=============================== CFJ 3770 ===============================
G. is currently permitted and enabled to award at least one
instance of the Patent Title Champion to Jason Cobb.
Barred: Jason Cobb
Called by G.: 13 Aug 2019 12:16:34
Assigned to Falsifian: 18 Aug 2019 22:15:15
Judged TRUE by Falsifian: 24 Aug 2019 20:32:06
Sequence of events (all within a couple days, apologies for not linking to
- Jason Cobb won the game (Spaaace) 1000 times in a single message.
- Then G. awarded em Champion, once (specifying "once").
- Then this CFJ was called.
The question is whether that "once" Champion award satisfied all 1000 of
the "authorizations" to award a patent title implied by R2449.
Argument for FOR: Each win carries its own authorization that is matched
to an award.
Argument for AGAINST: When a win occurs, the Herald enters an
"authorized-to-award" state ("is then authorized" in the Rule). A later
win (before the Champion award) doesn't alter this state, and the Herald
"exits the state" by a single award even if multiple simultaneous wins
There's a longer discussion captured here:
As the Herald, I assumed back in October (with D. Margaux's Apathy wins)
that FOR applies, but on reflection, the AGAINST argument is a bit more
like how we treat the Agoran gamestate (for things in general, not wins in
particular). Otherwise, no particular bias.
A discussion starting August 8, submitted by Judge Falsifian:
> I award Falsifian the Patent Title Champion.
> I award Jason Cobb the Patent Title Champion (once).
> (I'm sure these are fine, but I'll wait just a little longer in case
> someone wants to question the other 98).
> Arguably the text of rule 649 only allows you to award one copy of the
> patent title within the week after the player wins.
> Being authorised twice to award a patent title once doesn't strike me
> as having any different game effect from being authorised once to award
> a patent title once, and the time window is the same in each case.
> Oh - interesting!
> Other than how it's listed in the Scroll, the only, practical knock-on
> effect I see is whether staggered Champion awards extend the time window
> earning an Ultraviolet - which matters for Transparent, even if the
> gets the Ultraviolet from the 1st champion.
> Jason Cobb is trying to get a Transparent Ribbon right now so this might
> come up.
> If I write you a check for $50, then you are allowed to ask your bank to
> transfer $50 out of my account and into yours. You couldn't ask them to
> do it 20 times and end up getting $1000 from me. So, that check once
> authorizes you to take $50 out of my account.
> But, if I were to write you two checks for $50 and date them the same,
> each check is authorization to once take $50 out of my account. The fact
> that they are dated the same doesn't matter, each check is a separate
> authorization to perform an action once. Therefore, you are permitted to
> perform the action twice.
> I would argue that a similar logic applies here. Rule 2449 reads:
>> When the Rules state that a person or persons win the game,
>> persons win the game; specifically they win the Round that ends
>> with the indicated win. Agora itself does not end and the
>> remains unchanged. The Herald is then authorized to award those
>> persons the Patent Title of Champion once, by announcement.
> Each win was a separate event (even if they happened at the same time).
> Rule 2449 then, for each win, authorizes the Herald to once award the
> winner the Patent Title Champion.
A discussion starting August 18:
Arguments from G.:
> Addendum to the above argument: A place we treat it as AGAINST: If 3
> people call 3 separate CoEs on 3 different line-items in the same
> document (remembering that a CoE places a "doubt" on the document, not
> just the line-item), a single revised report satisfies all of the CoEs
> at once (at least we've always assumed that).
Gratuitous argument from Jason Cobb:
> Looking at R2201, that seems wrong to me unless the revision explicitly
> notes all three claims because of the requirement that the announcement
> "clearly cites the claim of error".
> I would assume that replying to the message in which the claim was
> created "clearly cites" that claim of error, but that doesn't really
> work if there's a bunch.
Response from G.:
> Oh that's right! Funnily enough, that actually supports my argument a
> bit better. The "clearly cites the claim of error" text was added
> just last month (it's not in this version:
> It was specifically added to fix the issue - which shows that in the
> default case (without such text explicitly spelling out a 1-to-1
> matching), we'd assume a single revision did the trick.
Judge Falsifian's Evidence:
> When the Rules state that a person or persons win the game, those
> persons win the game; specifically they win the Round that ends
> with the indicated win. Agora itself does not end and the ruleset
> remains unchanged. The Herald is then authorized to award those
> persons the Patent Title of Champion once, by announcement.
Judge Falsifian's Arguments:
The main issue here seems to be uncertainty about what the word "once"
means in Rule 2449.
Here are some other places where the rules use the word "once":
1. Rule 911: [When a judgement is entered into moot,] "the CFJ is
suspended, and the Arbitor is once authorized to initiate the Agoran
decision to determine public confidence in the judgement..."
2. Rule 2593: "In a timely fashion after a Space Battle is initiated, the
combatants SHOULD each once communicate to the resolver the amount of
Energy they wish to spend in the battle..."
3. Rule 103: "If at any time the office of Speaker is vacant, or when one
or more players win Agora, then the Prime Minister CAN once appoint a
Laureled player to the office of Speaker by announcement."
4. Rule 2496: "A player CAN ... earn the set of assets associated with a
reward condition exactly once ... each time e fulfills it..."
5. Rule 1367: "A specific degree CANNOT be awarded to any person more
Examples 1 and 2 give is some guidance here. If two judgements were
entered into moot in close succession, surely the Arbitor would be
authorized to initiate one decision for each. If a player owning two
spaceships initiated two Space Battles in close succession, surely e
should communicate the amount of Energy for each battle, not just one.
In fact, even under Rule 2449, after both Jason Cobb and Falsifian won
the game, we seem to have taken for granted that the Herald was
authorized to award the Patent Title of Champion at least twice: once
for Jason Cobb and once for Falsifian.
In each example, the word "once" is naturally interpreted to mean "once
per time the condition is fulfilled". The only difference here is that
the winner happens to be the same person in the multiple times the
condition was fulfilled, and I see no reason that should change our
The caller argues that our interpretation of Rule 2201 before the
"clearly cites the claim of error" text was added supports an AGAINST
interpretation, since a single revised report can satisfy a publisher's
obligations with respect to more than one claim of error. Another case
where one action is matched to multiple is that a a single announcement
of intent can be used multiple times. However, the use of the word
"once" distinguishes Rule 2449 from these examples. Just as Rule 2579
requires a player to "indicate intent to pay that fee for the sole
purpose of performing the action" (associating the payment with the
action) and Rule 2496 only allows a reward to be earned "exactly once"
per time the condition is fulfilled (associating the fulfilment of the
condition with earning the reward), the word "once" in Rule 2449
associates the winning of the game with the awarding of the Patent Title
of Champion: so this is a different situation than with claims of error.
I judge CFJ 3770 TRUE.