Index ← 3281 CFJ 3282a 3282 → text
============================  Appeal 3282a  ============================


Panelist:                               FKA441344
Decision:                               AFFIRM


Panelist:                               Roujo
Decision:                               AFFIRM


Panelist:                               Machiavelli
Decision:                               AFFIRM

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

History:

Appeal initiated:                       05 Nov 2012 14:57:55 GMT
Assigned to FKA441344 (panelist):       11 Nov 2012 19:20:50 GMT
Assigned to Roujo (panelist):           11 Nov 2012 19:20:50 GMT
Assigned to Machiavelli (panelist):     11 Nov 2012 19:20:50 GMT
FKA441344 moves to AFFIRM:              12 Nov 2012 10:01:35 GMT
Roujo moves to AFFIRM:                  13 Nov 2012 15:26:55 GMT
Machiavelli moves to AFFIRM:            15 Nov 2012 01:30:13 GMT
Final decision (AFFIRM):                15 Nov 2012 01:30:13 GMT

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

Gratuitous Arguments by Roujo:

I agree that "A player can do [action] without objection" translates
to "A golem can do [action] without objection", not simply "A golem
can do [action]" without the objection part. I think it's reasonable
to say that a Player announcing what eir Golem does can be considered
the Mechanism by which the Golem announces it's actions, otherwise the
Golem would be unable to do so which would mean they'd run afoul of
R101, and "no interpretation of Agoran law [...] may substantially
limit or remove a person's rights as defined by this Rule".

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

Panelist Machiavelli's Arguments:

What Rule 2361 says is, "The owner of a Slave Golem CAN cause it to
take actions that are not otherwise IMPOSSIBLE by announcement." I see
three reasonable interpretations of this statement:

First: "If it is, in fact, possible for a Slave Golem to take an
action, then its owner CAN cause it to take that action."

This interpretation is circular, so I won't consider it further.

Second: "If it is, in fact, possible for a Slave Golem to take an
action by a mechanism other than this one, then its owner CAN cause it
to take that action."

This interpretation appears to be literally correct, but since this
mechanism is the only mechanism by which Slave Golems can, in fact,
act, this interpretation implies that Slave Golems cannot act at all.

Third: "If it is permitted for a Slave Golem to take an action by a
mechanism other than this one, then its owner CAN cause it to take
that action", where "it is permitted" means that if the Slave Golem
attempted to take the action, the attempt would be successful. (In
Antony's words, "'something that's legally possible, regardless of the
situation of the guy on the other end of the line' rather than
'something that's practically possible for the person to do'".)

This interpretation gives the rule its intended behavior. This
interpretation is also literally correct, since, using the definition
in Rule 2152, "not IMPOSSIBLE" means "it is not true that attempts to
perform the described action are unsuccessful". This interpretation is
also highly ambiguous, though. Slave Golems cannot, in fact, attempt
to take actions, so what would happen "if a Slave Golem attempted to
take an action" is greatly up to the imagination. If a Slave Golem
attempted to take any of the following actions, would the attempts be
successful?

* Sending a specific message to a public forum
* Waiting a specific amount of time, and then sending a specific message
* Sending a message whose text is not specified exactly, but whose
meaning is (e.g. "I cause my Slave Golem to flip its activity")
* Sending a message whose text is completely unspecified ("I cause my
Slave Golem to send a public message")
* Sending a message whose text is specified, but requires some
computation ("I cause my Slave Golem to publish the decimal expansion
of 2 + 2")
* Sending a message whose text is specified, but requires a large
amount of computation
* Sending a message whose text is specified, but is uncomputable
* Sending a message whose text is specified, but only subjectively ("I
cause my Slave Golem to judge this case correctly")
* Making a decision, when the correct decision is obvious ("I cause my
Slave Golem to do whatever it wishes, with the goal of ceasing to be a
player")
* Making a decision freely ("I cause my Slave Golem to do whatever it wishes")
* Performing an action that would certainly, but not immediately, be
successful ("I cause my Slave Golem to, With Notice, destroy a
promise", where intent was not previously announced)
* Performing an action that would probably be successful ("I cause my
Slave Golem to deregister without 16 objections", where intent was not
previously announced)
* Performing an action that would probably not be successful ("I cause
my Slave Golem to, without objection, deactivate Murphy")
* Performing an action in the recent past ("I cause my Slave Golem to
have deregistered five seconds ago")
* Performing an action in the distant past
* Performing an unregulated action that requires mental activity ("I
cause my Slave Golem to fall in love with Kanaya Maryam")

I think the most reasonable interpretation is that the one and only
thing a Slave Golem "could do if it tried" is to publish a specific
message. (Although I admit that it would be neat if they were allowed
to have unregulated mental activities and make decisions freely.)
Something like "I cause my Slave Golem to deregister" is effective,
but not because if a Slave Golem attempted "to deregister", the
attempt would be successful (it would not), but rather because "I
cause my Slave Golem to deregister" is a synonym of "I cause my Slave
Golem to publish the message 'I deregister'", and attempting "to
publish the message 'I deregister'" would indeed be successful.

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