Gaming Social Software Technology

Self-reflexive rulesets in online communities…

It’s Tuesday morning and I’ve been in Seattle since Sunday evening at this year’s Microsoft Social Computing Symposium and frankly, I’m completely braindead through jetlag. I’m barely hanging on to intellectual coherence by my fingernails. Sunday evening I got about three hours sleep in total, last night a roughly similar amount. The quality of the event has been pretty high so far though, and I’ve met some fascinating people but I’m really not firing on all cylinders. Ross Mayfield’s taken a few chunks of notes, and most interestingly I’ve met some people with a similar interest to me in reflexive political models in online communities, including one guy who wants to build something very similar to the place I want Barbelith to become -online in an MMORPG. I finally got around to post up some of my earliest ideas around this subject on the Barbelith wiki a year or two back under the Tripolitica heading, but basically it goes a bit like this:

Imagine a set of messageboards, each with their own clear identity and each with a functioning moderation system based around a pre-existing political structure – one Monarchic, one Parliamentary Democracy and one Distributed Anarchy. Each of these political structures has been generated from one abstracted ruleset, and each component of that ruleset can be – in principle – turned on or off at will by the community concerned. Moreover, the rules are self-reflexive – ie. the community can also create structures to govern how those rules are changed. In other words, members of those communities can choose to shift to a different political model, or can develop their own by incremental improvements of changes to sections of the ruleset to allow moderators or administrators or normal users to create the ‘laws’ that govern how they inter-relate.

This self-reflexive component would operate with a bill-like structure – ie. an individual would be able to propose a new rule or a change to an existing rule that then may or may not require some form of wider ratification before it becomes ‘real’ and starts empowering or constraining the citizenry of that board.

When a new user joins the community, s/he is presented the current political structure of each one and from that point chooses a board to be affiliated with. S/he is then part of the population of that community and can rise up through the ranks (if there are ranks) and participate in the functioning of that political community. This goes right down to the creation of different parts of that commnuity, how the various parts of the community inter-relate with one another and who can post what and when.

Each community will have its own strengths and weaknesses – some will no doubt go horribly politically wrong and have power seized by mad administrators, but hopefully others will find their own kind of political equilibrium after a while – and maybe that political equilibrium could be a good model that could be genericised and used as a more common and rigid platform for new online communities that aren’t interested in the emerging rule-set component. That is to say, maybe we can evole a better system for handling debate, discussion and power relationships in messageboards and other online community spaces and games. Of course, for that to happen, the ruleset has to be sufficiently politically abstractable that new arrangements could emerge that didn’t initially occur to us during the creation of the ruleset and the reflexive process has to be comprehendible to real users.

Some sample bills:

  • Anna proposes a bill:

    Junior members to not be able to create threads

  • Bill proposes a bill:

    Administrators to not be able to change user roles.

  • Charles proposes a bill:

    Junior members to be able to create posts. Action will require ten ratifications from Moderators, Administrators, Normal Members. One disagreement can veto.

  • David proposes a bill:

    Moderators to be able to edit abstracts. Action will require three ratifications from Moderators or Administrators. Three disagreements will veto.

  • Edgar proposes a bill:

    The User Responsible to be able to change their own display name. Action will require no ratifications.

  • Fiona proposes a bill:

    Normal Users to be able to Unblock Users. Action will require 60% assent from Normal Users polled over 24 hours.

  • Gavin proposes a bill:

    Normal Users to not be able to propose bills. Action will require a 51% decision of all users polled over a 6 hour period.

I’d be interested in anyone’s thoughts around this stuff.

14 replies on “Self-reflexive rulesets in online communities…”

That’s interesting… I’d be interested in seeing the result – but less interested in participating as it would be group moderation for the sake of moderation; little incentive to participate unless you’re interested in the politics of it all.
I do think, however, such a cool idea would make an excellent educational tool on the different types of government. There was once a teacher in my school who, for a few weeks of every year, ran his classroom with a different governmental policy – to teach students the differences. Totalitarian week was local legend and many students hated it – until one year when the class banded together and assasinated their cruel leader with water balloons and shaving cream.
Anyway, I ramble. Point being – your concept might be the begining of an excellent tool to teach about governments. Without the mess.

Sounds fascinating, on a lesser scale, has a politics tab for it’s groups, whereby leaders can be overthrown by popular demand. I’m sure it would make a great experiment, but I’m not sure I would enjoy being part of it.

Definitely would be interesting to see it played out. But, in my experience with discussion-based communities, the more visible the rules are, the more the discussion will inevitably focus on the rules. Add a points system, and eventually all discussions will devolve into an argument about points. Give users the opportunity to propose rules, and eventually all discussions will devolve into an argument about the proposals.
Of course, that’s taking a pretty dim assumption of the better natures of the participants. With the right community, it could work. Somehow I’m not inclined to be as charitable about online community participants as Jefferson was about citizens of democracy.

one of the worst parts of policing a message board is the part where you have to resolve disagreements and ask people to leave. if there was the wide-spread political un-control that your talking about, my guess is that the majority of the message boards would tank but the few that made it through growing pains would be stellar. however, our culture shows that we prefer a leader, even a bad one. out of the boards that make it, my guess is that they all settle with someone(s) ending up the main moderator(s) that everyone else looks to for decision making approval. well, that’s my projection, anyway. i look forward to seeing how it really plays out.

Sounds very interesting. I think your example bills would need to be much more granular to allow real evolution of government. Simple concepts like roles, permissions, grouping, decision-making, etc. should be flexible and contextual. I’m imagining it should be able to handle something complex like ‘authors’ can create posts which are published if either 60% of ‘members’ (of more than 14 days( vote at least 60% favorably on a scale of 1-10, or 80% of ‘elders’ (of more than 1 year and more than 3 published posts) vote at least 50% favorably.

Perhaps stating the obvious, but you should definitely take a look at Nomic if you haven’t already, particularly the matter of “scamming”, which is likely to rear its head system with formalised rule changes. A bored or resentful user either exploits the existing ruleset in an unexpected way (perhaps using a load of friends or sock-puppet accounts to achieve something that the community wasn’t expecting to be easily possible), or wilfully misinterprets a careless wording (a wiki of mine saw a reviled candidate getting briefly promoted to moderator because the rules only stated that promotion required community discussion, not positive community discussion), or actively sneaks such ambiguous wording into the ruleset by hiding it in an innocuous-seeming bill.
In anything less than a dictatorship, these things can be difficult to clean up. Yes, multiple boards allow for the fun of migration, that a crazy scammer deposing the Monarchy is likely to find themselves with no subjects left to torment, but it could get a bit tiring.
Would be interesting to see what happened, though.

I be interested to see how the workability of social/polictal board moderation changes with community size. i.e. is their an idea size for this to work, to small and it lacks purpose, two big and it falls about.
Most forums I reckon seem to work up to a certain size, and then as they get bigger and bigger, their is too much chaff, and they fall apart, regardless of moderation.
Maybe it’s a factor of time (maturity) more than size of the population.
This is a bit like the life game, but with forum posts isn’t it?

Sounds fascinating, and I’m keen on evolving systems.
One thing to watch in the dynamics of such systems is how the focus switches between handling business as usual and extreme events. For every 1000 angry flame wars that can be resolved with standard well-known admin, there will be one mad obsessed stalker whose behaviour will be beyond the pale.
The amount of effort that goes into solving these extreme cases, and debate about changing laws etc, will presumably mirror political debate here about laws for terrorism.
Because these events are rare and often very different to each other, it is very difficult to evolve general solutions that cover them. Analogy: E.L.E.s for social systems . . .
I’d be interested to know if anyone has experienced such events?

Like other commenters, I think it’s Nomic, or would rapidly degenerate into Nomic. By which I mean that to be really interesting the communities would need to have a continuing purpose over and above talking about rules for talking about rules – but that, once the TARFTAR got under way, it would be likely to take over, and the people who wanted to keep the original purpose would either drift away or get converted to the new game.
Perhaps you’d end up with two communities thriving, one devoted to TARFTAR and the other consisting of all the people who didn’t want to be bothered with rules – and perhaps the ruleset of the second community would be a model of consensual minarchy (i.e. “Everything is permitted except screwing things up”, only codified). But I wouldn’t bank on it.

Not only is it a good idea but it has, to an extent, been done with very fruitful results. Unchat ( is real-time discussion software that allows the group to choose its own rules of conversation, embed those rules into the software and change the rules in order to create different kinds of group decision spaces. The group can decide, for example, whether to have an unmoderated, moderated or self-moderated discussion and the term and tenure for the moderators. Moderators can be chosen by selection, vote or rotation. The number of interruptions accorded to each participant can be selected. The rules of ostracism can be set. Etcetera. For more about the project, you can visit the (outdated) Unchat website or read the extended description here:

I have to say it sounds like a wonderful interweb utopia that may one day become reality. But after reflecting on such a possibility, I suggest that we not leave out the lessons that history has taught us – it has, afterall, been around a good while.
In any dynamical system (one that incorporates feeback) there is always the possibility of the system going “unstable” (turning anarchistic). As the laws of chaos dictate, it may only take a minor change to cause such an anarchistic outcome.
Almost all societies that are today democratic have got there only through many years of dictatorial rule followed by a period of civil war. I would imagine a ruleset-based online community would follow a similar evolutionary pattern.
I also find it hard to imagine that everyone in the community would vote on every decission, as the real-time updating of the rules you propose would mean. The UK can’t even get three quarters of the population to turn out for a general election, an action they need only perform every five years.
This may be because the UK economy is fairly stable at present, thus promoting apathy, and that new users of a rulesets-based online community would be more inclined to vote more often. But, at the end of the day, it is not feasable for everyone in a community to vote on every decision that must be made. Some sort of deferal system and hiarachy would inevitably come about.

Once upon a time, I came across a meta-game which had only one rule governing the sequence of turns. Everything else was up for grabs including allowing players to change the rules (apart from the one true rule). I believe there’s some game theory around this concept.
To make this more interesting you really need to set up some competition between competing rulesets which in this context probably means competing online communities. You could use Alexa! Except that it would be hard to filter out other reasons for success or failure.
There are many obvious parallels for the idea. And just like politics the most effective is the benevolent and wise dictatorship. But unfortunately, these always degenerate.

Comments are closed.