PerlNomic - An Experiment in Cooperative Coding

Anonymous Coward writes "PerlNomic is a game consisting of CGI scripts which allow you to submit proposals to alter ... the scripts themselves. All proposals must be approved by a voting process--at least for now. The game is styled after Peter Suber's Nomic. Deep knowledge of perl is helpful, but not required." Nomic is a really excellent game if you like mental puzzles, but somewhat difficult to get off the ground.

Comments

Collaborative regex expressions

[Anonymous Coward]

And this would be different from my cat walking across the keyboard how?

;-)

Re:Collaborative regex expressions

[eric_brissette]

Idunno, I'm too lazy to sign up for an account.

Contains political analogies

[KDR_11k]

Since Perl is notoriously "write-only", is the voting process on new codelines kinda like pushing an anti-terror bill through congress or an overbroad patent through the USPTO?

OK, I'll bite

[Sp1n3rGy]

I'm a perl coder but... What the hell is this? I've read the "rules" but still don't really get it. Anyone have a better explanation?

Re:OK, I'll bite

[Leadhyena]

It's REALLY weird but here's a quick summation... Nomic is a game where the game IS the collection of rules of the game, as sort of a metamathematical proof of concept of such a thing. There's no board or any game impelement involved (first set of rules I saw did have 1 six-sided die). People would present proposals and in turn you'd vote on them. If you vote for the winning side you would score points, and if you were in the minority you would lose points. If proposals were above a certain percentage they became new rules to the game and must be followed.

The rules for Nomic also have a distinction between immutable and mutable rules, kind of like a file-lock on really important system files, to enforce important restrictions (like a player may quit at any time, the game is not legally binding, you are never forced to perfore an action before being allowed to quit, etc) and keep them protected. However, it is possible to make immutable rules mutable in order for them to be changed, and that's when the game gets really weird.

Also, almost nobody wins by making it to 100 points (by the initial rules of the game). 9 times out of 10 the rules are ammended to allow another winning condition that is either easier or harder to achieve. A lot of times Nomic games are created just to keep them going for as long as possible. In many ways it's like a grown-up blend of AD&D and politics.

Nomic is really fun to mix with other games. Specifically my friends and I would play a mix of Nomic and Monopoly we'd call Nominopoly, and games would last for literally days or weeks. Mixing Nomic with a card game (essentially this is similar to Mao) is fun as well, especially if there's drinking involved.

Re:OK, I'll bite

[Short Circuit]

Read up on the Nomic link first. Then you'll understand the point.

Then follow the directions on the PerlNomic page to get a the game itself. (It requires modifying the URL...a technique at least one Ivy-League school calls hacking.) Look for the CGI module that's likely to get you an account.

Even without an account, you can look at the source.

Re:OK, I'll bite

[zudini]

A good starting place is suber's original nomic game:
http://www.earlham.edu/~peters/nomic.htm

The idea was to make a game out of making rules for the game. So each
turn, a player proposes a change to the rules, and people vote on it
and stuff. (This is usually done with people in a room writing on
index cards and posting to a bulletin board, though sometimes it is
played over email.) But when players disagree on the interpretation
of a rule, they call to a "judge" (who is just another player) to sort
it out for them.

Now think perl, and think self-modifying code, and think web forms
instead of index cards. No judges needed, because the script either
runs or it doesn't, and whatever the scripts allow are the "rules".

Now think obfuscated code, and hidden loopholes, and unfortunate
little bugs that allow you to get way more points than we expected you
would get when we all voted on your proposal.

That's the idea, anyway.

Re:OK, I'll bite

[snorklewacker]

Ever play bar games like quarters where when you get a quarter in the shotglass (or five into the tumbler) you get to make a new rule? And eventually the rules start referencing other rules, and then the game itself becomes more of a complicated meta-game where the rules themselves are the game?

That's nomic. Except it does away with the shotglass, everyone just makes a rule every turn. It's a parliamentary game about rule-making, where the one who can game the system to their advantage is the winner. Still, it's usually a lot more fun as a drinking game, so in-person games often get rules passed to that effect. Order decays quickly a few rounds after that.

Re:OK, I'll bite

[Sp1n3rGy]

AHH! Now I get it. Finally a concise answer. So basically it is a game of politics where you win by beating the system. Very interesting. I've signed up for an account. I'll check it out.

Re:OK, I'll bite

[DerekLyons]

See Nomic.Net.

Re:OK, I'll bite

[dr.badass]

Now I get it. ... So basically it is a game of politics where you win by beating the system.

For the sake of greater understanding I will point out two things, and then a very not-concise draft-quality discussion of Nomic, which you are free to ignore.

1) Trying to "get" Nomic usually causes one to grow dizzy and pass out. As with most things, you'll eventually discover something about it that you hadn't noticed before, and suddenly "get it" much better than you did before. This cycle of enlightenment won't ever really stop, unless you're, like, the Buddha, or Don Knuth, or something.

2) "A game of politics where you win by beating the system." is a reasonable explaination of what Nomic begins as, when playing by the original rules. It is neither a literal interpretation of the rules, nor a adequate description of the possibilities of what the game may become.

(I point these things out because I find Nomic deeply interesting, not because I think anyone else is wrong for not seeing it this way.)

For instance, winning is itself defined by a rule. In Suber's original Nomic, the initial winning condition is to score 100 points, and another rule says that the winning condition can't be changed to anything other than scoring a number of points.

The literal approach would be to get as many proposals passed as possible, by giving the other players reasons to vote for them, and to be on the winning side of every other vote. This, itself, is a deep game.

The "beat the system" approach (which never fails to excite people new to the game) involves trying to create rule paradoxes or very primitive combos, or invoking loopholes. Often, though, the game devolves into a kind of tug-of-war between people trying to craft incorruptable rules (patching every hole) and people trying to win by some clever master stroke (arguing over what the definition of "is" is). The game is still very "play-to-win", and winning usually means making a number bigger or 'killing off' the other players. It is still zero-sum.

It's beyond this, you start getting into the really fun part. It basically happens when people consciously or unconsciously agree to cooperate. Winning ceases to be the point (even if it is, in the rules) -- it's about continuing play. It becomes non-zero-sum.

This takes some getting used to, but it makes for the most interesting and stimulating games.

What happens when you : eliminate the rule that says the game is over when one player wins? eliminate winning altogether? eliminate voting? eliminate the first rule, the one that says players must abide by the rules? At what point does it stop being a game?

The last example should highlight the fact that what happens is entirely dependent on the players.

If you ever watch group improvisation (comedy, theater, jazz, or just children playing), you might notice that it works because each player accepts what the other players give them.

If Actor A says to Actor B : "Hi, Doctor B!", then B immediately becomes a doctor. If Doctor B says "Hi, Nurse A!" right back, then A becomes a nurse, even if he wanted to be a patient.

When it doesn't work, it's usually because someone rejects what the other players give, or tries to steer the act in another direction. If I touch you and say "Tag, you're it!" and you say "Oh, no I'm not!", well then we've got a pretty crappy game of Tag.

This isn't to say you should just except any change that another player offers -- but rather that your criteria for accepting or rejecting them becomes more about fun, and continuing the game, than than winning or losing. It's like having a conversation, not having an argument.

If you're interested in Nomic conceptually (rather than seeing it as "just" a game to be played) I can reccommend several books:

James Carse's Finite and Infinite Games, Herman Hesse's Magister Ludi, and Peter Suber's own The Paradox of Self-Amend

Re:OK, I'll bite

[snorklewacker]

> Herman Hesse's Magister Ludi

I believe this book is actually called The Glass Bead Game

Re:OK, I'll bite

[reverseengineer]

At least one major edition translated to English and published in the U.S. gave it the title "Magister Ludi," for whatever reason- in fact, it was until that title that many Americans first encountered it. You are absolutely correct that "The Glass Bead Game" is a more faithful translation of the German title, "Das Glasperlenspiel," and indeed, my copy of said book bears the title, "The Glass Bead Game."

Still, I wouldn't consider anyone who used the former title to be incorrect- I'm sure the book they read had that title on it, after all. It's all a question of how literally a given translator translates an original title- another example is Sartre's "Huis Clos," which is more familiar to many Anglophones as "No Exit," but more literally means, "Closed Door."

Re:OK, I'll bite

[Anonymous Coward]

We played a three-player Nomic Ping Pong (Pnong?) where two people would play at a time; at the beginning of each round, the two active players would create or modify a rule. After the round all three players would vote on whether to keep the rule. Then we would rotate.

It was a lot of fun, and quite effective as an engine for trying out rule variants without a major commitment. The biggest surprise was that the rule "you have to hit the ball twice before sending it back over the net" turned out to be really fun - it stayed active for the rest of the game session.

problems of nomic

[snorklewacker]

Nomic's not difficult to get off the ground at all. Ask enough people online, and they'll be more than happy to play. Sustaining a game past 10 rounds or so is virtually impossible however. You can guarantee you will lose half your players in round 1 from normal attrition of people who sign up and don't bother checking back. The first several rounds are pretty dull, and if people don't start creating subgames that are actually FUN really quick, everyone else gets bored and drops off.

The ideal environment for hosting a nomic backed with code is a MUD, such as MOO. However, not everyone wants to write MOO code (it's a good language compared to other MUD languages, but that's not saying much). With a virtualized server or usermode jail, one could confine the nomic server relatively safely however. However, even hosting a MUD, let alone a full server is still a fairly expensive hobby. Webhosting has gotten dirt cheap, but everything else is still going to cost probably as much or more than your monthly ISP bill.

Re:problems of nomic

[Anonymous Coward]

The first several rounds are pretty dull, and if people don't start creating subgames that are actually FUN really quick, everyone else gets bored and drops off.

I find that this happens usually because people starting online games diverge too far from Suber's rules from the start. The initial set is a fully-functional game, with a winning condition and everything -- and yet, most online games use badly-composed "untouchable" rules that can't be modified at all.

Yes, if you're running an online nomic-like game, you're going to have to modify the original rules -- but *MODIFY* them, don't rewrite them badly.

My advice (and the way I used to run games) was to modify the rules *by the rules* -- start playing with one player: yourself, and "build" rules around whatever platform you're using (website, mud, mailing list, whatever) by adding and amending rules as needed, paying attention to all existing rules, just as you would in a "real" game.

You can even include details like who gets control of the medium (i.e., who gets FTP access to the page, etc.) -- if there are real-world constraints (like that you're paying for it), then you can build some new category of rules that can't be modified by anyone else, etc.

There really is no reason to rewrite the original rules from scratch. That's part of the beauty of the game -- it provides for it's own destruction. Rule 101 says you have to obey the rules -- without that, is it even a game?

The problem isn't "subgames" -- it's people not knowing how to play Nomic.

Re:problems of nomic

[snorklewacker]

> The initial set is a fully-functional game, with a winning condition and everything

Suber's initial rules are completely unsuited for any online play, because they are serial. If someone quits and doesn't tell anyone, the game hangs forever. Or just doesn't log in for a week. Nothing gets people bored like one "round" in the initial game taking a month. Suber's ruleset also refers to "the player to the left" at least once...

Rule 101 doesn't say you have to obey the rules -- that's inherent in the definition of "rule", and you simply can't make that a rule without infinite regress. Its purpose is that it's a highest-precedence rule that says you must obey all the rules in effect, thus pre-empting a whole class of rules that repeal provisions of other rules without specifically amending or editing those rules themselves. Of course some games do have strange and amusing definitions of "in effect" (Calvinball anyone?)

Those poor guys...

[Anonymous Coward]

Only nine current voting members, and they just got slashdotted with seventy-five adduser proposals pending. And, according to the email I got back, more than half of the "active players" have to vote yea for each proposal for it to be decided favorably.

They've got a lot of work ahead of them.

Scheme vs Perl

[Taral]

There was a Scheme Nomic around a while back. For people who want to know more about Nomics, go search google a bit, then going Agora Nomic. We need new people. :)

Ha ha... Slashdot got soooo used...

[Leadhyena]

Either morbus is trying to use scare-tactics to slow the game down, or abliss got this posted on slashdot in order to use the resulting /. effect to win the game by just saying yes a lot of times. . I'd like to believe the latter; it seems too clever of a tactic not to be tried.

What a cool way to run a Nomic game!

Re:Ha ha... Slashdot got soooo used...

[zudini]

Actually, I (=abliss) posted it because I thought a bigger userbase would make for a funner game. I just put through a new proposal that will allow us to activate all the new users without awarding any points.

Had the idea too :-P Explanation

[RedLaggedTeut]

I played a few imperial nomics, and the game of Imperial Nomic usually proceeds by the players creating a boardgame(s) with their own rules, which get accepted or rejected by The Emperor. Usually some kind of victory condition, score, or money is involved. There can also be "Intellectual property" like owning letters of the alphabet(Unfortunately real life seems to copy the game).

You can initially propose rules, but later on the rules will allow you to give orders.
However, since there are a limited number of actions allowed per player, and every action is a lot of work for the moderator to work out because it is affected by all rules, there is no incentive to allow more actions for every player, and therefore there is not enough time to fully watch the effect of news rules, i.e. rules get proposed faster than taking effect.

A perl Nomic would solve some of these "bugs", e.g. you could give players twice as many actions each turn or so.

I think it would be cool to have a Physics Nomic(in perl or pseudocode), which would attempt to create interesting physics. However, the actual universe seems to be very complex, it takes quite a stack of particles to create laws like q*x^12-x^6 for the repellant force between atoms, quantum physics effects would either require lazy evaluation, or fourier transforms, or both.

--RedLTeut / Highlander+E2

Re:Had the idea too :-P Explanation

[Hamled]

Now all I need to do is submit a proposal for a perl script that will read all of the email-addresses out of the adduser proposals, and sell them to random spammers. *Pats himself on the back*

Re:Had the idea too :-P Explanation

[Anonymous Coward]

I think it would be cool to have a Physics Nomic(in perl or pseudocode), which would attempt to create interesting physics. However, the actual universe seems to be very complex, it takes quite a stack of particles to create laws like q*x^12-x^6 for the repellant force between atoms, quantum physics effects would either require lazy evaluation, or fourier transforms, or both.


Google for "Fantasy Rules Committee", a bloody brilliant and long-running Nomic-like game. Basically, each round they have a theme, and the goal is to build valid rules on top of each other using that theme. The winner of the round is the last person to be able to come up with a valid rule.

Something like that would make an interesting science-like game. Each player comes up with a Hypothesis which can turn into a Theory, which can turn into a Law, each subsequent Law must obey all previous laws.

Since science is based largely on observation, the Hypothesis->Theory->Law transition would have be backed up by some kind of observation in the real world or some other source -- a video game, popular culture, or better yet : Slashdot.

You could Hypothesize that people with 800,000+ UIDs are stupid, and every observation of a 800,000+ UID person saying something stupid is a point for that theory, and every smart thing they say a point against. Eventually, it becomes a Theory, and is somewhat "protected" -- it takes more evidence to mount a counter-Hypothesis against it or something.

Re:Had the idea too :-P Explanation

[Anonymous Coward]

Cool, and I could sell sphyerion.com as a great place to host kiddie porn. Who wants to fire the first shot?

Agora

[Sir Toby]

I'll second checking out Agora, especially if you are having difficulty finding an online Nomic that lasts a while.

From the nomic.net Wiki entry on Agora: Agora Nomic is the longest continuously running Nomic known. Agora started in 1993 and has been going on ever since.

I was a player in it for a while, however had to eventually drop out due to lack of time. If you have an interest in Nomic, it is definately a Nomic to check out.

For more information on Nomic in general, also be sure to check out nomic.net.

No, it is the other way around ..

[RedLaggedTeut]

Any Nomic is like Congress trying to pass bills by making them seem innocuous, like the fabled PAT
RIOT act, only maybe even more so than Congress because there is no pretense of morals to keep up.

Using perl doesn't exactly make things worse, since at least perl code has a somewhat well defined meaning, while normal language is ambiguous. Just consider the sentence "remote control flies" of /. lately.

I tried a Scheme Nomic...

[AvantLegion]

... but the first rule suggestion, "Don't use Scheme", was voted in unanimously.