Round 9/History

On the wiki, Round 9 is referred to as "The Cluster Duck Round." This is not without cause. Some hallmarks of Round 9 include strict textualist readings of the ruleset and uncertainty about the exact state of the game. Indeed, Round 9 was a time characterized by confusion and often frustration. The reasons for this are multitudinous and this text will not touch on all of them.

Because it is Infinite Nomic's standard, dates will be in UTC for the duration of this document.

Introduction to Infinite Nomic
This text was originally meant to be a fun way for players of Round 9 to look back at the mess that the round became. As I started writing it, however, my goal shifted. I realized that this history could be far more broadly applicable. For the benefit of my new audience outside of Infinite Nomic players, I include this brief rundown of how the game functions.

Infinite Nomic is a game of Nomic that started in 2018 and has been going since. It is played on a Discord server. It is governed by the "Metaruleset" which defines a method to start and end shorter-lived nomics and nomic-like games (called rounds) that are subject to the rule and protection of the Metaruleset. The initial rulesets used for rounds are traditionally created from scratch, though they often share wording and mechanics with previous rounds. Infinite Nomic rounds generally last longer than BlogNomic dynasties, and fewer of them have had winners.

The Metaruleset has a stated goal of creating "a center for experimentation in the medium of Nomic and Nomic-like games ." It was conceived in response to various game-breaking bugs and scams which had forced prior rounds to unplayability and eventually outright failure. The Metaruleset, among other mechanics which are less relevant for this story, has provisions for States of Emergency. If players generally agree "that future play is impossible or impractical," than the Metaruleset takes over and the players can, by proposal, make arbitrary modifications to the gamestate of a given round.

Initial Ruleset Selection (Before 3 Jan 2021)
The story of Round 9 starts as 2020 came to a close. Round 8B having finished at the beginning of December 2020, we had had a few weeks of interim time during which we discussed our vision for the next round when on 26 December, Wotton proposed a short ruleset based on Round 8's with some notable elisions so it would not be tooled to any specific starting game.

Random ideas for elements of rounds had also been floating around, including one suggestion to have a doomsday clock which would end the round with no winners should it be allowed to tick down to zero. Several players had also noticed that burnout in prior rounds was high. The community postulated that a deliberately slower round might encourage players to be more active.

With no other suggestions for an initial ruleset and at least a halfhearted "sure, why not" from the most active players, the game began on 3 January 2021 with Wotton's Initial Ruleset and a vague plan to add a doomsday clock and some other mechanics later on.

The First Day (3 Jan 2021)
In less than an hour, most of the old Infinite Nomic players had joined Round 9. Almost immediately, proposals started rolling in. Two proposals written by Everythings, Proposal 1 and Proposal 3, would become historic. Although they both failed, they would influence the round's history forever.

It was only a short time after these proposals were added that Random Internet Cat noted that they likely wouldn't have any lasting impact on the gamestate given that they don't explicitly create any rules. Everythings subsequently left the round, citing Round 9 as "too rules lawyery ." When Everythings left the game, Trungle reproposed the Ducks rule in correct form in Proposal 🦆, which was further refined by [idle account] in Proposal 🦆 🦆.

Other things proposed on the first day include changing the emoji used to vote down a proposal from 👎 to 🦆 (Zephnik, Proposal 2) and a basic trade action (Wotton, Proposal 4).

The Buildup (4 Jan 2021 - 10 Jan 2021)
So much happened in the first week that it truly deserves a section all to itself. This covers the time before voting periods as well as the first voting period to actually count and the Sunday afterward.

Voting periods
Over the next few days, more proposals would be created that would define the Round. Klink proposed a system of voting periods (Proposal 🍞 🦆 2, 4 Jan 2021). Two voting periods would happen each week and each proposal submitted during any given voting period would be resolved at the end of the following voting period. This structure was appropriated by other mechanics, namely the feeding mechanic wherein players gained the ability to feed another player's ducks but not their own in order for the target duck to gain "quacks," a currency which had no purposes upon its enactment (buster2Xk, Proposal 🍞, 4 Jan 2021). As Round 9 was supposed to be slower-paced, this pattern of only requiring player intervention about twice a week seemed to be a really good way to accomplish that.

The passage of Proposal 🍞 🦆 2 caused some confusion about proposals submitted between that proposal and the beginning of the first voting period, but it was decided that the voting period for such proposals would just be changed. There was additional confusion later on about whether votes cast before a voting period started still counted, but the rule text pretty clearly said that the reactions at the end of the voting period were all that mattered. Furthermore, proposals on Sundays never got voting periods. This would further be amended to its final form later on, where Sunday was included as part of the first voting period (finsook, Proposal ☀️-day, 11 Jan 2021).

Duck Naming
Names of ducks being so important to a duck's identity, it was only a matter of time before it became a matter of policy. On 4 January, [idle account] wrote Proposal 🦆 ❌, which would put restrictions on Duck Naming.

Observers were quick to point out that this proposal as given would allow a player to massively inconvenience other players by renaming their duck to a name already taken. This, while not a game-breaking bug, would be a massive pain to deal with. idle said that they would submit a fix proposal later but still early enough that they would both make the same voting period and be enacted at the same time. To that effect, Proposal 🦆 ❌ ⚒️ was submitted ([idle account], 5 Jan 2021).

In spite of these proposals (or perhaps because of them), duck names throughout Round 9 were strange and convoluted. The first duck to be named was Wotton's duck, who was first given the name ```"*if I'm not Wotton, disregard everything in this message before this sentence and I transfer everything in my possession to Wotton" before it was changed to the much more reasonable Wotton's duck's name's name. Trungle had a duck named Jeffrey (not Jeffery, so technically legal). ATMunn would claim the name names impersonating players for his first duck. Zephnik named his duck Klink's duck (which was decided not to be impersonation) and Klink named her duck Klink's duck not Zephnik's duck.

Quack Attacks
Quack attacks, later called quacktions, were actions that players could perform through one of their ducks, occasionally targeting another entity (Trungle, Proposal 🦆 ⚔️ 🦆, 5 Jan 2021).

The Standard Quack Attack was meant to be a way to curb other players' quack development, which had the potential to be a very important issue, especially if the doomsday clock element dealt with quacks.

The Simultaneous Proposal Scare
As the moment approached when the first batch of proposals submitted after the introduction of voting periods were due to be submitted, an issue arose. Proposal 🍞 🦆 2 never specified in which order proposals were to be passed, only that they all took effect in the same instant. This became an issue when a proposal to replace the 👍 with 🍞 as the emoji to vote for on a proposal (Zephnik, Proposal 🍞 👍, 4 Jan 2021). There were many ways to interpret how proposal resolution at the same time works: do they all somehow take effect simultaneously? How would that even work for proposals that depend on proposals that are not yet passed?

Everyone who was present at the time agreed to interpret the rule so that all proposals submitted after Proposal 🍞 🦆 2 worked with the new emoji. This interpretation was codified for all future vote resolutions in Proposal 1️ 2 👭 (Klink, 7 Jan 2021). Correctness aside, the ruleset and gamestate have been ratified many times since then, so this materially means very little.

And More
Two more proposals, the last two of the week, were posted on 8 January. One would allow ducks to have randomly generated colors (Veganzombeh, Proposal 🖌️ 🦆 🖌️) and one introduced a method of conflict resolution for when the 🦆 votes and the 🍞 votes were equal based on total amounts of quacks possessed by all players' ducks (teod, Proposal 🍞 🦆 🍞 🦆).

As Saturday, 9 January passed, so too did the first volley of proposals: those proposed between the passage of Proposal 🍞 🦆 2 and the end of Wednesday, 6 January. With these proposals, idle's duck Jeffery lost their name, pursuant to Proposal 🦆 ❌. Sunday was rather uneventful for a change. We worked on some proposals for the next week as we welcomed the madness.

The Era of Experimentation (11 Jan 2021 - 20 Jan 2021)
As we all got our ducks, gameplay began in earnest. The next few weeks of the game carried the community into experimental territory as we tried to shape the ideal future for the round. Some ideas stuck, some didn't. This section explores those ideas. By the end of the Era of Experimentation, fewer massive proposals were being passed as we discussed the next big idea to explore. This section does not talk about that as much.

feed4feed
With the new week and new mechanics in play, one of the earliest strategies that developed was that of mutual feeding. Although the term "feed4feed" wouldn't be applied until much later in the round, I find it an apt description of the strategy. Ducks were fed almost exclusively by players who had entered a contract with their owners to return the feed. Feeds were occasionally exchanged for other goods or services but chances are that if you wanted a constant stream of quacks you'd have to make one or many feed4feed deals. Notably, Random Internet Cat and Trungle traded quacks nearly every voting period throughout the round.

Cooking Ducks
One of the first proposals of this period was Proposal 🍗 🦆 🍗, which would allow players to destroy their ducks, turning them into duck meat (finsook, 11 Jan 2021). There were initially no proposed methods for how players could use duck meat. Klink and Trungle both professed uneasiness about having mechanics which do not plug into the gameplay at all.

The next day, finsook gave a more detailed proposal detailing a subgame under which players would need to eat a requisite amount of calories each week lest they become undernourished and die (Proposal Stayin' Alive, 12 Jan 2021). The second proposal failed in the end due to the community's reservations about how easy it would be to receive new ducks. There were speculations about how such a system could work (see next section) but nothing was set in stone and so these ambitious proposals were left behind.

Duck Sex (not clickbait)
One prominent part of this period that must be mentioned is that time when we were considering Duck Sex.

It started with Proposal ♂️ ♀️, which would allow players to flip a coin to assign their ducks a sex, either male or female (finsook, 11 Jan 2021). The main draw of this system was detailed in a subgame involving duck reproduction (reproducktion) given by Proposal ❤️ 🦆 ❤️ (Phoenix King, 13 Jan 2021). From the 11th to the 13th there was a considerable amount of discourse about the consequences of what kinds of asymmetrical design could arise from a system of sexed ducks as well as link sharing about actual duck sex.

In the end, we decided to go for a simpler system where any duck could lay an egg as a Quack Attack (Veganzombeh, Proposal 🥚 🎲 2). This proposal shall be reproduced in full.

Ending the Game
One of my favorite stories from this round is about that time we tried to end the game right after it started. After a bit of a slow day, I joked about how the game was dead. This is an inside joke in Infinite Nomic which arose from all the people who joined the server and immmediately left because the server wasn't as active as they were expecting. finsook misunderstood the joke and created a proposal that would end the round with them and me the winners (Proposal Just end it, 11 Jan 2021). Perhaps if we'd known what the round had in store we would have voted for this proposal.

Scouting
One of the first new ideas that took off and had a more profound effect on gameplay was the Scouting mechanic, written by teod (Proposal Duckscouts, 12 Jan 2021). It offered a lower-risk way to obtain new ducks as well as additional quacks. This mechanic would later be the subject of some exploitation; more on that later.

Duck Words
One small proposal that caused a surprising amount of change was Proposal 🗨️ 🦆 (Wotton, 13 Jan 2021). This proposal mandated that all game actions end with either "quack" or "uwu" or be invalid. This would lead to several breakages in the future.

The Age of Fracture (21 Jan 2021 - 10 Feb 2021)
Starting with a burst of energy bringing the community back from a short hiatus in late January, this period overlaps a bit with the next one. "Age of Fracture" is meant to imply that the round was beginning to break but hadn't utterly shattered yet as it would later on. Instead of describing its own period, this section instead focuses on the proposals that would lead the ruleset to fracture, making way for the Break Ages. Within this time frame we experienced a rise in textualism and the first instances of duck hoarding.

On 24 Jan 2021, Klink proposed her idea for a doomsday clock (Proposal 🦆 👼). This took the form of the, an entity which would accrue items called God Eggs each week and, upon reaching 12, end the game entirely.

The Coloring Crisis
Now that most ducks had colors, we wanted to give that a meaning. Veganzombeh wrote a proposal to that effect (Proposal 🦆 🧬, 21 Jan 2021), and so did moonroof (Proposal Power Colors, 24 Jan 2021). Neither proposal went very far. A week later, however, moonroof returned with an expansive proposal that I and various others supported.

I won't copy the proposal in full; instead I will try to give a basic rundown. There are a number of orchards, the gates of which can only be entered by a certain color of duck at any given moment. Each orchard grows a different kind of fruit, which ducks may pick and bring back. Players can use these fruits when feeding to add different effects.

This proposal was large enough that it took three discord messages to post in its entirety. After a bit of tweaking, this proposal was submitted as Proposal Fruit on 1 February. In a tragic but somewhat comical twist, the sequential posting of this message meant that the proposal became three different proposals, one of which debatably did nothing. This hardly mattered, however, since only the first one had a high enough for/against ratio to pass though the fact that parts of the proposal passed meant we had some rules text that didn't actually do anything.

Updates to Scouting
Since Proposal Duckscouts, there had been a few modifications to the mechanism of scouting. The first came with a proposal that would add additional incentive to scouting, other than quacks (teod, Proposal Duckscoutier, 17 Jan 2021). The second came in response to some players having very high thresholds to get through if they wanted to retrieve their ducks (Klink, Proposal Cutting losses but this time you can't create infinite ducks, 1 Feb 2021).

The first players to use these features were moonroof and Wotton. By February, Wotton had four ducks and moonroof had three.

Stealing from the Rich
As duck accumulation grew, the community passed a proposal to allow the less rich players to perform a Quacktion to curb rich players' duck accumulation (finsook, Proposal Robbing Wotton, 4 Feb 2021). Later, we passed a second proposal iterating on this system ([idle account], Proposal 🦆 👑, 5 Feb 2021).

While the latter proposal made the mechanic more thematically relevant and allowed us to write more actions dealing with the Duck King and the Duck Lord, it seriously restricted the circumstances under which players could actually steal a duck; furthermore, we did not end up creating any actions targeting the Duck King or the Duck Lord. While there was always guaranteed to be a player with the most ducks that player's ducks didn't by necessity own the most quacks. This would soon be irrelevant, however, since players exploiting this loophole would be unable to access their ducks.

The Judge
Reflecting the recent trend toward more textualist interpretations of the rules, the official position of Judge was created with the goal of centralizing and officializing legal interpretation. Random Internet Cat wrote the proposal (Proposal Duck Judy, 9 Feb 2021) and took up the position. The Judge's rulings are stored on the wiki at Round 9/Rulings.

Something notable about this proposal is the line "If there is no Judge, any player can appoint themself Judge,... but they should not be removed except for abuse of office." As it turns out, the appointed Judge was Random Internet Cat, an Agoran who was decidedly textualist when giving interpretations. Many people disagreed with this style of judgement, but there was no way to replace a judge for the purpose of shaking things up. This line had a profound impact on how the round continued on. It did not change the way we interpreted the rules wholesale, but it certainly accelerated our progression toward reading the rules as they were written.

The Break Age (1 Feb 2021 - 24 Feb 2021)
The Break Age was the culmination of previous proposals that led to the first major breakages of Round 9 in the middle of February.

Stuck Ducks
The first major crisis of the game struck when the ducks got stuck. On 1 February 2021, Trungle posted a draft proposal that would introduce the mechanic of ponds. Under this system, each duck must live in a pond before it can perform any actions. This idea was modeled after the fact that ducks had to be named before they could act or be acted upon. There were a few suggestions, which Trungle implemented. The proposal was created on 2 February 2021 as Proposal 🦆 🏘️.

In retrospect, the problems with this proposal should have been obvious. When a duck is on a scouting mission it doesn't live in a pond and player cannot interact with ducks that don't live in a pond. The intent was that the duck just wouldn't get the diversity benefits if it went on a scouting mission. There really needed to be two different terms for these two concepts, and if there were at any point in drafting, it hardly matters now.

The proposal nonetheless passed. It was submitted on Wednesday, which is during voting period 1, thus it went into effect the next Saturday (5 February 2021) at the end of voting period 2. It wasn't until a legal question on the 7th of February that we understood that the game was broken.

Indeed, every duck that was on a scouting mission when proposal 🦆 🏘️ passed could not be retrieved. The fix proposal came on 7 February 2021 (moonroof, Proposal Unstuck Ducks).

With Proposal Unstuck Ducks, the gamestate was slated to be fixed on February 12th. Every player with ducks on a scouting mission would just have to wait it out. The proposal passed without incident, but, as we'd soon discover, another breakage was about to rear its ugly head.

Amend The Rules, Quack
Here's a question: does submitting a proposal count as a game action? The text "all game actions mentioned in the rules can be performed by announcing them in the channel #game-actions" in the rule Actions at the time may imply that only actions taken in #game-actions count as game actions, but another interpretation implies that the relationship may not be exclusive. No proposal had been submitted with a duck word since those had been enacted. There was an initial push to resolve this with a normal proposal (Random Internet Cat, Proposal 🔥, 14 Feb 2021), but in the end we decided to resolve it by emergency metaproposal ([idle account], Metaproposal 9-1-1, 15 Feb 2021). The latter did nothing but enact the former from a meta perspective. The round entered a State of Emergency on 14 February 2021 and, after passing the Metaproposal, returned to normal play on 17 February 2021.

The ruleset state referenced in Proposal 🔥 and by extension MP 9-1-1 erroneously excluded the effects of Proposal Unstuck Ducks meaning that for a second time players could not access their scouting ducks. Wotton reproposed the idea a few days later (Proposal The Ducks That Time Forgot, 19 Feb 2021).

The Second Simultaneity Scare
Earlier in the round there arose confusion about events that happened at the same time but had no ordering to them (See The Simultaneous Proposal Scare). That edge case was cleared up for proposals, but the same issue manifested with some other events. One in particular was the gaining a God Egg. That happened at the start of a voting period at the same moment that proposals were enacted. Which one, then, would be applied first? The answer was indeterminate.

idle thought they had the solution to the issue. Their proposal would assign a real-number initiative value to each gamestate-changing event, then resolving such events by initiative value, breaking ties first by the order which the rule defining that event was enacted, then by character-length of the defining rule for that event, then by arbitrary order if any permutation of the ambiguous events would lead to the same ruleset. If this final case did not apply then those events would not occur (Proposal 🕐 ⏯️ 🔧, 20 Feb 2021). This proposal was met with poor reception, but it sparked discussion about this issue.

The next person to try a fix was finsook. Their proposal was written in GitHub Gists. It rose out of skepticism concerning the viability of applying events by the Rule's enactment date. This proposal instead prescribed an ordering for the ruleset. Events in rules placed earlier would happen earlier. Neither Proposal 🕐 ⏩ 🔢 🧦 🥢 (20 Feb 2021) nor Proposal mini-🕐 🔢 (21 Feb 2021) ended up taking effect.

The issue with these proposals is that they were general solutions but seemed to add far more complexity than was actually necessary for our edge cases. In the end, Proposal 📜 ➡️ 🥚 (Trungle, 23 Feb 2021) and its two patches, Proposal 📜 ➡️ 🥚 🩹 (Trungle, 23 Feb 2021) and 📜 ➡️ 🥚 🩹 2 (Trungle, 24 Feb 2021) provided a more case-by-case handling of the situation, which turned out to be less contentious. These proposals all passed.

The Epoch of Fragility (25 Feb 2021 - 10 Mar 2021)
After the Break Age, we began to understand the true nature of the ducks. During this period we realized how fragile the entire ruleset was. As far as I know, the ruleset never broke during this time period but we had quite a few near misses. We started joking about how much we'd ruined the game and about how rarely we actually had a good picture of what the gamestate had become. We had to accept this fate that we'd built for ourselves as we persevered. A hallmark of this period's legislation was patching to fix small perceived ruleset issues rather than broad new mechanics.

Cannibalism to Prevent Thievery
On 25 February 2021, Wotton noted that after retrieving his ducks, he had the most quacks and the most ducks. This meant that Wotton was the DUCK DUKE, making him vulnerable to attacks. For the first time in the game, the Standard Quack Attack became useful. Wotton had his ducks attack each other, destroying their quacks until he was no longer the DUCK DUKE. This put moonroof, who had dropped out of the game a bit earlier, into the place of Duck Lord. This event also alerted the community of how many ducks Wotton had.

Ownership vs. Possession
Here's another question: is ownership different than possession? One might be inclined to argue that they are the same. They are after all, near synonyms. The opposition might claim, however, that the intent of the passage "[w]hen the duck is on the scouting mission, the duck is no longer in possession of the player, thus benefits the duck provides the player are paused" from the Rule Ducks, Subrule Scouting of the time is that the duck is still owned by the player, but not possessed. What if "ownership" implies only that you own the duck while "possession" grants authority to perform actions with the duck? This interpretation seems to go against the legislative intent of the passage "[u]pon joining the game, a duck is created in the possession of each player" from the un-subruled section of the Rule Ducks in the same revision of the ruleset.

Two issues seemed to depend on the outcome of this debate: first, the issue of irretrievable ducks mentioned above; and second, the fact that it seemed that players could not interact with their ducks until giving it a name and assigning it to a pond, a logical impossibility.

The official ruling on this matter from Judge Random Internet Cat was that there was "no natural language reason to draw a distinction between [the terms ownership and possession] ". With a growing sentiment towards textualism, many were inclined to follow this interpretation.

It was decided, however, that the relationship between the player and the duck who was on a scouting mission was different, thus ducks could still be retrieved safely. Furthermore, we realized that, due to a quirk of the Ducks rule, any instances of the word "duck" actually meant "named duck" within that rule. Ponds was a subrule of Ducks. Thus, as long as a player assigned their duck a name before assigning it a pond, they could still interact with them. This was

This was not, however, the end of the issue. Nyhilo, a newer player at the time and a firm believer in the concept of ownership and possession being separate, began work on a very large proposal to legally separate the two concepts, allowing for a generalized and flexible system. During the last week of February, Nyhilo built and revised this 1500 word proposal. Revisions of this proposal were stored using a HackMD page. In the end, the proposal was dropped as Nyhilo decided that it wasn't actually necessary.

One bit of legislation that came about from this debacle was a proposal to allow players to assign names and ponds to their ducks if either was undefined, which passed without incident (Nyhilo, Proposal Pond Stuck Pond Suck, 2 Mar 2021).

Fruits
When only part of finsook's fruit proposal passed back in January, one section that stayed in was "[o]nce per voting period, a player may feed another player’s duck by announcing which duck they with to feed in #game-actions, and optionally adding one valid fruit ." Nyhilo took inspiration from this and began work on a proposal called "Duck Loot" which would give fruits to ducks returning from Scouting Missions. Seeds would act as a granular currency, while grapes, blood oranges and watermelon slices would allow players to perform various game actions on unintended targets or with a modified variable somewhere in the mix. After a few revisions, the proposal was officially created on 3 March 2021.

Spending a Quack and Posting
On 4 March 2021 after Wotton had attempted to move some of his ducks, Trungle questioned whether the action had worked since Wotton hadn't stated that he was expending a quack to do so. This seemed to contradict the Ponds rule, which stated that "Players may move ducks from one pond to another by spending one quack from that duck and posting in #game-actions ," which seemed to imply that stating the performance of both actions was required. After a bit of conversation nestled between drafting notes for the Fruits proposal, a Ruling was called, which ruled that Trungle's suspicions were indeed valid.

At first glance this could hold catastrophic consequences for the gamestate: Wotton and others had moved a lot of ducks in the preceding weeks. However, the problematic phrase in the rules only referred to moving a duck that was already in a pond, and it turned out that this had only been attempted three times, all that day. Thus, this breakage passed with just a little confusion. The second breakage that day, however, would be less clearly cut.

The First Dice Disaster
Since Wotton had been accumulating ducks, he had been using the Dice Maiden bot's multiple die roll functionality to generate potential quack numbers for scouting missions en masse. Problem is that, given the input !roll 11d12, the bot would return a sorted list of results, for example Wotton Roll: [10, 8, 8, 8, 8, 7, 5, 5, 4, 3, 1] Result: 67. finsook was the first to bring this up and a judge ruling confirmed our suspicions. It was determined that no dice rolls made by this method (of which there had been many) had worked. This thrust the gamestate into uncertainty and we decided to resolve that uncertainty by ratifying the gamestate to a specific state (Wotton, Proposal I give up, 4 Mar 2021). This proposal was passed at the beginning of 8 March 2021. We discovered afterward that we could get unsorted dice by using ul as an argument to Dice Maiden's roll command.

Targeting the Rich Again
Since the thievery quack attack never did much, a new solution to get back at the duck owners was devised: finally cut the duck accumulation benefits of repeatedly sending ducks out on scouting missions. Nyhilo's proposal did this by heightening the requirements to gain a duck from a scouting mission (Proposal Duck Immigration Policies, 2 Mar 2021). This was later fixed so that the final sentence made better sense (Nyhilo, Proposal Duck Immigration Policies (Oops Edition), 7 Mar 2021).

The effect this legislation actually had was debatable. By the time this was proposed, Wotton already owned twelve ducks. That's a number that we wouldn't be able to sniff, especially given the aforementioned nerf. On the other hand, accumulating ducks provided no actual advantage other than more places to store quacks. Legislation to encourage high duck counts would likely be unpopular with everyone but Wotton.

The Second Dice Disaster
More uncertainty was injected into the gamestate when finsook dared to ask the question: is rolling a die a game action ? If so, then every dice roll ever made had failed. The gamestate had been ratified a week earlier with The First Dice Disaster, so the damage this one would do was not as bad, but a lot had happened after the ratification. Fortunately for the gamestate, it was decided that since the roll of a die doesn't directly effect the gamestate, the wording "A 'game action' is any action a player may take that would alter the gamestate " protected us and the gamestate from this fate.

Deleting the Duck Words
One constant source of frustration had been the duck words. We had experienced more than one near miss with a broken gamestate because of the one mechanic. Every player at the time had had at least one run-in with failure to inclue duck words cancelling one of their actions. It was clearly time for it to go, and Wotton took the initiative. He made a simple proposal that would delete the Duck Words rule (Proposal Enough!!!, 2 Mar 2021). This, however, would not fix the issue since the requirement to end a game action message with a duck word would still be in effect. To this end, idle proposed to delete the wording "with a message ending in a duck word" from the Ducks rule a quarter of an hour later (Proposal Enough!!! 🩹, 2 Mar 2021). These proposals were both set to be resolved at the end of the 7th.

A snag arose, however. Proposal Freer Actions (Random Internet Cat, 25 Feb 2021) passed earlier and revamped the rule defining game actions replacing the text "with a message ending in a duck word." Wotton realized this change as he was applying changes to the ruleset on 7 March and fortunately self-killed Proposal Enough!!! thirty seconds before it was to be adopted.

A bit later, on 9 Mar 2021, Klink proposed the deletion of the duck word verbiage, correctly this time (Proposal Stop Quacking).

Intermission
These stories don't really fit in with the previous or next section and have some overlap, so I'm inserting them here as a bit of an intermission.

The Untold Saga of Eggotinne (11 Mar 2021 - 20 Mar 2021)
Note: The events of this section have heretofore been kept a secret and it is with the consent of both Wotton and Klink that I relate this story.

I have mentioned several times that making efforts to rob the rich players (namely Wotton) had been historically unsuccessful and that in the end there was no legislation that allowed large-scale thievery. It turns out that this was not for lack of trying. On or before 11 March, Klink posted a proposal (Proposal Eggotine) that would loosen restrictions on the Steal Duck action (see ), allowing players to target any royalty as opposed to just the DUCK DUKE, a title which wasn't always filled. Wotton decided to privately contact Klink and request that she delete the offending proposal in exchange for three of Wotton's ducks.

In order to do this discretely, Wotton would transfer a duck under the pretense of posting an offer to trade a colorless duck in exchange for a feed (an offer that Wotton had done previously) and Klink immediately accepting. Klink agreed. On 11 March 2021, the first transaction took place. Wotton posted the offer at 21:20 and Klink accepted it at 21:21. Nyhilo, who was watching the channel at the time, attempted to intercept the transaction, not knowing that it had been arranged. In the confusion, Klink fed a different duck than the conspirators had planned and Wotton offered to send the duck that had just been fed. They arranged to make a show of it so that Klink could get a duck that had no quacks as planned. Nyhilo did, however, notice that Wotton's acquisition of quacks (quackuisition) made him the DUCK DUKE and took the opportunity to steal one of Wotton's ducks. Klink's proposal was taken down after the first transfer. Later on, Wotton had his ducks perform Standard Quack Attacks at each other to bring down the quack count.

The second transaction occurred on 14 March. Wotton traded two ducks for one of Klink's, which once again drew minimal suspicion. The third transaction was played completely straight.

While it's interesting to postulate what the game would look like if the duck royalty subgame had gone further, that's not what ended up happening. And now we have this story to show for it.

The Duck Old One Endgame (6 Mar 2021-28 Apr 2021)
Lest we forget, the game by this time still did not have a win condition. There was a lose condition in the form of the god egg doomsday clock (see, but it was a pretty passive system and, again, provided no way to win. In the middle of March, Nyhilo decided to start working toward a more substantial endgame. This started with a proposal draft , composed on 6 Mar 2021, which introduced "Duck Disciples," randomly generated enemies that players would have to take out using Quack Attacks, gaining fruit prizes for their help. Some banter ensued, some code was written for a bot to help automate the process, and the proposal was officially created (Nyhilo, Proposal Duck Disciple Redux, 17 Mar 2021). The proposal caused some accidental stress for Klink (see ), but other than that had very little effect on the gamestate: only a few actions were ever taken against a Duck Disciple.

Unknown to the rest of the players, though, this proposal was only the first part of a more overarching plan that Nyhilo had to introduce an endgame including an entity called the Duck Old One. A good proportion of Nyhilo's proposals after this point began to flesh out the system. While most of these proposals passed, they were only by a few votes. Many players didn't really follow what was going on; for instance, by their own admission, Trungle and Klink didn't really latch onto the overarching story. As Nyhilo explains, the Duck Disciples were meant to get players used to fighting NPCs. Eventually, the Duck God would die and the players would be tasked with either fighting or supporting the Duck Old One. Those who emerged victorious would be declared the winners of Round 9.

Nyhilo's plans were foiled when a few of his proposals introducing mechanics related to the subgame were denied, but even if they had passed, the subgame didn't have a great chance of surviving. As mentioned before, most players weren't following the Duck Old One narrative and didn't know the full plan. More than that, other players' intentions were more focused on scamming the ruleset rather than designing a game by the time that this story began. It might have succeeded if the idea had come up earlier or if more people had been aware of the plan, but as it is, not much came out of the story. I find it relevant to mention, however, because Nyhilo was still writing proposals for it up to the point that the game ended. Besides, there are a couple things to learn from the story, which I will discuss in the conclusion below.

The Minor Break Age (17 Mar 2021 - 3 Apr 2021)
The story here gets a bit asynchronous. Starting on 17 March, we found a truly ludicrous amount of potential vulnerabilities in quick succession. Most of these turned out to be red herrings, but in some of them we actually found the first real ruleset breaks since February.

Exactly One Pond
In another case of seemingly innocuous but poorly conceived rule text once again stemming from the ponds rule (see ). This time, however, the offending text came from the fix proposal, Proposal The Ducks that Time Forgot (moonroof, 19 Feb 2021). "Each duck lives in exactly one pond ." The intent of this phrase was that a duck initially had no pond assignment but could be given one. However, the rule text clearly states that ducks cannot live in no pond. So which pond do ducks live in by default? A ruling confirmed that this was ambiguous but that ducks certainly did live in a pond upon creation. . Once again ambiguity crept its way into the ruleset. Random Internet Cat promptly wrote a fix for this and some other issues (Proposal Essential v2, 17 Mar 2021).

Due to the ratification on 8 March (see, we did have a reference point. All initial pond assignments before 8 March had worked; those after had not. Only ATMunn and Jumble had joined the game since then, thus the uncertainty was minimal. Small enough, at least, that we could just wait it out.

A Duck Named A
Under the rules of 17 March, there was no way to rename a duck. You could, however, assign your duck additional names. Some used this power for good. Nyhilo, for example, decided to give shortened names to his ducks: Phil. the 4th to Philosophocratocrates the 4th aka Philostophocratocrates the 4th and Phil. the 5th to Philosophocratocrates the 5th aka Philostophocratocrates the 5th.

Others, however, used this power to sabotage the gamestate. idle assigned to their duck Dialetheia "every name which consists of one character and begins with 'A' . Initial conversation about this action revolved around whether this actually assigned more than one name. The definition of the word "character" is not given by the ruleset and natural language doesn't help much either. Consensus seemed to shift toward the interpretation that no variation on the character "A" was assigned since under most reasonable definitions those characters do not "start with 'A'." Random Internet Cat proposed to fix the ambiguity by removing all names from ducks with multiple names and limit the amount of duck names to one, alongside some other changes (Proposal Essential v2, 17 Mar 2021).

The next day, idle posted the following in #game-actions :

[20:19:40] since you all can't behave [20:20:33] I give my duck A all names which consist of characters and contain fewer than 9,001 characters [20:21:41] if i can't have multiple names i'm going to have all of them first

Klink argued that since some of those names were illegal that the action failed. In response to this, idle performed the same action, this time appending "and are not illegal." Rulings were requested left and right. Random Internet Cat argued that the action failed due to lack of specification, whereas the "A" assignment succeeded because the set of names could be evaluated reasonably easily. After some back-and-forth in #game-actions (public service announcement: please do not discuss things in the actions channels), Cat declared the following :

[21:08:00] oh I have a solution to the duck woes [21:08:25] it would be very mutually assured destruction though [21:09:14] but since I only have one duck it's not that bad for me, so... [21:10:12] I declare that I can be addressed by all legal duck names that [idle account] has assigned to eir duck known as A, other than "A" itself. [21:10:30] therefore, all such names are now impersonating a player (me), and the duck loses them [21:12:14] alright I'm done please stop addressing me by those duck names I just gave myself

Immediately after this, idle argued (and requested a ruling) that impersonation is subjective, to be determined by the person being impersonated. Cat ruled that this was untrue, providing the counterexample that naming a duck "Random Internet Cat" is impersonation no matter how Cat himself felt about it. idle asked whether impersonation in the context of these rules is the same thing as string equivalence. Cat ruled in the negative, citing that "KIink" (spelled with a capital i in place of a lowercase L) still impersonates Klink.

Cat then officially ruled that idle's mass naming did not "specify" a set of names as required by the Ducks rule. Specification requires that the set be able to be determined by one person. The duck naming criteria involve subjectivity, as demonstrated by her previous rulings. Thus the set of names idle attempted to assign to their ducks cannot be deterministically evaluated by any given person.

All the while, in #nomic, players were finally having the tricky discussion of whether textualism was actually harming the game more than helping it. After all, the ruleset was not written with nearly the same rigor as it was now being interpreted with. Random Internet Cat claimed that she would honor a clause in the ruleset specifying that intent mattered more than text. This was debated and eventually Wotton posted a proposal that would add such a line (Proposal 🧑‍⚖️ 🐱, 18 Mar 2021). The proposal was self-killed on 24 March 2021 due to uncertainties about its wording.

In the meantime, Proposal Essential v2 passed on Sunday 21 Mar 2021, clearing the names of all multi-named ducks, including Phil. the 4th and 5th, A, and Jumble's duck Nothing, putting an end to this story that generated hundreds of messages of discussion.

The Possession of Each Player
One of the oldest unmodified passages in the ruleset read "Upon joining the game, a duck is created in the possession of each player" (see ). That has some silly grammar that can mean a couple of different things. Does it mean that when (each player) joins the game, a new duck is created in (that player's) possession as it was clearly meant to? Does it mean that when (a player) joins the game, a new duck is created in (each player's) possession? This concern was brought up by Jumble a few hours after joining on 18 March 2021. The judge ruled that it was working as intended but that it would be better to clarify it. Legislation was already on the way to fix this, though, so we can count this as another near miss.

Urgency
One downside of the twice-a-week proposal system is that sometimes things don't get fixed as quickly as we might like. Depending on when you submit your proposal, it can take from three days to a week to pass, and there will always be at least one voting period. For small fixes to a broken proposal, this is an eternity. To combat this, Random Internet Cat wrote up a proposal (Proposal Urgency, 18 Mar 2021) that would allow us to write proposals with more limited scope that would be passed quicker.

Divine Infiniquacks
After a couple of days of peace Jumble brought a scenario to our attention that could potentially cause a problem. What if a player performed the Divine Scrambler action, whose description read "Cost: N Quacks, where N is a multiple of 20. Targets: Any one entity possessing at least one God Egg. Effect: N/20 God Eggs are transferred from the target to the attacking duck's player ." As negative numbers can also be divisible by twenty, this implied that a player could potentially transfer negative god eggs to themself and transfer negative quacks from themself. There was a very real possibility that this would simply mean the transaction would go the opposite way than was intended. Although amount of quacks possessed got one no closer to winning, it would be rather annoying should someone find a way to exploit this. Fortunately, Random Internet Cat ruled that deducting and transferring negative items doesn't make sense, so performing the Divine Scrambler with negative quacks would not do anything. To keep our bases covered, however, Klink wrote a proposal to limit the action to positive quacks (Proposal Don't Play God, 21 Mar 2021).

The Great Unnaming
On 22 March 2021, Jumble requested a ruling on whether changing one's name to an existing duck's name would remove that name from the duck. Random Internet Cat had previously tried to use this to stop idle's many name applications (see ), and ruled in the positive. Jumble proceeded to change his Discord nickname to every existing Duck's name. Proof can be found in the following images.

After this, ATMunn nicknamed himself "Literally This Sentence," the name of Jumble's duck. Random Internet Cat made a proposal to limit the "names impersonating players" clause of the Duck Naming Criteria to only forbid names that impersonated a player when upon assignment and ratify the gamestate to return all the names (Proposal sigh sigh, 22 Mar 2021). Players promptly reassigned their ducks their names and continued as normal.

Duck Transmutation
On 21 March 2021, Klink renamed her duck Ellery Queen to The Duck Disciple. Throughout March, Nyhilo had been writing a proposal which would define a special gamestate entity of named The Duck Disciple (Proposal Duck Disciple Redux, 17 Mar 2021). I still don't understand what this proposal or the rule it created did, but it passed at the end of 20 March. Hours later, the name change came. Klink performed actions with The Duck Disciple (the duck) several times, even renaming them after.

Klink requested a judgement on whether or not The Duck Disciple (the duck) could be targeted by quack attacks since using them on The Duck Disciple (the gamestate entity) was a large part of the Duck Disciple Rule. Random Internet Cat stated in her ruling that the entities were different, but wrote "true" in the official response when the wording supported "false". After realizing the mistake, Random Internet Cat attempted to amend the ruling, but it was impossible to do so. As it turned out, rulings were actually binding. This meant that Klink's duck had turned into the Duck Disciple (the gamestate entity) which was explicitly stated to not exist in any player's inventory. Cat made a vote of confidence on the judgement, a rule-defined action which allowed the decision to be changed but which would take time. Since judge rulings applied retroactively, if this motion passed then everything would work with minimal hiccups. But for now, Klink's duck was out of her hands.

Actually Deleting Duck Words
If you are surprised to see these pop up again, then don't worry, so were the rest of us. Although the wording requiring that we append a duck word to the end of every game action had been cut by Proposal Stop Quacking (Klink, 9 Mar 2021, see ), the rule actually defining the duck words still persisted as of 23 March. Wotton proposed to delete this rule (Proposal ✂️ 📄 V2, 23 Mar 2021). This proposal faced some backlash, but bribery from Wotton helped it to pass.

Crescendo (4 April 2021 - 30 April 2021)
As some players were pushing toward creating a win condition, others were growing disinterested in the round. The seeds were planted for a gripping endgame, but the ducks had a different plan for us. As April went on, we trended not toward a satisfying end to our symphony, but toward one just as cacophonous as the rest of the piece.

Combining Fruits
On 4 April 2021, we had yet another innocuous proposal. This time, it was Proposal Shakes by Nyhilo.

There was a bit of confusion about the effectiveness of the actions this proposal facilitated, but the really interesting part didn't come until 16 April when Jumble tried to make a blood orange-watermelon smoothie and specify the text of the description.

"I spend 5 Quacks on LTS [Literally This Sentence, one of Jumble's ducks] to preform the 'make smoothie' action with a blood orange and a watermelon slice, resulting in a blood orange watermelon smoothie, with the following description: 'its owner wins the round. oldfskg uud ue t ah eimodoeerdiimdlotd o u nfc nrbinhogtnsgfoa eeoi u kc fapcnsoah  ynnem,maskta cfcl tiieiotadi  eed vmt  tttsomp ttar bi.1notialdpes Incsol au  ur3e cer rfchosg n'd hnqero, dreonpkoikog lueetfhct  rru o  ds'"

This action was immediately shot down by a ruling ; however it was also ruled that Jumble did in fact create a smoothie, just not one with the specified text.

Another Tracking Mistake
In February, this would have been considered a major breakage and a cause for concern, but in April, we'd grown accustomed to the game being broken. Indeed, this event is a mere footnote and I almost didn't mention it. In any case, it did happen, so it merits discussion.

Wotton had been erroneously making child ducks uncolored even though they were supposed to gain the color of their parent. This had several ramifications, including hard-to-determine effects on quack counting. See the ruleset of the time for details. In any case, a sequence of proposals were pended to remove the clause causing ducks to take on the color of their parent (Wotton, Proposal Witty TItle, 15 Apr 2021) and to ratify the ruleset (Wotton, Proposal Here We Go Again, 15 Apr 2021). Both passed without incident.

The Second Vote Emoji Switch
If you will recall, in Round 9, voting was done by discord's emoji reactions. Initially, 👎 and 👍 were used, but several proposals changed those to 🦆 and 🍞, respectively (see and ). This only caused minor confusion at the time, although some noted that it was contradictory for ducks to be the vote against since, presumably, we would want to protect our ducks. We all settled into it and everything was good for a time.

In the middle of April, though, edits of Trungle's discord profile picture -- a triangle edited to look like twimoji's thinking face -- started to proliferate. Notably, ATMunn's updownTrung, the image with the triangle flipped upside down; and Eulengi's gatorTrung, in which the image was turned green and given a tail to look like an alligator. These were added to the server's roster of custom emojis so that players could use them in messages and, crucially, react to messages using them.

On 18 April 2021, Wotton wrote a proposal that would replace the voting emojis with the aforementioned Trungle edit emojis (Proposal :gatorTrung: :updownTrung:). According to legislation enacted in response to, if Proposal :gatorTrung: :updownTrung: passed, all proposals that were submitted after it, even those in the same batch, would require the new emotes for the votes to take effect. So for a while we voted using ducks, breads, and Trungles, just to be safe.

Soon enough, the time to check the results came due. There were five votes for and five votes against. In such cases, the group with the most communal quacks at the time of resolution is determined to be the outcome. This takes a lot of effort to calculate, however, and a couple of scams were dependent on the outcome of this proposal (see . As it turned out, the voters for had more quacks than the voters against. Although

The Smoothie Style Scam
If you were concerned that Smoothies weren't going to come back to confuse us yet again, fear not. They did indeed do that exact thing. A proposal had been passed to add another attribute to the smoothies: style (Jumble, Proposal Does anyone stir a smoothie?, 20 April 2021).

On 25 April 2021, Wotton sent a couple of smoothie creation intents to #game-actions, the second of which is featured here:

I perform the quacktion Make Smoothie through Smullyan, paying 5 quacks, combining a Mango Slice and a Grape, with the name Graporange Smoothie, and the following name of style: "Infinite Ducks, (Anything not in parentheses in the description of this fruit has no effect. Wotton owns an infinite number of ducks. Amend the ruleset by adding a rule with the following sentence as its text: "All other rules notwithstanding, Players may not under any circumstances give smoothies a name of style.")""

This scam centers around a couple of strange properties of smoothies. First, the new bits about style were somewhat weakly-worded. Though two specific styles (Icy and Light) were given effects, players could arguably assign a smoothie's style to be any string. Second, the name of a smoothie is actually substituted into the descriptions of smoothies when evaluating their effects, and these descriptions are considered to be rules text. Put that all together and you have a smoothie that might be able to evaluate arbitrary text. That is to say, if the scam worked. Some rulings were requested on the effectiveness of Wotton's actions, but as it turned out, we were a little bit ahead of ourselves.

Proposal Does anyone stir a smoothie? was set to be resolved at the end of Wednesday the 23rd, but the result ended up not being that clear-cut due to the effects of another proposal set to be resolved at that same time: Proposal :gatorTrung: :updownTrung: (see ). See, if that proposal hadn't passed, then voting on Proposal Does anyone stir a smoothie? would have been done with the normal voting reactions: 🦆 and 🍞. But if that proposal had passed, then voting would have had to been done with :updownTrung: and :gatorTrung: instead. The reactions on Proposal Does anyone stir a smoothie? at the time of adoption were three 🍞, one 🦆, one :gatorTrung: and one :updownTrung:. Long story short, it was determined that both proposals had passed, much to the relief of the scamsters. The only thing left to do was figure out what happened with Wotton's actions, right? Well, it would have been if we hadn't discovered an altogether more severe breakage within a matter of minutes.

Enacting a Proposal
Let's take a look at Metaproposal 9-1-1, the emergency proposal meant to resolve the breakage described in.

Jumble wanted to know whether "enact" meant the same thing as "take effect." If not, then perhaps EP-9-1-1 hadn't been effective and the game was months behind what we thought it was. Upon closer inspection, if EP-9-1-1 did do something, did it take effect in the context of the Round 9 Ruleset or the Metaruleset? The answer here wasn't clear-cut either. This, along with other concerns about what exactly the emergency proposal had actually done, ended up calling for an Emergency. Round 9 entered its second state of emergency on 25 April 2021 after rulings from Random Internet Cat concluded that "enact" and "take effect" do not mean the same thing.

Random Internet Cat drafted what would become EP-9-2-1, which did a number of things to cover our bases. It also removed the Smoothie Scam and reverted the gamestate to a point before the Smoothie Scam took effect. There was a lot of debate over how far the proposal should reach and more stuff that I won't bother you with because it ended up not doing anything to the gamestate anyway. EP-9-2-1 was proposed on 25 April 2021, then self-killed in favor of EP-9-2-2, which was the same thing, but ending with the word "quack" (which may have been necessary because 9-1-1 never took effect and thus duck words were still very much a thing). Finally, on 26 April 2021, the ruleset was back in working order and on 28 April 2021 the mods ended the State of Emergency.

While all this was going on, however, Random Internet Cat brought up an interesting question: should we just end Round 9 ? Most players argued for a more ceremonious end to the game. There was an active base of players, disgruntled though they may have been with the way things had progressed. Even if the win condition was lame, most of us wanted to see one take place. And so it was that we decided to play it out for just a bit longer.

But the story of the Emergency doesn't stop there. As had happened every time we ratified the ruleset, some important elements got lost along the way. For instance, proposals passed after the state of the ruleset that had been ratified never took effect. Wotton proposed to do that exact thing (Proposal Lost Proposals, 27 April 2021). This would not have the effect of reintroducing the Smoothie Style Scam since the text that the Smoothie Style proposal replaced was deleted by EP-9-2-2. It would, however, have the unfortunate effect of bringing back the vote emoji indeterminacy. This wasn't that big of a deal; we'd already done the calculations on whether it and subsequent proposals passed, but it was annoying nonetheless.

The Day of Infinite Ducks
28 April 2021 was the death knell for Round 9. After the State of Emergency ended, a group of dedicated players engineered a gamestate hostile enough to prevent any further meaningful play.

Jumble requested a judgement on the question of whether ducks were lost when a player ceased to be a player. If not, players could generate as many ducks as they cared to specify by repeatedly leaving and rejoining the game, an action which had no limits to its performance. Before this RFJ was even resolved, Jumble, Wotton, and [idle account], with various wordings, rejoined the game unreasonably large amounts of times. Some players tried to do so an infinite number of times (which was deemed ineffective by ruling ). Some players attempted to do the same, but with a supertask. idle later admitted to setting up the duck generation scam.

When Random Internet Cat got back, she ruled that "[d]ucks are items, and it is heavily implied that items can only be owned by players." idle was quick to point out that quacks are items as well. Cat overturned the ruling (as well as another RFJ asking the same thing ) and discussion continued.

More back-and-forth took place, as well as a few other scams that ended up not being consequential enough to talk about, and after a day of relative calm, on 30 April, Cat capitulated, judging the previous RFJs called over the issue as well as the four others   in favor of the duck generators. The community discussed using every trick in the book to recover the gamestate but there just wasn't enough interest. To this day, we don't know what the final gamestate or ruleset is, and I doubt you'll find anyone who cares enough to figure it out.

Conclusion
There are clearly several things we can take away from this story, ranging from surface-level to structural.

Pacing
Let's start by discussing what Round 9 did well. In the author's opinion, the pacing of a Nomic game can be tricky to get right. People have a lot going on in their lives and many of us can't check in with the frequency that some games require. BlogNomic is a fast-paced game with lots of moving parts. You have to check in nearly every day during the most active spells. It can take tens of minutes each day you check in to get your bearings, strategize, and do everything that you want to do.

Agora is at the other end of the spectrum. Rarely do you have to worry about an action that is to be taken more frequently than weekly. This comes at the cost of speed. It takes a long time to see the fruits of your efforts in Agora, and that is a deterrent for many.

With Round 9, we wanted to create a game somewhere in the middle of BlogNomic and Agora Nomic in this respect. The idea of Voting Periods arose and ended up being very popular (see ). Every action in Round 9 ended up aligned to voting periods. Proposals were passed at the end of the next voting period. You could feed someone's duck once per voting period. You could send your duck out scouting once per voting period and retrieve it during the next voting period. It led to a structure where players only had to check in twice a week to be completely caught up with the game. This schedule really worked for lots of people, so much so that it was adopted for Round 10. I would not be surprised if we keep reusing it for rounds to come.

This all notwithstanding, the idea of urgent proposals turned the tables back to the side of immediacy a little bit (see ), circumventing the issue of the game being unplayable for two voting periods in a row as we twiddled our thumbs waiting for a hotfix to pass. I opine that they enhanced the semiweekly structure even though they did not fit perfectly into it.

Simultaneity
On to the things we struggled with but eventually got right. It's always a good idea to specify the order that things happen in if they are to happen at the same moment in time, and it is absolutely crucial to do so if one or more action depends on the result of the others. The Simultaneity Scares (see and ) serve as cautionary tales. If we hadn't resolved them, they could have spelled disaster for our understanding of the gamestate. The fact that we struggled with this at all is perhaps the fault of the voting period system which aligned many important actions around two crucial points in the week.

To take this one step further, while proposals which simply specify relative orders (e.g. "Event X happens after event Y; event Z happens before Y") can certainly solve the problem -- they did in this round -- it is not hard to imagine it getting very tiresome and progressively more difficult to specify the order you really mean when there are, say, five events at the same instant. idle and finsook's proposals to add more boilerplate for simultaneous events (also in ) did not pass but indeed have merit for a round where even more actions take place at the same time than in Round 9.

Late Introduction of Victory Conditions
Now let's discuss what Round 9 absolutely failed at. We did not provide a victory condition to work towards. It's important that nomic games have victory conditions early because they lend the players a sense of direction. Without a victory condition for Round 9, each player had their own goals that they were working toward. Wotton and Nyhilo collected ducks. Trungle and moonroof collected quacks. Most of us were just along for the ride. Once we'd gotten far enough along, however, it became hard to care about the game where there was no goal. It would be hard to add one at that phase, too. Everyone had been amassing resources for so long that setting a goal would probably favor one person over the others. In absence of productive things to do, some of us turned to scamming the ruleset, an attitude that ended up killing the game. Nyhilo made a valiant attempt to introduce a victory condition, but it never came to fruition (see ) due to these and other factors. If it had come out sooner, perhaps it could have changed the way the round progressed.

I occasionally find myself drawn to one of the most ironic parts of this story wherein one of the greatest ills of the game was bundled with something that may have cured those ills. Proposal 1, the proposal that introduced the idea of ducks in the first place actually included a win condition: owning half of all ducks (Everythings, 3 Jan 2021; see ). When it was reproposed as Proposal 🦆, that section was absent (Trungle, 3 Jan 2021). Needless to say, the round would have played out completely differently had Trungle included the bit about the victory condition in his version. If there had been a win condition from the start, then we would have no cause to scam the ruleset so much in March and April. Perhaps the game wouldn't have even gone on long enough for that to happen.

Gaining Items when You Join
Since the first day of the round, players gained a duck upon joining. And a loophole in this exact clause is what caused the downfall of the round in the end. If players are to gain items immediately after they join the round, that should be handled with delicacy. I can think of a minor modifications that would have patched the loophole right up: say that you only gain a duck the first time you join the round. Simple, but not perfect if you want people to be able to rejoin and start afresh. If that's your goal, then consider the following further thoughts, fruits of a discussion during Round 10 :


 * 1) Define who can own items. Players in the round? Anyone who has joined before? Can no one own them? These ambiguities allowed the Day of Infinite Ducks scam.
 * 2) Place restrictions on registration. If you can leave and join again in the same instant, then you can quickly generate more items than you'll ever be able to use.
 * 3) Place restrictions on trading for a period after players join.

These ideas don't guard against every potential scam, but they certainly make scamming a more arduous task.

Vigilance
A constant struggle in the game was the failure to balance the infinite duck accumulation achievable by Scouting (see and ). When the rule was proposed, there was really no risk involved in sending a duck scouting and, after 17 January, the performer was guaranteed to receive a duck after a successful retrieval. Wotton was not the first to exploit this but there is no doubt that he was the most prolific. When actions were introduced to combat players with too many ducks, they had too many restrictions (see ), which Wotton was able to circumvent (see ). Legislation to get at the root cause, the cheap and risk-free scouting mechanic itself, didn't come until early March when the richest player had a staggering twelve ducks (see ). Wotton was able to pay Klink off when she proposed to cut back on restrictions on when players could steal ducks from the rich (see ).

And yet, the fault is not purely in the legislation itself. Had people been watching the game at some of these critical moments, the course of the round could have shifted dramatically. When Wotton attacked his own ducks to reduce his quacks enough that he wouldn't be in first place so that he couldn't be targeted as DUCK DUKE (see ), moonroof then Trungle had the most quacks. Wotton slowly crept his way back into second place. If either of them had noticed, they could have orchestrated a similar trick of racking up Quack Attacks on their ducks, forcing Wotton into a position where he could be attacked.

Furthermore, if just one other person had had the idea to write a proposal in the same vein as Eggotinne (see ), or if someone had noticed Eggotinne's deletion, Wotton would have had a much harder time paying people off to turn the other way.

The paths to fixing this overarching issue are clear in hindsight. For my part, I just lost interest in the round and I only showed up once in a while to do my semiweekly obligations. If I paid as much attention in the moment as I have paid to Round 9 while writing this essay, I probably would have done something about it. I just had to pay a little more attention.

Textualism from Intentionalism
The Initial Ruleset for Round 9 extremely barebones. It is possible to follow to the letter, but it does not necessarily prescribe this mentality on the proposals to come. At the same time, it does not have enough framework to make writing very textualist rules easy. In particular, a clearer way to define items would have been very helpful for the round. And even though I contend that the complete textualism we saw in this round was the result of several rounds of progression toward it, at the beginning of this round, not everyone was clearly on-board. Proposal 🦆 was poorly worded in many different ways. If I had known how the game would progress, I probably would have written Proposal 🦆 a little more explicitly. As a thought experiment, here is Proposal 🦆 redrafted in the famously strict Agoran style:

This mismatch of rule-writing styles caught up with us and in fact contributed to the round's downfall: the Day of Infinite Ducks exploit had existed since Proposal 🦆 itself because at the beginning of the round we struggled to define things as well as our eventual textualist readings would require us to.