Sunday, January 05, 2025

Something that’s bothering me

This dynasty has been full of incorrectly performed actions. We collectively (and I in particular) have been trying to fix the tracker as we notice them (the Glossary permits you to “alter the representation to match what [you] believe to be the correct application of an incorrectly-applied alteration. This may include completing incomplete actions”), but given how many mistakes we’ve already found and caught, it seems likely that there are still some uncaught mistakes in the tracker that cause it to not properly reflect the gamestate.

And that has got me worrying a bit. At BlogNomic, the normal way you perform actions is to directly edit the tracker to reflect the result of the action. However, what if the tracker was wrong to start with? If you’re working from the wrong starting value, you produce the wrong result – and by the current reading of the appendix, that probably implies that the action never happened and/or is still in progress. (The exact wording is “If authorised by the rules as a result of a Snail’s action, changes to gamestate which is tracked in a specific place (such as a wiki page) do not take effect until the representation of that gamestate has been updated to match the authorised change.”, which is ambiguous based on the meaning of “updated” – but it probably means “the tracker has to match the gamestate after the change, at least with respect to the values that changed”. There’s potentially a more lenient reading, “the tracker has to be changed in a way that simulates the effect of that change on the tracker, even if this is different from the affect the change has on the gamestate”, although I don’t think that’s consistent with the usage of “authorised”.) This means, in effect, that a mistake by someone else can cause an apparently legal action to not take effect. (This is a particular problem if the action is a step in an atomic action; in that situation, you have to revert it due to being unable to continue, but it is unclear what it means to revert an action that was impossible in the first place.) All this is particularly likely to cause a snowballing effect – if one action fails due to the tracker incorrectly reflecting the starting state of the action, the tracker will be in an even wronger state, making further actions likely to fail, and it is quite possible that we end up in a state where all dynastic actions have been illegal for weeks.

For what it’s worth, I consider that taking advantage of this situation would essentially be a core rules scam (and thus barred by Fair Play) – if we are making a best-effort attempt to maintain the gamestate, we should probably treat that as the actual gamestate for the purpose of determining whether actions are legal, rather than letting someone point out an error early in the dynasty as a late-reveal scam. I’d be more comfortable if the rules actually reflected that intuition, though (and yet it’s a hard thing to express unambiguously in ruletext, something I know from experience at other nomics). This is particularly important for knowing how much to revert if an action is performed incorrectly – say a Snail attempts to perform an action near the end of their Waking Hour, and notices there was a mistake in it after the Waking Hour is over, are they allowed to fix it, do they have to revert, or do they have to wait for someone else to fix it during their Waking Hour? The core rules don’t really help us to figure out the details of a situation like that.

I’m not sure what we should do about this (or even whether we should do anything about it), but I thought it was worth letting other players know what I was worried about rather than worrying alone, especially if there’s a simple way to fix the situation.

Comments

JonathanDark: he/him

05-01-2025 16:20:39 UTC

I feel like this is inherently a part of the set of core rules that allow any player to edit the gamestate wiki page to reflect the current gamestate. The immediate resolution would require all gamestate changes to go through a neutral authority for validation. This would most likely the current Emperor if they are not also playing, which puts a lot of burden on them.

A third option would be to make gamestate changes their own posts, subject to review and approval, before they became official. That would really drag down gameplay with a lot of “paperwork” for the benefit of ensuring agreed-upon accuracy, and probably wouldn’t really guarantee that as people would naturally start just blindly signing off on changes for expediency.

I’m not sure what other options there are.

ais523:

05-01-2025 16:26:14 UTC

I agree that it’s probably unrealistic to add mechanisms to force all the changes to be correct (and might not be beneficial even if we could). I think, however, it might be beneficial if we can come up with some way to formalize the notion of “yes, we might have made mistakes, but if nobody publicly catches them we’re just going to plow ahead from what we think the gamestate is” – but you can’t say that directly in case players start making intentional mistakes to their own benefit.

ais523:

05-01-2025 17:47:56 UTC

Oh dear – I just realised (because it actually happened) that dynastic rules can conditionally make unidling illegal based on the dynastic gamestate, and that in turn raises the possibility that (if we’re incorrect as to what the gamestate is) the uncertainty could spread to the core gamestate and possibly even between dynasties (e.g. if we’re incorrect about who is and isn’t idle, we may think a DoV was legally enacted when it wasn’t, meaning that the “DoVs uphold everything” clause never triggers).

Josh: he/they

05-01-2025 19:02:53 UTC

Yes, this ruleset is very messy and it’s been bothering me too.

I suspect that the fix is to make the comment on the race post informational rather than a core part of how the action is performed. Being able to say “oh, actually, it was in this order” is fine if the narrative post (and its associated chronology) are superceded by the wiki page, but that currently isn’t the case. Maybe we need Races to uphold on completion as well, just to belt-and-braces it.

Habanero:

05-01-2025 19:22:01 UTC

I find it easy to believe that mistakes have been made in the majority of past dynasties which by knock-on effects rendered their gamestate completely invalid; even in the simpler dynasties it’s easy to be careless, slip up, and have no one notice. The uphold on DoV makes things sort of fine, but honestly I think it’s a fool’s errand to maintain some magical platonic ideal gamestate. The core rules and appendix make about a million contradictory (and in my opinion mostly unnecessary) statements on what the gamestate means, how one modifies it, and whether what you see on the wiki page is the actual gamestate. If we want to play something even remotely resembling a game, I think we’re sort of forced to treat things on a best-effort basis and consider unnoticed errors as not being errors at all.

FWIW if someone were to consciously refuse to point out an action they perceive as illegal to get some advantage later and the group agreed it was ok, I’d probably idle out; to me that is a breach of the social contract we agree to when we play this game that we will all enforce the rules to the best of our ability. I very strongly believe that it would be no different from cheating in a board game.

Josh: he/they

05-01-2025 19:33:31 UTC

I agree with some-to-most of what you say, Habanero! The bit that I don’t agree with is the ‘fool’s errand’ bit - some tasks attain meaning through their execution, and the bit of nomic that’s an exercise in form demands that we at least attempt to make the rules iron-clad and the gamestate coherent, even if the act of play makes things messy.

To put it another way: there’s a lot of ‘right’ ways to play nomic, and the version of it that attempts to make rulesets perfect is as noble as the one that attempts to exploit scams, or the one that wants to win conventioinally, or the one that wants to treat it as game design improv. ais and I aren’t always on the same side of the tracks on the question of what the best way to play nomic is, but I respect his side of it.

For what it’s worth I don’t think the spectre of someone truly holding the game to ransom is likely, but the disordered stack of comments in the Race threads is enough to make even me feel a little queasy, and as a matter of good practice we should probably fix that rather than shrugging it off.

Habanero:

05-01-2025 19:53:03 UTC

I see what you mean. I may not personally see value in it but I do agree one could find entertainment and meaning in trying to construct as robust a ruleset as possible. I think I owe a bit more respect for the approaches of others towards the game

ais523:

05-01-2025 21:10:02 UTC

“For what it’s worth I don’t think the spectre of someone truly holding the game to ransom is likely” – it’s been attempted at least twice in the past. (On one of those occasions, it was discovered that the dynasty had accidentally held itself to ransom ever since an early proposal – once it was enacted, proposals, CFJs and DoVs all became illegal to make and nobody noticed until someone attempted to abuse it. We probably managed to recover from that state, but it took a core rules scam to do it, which incidentally might still work in the current ruleset.) I think it’s much less likely nowadays than it was in the Age of Scams, though (especially as game culture is clearly on the side of “this is unacceptable behaviour”).

I think one thing that would help would be to make it possible to create CFJs while idle – that would have come in useful twice already this dynasty, and also seems useful in recovering from “we don’t know who is and isn’t idle” problems.

As for fixing the mess with Races, automatically changing the gamestate to match the tracking page immediately after each Award Ceremony seems like it might be helpful, with a time delay to catch (accidental or malicious) errors in the Award Ceremony itself. I’ll try to see if I can design a proposal to do that – it might even make a good Building Block if we can find good wording.

You must be registered and logged in to post comments.