The below CFJ is 3819. I assign it to R. Lee.
status: https://faculty.washington.edu/kerim/nomic/cases/#3819
=============================== CFJ 3819 ===============================
Gaelan submitted a bid in the February zombie auction, or would
have done so if the auction existed.
==========================================================================
Caller: Gaelan
Judge: R. Lee
==========================================================================
History:
Called by Gaelan: 20 Feb 2020 19:35:09
Assigned to R. Lee: [now]
==========================================================================
Caller's Evidence:
My messages:
https://www.mail-archive.com/agora-business@agoranomic.org/msg35639.html: {
I bid n coins in this auction, where n is 1 higher than the current
highest bid. (AFIAK, this is 16 if CB’s bid worked and 13 if it didn’t.)
}
https://www.mail-archive.com/agora-business@agoranomic.org/msg35640.html: {
Crap, Jason’s bid came in as I was writing this message. […]
I retract my 52-coin bid, and bid n coins, where n is one higher than the
current SECOND highest bid.
}
Falsifan’s arguments (submitted by the Caller):
Note that I am assuming here that Gaelan did not bid in the auction,
since I believe e did not meet R478's standard of "...clearly
specifying the action", since the precise values of eir bids depended
on whether Cuddle Beam successfuly bid, and also on realizing that
Jason had not successfully bid, which we only figured out later. If
anyone is interested in calling a CFJ on this, e may want to link it to
the CFJ I called today ("Rance's master switch is set to Agora.”).
Caller's Arguments:
Falsifian claims that my actions are invalid due to their conditional
nature. I believe this is a dangerous reversal of precedent.
As Falsifian notes, R478/37 states that a person performs an action by
announcement "by unambiguously and clearly specifying the action and
announcing that e performs it.” “Clearly” is a relative term, so therefore
the rules are (ironically enough) “unclear” (R217/12) as to what exactly
that means, and therefore we fall back to the four factors.
I believe my actions were straightforward conditional actions, conditional
on the current gamestate (even if the current game state is not currently
known), of the sort that Agorans use all the time, usually without any
trouble. Therefore, game custom supports TRUE.
The use of actions conditional on unknown state are a valuable tool for
resolving ambiguities in the game state. Therefore, the best interests of
the game support TRUE.
The ability to make conditional actions is admittedly not obvious from the
text of the rules, so there is a “common sense" argument to be made for
FALSE. However, game custom and the best interests of the game are so
overwhelmingly in support of TRUE that I believe common sense loses out here.
“Past judgements” are surprisingly hard to find—I looked through most
judgements of the past decade containing the keyword “conditional”, and
the few that addressed the legality of conditional by-announcement actions
simply acknowledged their legality in passing. Specifically:
* The judgement of CFJ 3505 implies that by-announcement actions
conditional on the past are fine, and even goes on to rule that sometimes
future conditionals can be OK.
* The judgement of CFJ 3486 states (in passing) that {Actions may be
conditional, but the conditions must be resolvable using the gamestate at
the time the action was taken.}
* The judgement of CFJ 3479 states that "I submit the following proposal
if it has no formatting errors” doesn’t work, but implies that conditional
actions generally do.
* The judgement of CFJ 3270 states {
However, it has generally
been accepted that a player can, in reasonable circumstances, indicate that
they wish to perform a sequence of actions conditional on the entire sequence
succeeding; *this is especially convenient if the gamestate is in question.*
} (emphasis mine)
* The judgement of CFJ 2971 is an exception, stating that { "I register if
some player is currently Rulekeepor", where the occupancy of the office in
question is under CfJ, should not work since it is not clear. }. There are
several things worth noting about this:
* It does state that conditional registration generally works.
* The actual question of the CFJ was about a different conditional,
which was judged to work; the “Rulekeepor” example was simply part of the
arguments. Therefore, it may not have faced much scrutiny.
* Registration is not by announcement (then or now), but it uses similar
phrasing that at least appears to be more permissive than normal by-
announcement actions.
* This case is from 2011, which predates my Agoran involvement;
therefore, I have no idea what the customs around conditional actions were
at the time.
Therefore, I believe that past judgements broadly support TRUE as well.
==========================================================================
