UNDETERMINED is also inappropriate. UNDECIDABLE is appropriate
because
as has already been explained, the statement of the CFJ is not capable of
being
accurately described as either true or false.

========================================================================

Caller's Evidence:

ais523 wrote:
> I agree to the following contract (named "The Gnarly Contract"), and
join it as a party:
> {{{
> 1. This is a public contract.
>
> 2. This is a pledge.
>
> 3. Any player may join this contract by announcement.
>
> 4. There is a property associated with each party to this contract,
known as eir
> gnarliness; it acts much like a switch possessed by parties to this
contract with
> values in the set {slippy, spiky, twisty, mixy}, with slippy being the
default.
> (It is not a switch because it is not defined by the rules, and it is
not tracked
> by an officer.) A party's gnarliness CAN be flipped as described in
this contract,
> and by this contract, and CANNOT be flipped except as allowed by this
contract.
>
> 5. Any party to this contract may flip eir gnarliness to either slippy
or twisty
> by announcement.
>
> 6. Whenever a party's gnarliness becomes twisty (including, but not
limited to,
> being flipped to twisty), it is flipped to mixy.
>
> 7. Whenever a party's gnarliness becomes mixy (including, but not
limited to,
> being flipped to mixy), it is flipped to spiky.
>
> 8. Whenever a party's gnarliness becomes spiky (including, but not
limited to,
> being flipped to spiky), it is flipped to twisty.
>
> 9. A party CAN leave this contract by announcement, but only if eir
gnarliness
> is mixy.
>
> 10. When this contract has been in existence for at least a week,
paragraph 9
> changes to "A part CAN leave this contract by announcement."
> }}}
>
> I flip my gnarliness (as defined by The Gnarly Contract) to twisty.

========================================================================

Judge Pavitra's Arguments:

Because there is no time delay between gnarliness becoming a given
value and it being automatically flipped to the next value, the waiting
period for any non-slippy gnarliness to become mixy is zero.

Thus, at any given moment, gnarliness is, if not slippy, effectively
each of spiky, twisty, and mixy. It cycles through these in sequence,
but that sequentiality passes perpendicularly to time rather than along
it as a timed cycle would, so that although gnarliness passes through
each value individually and distinctly, as if "one at a time", in
practice all three values are "available" in any given chronological
instant.

Therefore, at the time the CFJ was called, ais523 was in fact mixy, as
well as spiky and twisty, and thus could leave the Gnarly Contract by
announcement.

========================================================================

Appellant omd's Arguments:

So... observation of the switch having any one value would collapse
the wave function, and prevent any further attempt in that message
that would assume gnarliness to be any other value?

Well, that makes sense, but then again, not really.  Quantum mechanics
confuses me.  And I'm not sure it's been established to exist in
Agora.

========================================================================

Appellant Wooble's Arguments:

This whole case is a great argument against Platonic gamestate that
automatically changes and in favor of the Agoran tradition of
pragmatism. I believe self ratification is really the only automatic
gamestate change we've got right now, and preventing contracts that
create bits of gamestate that automatically make changes to themselves
would probably be a good thing (especially in cases, unlike this one,
where someone is actually responsible for tracking those changes, and
even moreso if those changes involved, say, changes to assets without
a default recordkeepor).

However, I don't think any of this matters here; I believe this case
falls firmly under Rule 2197's "permissibility cannot be determined
with certainty at the time it is attempted", which argues for FALSE.
I support comex's intent to appeal.

========================================================================