{ "key": "%gfMst8/+59xeLZNhWIUvlrd+bvpRKUsk85sDx/OtCp0=.sha256", "value": { "previous": "%xRRiYQhkHu6mk5Ed5kkprQgUQbCbh/ZPZZT7LIRDyiA=.sha256", "author": "@3PGWnoMIY5ebs6sJjkwc7DKjxrgcrOeIUC2YegkMtIY=.ed25519", "sequence": 194, "timestamp": 1502233595932, "hash": "sha256", "content": { "type": "pull-request", "repo": "%gW+ySFcMRjdXTf0hCEvbRrS3NcK7V6Ex0ZPTcptkJTE=.sha256", "project": "%gW+ySFcMRjdXTf0hCEvbRrS3NcK7V6Ex0ZPTcptkJTE=.sha256", "branch": "master", "head_repo": "%gW+ySFcMRjdXTf0hCEvbRrS3NcK7V6Ex0ZPTcptkJTE=.sha256", "head_branch": "r301" }, "signature": "opJvOfwMInJjfthx1zqF7HUYMTFkcDGtvrNrRz9hqJCM6R5Bqn2YuQzKHYpfyU0pBoZCZb6/lBfN7xzCCfvMAQ==.sig.ed25519" } }

I support the intent of this proposal, but as I noted in %dWoLSSS... believe this proposal is illegal. however, rule 111 also states:

the other players may suggest amendments or argue against the proposal before the vote. A reasonable time must be allowed for this debate. The proponent decides the final form in which the proposal is to be voted on and, unless the Judge has been asked to do so, also decides the time to end debate and vote.

I think you could legally implement your proposal as an ammendment to rule 203 instead of a repeal, look at the form of rules 201 or 210 for example forms.

I think I understand. You are suggesting that rule 301 could be rewritten to amend rule 203 with the text "except after the passing of this amendment, whereby..."?

note that the other rules say things like "the <section of above rule> is ignored in <restricted circumstance>"

what about something like "if at least 80% of eligible voters have voted for a proposal after 24 hours since the vote began it is considered unanimous"

note any ammendment to rule 203 would trigger the second sentence, removing simple majority.

@Dominic I think we are closer to a solution and I like your idea of adding the clock. However, I want to ensure that the entire intent of my proposal is preserved as well avoid any conflict with the 100% participation requirement of rule 105.

What about "If at least 80% of current players have voted for a proposal after 24 hours since the vote began it is considered unanimous. Any player who hasn't voted after 24 hours automatically votes against the rule change."?

Also what are your thoughts on how amendments should be styled in the document?

I like the 24 hour rule to work out active voters, maybe make it whatever the majority of those active voters reach - that way to drift inactive just limits your control, but lets the game continue unaffected really

@dangerousbeans the benefit of not doing it that way is that it is currently written to also further protection against preventing players from participating by changing the definition of an eligible voter in rule 105 by requiring at least 80% of players to even have the ability to vote for a rule to be changed. I agree we need a way to keep the game moving forward in the event that players drop off, but I think that should be a different rule with a more nuanced process.

hmm, so @soggypretzels are you saying that 80% within 24 hours counts as unanimous, but otherwise, non-voting players are counted as votes against.

I suggest: " If at least 80% of current eligible voters have voted for a proposal after 24 hours, that is counted as a unanimous passage of that rule. If the number of voters falls short of 80% any who did not vote are counted as votes against.

Where other rules allocate points via voting results, only explicitly cast votes shall be used "

Since some rules have point rewards for the number of votes, I think your ammendment better clarify how that is interpreted.

Also note, I believe this means 80% of eligible voters vote, the vote stays open until either all voters have voted or 24 hours. This also means that eligible voter can veto a proposal within 24 hours. I guess this means we might as well move on after the first veto.

(although, for the purposes of byzantine generals, we should really want until >50% of voters have acknowledged the veto) but lets make rules about this later.

Just for clarity's sake, does the 24 period start the moment a pull request is created? And so, if a turn is defined in #202 as 1.) proposing a rule and voting 2.) calculating points, then would a player's turn(and the 24 hours) begin the moment they submit their pull request? In this case (if I'm reading the timestamp correctly) it looks like we are about 4 hours into our 24 hours.

(this is my first time using git-ssb. I apologize for that accidental close!)

@Dominic I expected the automatic nays to trigger point allocation, but I wasn't bothered by it. I am equally not bothered by not allocating points for them and thus am happy to include that amendment if there is no one who feels otherwise.

I agree that your new suggestion does imply that a single nay before 24 hours would kill a proposal which is not the goal of my proposed rule. Does this help with some of that:

"24 hours after voting has begun on a proposed rule, voting ends. If at least 80% of players voted for the proposed rule, that is counted as a unanimous passage of that rule. Players who did not vote are counted as having voted against the proposal. Where other rules allocate points via voting results, only explicitly cast votes shall be used"

@Zach! according to rule 111

The proponent decides the final form in which the proposal is to be voted on and ... decides the time to end debate and vote.

@soggypretzels but if 2/10 voters neglect to vote, it's both passed unanimiously and has two votes against. That doesn't make sense to me. Or are you saying that if 20% of eligible voters do not vote then it is an automatic veto?

Prehaps it would help if you described what you effect you intend this rule change to have? I'm not entirely sure if you intend for this to speed up play, or make it harder to pass a rule.

@Dominic The goal of my rule is to change the bar for passage of rules from 100% approval of eligible voters to 80% of players. If by some later circumstance more than 20% of voters are unable or ineligible to vote on a proposal, my rule will prevent it from passing while still allowing proposals with broad but non-unanimous support from passing. Do you have a more clear way to state that?

OK, a couple of suggestions:

1) The way I interpret the word "unanimous" is ALL players vote FOR/AYE.

As soon as we start talking about percentages, to allow for inactive players, which I totally agree with, then we should drop the "unanimous" from the rule.

2) IMO players who do not vote should be called ABSTENSIONS, not AGAINST/NAY.

What I mean is that the 80% figure you suggest should be 80% of all votes received in the allowed period of time. Foe example:

10 players: 6 AYES / 1 NAYS / 3 ABSTENTIONS

Of all 7 votes received, AYES got 6 (85.7%) and NAYS for 1 (14.3%).

If ABSTENSIONS are counted as NAYS then obviously the result would be very different: Of all 10 votes (including abstensions), AYES got 6 (60%) and NAYS/ABSTENSIONS got 4 (40%).

If we take the second route, and count ABSTENSIONS as NAYS, then I suggest we lower the passing percentage to a simple majority (50%+).

Given the potentially fluid concept of "players", perhaps we should first make a roster of players before each vote, and require an acknowledgement from each that they will participate.

Then base the vote percentage on that opted in set.

Which I guess in the limit comes down to percentage of votes actually cast.

It's still debatable if ABSTENSIONS should incur a point penalty. I'm in two minds about that. But I do agree that we should encourage participation and discourage sitting on the fence strategies. My main reason for supporting the concept of ABSTENSIONS not counting as NAY votes is to prevent situations where good rules aren't passed due to player inactivity.

@alanz you might still run into a situation where a player responds to the call for roster but then goes inactive when the actual vote starts, and it would add another layer of latency to the rounds.

I think the vote is the active roster. (inactive players are still in the game roster of course).

For new players who have not yet participated but want to join, we can probably add them to the roster for the following round being played.

@ktorn I agree. And I think we should allow FOR / AGAINST / ABSTAIN to count for total players, and ignore any player who does not provide input during the voting period.

i.e. allow a player to actively ABSTAIN

I agree with @ktorn that there may be a need to discourage abstention, though I hesitate to add that into this legislation so as not to expand its scope too much.

Regarding your suggestion of using a simple majority if counting abstentions as nays, my thinking here was that, at least in the early stages of the game, we would want to set up a system where rules could be changed even if a small minority do not vote in favor (either due to disagreement or absence) while still ensuring that all successful proposals have near unanimous support.

Do you believe this is too high a bar for consensus even this early in the game?

@alanz +1 for allowing explicit ABSTAINs (not the same as, but more or less aligned with the proposal I have in mind for my turn).

With regards to the proposal under discussion, I also agree that we should lower the voting threshold. And even if the proposal is not as radical as my suggestion, it is still a step in the right direction.

Having said, not touching rule #203 for 2 rounds would actually automatically bring down the threshold to what I suggest. So I'm in two minds about supporting this.

Damn, this game is already awesome.

simple majority

How about quadratic voting?

BTW, I know we don't have the 24 hour thing in place yet, and looks like it's good that we don't.

We have been discussing this for about 18 hours, which leaves around 6 hours, and I'm about to go to sleep. I wouldn't want to ABSTAIN by inactivity due to missing the voting deadline by just a few hours.

Just noting that we need to think about this when adding timing thresholds. Perhaps explicitly lock the proposal for voting and only start counting from that moment. But in that case we could also use an informal limit on pre-vote discussion.

PS: A benefit of qudratic voting is that it encourages participation (and another that it prevents tyranny of the majority).

Given that discussion is to and fro, and the players are dispersed across multiple timezones, perhaps the discussion period should be longer?

@ktorn according to my understanding of 111, voting doesn't start until the proponent "decides the final form in which the proposal is to be voted on and ... ends the debate." You shouldn't run into an issue where a proposal is made and debated for 23 hours before you sleep, only to wake up to find a 1 hour voting period started and stopped without you. As long as you check in a minimum of once a day it should be impossible for you to miss a vote.

One problem with only counting players who actively voted is that it could encourage the calling of rush votes. This could be especially dangerous when we start modifying the powers of Judges.

@alanz so far I have no issue with the proponent of a rule-change having the power to end debate as quickly as they want as long as the vote itself is given a proper durration.

@Soggypretzels, yes you beat me to it. Just re-read the rules and was going to mention #111.

Actually, as it stands that rule gives a lot of power to the player whose turn it is. Of course it's not in your interest to rush the discussion and/or voting, but you could also keep it going for quite a while. It would require the involvement of a Judge to overrule that for the round.

@kas, I'm unsure what the cost is that is incurred with quadratic voting. Would we be putting up points to vote?

In my opinion 24 hours appears to be too short of a timespan. Specially if the wording of the rule can change at the last minute while people are offline (asleep). I'm also unsure how one leave a vote in our current setup. The gameplay at this point is still a little unclear to me.

@nanomonkey,

I'm unsure what the cost is that is incurred with quadratic voting. Would we be putting up points to vote?

I'm sure it can be implemented in many ways. What I imagine is something like: At the beginning of time, each voter is given a number of vote rights (let's say a hundred) for free that they can use as they see fit. 1 vote on an issue costs 1 vote-right, 2 votes cost 4 vote-rights, 3 votes cost 9 vote-rights, and so on. So with 100 vote-rights I can vote 1 aye or nay for each of 100 issues, or I can choose to put 10 votes on one issue. Once the ballots have been counted, all spent vote-rights are divided equally between those who have voted. Those who didn't vote don't have a part in these vote-rights, but they still have the vote-rights they had before.

To me is seems quadratic voting is something that needs to come in as a rule amendment. i.e. part of the game. So is not relevant right now.

I believe some nomic games end up with all sorts of economies, I can see how that can come about if you need to accumulate and manage resources to maximise voting effectiveness.

@nanomonkey The period of time for which the wording of the rule is locked in place is equivalent to the length of the voting window. How long do you feel players should be given from the point the debate ends and the final form of the proposal is submitted (given that this can end at any time) to when they must cast their vote?

@ktorn I support a policy for an abstain move. I think there is quite a bit we use that concept for (in other rules).

@kas Quadratic Votes sounds interesting, although I do not support it at this early stage (and remember, until 203 is ammended rule changes need unanimous support). Btw, you are not officially in the game - would you like to play?

I have also wondered if 24 hours is too fast. It's possible that it becomes hard to get a vote though, although that is hypothetical at this point.

@soggypretzels oh! maybe a better way to look at this is that your proposal is to lower quorum (minimum number that makes the vote valid) to from 100% to 80%...

Oh, sorry! call everything off. this proposal is in conflict with 105:

Every player is an eligible voter. Every eligible voter must participate in every vote on rule-changes.

Although, saying "ABSTAIN" would count, especially if we define "participate" as including it.

Also, note this 105 is immutable, so it needs to be transmuted before it can be ammended.

oh @ktorn already spotted the problem with 105 in the other thread

There are ways you could work around this without transmuting this rule, for example we could define "player" in a way that allows them to passively drop out - until the number of players becomes the number who have voted.

@dominic,

Btw, you are not officially in the game - would you like to play?

No thanks, not at this time.

(I am scuttlebutting via patchfoo, so I see all sorts of messages that are not visible in patchwork, also from people I don't follow, and since your nomic thread looks like a normal discussion thread to me, I just comment whenever I feel like I have something to say.)

How long do you feel players should be given from the point the debate ends and the final form of the proposal is submitted (given that this can end at any time) to when they must cast their vote?

My problem isn't so much with a 24 hour voting window, it was with the idea that wording could change or amendments could be implied during the 24 hours while players were offline. Rereading the discussion, I see that this may have already been part of the suggestion. Is there a finalized wording that you have in mind?

@Dominic I don't think it is in violation 105 if my rule automatically votes for players who fail to vote. Would you have any problem with a rule amendment that negates all of the effects of of the exiting rule as one of the things it it does?

Alternatively, according to rule 211 I can make a rule that claims precedence over another, so rather than making 301 an amendment to 203 how would you feel about something like:

"Rule 301 takes precedence over rule 203. A rule-change is adopted if and only if the 80% of players vote in favor of it within 24 hours of the beginning of the vote"

This would not effect the triggering of simple majority, however 301 would still take precedence over it. At that point it would only take a repeal of 301 to bring simple majority in. Any disagreement with my reading?

@soggypretzels on precidence, you missed rule 110:

In a conflict between a mutable and an immutable rule, the immutable rule takes precedence and the mutable rule shall be entirely void. For the purposes of this rule a proposal to transmute an immutable rule does not "conflict" with that immutable rule.

rule 301 can override 203, but it can't override 105.

There are many words left undefined in the rules "participate" is one of them. I think we need to use common-sense interpretations for these, and ultimately, it's Rule 212, Invoking Judgement, that will settle these matters. As it stands, I wouldn't consider doing nothing to be "participating" but if there was a rule that guided my interpretation of that word, I'd judge it according to that rule. I believe we'll be able to build more interesting rules if we take them seriously.

In any case, to pass the first new rule, you'll need to find something everyone can agree on, because we require a unanimous decision.

"participate" could be defined as casting a vote within the voting period.

Given players can come and go, this is a way of finding out who the players are at the same time.

@Dominic would you be opposed to my proposed rule clarifying that for purposes of the new rule, votes cast for players automatically due to failure to vote, counts as participation?

As you point out, the bar is set very high for new rules currently which is why I chose this as my first rule proposal. Because of the required unanimity I would be interested to hear if anyone had immediate disagreements with my suggestion of an 80% quorum so those issues can be resolved first before finalizing the details of the implementation.

@soggypretzels as this is the first rule-change, I intend to support it provided it 1) seems like a reasonable idea and, 2) is a legal rule change.

Your proposal already satisifies 1. I believe 2. is a matter of edgecases, interpretation and phrasing. I'm confidant that the intent of your proposal can be legally realized.

It's not required by any rule, but I think this a good policy for rules to become more explicit and thus easier to reason about.

@dominic

There are ways you could work around this without transmuting this rule

True, as @alanz already mentioned, we could simply ignore non-voting players. In other words, define 'player' as anyone who votes on the current turn (a non-voter is not a player), which could be interpreted as paradox vs #105, or simply working around it.

@soggypretzels' last suggestion of automatically attributing a 'vote' (let's call it ABSENT) to non-voting players is basically a similar approach, but more explicit. We could find ourselves having ever increasing ABSENT votes (voting crud) as players abandon the game and never return. It could also have an undesired impact on the quorum calculation.

The cool thing about keeping it implicit (just ignoring non-voters) is that we would silently drop players that abandon the game. No crud.

I think that ABSTAIN votes (explicit votes) should still be allowed and count towards the quorum.

The question then is, when a player misses a vote and becomes a non-player, what is the procedure for re-joining as a player. Would they need to re-apply, or simply voting again is an implicit re-join? What about their turn? If a player misses a vote the turn before their turn, are they still expected to take their turn or does it auto-skip to the next player?

BTW, my main concern early in the game is to try and keep things both fluid and flexible. Allow people to join and leave at will without affecting the experience for others (i.e. without getting stuck on votes for too long or rejecting good rules due to inactive players). I'm happy that @soggypretzels already started down this path.

PS: what's the trick to do proper @nick links in these comments when running git-ssb-web from the terminal?

The cool thing about keeping it implicit (just ignoring non-voters) is that we would silently drop players that abandon the game. No crud.

@dominic, just to make it clear. I agree that rules should be explicit. I think a rule should explicitly state that non-voters are not players! :)

I feel like we're trying to prototype something as perfect from the start which is very very difficult

Maybe we should just accept this rule with whatever details and tweak it later?

There's nothing particularly at stake yet and we've not even had a voting round to see if any of these problems actually exist

@ktorn sorry, git-ssb-web doesn't do @nick links properly

I hope I'm wrong but seems like the initial set of rules are too restrictive by requiring both that all players vote and that the vote is unanimous. All it takes is one player not voting and it all gets stuck. I will probably support this proposal (even if it falls short of my preferred simple majority) just because it's a step in the right direction.

In the next turn (mine) you can expect that much needed workaround for #105.

BTW, not sure if you noticed that rules #105 and #111 are in conflict from the very begining.

105

[...] Every eligible voter must participate in every vote on rule-changes.

111

[...] The proponent decides the final form in which the proposal is to be voted on and, unless the Judge has been asked to do so, also decides the time to end debate and vote.

Surely according to #105, which has a lower ordinal number and therefore (according to #211) takes precedence, the "time to end vote" is only reached when the last elegible voter has cast their vote.

@dangerousbeans I agree with you, which is why I think it's ok for us to try that slightly messy workaround to #105 (i.e. redefine 'player') than a more tidy transmutation + amendment of the rule. Surely that can come later.

That rule has it's days counted ;)

PS: @cel no worries.

I agree with @dangerousbeans, We don't need to fix every problem with the first rule, and this first rule will make finding consensus on more complex topics easier for following rules. Trying to address these complex topics now, in this proposal, will only make it harder to come to unanimity leaving us in this state of confusion for even longer.

Also, while I agree that we should be able to explicitly vote "abstain", I believe that is already allowed under rule 116.

Does anyone have any objections to the following based on legality, ambiguity, merit, etc:

"Rule 301 takes precedence over rule 203. A rule-change is adopted if and only if at least 80% of active players vote in favor of it within 24 hours of the beginning of the vote. If after 24 hours of the start of a vote fewer than 80% of active players have voted in favor of the rule change, it is defeated. An active player is a player who voted on the previous rule-change within the 24 hours of the start of it's vote."

I'm headed to bed and look forward to reading your arguments in the morning!

@ktorn Regarding rule 111's "decides the time to end debate and vote." I read that as "Decides when to stop debating and start voting" not "decides when both the debate and vote ends". Thoughts on that as well?

So, I opened up chapter 4 in Metamagical Themas where Hofstadter describes and writes about Nomic, it is interesting to me that so many of the rules were later changed for playing the game by mail or computer. Rule 210 explicitly states that players cannot conspire or consult on the making of future rules, and ends there. So originally all of this conversation would not have been allowed. Simple changes like this change the gameplay so much. I can't imagine playing this by mail, it must have taken ages.

I'm 100% in support of defining player (not _re_defining it, because it's not currently defined!)

@dangerousbeans I hope that one day there is a space faring civilization who's origin story begins with this game of nomic, the wisdom of our decisions now might have tremendous implications later.

@soggypretzels yup! I'd vote for that! I like that since the first rule must be unanimous, it means every player voting on the first turn (if this is passed) will be active in the next turn.

also decides the time to end debate and vote.

it's actually ambigious. if you parse "end debate<noun> and vote<verb>" it means they call the start of the vote. if you parse it as "end debate<noun> and vote<noun>" it means they call the end of the vote too, but both of these are gramatical english! (pretty sure they put things like this in just for fun) However, since the other rules about what constitutes a successful vote are quite clear, the only new power this gives them is really to abort the round, defeating their own proposal immediately.

I also notice that your "active player" concept allows "new players" to squeeze in - they must vote in a previous round, where they are not counted as active (as so their vote doesn't seem to count?) then in the next round they will be active. But whatever, lets vote for it!

I'm working on the definition of player for the next rule, which is basically similar to @soggypretzels' active player but a bit more flexible.

So you can either just refer to 'player' in this one, meaning it may be re-defined in the next turn (if it passes), or I can add the definition in the next turn, in which case we will have both definitions for 'player' and 'active player' to mean slightly different things.

That could be interesting! :)

I blame @dominic for this! The initial PR was nearly there! :D

Actually, my definition of (active) player may be overly flexible. We'll see in the next round.

I'm happy to vote on this one as an ammendment of rule 203.

If an ammendment replaces the old rule with the new one (and assigns the new rule number), does the git commit also need to remove the old rule from the list?

I think the original rules were formulated for paper, where append-only made sense. This is no longer the case.

And I guess via git you can see the history. But how would it work if we have rule 309 that says "repeal rule 301", would 301 be removed as part of that commit, or as a separate one?

@ktorn With the latest proposal I don't believe it would count as an amendment of 203 but rather simply a new rule that takes precedence over 203. I like your idea of firming up 'player' after this as my proposal allows for players to transition in and out of being active players, but does not entirely solve the problem of play halting for single absent players.

I'll leave the floor open for another few hours to hear any final objections while I formally submit the final version of my proposal.

OK, if everyone is happy with that I'll go along too.

BTW, I started a discussion for # 302.