Thursday, October 13, 2011

More discussion about ais523’s DoV

(since the other one has been locked, and the CfJ is a separate topic)

Comments

omd:

13-10-2011 08:59:55 UTC

I’ll start.

Darknight, my comment was a somewhat snarky reference to an old game of Nomic: http://a.qoid.us/game1.txt

ais523:

13-10-2011 09:29:54 UTC

I think that, regardless of whether people think the scam worked, we should work together on a set of CFJs to make sure that we have a known non-broken gamestate (and we can worry about what to do about the current/next dynasty afterwards). I suggest that I should make a fix CFJ, and that someone else should idle out, then once the timer expires make a post, unidle back in, and edit it into a CFJ saying the same thing. Both CFJs should be uncontroversial fixes, to make sure they pass, and we should agree on the wording in advance.

The changes I’d like to make are: set the precedence to core > dynastic unless the dynastic rule specifically overrides the core rule (e.g. the opposite of what we have at the moment); state specifically that posts in the CFJ category are always CFJs regardless of any other rules, probably somewhere in the glossary (this is less important for DoVs and proposals, but is an important safeguard to avoid a locked gamestate); and ideally, require rules that define sorts of Official Post to specifically state that they’re defining a type of Official Post, rather than trying to infer it from the text of the rule. (We should probably reform the whole Story Post thing while we’re at it.) Anything else that needs fixing?

bateleur:

13-10-2011 11:52:26 UTC

ais523 - First paragraph of that post sounds like a good plan.

Regarding the rest, I’d recommend unbundling the Dynastic vs Core issue from the stuff that actually needs fixing. As you say, the fixes in the CFJ in question must be uncontroversial and that really isn’t at all.

Josh: Observer he/they

13-10-2011 15:57:38 UTC

Alternatively, we could just collectively agree to start a new game of BlogNomic with the same players, on the same website, with the same gamestate and the same, current ruleset, albeit with a few minor changes to prevent this from happening again.

Klisz:

13-10-2011 17:10:31 UTC

I’d prefer to think we haven’t set the chessboard on fire quite yet, Josh.

ais523:

13-10-2011 19:06:34 UTC

It does remind me a little of B Nomic, though, where standard gameplay was to set the chessboard on fire repeatedly and attempt to get a fire extinguisher there in time.

The real problem with this sort of scam is that you can’t discuss the implications through with players as a whole in advance, as they’ll simply either steal or block the scam.

southpointingchariot:

13-10-2011 21:36:49 UTC

As a complete newcomer to this nomic, I do not have sufficient knowledge of the relevant precedents to issue but so useful an opinion. Feel free to tell me if my comments are out of whack.

In the previous nomics I have played, the rules typically have stated that if a player attempted to make a legal move to create or demonstrate an unavoidable paradox in the ruleset, the game would end, and the player would be declared the victor. I understand Rule 1.9 has been claimed to forbid these types of actions, though the example included with the rule seemed to imply it applied to technical, not legal, “bugs.”

In general, there is an important question in nomics about what type of judgement interpretation is required - does a “reasonable” interpretation mean that one should assume, when possible, that a correct interpretation of the ruleset allows the game to continue? I’m not sure if that question is relevant here, but as someone coming into this with fresh eyes, I wonder what the answer would be.

As to ais523’s specific argument: perhaps I am confused, but the phrase “unless otherwise specified by dynastic rules” in Rule 3.2 would seem to clear up some of the inconsistency. Though Rule 2.8 might add another category for posts to be placed in (or might not), as part of the dynastic ruleset, it is explicitly allowed. I suspect there is more to this than I see, so please feel free to show me if I am missing something.

southpointingchariot:

13-10-2011 21:38:52 UTC

I would also say, from a purely selfish perspective, I would probably support efforts to start a new dynasty, as this would make entering play as a newbie far easier and more enjoyable. Stupid democracy getting in the way ;)

Darknight: he/him

13-10-2011 21:58:17 UTC

@com: Ah.

As for the issue at hand, I’m gonna let the nomic vets handle this. I myself don’t really know how to handle this for what its worth lol.

ais523:

14-10-2011 09:03:16 UTC

@southpointingchariot: the “otherwise specified” is generally interpreted as requiring the rule to specifically say something along the lines of “a Faux Pas can also be a Proposal”, in nomics. So the issue is that the dynastic rule is overwriting the proposal/DoV/CFJ designation of a post, rather than combining with it.

There definitely isn’t a paradox here, incidentally. True paradoxes tend to be quite hard to construct. The issue is just that the consequences of the rules are not what people would really want them to be; not paradoxical, just wrong.

Kevan: he/him

14-10-2011 09:15:57 UTC

Requiring dynastic rules to specifically overrule core ones seems like something we’re going to trip over without realising. And in scammable ways - if “a player with fewer than three clues cannot declare victory” lacks an explicit “this overrules Rule 1.8”, then whoever wrote or noticed the loophole can go ahead and declare victory ignoring the restriction.

Similarly, I think we need Official Posts to be inferred from text, for the same sort of reason - without it, someone will occasionally forget (or scam it) and add a rule that creates a type of blog post which affects the game but ignores the Official Post restrictions. (Although this may be covered by “Gamestate can only be altered in manners specified by the Ruleset”, if we bump up the definition of Gamestate to include “creation”.)

Remind me why “cannot simultaneously be any other type of official post” is important at all, though? (I thought it was to prevent scams where a post is made both a “Story Post” and a “Proposal”, causing it to show up just as “Story Post: Title”, with the author unmasking and enacting it as a successful proposal when it times out. But testing that, the title comes up as “Proposal: Story Post: Title”.) Does anything actually break if I make a post which is simultaneously a CfJ and a Proposal?

Josh: Observer he/they

14-10-2011 10:14:18 UTC

It was definitely put in as a result of scam. Something tells me that ais was involved in that one, too.

How about just distinguishing between core rule official post types and dynastic post types? Or distinguishing between post types and post categories? It seems that there are simpler resolutions that changing prioritisation, which is a whole separate issue anyway.

Prince Anduril:

14-10-2011 10:35:17 UTC

I like Josh’s suggestion here. We could simply restrict ‘official’ posts to Proposals, CfJs and DoVs. Then, other people can make another type of post in Dynastic rules, called, say a ‘formal’ post which allows alteration of the gamestate. Then, we are free to have posts which are Works of Art, as a ‘formal post’, allowing them to be both ‘formal’ and ‘official’ at the same time, but not having the overlapping consequences, which have hijacked this dynasty.

omd:

14-10-2011 11:42:06 UTC

Even if core and dynastic official posts are split out, you can still easily have conflicts- for example, in the current dynasty, Works of Art and Critiques might have been considered mutually exclusive types of dynastic-official posts (if not with the current wording, definitely if slightly different wording had been used); yet considering them official posts would be the right thing to do, as they should be subject to the rules about immutability.  Perhaps just move the exclusivity requirement to only apply to Votable Matters.

Kevan: he/him

14-10-2011 12:07:26 UTC

Combing the archives, Galtori added the “cannot simultaneously be any other type” sentence in May 2011; before that we just had scattered “proposals may not be CfJs, DoVs may not be anything else” clauses (which
Purplebeard cleaned up following Galtori’s addition). Skimming posts from that dynasty, it’s not clear what this addition was actually for, but perhaps it was about making Proposals which were also secretly gameplay-affecting Story Posts.

(And yes, Story Posts need fixing. They’re defined as “official posts which aren’t made by Core Rules”, but - and I can only find one exception - we never actually require such posts to be made in the Story Post blog category. I think someone just added the category to ExpressionEngine to be helpful, without modifying the ruleset to enforce its use.)

bateleur:

14-10-2011 12:29:55 UTC

Reading the discussion on this topic it seems to me that “official posts” just need properly defining and then we’re all good. In particular, a particular category of post should only be Official if something says so explicitly. A subrule of the rule defining them can then mention that Proposals, CFJs and DOVs are Official (which the base rule should not, for clarity).

Prince Anduril:

14-10-2011 14:02:04 UTC

bateleur: The problem is also to do with prioritisation. Currently (for this context) we have:

1. Dynastic rules which say a post can be more than one type of official post
2. Appendix
3. Core Rules which say they can’t be overruled by Dynastic
4. Dynastic
5. Core

This seems a bit of a mess to me. Firstly, why are we allowing Dynastic rules to specify posts being of more than one type? Appendix clearly should trump everything else. But the only Core Rule that adheres to point 3 is Calls For Judgement.

Wouldn’t it be simpler to have the rules where they are meant to be? Thus, putting Proposals, CfJ’s and DoVs (or at the very least just CfJs) in with the Appendix, meaning they can never be overruled by another rule.

That way we could just have:

1. Appendix
2. Dynastic
3. Core

And everything would be much simpler.

bateleur:

14-10-2011 14:07:13 UTC

Putting appropriate things in the Appendix may be a good idea, but it’s a separate issue. Properly defining Official posts completely removes all concerns about Dynastic stuff at a single stroke, along with issues of relative priority, because none of the Dynastic stuff would (yet) be Official.

Bucky:

14-10-2011 16:24:38 UTC

The Story Post category is older than the current definition of ‘Story Post’.  It was implemented in the 2nd dynasty of Angry Grasshopper (which had its own definition of Story Post in dynastic rules) and widely reused later despite being undefined.  Its current definition is a formalized version of what we were already doing.

The reason that the glossary protects proposals from also being other kinds of story posts comes from the end of the Fourth Dynasty of Purplebeard, where a scam involved making a proposal that was also a type of story post on which voting icons were meaningful.

southpointingchariot:

14-10-2011 18:18:21 UTC

Just want to see, I’ve been reading through the wiki, and aside from being very pleasantly surprised at the quality of the game and community, I just got the chessboard reference between Josh and Darth Cliche. Good times.

redtara: they/them

15-10-2011 13:47:19 UTC

“I think that, regardless of whether people think the scam worked, we should work together on a set of CFJs to make sure that we have a known non-broken gamestate (and we can worry about what to do about the current/next dynasty afterwards). I suggest that I should make a fix CFJ, and that someone else should idle out, then once the timer expires make a post, unidle back in, and edit it into a CFJ saying the same thing. Both CFJs should be uncontroversial fixes, to make sure they pass, and we should agree on the wording in advance.”

This seems good to me. The remaining problem with precedence will be tricky. I am personally in favour of precedence being the following:
1. Dynastic rules which explicitly state to take precedence over core rules.
2. Core rules.
3. Other dynastic rules.
4. Appendix.

The appendix should be, in my opinion, merely supplementary information. Terms, clarifications, etc.

I think it is useful to allow some dynastic rules to have priority over core rules. For example, if it makes sense to allow an extra proposal per person (for some reason) at a time, I think there should be some way to do that.

Core rules are, obviously, the core of the game. I think they should take precedence in most cases.

Prince Anduril:

15-10-2011 20:16:32 UTC

Appendix has to be on top, otherwise people can define things however they want. Is there a way of calling dynastic rules that take precedence over core rules something different, putting them in a separate category?

bateleur:

15-10-2011 20:27:43 UTC

I still think we shouldn’t be worrying about precedence stuff until after the immediate problem is fixed. It’s a complex issue which the above suggestions seem too simple to fully fix.

If nobody else drafts an attempt at addressing the Official Post vagueness by Tuesday then I’ll do it myself, although it seems to me that someone who knows Blognomic better might be a better choice.

ais523:

15-10-2011 20:34:22 UTC

@Kevan: I think the ban on posts being two sorts of official posts at once followed (although not directly) a controversial scam by me where I edited a story post (with no comments) into a proposal, then claimed it had already been open for long enough that it could be adminned immediately (I’d given people the opportunity to vote on it and even requested they did so, but the post was deliberately boring and so got no comments, and IIRC there was a veto speed change proposal there at the same time to deflect attention). I don’t think an exclusivity clause is the best fix to that sort of scam, though; simply put better restrictions on edits (which I think we’ve already done).

Also, “may not declare victory” is really really bad wording, as it doesn’t prevent someone winning; just claiming they’ve won. “may not achieve victory” would avoid all the issues you mentioned with a core > dynastic unless otherwise stated rule.

(It’s worth mentioning that equivalent-of-core > equivalent-of-dynastic is a rule in pretty much every other nomic I have reasonable knowledge of the rules of, except B Nomic, which broke really frequently as a result.)

bateleur:

15-10-2011 20:42:32 UTC

The trouble with Core > Dynastic is that it creates situations where Dynastic rules don’t do what they seem to.

For example, if a Dynastic rule said “Proposals satisfying such-and-such may not be Vetoed” this would be like a blank clause because the Voting rule says that Proposals can be Vetoed.

The good thing about Dynastic > Core is that Dynastic Rules are (almost always) more recent and it’s natural for recent rules to override older ones.

Uristqwerty:

15-10-2011 21:39:05 UTC

Hello, I’ve just been lurking around occasionally, but from my computer programming experience, I see that the wording of the official post limitation is poorly worded and desperately wanting to be something more official. To me, it is clearly an attempt at a hierarchy of some sort, but without explicitly stating so, it opens the possibility of conflicts like this.

I see two forms that it could be (sorry for the space-consuming ASCII “art”)...

Each post could have exactly one type, and types are arranged into a tree
post
|-official post
...-Votable Matter
....|-Proposal
....|-Call for Judgment
....|-Declaration of Victory

In that case, Work of Art and Treacherous posts would, unless explicitly added to the type tree, could be treated as properties of a post, independent of the type (unless otherwise restricted by a rule).

Alternatively, rather than a single type, each part of the tree could specify how many of it’s direct subtypes a post may/must be, and it could then be structured as

Post (may be any number of subtypes)
|-Official Post (may be at most one subtype)
|.|-Votable Matter (may be at most one subtype)
|...|-Proposal
|...|-Call for Judgment
|...|-Declaration of Victory
|-Work of Art
|-Treacherous post

ais523:

15-10-2011 22:00:38 UTC

@bateleur: It’s easy enough to notice that you’re explicitly overriding a core rule, though, surely? What’s intended is “dynastic rules cannot override core rules by mistake”, but you can’t write that in a ruleset as it’s far too ambiguous. Wording it as “dynastic rules do not override core rules unless they explicitly say they do” comes to much the same thing, with care in writing proposals.

Bucky:

16-10-2011 02:27:16 UTC

I think narrowing it down to “An instance of one type of Votable Matter may not also be any other type of Votable Matter except where the rules regarding one of the types of Votable Matter explicitly say otherwise.  ” is a good idea.

Prince Anduril:

16-10-2011 06:14:18 UTC

I like Bucky’s idea, but we need to say first. “There are a number of types of Votable matters, listed in the following sentence: Proposal, CfJ… etc. No other type of Votable matter can be created without their addition to the following list. This rule cannot be over-ruled by Dynastic rules.” I think this might have tied down ais’s scam of implicit definition.

Bucky:

16-10-2011 18:19:10 UTC

The list would have to include Dynastic Vote or something similar, or else it defeats the original point of having Votable Matters.

Kevan: he/him

16-10-2011 20:09:52 UTC

[ais523] It’s also easy enough to notice that you’ve forgotten to specify that a new wiki page is created blank, or that a new GNDT stat should have a default value, but we do that all the time. I agree with Bateleur that it’s too unintuitive if dynastic rules of “players may make extra Eureka proposals” or “players with less than three clues may not declare victory” have no effect if they omit the “this overrules core” clause - people are going to forget, and people are going to scam it.

(And “player X may not achieve victory” seems a stranger wording than “may not declare” - if the victory condition is “score 10 points”, does that mean player X can’t take any action that would award the tenth point?)

It looks like the real problem is that we voted in a rule back in May that said a CfJ/Proposal/DoV “cannot simultaneously be any other type”, rather than “if it’s multiple types, those other types are ignored and it’s just a CfJ/Proposal/DoV” or something. I’m not sure that precedence is particularly relevant.