status: https://faculty.washington.edu/kerim/nomic/cases/#3857
(This document is informational only and contains no game actions).
=============================== CFJ 3857 ===============================
If the person who sent the above message is a player, e cast a
vote on Proposal 8442 in that message.
==========================================================================
Caller: Jason
Judge: R. Lee
Judgement: DISMISS
==========================================================================
History:
Called by Jason: 23 Jun 2020 16:22:34
Assigned to grok: 25 Jun 2020 21:58:41
grok recuses emself: 29 Jun 2020 16:20:33
Assigned to R. Lee: 29 Jun 2020 16:24:11
Judged DISMISS by R. Lee: 30 Jun 2020 10:57:57
==========================================================================
Caller's Arguments:
The "above message" [0]:
On 6/22/20 9:44 PM, Unspecified Behavior via agora-business wrote:
> I do not register, as I am already a player.
>
> I withdraw any vote I may or may not have cast on Proposal 8442. I
> vote FOR on Proposal 8442.
[0]:
https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-business/2020-June/043565.html
Rule 478/48 [Excerpt]:
> A public message is a message sent via a public forum, or sent to
> all players and containing a clear designation of intent to be
> public. A rule can also designate that a part of one public
> message is considered a public message in its own right. To
> "publish" or "announce" something is to send a public message
> whose body contains that thing. To do something "publicly" is
> to do that thing within a public message.
>
> Where the rules define an action that a person CAN perform "by
> announcement", that person performs that action by unambiguously
> and clearly specifying the action and announcing that e performs
> it. Any action performed by sending a message is performed at the
> time date-stamped on that message. Actions in messages (including
> sub-messages) are performed in the order they appear in the
> message, unless otherwise specified.
Caller's Arguments:
Does the actor of the action need to be clearly specified for it to be
performed by announcement? I don't see any text in the rules that
necessarily implies so.
--------------------------------------------------------------------------
Gratuitous Arguments (conversation summarized by Jason):
omd
(https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-business/2020-June/043634.html):
> On Tue, Jun 23, 2020 at 9:42 AM Jason Cobb via agora-business
> <agora-business at agoranomic.org
> <https://mailman.agoranomic.org/cgi-bin/mailman/listinfo/agora-business>>
> wrote:
>> If there exists no CFJ with the statement "If the person who sent the
>> above message is a player, e cast a vote on Proposal 8442 in that
>> message.", I initiate one.
>
> Arguments:
>
> I thought I remembered a precedent that anonymous actions didn't work,
> because for a statement of "I perform an action" to be unambiguous,
> you need to identify who "I" is.
>
> ...But apparently my memory is faulty. This pair of cases is all I
> can find, and it suggests that anonymous actions do work, or at least
> did under the rules of the time:
>
> https://faculty.washington.edu/kerim/nomic/cases/?2179
> https://faculty.washington.edu/kerim/nomic/cases/?2180
Response to omd from G.
(https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-discussion/2020-June/059219.html):
> On 6/23/2020 11:46 AM, omd via agora-business wrote:
>> On Tue, Jun 23, 2020 at 9:42 AM Jason Cobb via agora-business
>> <agora-business at agoranomic.org
>> <https://mailman.agoranomic.org/cgi-bin/mailman/listinfo/agora-discussion>>
>> wrote:
>>> If there exists no CFJ with the statement "If the person who sent the
>>> above message is a player, e cast a vote on Proposal 8442 in that
>>> message.", I initiate one.
>>
>> Arguments:
>>
>> I thought I remembered a precedent that anonymous actions didn't work,
>> because for a statement of "I perform an action" to be unambiguous,
>> you need to identify who "I" is.
>>
>> ...But apparently my memory is faulty. This pair of cases is all I
>> can find, and it suggests that anonymous actions do work, or at least
>> did under the rules of the time:
>>
>> https://faculty.washington.edu/kerim/nomic/cases/?2179
>> https://faculty.washington.edu/kerim/nomic/cases/?2180
>
> I vaguely remember arguing the same thing during the Annabel crisis and
> didn't make headway - the assumption then was it worked (albeit that was
> only retroactive uncertainty).
>
> Gratuitous based on current rules:
>
> In this R78 text:
> unambiguously
> and clearly specifying the action and announcing that e performs
> it.
>
> the "e" should not imply the ability to simply say "I" when the identity
> of the message-sender is ambiguous, but rather should include the notion
> that the pronoun must have a clear referent (that is usually implied by
> the email address if the message isn't signed), otherwise e's not saying
> that e's the one performing it. This is a "for the good of the game"
> argument where the rules are silent (it would definitely be better if the
> self-identification requirement was clearer in the rule).
Response to G. from Aris
(https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-discussion/2020-June/059223.html):
> On Tue, Jun 23, 2020 at 1:28 PM Kerim Aydin via agora-discussion
> <agora-discussion at agoranomic.org
> <https://mailman.agoranomic.org/cgi-bin/mailman/listinfo/agora-discussion>>
> wrote:
>>
>>
>> On 6/23/2020 11:46 AM, omd via agora-business wrote:
>>> On Tue, Jun 23, 2020 at 9:42 AM Jason Cobb via agora-business
>>> <agora-business at agoranomic.org
>>> <https://mailman.agoranomic.org/cgi-bin/mailman/listinfo/agora-discussion>>
>>> wrote:
>>>> If there exists no CFJ with the statement "If the person who sent the
>>>> above message is a player, e cast a vote on Proposal 8442 in that
>>>> message.", I initiate one.
>>>
>>> Arguments:
>>>
>>> I thought I remembered a precedent that anonymous actions didn't work,
>>> because for a statement of "I perform an action" to be unambiguous,
>>> you need to identify who "I" is.
>>>
>>> ...But apparently my memory is faulty. This pair of cases is all I
>>> can find, and it suggests that anonymous actions do work, or at least
>>> did under the rules of the time:
>>>
>>> https://faculty.washington.edu/kerim/nomic/cases/?2179
>>> https://faculty.washington.edu/kerim/nomic/cases/?2180
>>
>> I vaguely remember arguing the same thing during the Annabel crisis and
>> didn't make headway - the assumption then was it worked (albeit that was
>> only retroactive uncertainty).
>>
>> Gratuitous based on current rules:
>>
>> In this R78 text:
>> unambiguously
>> and clearly specifying the action and announcing that e performs
>> it.
>>
>> the "e" should not imply the ability to simply say "I" when the identity
>> of the message-sender is ambiguous, but rather should include the notion
>> that the pronoun must have a clear referent (that is usually implied by
>> the email address if the message isn't signed), otherwise e's not saying
>> that e's the one performing it. This is a "for the good of the game"
>> argument where the rules are silent (it would definitely be better if the
>> self-identification requirement was clearer in the rule).
>
>
> A ton of people are taking this reading, and I just don't get it.
> You're reading an awful lot into that e. I think the obvious and most
> sensible reading of that provision is that a person doesn't just have
> to specify an action (people do this all the time even when they don't
> want to perform an action). E also has to specify that e wants to
> perform the action, which stops people from accidentally triggering
> the provision by mistake.
>
> As further exposition for this, I'll note that if I meant the
> interpretation I just wrote out, I'd have written the rule exactly the
> way it's written now. I'd figure that anyone who says "I do X" is
> announcing that e is doing X (practically by definition). If I'd
> wanted em to have to specify eir identity, I would have written that
> as a separate thing e must specify.
>
> Imagine, for a second, that you're a detective and you hear a man in
> another room say "I stole the painting!". You might whisper to your
> assistant "he's confessed that he's the thief!". Now imagine that your
> assistant told you "no, he hasn't, because we don't know what the
> referent of the word "he" is, and that sentence doesn't make sense
> without a referent".
>
> Are you seeing my point? The argument that you need to know someone's
> identity for the person to announce that they're doing something is
> ridiculous. It borders on the downright absurd. That's... just not how
> words work. You can't go and take perfectly clear rules text and make
> it say whatever you want because what it actually says isn't in the
> best interest of the game, and I think that's what's happening here.
>
> There's another way to argue this for by announcement actions though.
> You could argue that the actor is intrinsically part of the action, to
> the point where any unspecified actor voids the action. I disagree.
> Voting for a proposal is an action, and it's the same action whether I
> do it or G. does it or someone else does it.
>
> A final point is that this entire discussion is irrelevant because
> voting is done by notice, not by announcement. The exclusive criteria
> for valid ballots are in Rule 683.
>
> Rule 683/26 (Power=3)
> Voting on Agoran Decisions
>
> An entity submits a ballot on an Agoran decision by publishing a
> notice satisfying the following conditions:
>
> 1. The ballot is submitted during the voting period for the
> decision.
>
> 2. The entity casting the ballot (the voter) was, at the
> initiation of the decision, a player.
>
> 3. The ballot clearly identifies the matter to be decided.
>
> 4. The ballot clearly identifies a valid vote, as determined by
> the voting method.
>
> 5. The ballot clearly sets forth the voter's intent to place
> the identified vote.
>
> 6. The voter has no other valid ballots on the same decision.
>
> A valid ballot is a ballot, correctly submitted, that has not
> been withdrawn. During the voting period of an Agoran decision,
> an entity CAN by announcement withdraw (syn. retract) a ballot
> that e submitted on that decision. To "change" one's vote is to
> retract eir previous ballot (if any), then submit a new one.
>
>
> -Aris
Clarification from Aris
(https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-discussion/2020-June/059225.html):
> On Tue, Jun 23, 2020 at 1:57 PM Aris Merchant
> <thoughtsoflifeandlight17 at gmail.com
> <https://mailman.agoranomic.org/cgi-bin/mailman/listinfo/agora-discussion>>
> wrote:
>> There's another way to argue this for by announcement actions though.
>> You could argue that the actor is intrinsically part of the action, to
>> the point where any unspecified actor voids the action. I disagree.
>> Voting for a proposal is an action, and it's the same action whether I
>> do it or G. does it or someone else does it.
>
> Just to make it clear, I think this is the weak part of my arguments.
> I think this is right, but if the judge disagrees with me I'm not
> going to complain. For the "e" thing though, I just literally cannot
> understand it. People tried to explain it to me several times on
> Discord, and it flatly does not make sense in my head.
>
> -Aris
Response to Aris from nch
(https://mailman.agoranomic.org/cgi-bin/mailman/private/agora-discussion/2020-June/059224.html):
> On 6/23/20 3:57 PM, Aris Merchant via agora-discussion wrote:
>> Imagine, for a second, that you're a detective and you hear a man in
>> another room say "I stole the painting!". You might whisper to your
>> assistant "he's confessed that he's the thief!". Now imagine that your
>> assistant told you "no, he hasn't, because we don't know what the
>> referent of the word "he" is, and that sentence doesn't make sense
>> without a referent".
>>
>> Are you seeing my point? The argument that you need to know someone's
>> identity for the person to announce that they're doing something is
>> ridiculous. It borders on the downright absurd. That's... just not how
>> words work. You can't go and take perfectly clear rules text and make
>> it say whatever you want because what it actually says isn't in the
>> best interest of the game, and I think that's what's happening here.
>
> This is more like finding a note that says "I stole the painting" and
> then claiming "this is definitely an admissible confession."
>
> --
> nch
> Prime Minister, Webmastor, NAX Exchange Manager
--------------------------------------------------------------------------
Judge R. Lee's Arguments:
An anonymous email account purporting to be a player who already voted on a
proposal retracted their vote and cast a new vote FOR the said proposal. A
link to the message is above. The question in this CFJ is whether, assuming
that anonymous account was indeed a player, their vote counted. I hold that
this is impossible to decide, as their vote MAY have been valid, but their
retraction was invalid, which makes it impossible to tell whether or not
they had already cast a valid ballot on that Agoran decision.
Firstly, the method of casting a vote. I will remind you all that by the
CFJ's statement, we are assuming that this anonymous individual was a
player, and there is therefore no need to question whether e may validly
take game actions that a player may take. For a vote to be valid, it must
satisfy six conditions, which are very different from the one condition of
a by announcement action. Therefore, rule 478's definition of "by
announcement" actions has no effect on the ballot itself. Let's go through
the six conditions and see which are met by the anonymous ballot (I am
quoting rule 683 here, almost in full)
"An entity submits a ballot on an Agoran decision by publishing a
notice satisfying the following conditions:
1. The ballot is submitted during the voting period for the
decision."
This condition was satisfied (there is no argument on this point)
" 2. The entity casting the ballot (the voter) was, at the
initiation of the decision, a player."
It is impossible to tell whether the ballot was cast by a player.
However, the CFJ statement specifically instructs me to accept that it
was, so we can assume that this condition was satisfied. Note, if it
were not for the statement of the CFJ, this would be an undecidable
question and lead to a judgement of DISMISS
" 3. The ballot clearly identifies the matter to be decided."
The ballot clearly identifies proposal 8442. Condition satisfied.
"4. The ballot clearly identifies a valid vote, as determined by
the voting method."
FOR is a valid vote under the applicable AI majority system, condition
satisfied.
"5. The ballot clearly sets forth the voter's intent to place
the identified vote."
Condition satisfied. E said that e wanted to vote that way.
"6. The voter has no other valid ballots on the same decision."
Whether or not the ballot satisfied the sixth condition for a valid
ballot is impossible to decide. This is because the anonymous player
was unsuccessful in retracting eir previous vote if e had cast one. To
quote rule 683 "an entity can by announcement withdraw... a ballot".
For an action to be taken by announcement, rule 478 tells us that the
actor "performs that action by unambiguously and clearly specifying
the action and announcing that e performs it." In this case, the actor
was taking the action of withdrawing a specific player's previous
ballot. For an action to be "unambiguous" and "clear" it must, at the
bare minimum, be possible to resolve, but this retraction is a
gigantic mystery. The recordkeepor simply doesn't know who to withdraw
the vote from, and no other Agoran knows either. I don't hold that
anonymous actions can never be ambiguous and clear, I simply hold that
in the specific case of withdrawing a specific person's ballot on an
Agoran decision, part of that action is obviously and integrally the
actor and their previous vote on the agoran decision (after all, a
vote never cast can't be withdrawn). So this action of withdrawing
fails.
This doesn't quite end this case, however. If the anonymous poster had
never cast a vote on that decision before, this vote was valid. How do
we know whether e did so or not? We can't because the voter
intentionally kept eir identity entirely secret. "Insufficient
information exists to make a judgement with reasonable effort", under
rule 591 (this is DISMISS as opposed to INSUFFICIENT because the case
was ably argued by several H. Agorans, thanks especially to Aris)
For the above reasons, DISMISS is the most appropriate judgement for
this CFJ.
==========================================================================
