I’m finding this discussion so very confusing; I’m sure I’m missing something
obvious.
To me, the important point is that, by default, ratification changes the
gamestate so that the ratified document is as true and accurate as possible on
the day of PUBLISHING, not the day of RATIFICATION.
That means that there can be retroactive changes to the PRACTICAL effect of
certain actions taken in between publication and ratification. For example, an
action might be EFFECTIVE under the pre-ratification gamestate, but the
post-ratification gamestate might go forward AS IF that action had been
INEFFECTIVE.
We could ask, is that action “EFFECTIVE” or “INEFFECTIVE”? I guess that
question is ambiguous, and so should be DISMISSED as a CFJ.
Ultimately I think we maybe all agree on the practical consequences under most
situations—though there may be special cases, e.g., how we interpret the
restriction that CANNOT change the rule set (as omd pointed out).
To test whether we agree in practice, let’s imagine this timeline:
Day 0 -
All players have 0 blots.
The Promotor distributes Proposal 1000, which has AI=10 and would change a
Rule.
Day 2 -
Referee publishes a report stating <Fugitive has 100 blots. Blotted has 100
blots>.
Day 3 -
Referee attempts to exile Fugitive.
Referee casts a contingent vote on Proposal 1000: <FOR if Blotted has 100
blots; otherwise AGAINST.>
Fugitive and Blotted attempt to cast ballots AGAINST Proposal 1000.
Day 7 - Close of the voting period for Proposal 1000.
Day 9 - The Referee report self-ratifies without any CoE.
Day 10 - The Assessor resolves Proposal 1000.
***
Here are the hypotheticals and my answers:
1) Is Fugitive in exile on Day 10?
— Yes. Upon ratification of the Referee’s report on Day 9, the gamestate
is changed to what it would have been if Fugitive has 100 blots on Day 2. Under
that hypothetical gamestate, Referee’s attempted exile would be EFFECTIVE.
Therefore, Fugitive is in exile on Day 10.
2) At the time of assessment on Day 10, does Hapless have any voting strength
on Proposal 1000?
— No. Upon ratification of the Referee’s report on Day 9, the gamestate is
changed to what it would have been if Hapless had 100 blots on Day 2. Under
that hypothetical gamestate, Hapless would have 100 blots at all times after on
Day 2, and that means 0 voting strength when assessed on Day 10.
(Incidentally, are the rules clear about when voting strength is determined? At
the time of end of voting period or time of assessment?)
— The ratification still changes the voting strength even though that affects
the outcome of Proposal 1000. Rule 1551 says that ratification generally CANNOT
“include rule changes,” but the ratification here did not directly change any
rule or “include any rule changes.” Instead, it changed a player’s voting
strength on an Agoran decision. The Assessor’s resolution of Proposal 1000 is
what does or doesn’t change the rule, not ratification itself. That has to be
true, otherwise the Assessor wouldn’t be able to count on Referee reports as
reflecting accurate blot totals, and there could be divergent blot counts for
assessment purposes!
3) Should the Assessor count Referee’s vote as FOR or AGAINST?
— FOR. Upon ratification of the Referee’s report on Day 9, the gamestate is
changed to what it would have been if Hapless had 100 blots on Day 2. Under
that hypothetical gamestate, Referee’s ballot would have resolved to FOR. When
the Assessor resolves Proposal 1000, e should use that hypothetical gamestate.
4) What would happen if the Assessor resolved Proposal 1000 before ratification?
—If resolved prior to ratification, Proposal 1000 would be REJECTED, with
Fugitive, Hapless, and Referee all casting votes AGAINST with 3 strength. BUT
after Day 9, the Assessor should change the result to ADOPTED if there is a CoE.
>>> On May 28, 2019, at 8:01 PM, Kerim Aydin <ke...@uw.edu> wrote:
>>>
>>> On Tue, May 28, 2019 at 1:26 PM omd <c.ome...@gmail.com> wrote:
>>> On Tue, May 28, 2019 at 7:05 AM D Margaux <dmargaux...@gmail.com> wrote:
>>> Additionally, I do not think the conditional vote “required the report
>>> ratification to go through before the voting period ended”; did it? If the
>>> empty reports self-ratify tomorrow, wouldn’t your vote still resolve to
>>> FOR? That is because, upon self-ratification, the Clork and Astronomor
>>> switches would revert to their default values at the time of the report
>>> publication, which would be before the end of the voting period.
>>>
>>> So I could have waited until report self-ratification and assessed the
>>> votes the same way on Wednesday (but didn’t have to because of my
>>> ratification without objection).
>>
>> That's another case that depends on whether ratification creates a
>> legal fiction about the past. If it does, that would work. If it
>> doesn't... well, it depends on whether you read R2127's "a conditional
>> vote is evaluated at the end of the voting period" as (a) "a
>> conditional vote is evaluated at the time of resolution based on
>> circumstances at the end of the voting period", or (b) literally "a
>> conditional vote is automatically evaluated at the end of the voting
>> period, and that value is then stuck into the gamestate, waiting to be
>> used when it's resolved". In the latter case, it also works, because
>> ratification would change that value stuck in the gamestate. In the
>> former case, ratifying after the end of the voting period but before
>> assessment wouldn't work; ratifying *after* assessment would work if
>> not for the prohibition on ratification implicitly modifying the
>> ruleset.
>
> One of the things that biases me towards omd's counterarguments (a
> bit, both are reasonable IMO) is that conditional votes work by
> defining when a voting value was clearly specified, which links to the
> requirement in R683.4. In general, when we require clear
> specification, we don't allow a person who communicates ambiguously to
> go back and clarify with retroactive effect. For example, we don't
> allow someone to say "now that ratification has clarified the game
> state 3 days ago, I hereby clarify that my announcement of 2 days ago
> meant X instead of Y".
