status: https://faculty.washington.edu/kerim/nomic/cases/#4032
(This document is informational only and contains no game actions).

===============================  CFJ 4032  ===============================

      There are some persons right now who have more than 0 Rice.

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

Caller:                        Yachay

Judge:                         ais523
Judgement:                     TRUE

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

History:

Called by Yachay:                                 25 May 2023 19:17:23
Assigned to 4st:                                  25 May 2023 23:36:45
Judged TRUE by 4st:                               27 May 2023 17:40:02
Motion to reconsider group-filed:                 28 May 2023 03:15:50
4st recused:                                      08 Jun 2023 22:59:57
Assigned to ais523:                               13 Jun 2023 18:42:28
Judged TRUE by ais523:                            16 Jun 2023 23:56:48

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

Caller's Arguments:

[none provided so far]

--------------------------------------------------------------------------

Gratuitous Arguments by G.

This CFJ was called on 25-May-23, when only one Rice Harvest had occurred,
and Rule 2682/0 was in effect.

The report in question is here:

https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-official/2023-May/017077.html

No factual errors have been noted in this report, in terms of who sent
messages attempting to either make Rice Plans or to sign rice plans
either via direct consent or contract.

The question is wholly interpretive, in that all signatures allegedly
applied to the Rice Plans were governed by this clause of R2682/0:
>      A Rice
>      Plan has an active player's Signature as long as that player is
>      consenting to it.

This clause is written passively, without our general standards like
"CAN sign by announcement" etc.  And R2519/2 covers consent for "actions"
not continuous states so it's unclear how that applies.

Basically, if signing Rice Plans works as generally intended by the
rule's author, then this CFJ is true, with the persons with more than
0 rice indicated in that report.

Judge 4st initially judged this CFJ and eir original judgement is
included below, but a Motion to Reconsider was filed for that judgement,
and 4st was later recused.  The players supporting the motion to
reconsider generally gave the reason that Judge 4st's arguments called
"having a signature on a plan" a kind of "continuous action" that
players were "continually agreeing to" which didn't generally match
Agoran conceptions of actions as instantaneous events (sorry if this
is a very coarse summary of the objections).


------------------------------------------------------------------------
Rule 2682/0 (Power=1)
The Rice Game

      The Ricemastor is an office, in charge of tracking Rice, Rice
      Plans and Signatures. Rice is a fixed asset, ownable only by
      players. Any active player can create a Rice Plan by announcement,
      if e hasn't done so yet in the current week. Rice Plans can have
      Signatures, and each Signature must be of an active player. A Rice
      Plan has an active player's Signature as long as that player is
      consenting to it. An active player can destroy a Rice Plan that e
      has created by announcement.

      A Harvest occurs at the beginning of each week. When this occurs:
      - If there is only one Rice Plan with the most Signatures, that
        Rice Plan is Harvested.
      - If there is more than one Rice Plan with the most Signatures,
        the one that was created earliest is Harvested.
      - In all other cases, nothing happens.
      And then all Rice Plans are destroyed and the Harvest ends.

      Rice Plans consist of two lists of players, with each list having
      no repeated players, and the lists can be empty. One of these
      lists is its Rice Up list, and the other is its Rice Down list.
      When a Rice Plan is Harvested, for each player listed in its Rice
      Up list, if that player is active, e gains 1 Rice; and for each
      player listed in its Rice Down list, if e has at least 1 Rice then
      e lose 1 Rice.

      If after a Harvest there is a single active player with at least 2
      Rice and more Rice than any other player, then that player wins
      the game, and all Rice is destroyed. When the game has been won in
      this manner three times, this rule repeals itself.

------------------------------------------------------------------------

Judge 4st's Arguments:

Evidence:
Rule 2682/0 (Power=1)
The Rice Game

      The Ricemastor is an office, in charge of tracking Rice, Rice
      Plans and Signatures. Rice is a fixed asset, ownable only by
      players. Any active player can create a Rice Plan by announcement,
      if e hasn't done so yet in the current week. Rice Plans can have
      Signatures, and each Signature must be of an active player. A Rice
      Plan has an active player's Signature as long as that player is
      consenting to it. An active player can destroy a Rice Plan that e
      has created by announcement.

      A Harvest occurs at the beginning of each week. When this occurs:
      - If there is only one Rice Plan with the most Signatures, that
        Rice Plan is Harvested.
      - If there is more than one Rice Plan with the most Signatures,
        the one that was created earliest is Harvested.
      - In all other cases, nothing happens.
      And then all Rice Plans are destroyed and the Harvest ends.

      Rice Plans consist of two lists of players, with each list having
      no repeated players, and the lists can be empty. One of these
      lists is its Rice Up list, and the other is its Rice Down list.
      When a Rice Plan is Harvested, for each player listed in its Rice
      Up list, if that player is active, e gains 1 Rice; and for each
      player listed in its Rice Down list, if e has at least 1 Rice then
      e lose 1 Rice.

      If after a Harvest there is a single active player with at least 2
      Rice and more Rice than any other player, then that player wins
      the game, and all Rice is destroyed. When the game has been won in
      this manner three times, this rule repeals itself.
Rule 2519/2 (Power=3)
Consent

      A person is deemed to have consented to an action if and only if,
      at the time the action took place:

      1. e, acting as emself, has publicly stated that e agrees to the
         action and not subsequently publicly withdrawn eir statement;
      2. e is party to a contract whose body explicitly and
          unambiguously indicates eir consent;
      3. the action is taken as part of a promise which e created; or
      4. it is reasonably clear from context that e wanted the action to
         take place or assented to it taking place.
Rule 2125/13 (Power=3)
Regulated Actions

      An action is regulated if: (1) the Rules limit, allow, enable, or
      permit its performance; (2) the Rules describe the circumstances
      under which the action would succeed or fail; or (3) the action
      would, as part of its effect, modify information for which some
      player is required to be a recordkeepor.

      A Regulated Action CAN only be performed as described by the
      Rules, and only using the methods explicitly specified in the
      Rules for performing the given action. The Rules SHALL NOT be
      interpreted so as to proscribe unregulated actions.

This judgement was called due to this sentence:
"A Rice Plan has an active player's Signature as long as that player is
consenting to it."

Specifically: whether that sentence allows players to consent to rice plans.

So firstly, let's go over "consent". Consent, in natural language and in
Agora, is undefined when applied to things that are not actions. In natural
language, we don't consent to contracts: we consent to things the contract
will do. We don't consent to sex, we consent to the act of sex, and
specifically, we consent to particular actions of it. So, just to be
perfectly and absolutely clear and unambiguous: consenting to non-actions
is undefined and unreasonable to assume, firstly based on Agoran lack of
definition, and secondly based on natural language lack of definition.

So: this sentence is written poorly based on current Agoran culture and
tradition. There are three primary ways to read this sentence atomically.
The first is that the "it" in the sentence refers to "that rice plan". A
rice plan is not an action. Thus, it makes no sense to be consenting to a
non-action, and this is unreasonable to assume.

The second is that the "it" in the sentence refers to "that rice plan has
that player's Signature." However, this is a state of something: how can we
consent to a state of affairs? However, similar to previous examples, we
DON'T consent to states of affairs: we don't consent to whether sex
happened, but rather, whether it is going to happen, and we don't consent
to contracts, but rather what the contract will do.

Thus, we reach that third interpretation: we are consenting to that rice
plan obtaining or not obtaining your signature. You consent to the change,
the action that occurs, or you reaffirm consent that inaction is
acceptable: that no change is acceptable. (It is not strictly necessary to
reaffirm consent to Inaction, but you can say no or yes to something many
times if you'd like to get your point across. Note that inaction can be
interpreted as an action in of itself. EG this is how tardiness works: we
don't apply the blots to Time, we apply the blots to the player.)

Since players are consenting to actions, I should clarify: what does it
mean to say "I consent to this rice plan."? This is shorthand for the
action: much like "I consent to this contract" and "I consent to sex".
Again, it is unreasonable to assume consent to a non-action.

Now: with regard to "is consenting to it". This is the final nail we should
hammer down. What does it mean to be consenting to something? In R2519,
consent is treated like an action in clauses 1 through 3. However, in
clause 4, we see some of the natural language definitions appear. This
makes consent work in most reasonable situations if it could otherwise not
work. However, most of the consent presented for rice plans currently falls
under clause 1: we publicly agree to rice plans. If there's any concern
about whether we have consented to rice plans, then we can also apply
clause 4: it has been reasonably clear that we assented to said rice plans.

The full process goes like this for consenting to rice plans, and the
signatures being applied:
1. A player consents to a rice plan.
2. The signature is added to the rice plan, as the player consented to this
step in 1.
3. That player does nothing with regard to that rice plan.
4. The player is already consenting to that rice plan having that signature
(under clause 4 of R2519), so the signature is not removed. If for some
reason, the signature was removed, then the signature would be added back
to the rice plan, as the player has already indicated reasonably and
clearly that e would like that to occur (again, clause 4 of R2519).
5. That player consents to that rice plan (again).
6. If for some reason, the signature was removed, then the signature is
added to the rice plan, otherwise, the rice plan already has the signature
and no change to consent to occurs, and this is fine. The player has now
indicated reasonably and clearly that e'd like the signature to be added, a
second time, and this violates no rules.
(similar steps occur for withdrawal of consent.)

Finally, to close this case: in general, as a point of law, clear public
attempts to consent work. The Ricemaster Report is reasonable evidence for
who tried to consent, and since I'm [for independent logical reasons]
judging that clear attempts work, we can assume the ones in the report
worked.

THUS, I judge the CFJ to be true.

--------------------------------------------------------------------------

Judge ais523's Arguments:

The main question at issue in this CFJ is how to interpret "A Rice Plan
has an active player's Signature as long as that player is consenting
to it".

The first thing to note is that the use of "as long as" means that this
is clearly a continuously evaluated condition; there's no wiggle room
or ambiguity in that wording, and no need to apply rule 217 tests. In
other words, this part of the rule is creating a definition: in rule
text that rule 2682 is sufficiently powerful to define the meaning of,
"a Rice Plan has X's signature" is a synonym for "X is consenting to
{the Rice Plan / the Rice Plan being signed}". There is some doubt
about whether the thing being consented to is the Rice Plan itself, or
the status of the Rice Plan being signed, which will be covered later
in this CFJ judgement.

In Agora, consent is normally given in terms of consenting to an
action. However, the concept of consent is a concept that's meaningful
for states rather than actions. (A real-world example is that you can
"consent to have your data stored by [a particular company]"; the
initial storing of the data is action-like, but what makes this a state
is that (in some jurisdications) if you withdraw your consent, the
company then has to delete the data, and thus consent is continuously
required for the data to be stored.) The plain language of the rule is
that in the context of Rice Plans, consent is being evaluated as a
state rather than as an action, so it's necessary to interpret what it
means at Agora to consent to a state.

Rule 2519 gives a definition of "consent" as it applies to actions.
This definition does not directly apply to consenting to a state,
meaning that the rule 217 tests apply (the text is "silent", in the
language of rule 217). It seems clear that the relevant definition
should either be a) the natural-language definition of "consent" or b)
the rule 2519 definition adapted to apply to states rather than
actions. These possible definitions are hardly in conflict, in this
case: the rule 2519 definition is pretty close to a natural-language
definition of consent as it is:

>   A person is deemed to have consented to an action if and only if,
>   at the time the action took place:
>
>   1. e, acting as emself, has publicly stated that e agrees to the
>      action and not subsequently publicly withdrawn eir statement;
This definition can be applied equally well to actions and to states,
and is compatible with the natural-language definition in the case
where the player is telling the truth in their speech actions. Rule 217
strongly comes down on the side of "assume that a player's granting /
withdrawal of consent via public statement is truthful": we need to
have some sort of concrete evidence of consent / its withdrawal in
order to make the game playable (speaking to "the best interests of the
game" and "common sense" factors), and it is already against the rules
to lie in a public message, with game custom being that public messages
are the standard way to inform players about a fact and have the game
take notice (thus the "under penalty of No Faking" that is sometimes
seen when a fact that wasn't previously relevant to the game is needed
to, e.g., resolve a CFJ). Thus, the rule 217 tests conclude that we
should probably apply this one to states in addition to actions.

>   2. e is party to a contract whose body explicitly and
>      unambiguously indicates eir consent;
This one is a little trickier: it expresses the principle that the
binding nature of contracts takes precedence over attempts to withdraw
consent, something that isn't obvious from the natural-language
meaning. As such, there's some scope for interpretation available as to
whether this overriding happens *in general* or whether it happens
*specifically for actions rather than states*. However, the two
possibilities are not at odds with each other in the case of a contract
that can be freely amended (e.g. a contract, as opposed to a pledge,
that has only one party); such contracts effectively serve as a record
of consent that can be revoked at any time.

It is not necessary to judge this CFJ to worry about what would happen
if a player were bound, by unamendable contract e could not escape
from, to consent to a Rice Plan that e did not want to consent to; that
would only be relevant if an attempt of that form affected whether a
rice-yielding or non-rice-yielding Rice Plan were harvested, a
situation which doesn't seem to have occurred here. The normal case, in
which a contract serves as a record of consent, clearly works under
both the rules-based and natural definitions.

>   3. the action is taken as part of a promise which e created; or
This one isn't relevant in the case of Rice: the Harvest occurs
automatically/platonically, and Promise cashing happens
instantaneously, so there's no way that a promise could be performing
an action at the time of the Harvest.

>   4. it is reasonably clear from context that e wanted the action to
>      take place or assented to it taking place.
This one is basically stating that consent from the natural-language
point of view can be treated as consent from the Agoran point of view,
if there's evidence of it. The reasoning here is analogous to that from
definition 1.

As such, the vast majority of the potential ambiguity turns out to be
inconsequential; common sense and the best interests of the game
dictate that we should have some clear way to determine whether players
are consenting to a Rice Plan or not, such a method should logically be
the natural-language definition of "consent" modified to be based
purely on evidence that's observable to Agora, and that is pretty much
what the rule 2519 definition is. Even if rule 2519 didn't exist, the
resulting definition of "consent" for Agoran purposes would be pretty
much what rule 2519 says anyway, and thus when checking consent to a
Rice Plan, we effectively let those definitions guide us. (The only
potential ambiguity is as to what happens if a player attempts to
withdraw consent when a contract is forcing em to consent, but that
situation is not at issue in this CFJ.)

Does it matter whether a player consents to a Rice Plan, to the
harvesting of the Rice Plan, or to the signature of the Rice Plan? The
preceding discussion implies that it doesn't, because both the natural-
language definition and rule 2519 item 4 imply that what is important
is making it clear from context that the consent exists, rather than
using the exact correct sequence of words; there's only one context in
which Rice Plan consent is relevant, and thus making it clear that
you're trying to consent to that Rice Plan being selected for the
Harvest is all that's required for the rules to consider the consent to
exist.

Based on this reasoning, snail's Rice Plan was the plan harvested on 22
May 2023 (the only occasion on which a Rice Plan could be harvested
prior to the calling of this CFJ). It had at least three players
consent to it (the below-quoted messages each apply to the plan in
question):

On Thu, 2023-05-18 at 23:51 -0500, secretsnail9 via agora-business
wrote:
> I create the following contract:
> {
> Parties to this contract consent to all Rice Plans with snail in the
> Rice Up list.
> }

On Fri, 2023-05-19 at 09:23 +0200, Yachay Wayllukuq via agora-business
wrote:
> I consent to that Rice Plan having my Signature.

On Fri, 2023-05-19 at 13:19 -0400, Beokirby via agora-business wrote:
> I create and become party to the following contract:
> {
> Parties to this contract consent to be signatories to all Rice Plans
> with beokirby in the Rice Up list
> }

There was only one other Rice Plan that could have received three or
more Signatures, juan's. That plan was unambiguously signed by juan,
Janet, and me. There was one potentially ambiguous signature, by 4st;
however, that ambiguity applies equally to both juan's plan and snail's
plan (4st either signed both, or neither). As such, it is unambiguous
that snail's plan had at least as many signatures as every other Rice
Plan, at the time of the relevant Harvesting; and it was submitted
earlier than juan's, thus would beat it on a tiebreak.

Because snail's plan caused at least one player to gain Rice when
harvested, there were indeed persons who (at the time CFJ 4032 was
called) had more than 0 Rice.

I judge CFJ 4032 TRUE.

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

Reply via email to