Dunbar Number Presentation at MeshForum 2006
Last May I did an abbreviated version of my Dunbar Number talk at MeshForum 2006. A MP3 podcast of that talk is now available at IT Conversations or can be downloaded here (10mb).
If you'd like to follow along, here is a pdf copy of my presentation sides (10mb).
Biggest addition to what I've written about before is some discussion about different kinds of social software and what what size groups they seem to be appropriate for.
Some other posts about the Dunbar Number and group size issues:
Posted on August 31, 2006 at 11:52 AM in Science, Social Software, Web/Tech | Permalink | Comments (0) | TrackBack
Future Topics
I've been working on an ambitious list of topics that I'd like to cover over the next year. I offer them to you here so you can have some idea the areas that I am thinking about.
Office Architecture for Innovation -- Over the years I've built or converted three offices to my specifications. From this I have learned a number of things about about how to create a productive environment innovation-oriented businesses. These include some of the obvious suggestions such as fresh air and natural light, but also include not so obvious ideas such as using magnetic paint and providing a small washer-dryer.
Requisite Variety -- This concept from cybernetics applies to social systems as well. "The larger the variety of actions available to a control system, the larger the variety of perturbations it is able to compensate." More people, as well as a diversity of thinking styles and experience, give social software more "variety of actions", thus this is part of the reason why social software can be so effective.
The Art and Craft of Meme Design -- We are learning more about how to create an effective meme. Creating memes has always been an art performed by publicists, marketers, politicians, the press, and to a lesser extent by scientists and other academics. Have we learned enough to turn this art into an explicit craft?
Wiki Editing Dichotomy -- One interesting possible barrier of entry to active participation in a wiki is what I call the "wiki editing dichotomy". You have to be proud enough to believe what you are contributing is generally worthwhile to others (or at least worth your effort), but you also have to be humble enough to understand that others can improve it. I don't know of many other collaborative media that requires both pride and humility.
Choice & Neuroeconomics -- There are some that say at the root of every decision is emotion. Even a 'rational' decision appeals to a sense of balance or beauty. Recent studies using PET are establishing a neurological basis for emotions, and some reveal interesting facts about how we make choices.
Assessing Risk -- There are a variety of areas where we as humans have a difficult time being completely rational. One of these is risk assessment. It turns out we may be hard-wired to not be able toeasily understand risk that is greater then one in a hundred or so. Thus a very rare risk, say one in a thousand, will often be emotionally interpreted as having a much higher risk.
Persuasive Computing -- BJ Fogg's group at Stanford has done some interesting research on how computers can persuade you to do something. There are a number of useful ideas that come from this research. There are also some ethical considerations that should be discussed.
Cognitive Dissonance -- This technique is central to many forms of persuasion intended to change beliefs, values, attitudes and behaviors. It is used by facilitators, businesses, military organizations, and even cults. It can be used positively or negatively. How might it be used in social software?
The Dark Side of the Force -- The same social tools that we use for good, can also be used for harm. How do we ethically use what we are learning about social software? Some say that almost by definition social software attracts spammers, trolls, and innappropriate sexuality. What can we do to prevent these misuses of social software?
Conversation vs Communication -- Update and rewrite of my 1990 essay on how social software design needs to balance conversation vs communication.
Social Emotions -- We appear to have evolved a number of emotions that appear primarily to exist to support a common good, rather then to ensure our individual success. These include schadenfreude, mirth, naches, revenge, shame, pride, outrage, approbation, admiration, elevation, etc. Studies by Eckman on unconscious facial gestures, studies on the neurological basis for emotions, and studies on emotions in games, are proving that these social emotions exist. A number of them have interesting implications on social software.
Glances & Strokes -- There is some old work on the amount of eye contact we make with others in small groups, as well as some research from transactional analysis on strokes, which are the amount of recognition given to others through words and deeds. Is there a neurological basis for needing a certain number of glances and strokes each day? How does this concept apply to social software?
Weak Links -- There are some interesting social implications behind what we've learned about weak links in social networks. How do we identify and encourage weak links in our social software systems?
Negativity vs Positivity -- It is far easier for someone to respond negativel than positively. In political systems it is far easier to say no rather then yes. What social software encourages positivity, and is it possible to design social software to do so?
Time Economy -- Our ultimate most unrenewable resource is time. How time and attention are a basic economic unit that should be considered when looking at social software.
Group Life Cycle -- We often focus on how groups form, emerge, and grow. Yet there are many lessons to be learned from how groups die, including that it isn't necessarily a bad thing and that keeping a group from death can be dysfunctional.
Groupthink -- What causes groupthink? When is it good and when is it bad?
Two Thresholds in the Value of Knowledge -- In order for knowledge to be valuable, it must at minimum be more valuable then the costs to find and absorb it (costs which include bandwidth and attention). Tools like Google have tremendously increased the amount of knowledge that is worth the time and attention to find it. What types of knowledge fall below this threshold of value? Is there a limit to how much we can lower this line? Furthermore, there is another threshold where the knowledge is significantly more valuable than only bandwidth and attention. How much have internet tools impacted this second threshold? Many internet business models, in particular content models, require some ability to offer value in this upper threshold -- can they survive as this upper threshold changes?
Intimacy and Social Networks -- Social Networks Analysis tends to focus on the connections between people, either explicitly through acknowledgement of connections (LinkedIn, Friendster) or implicitly through analysis of your communication (Spoke). None are able to measure intimacy. Yet our intimate social networks are an important component of our overall happiness and contentment within both our professional and personal lives. How does intimacy work in social networks? Also there are some concepts in psychology known as communal vs exchange relationships by Clark & Mills from late 70's that may apply here.
Social Games -- A recap of Shannon Appelcline's and my analysis of the basic forms of social games types. These include relatively well understood ones like majority control, voting, meta-voting (Nomic), auctioning, etc., but also include less well understood games like playing of roles, dominance and submission, etc. There are also links to social emotions, such as mirth and schadenfraude.
Lessons from Castle Marrach -- We released the Castle Marrach online game in September of 2000, and it was designed from the beginning to be a game for the Bartle-type known as "the socializer". What lessons did we learn in the five years since the release of the game? What tools are in my social software toolbox today that might have helped with the design?
Lessons from F2F Facilitation -- There are many skilled practitioners of face-to-face facilitation, some of which are paid very high fees for their skills. What lessons can we learn from their experience that we can apply to social software? Why have so many of these facilitators failed to have success online?
More Human vs More Than Human -- Many futurists seek to offer us augmentation of our minds and bodies through technology. Many of these ideas may fundamentally change what it is to be human, and may even have unforeseen complications unanticipated by their creators. One interesting approach to looking at these technologies is to examine which make us "more human", rather then "more than human".
Lessons from Mental Disorders -- Most, if not all, mental disorders have their roots in survival strategies; however, they are over-expressed because of genetic or other causes. Examining the healthy behaviors hiding behind depression, autism, mania, schizophrenia, paranoia, etc. offer a number of insights on how we think.
Joy of being a Primate -- If you scan the surface of my writing, you may observe that I have a strong belief that our animal nature and genetics form an essential and often unconscious part of what it is to be human. You could interpret these "nature" over "nurture" ideas as limiting us to being just animals. Instead, I believe that by becoming aware of our primate nature, and choosing to leverage it or suppress it by conscious choice rather then letting it drive us unaware, is what makes us more powerful.
Smart Contracts -- Nick Szabo popularized the idea of using some of the primitives of cryptography in unique ways to create what he calls "smart contracts". He hypothesized approaches to cryptographic handling of collateral, bonding, delineation of property, bearer certificates, and much more. Others have proposed various auction protocols using these concepts. One of the fundamental atomic elements of many of these smart contracts is something called a "reusable proof of work", which Hal Finney recently demoed a version of at CodeCon 2005. What are the possibilities offered by smart contracts? What are the barriers to implementation?
Club System -- In the 80's, development of Ted Nelson's Xanadu vision was being financed by Autodesk. Unfortunately, for a variety of reasons Xanadu failed to deliver any technology. However, a couple of their ideas could be valuable today, one of which is the concept of an alternative to the "user and groups" metaphor for computer security. Xanadu turned that idea upside down and called the result the "club system". The club system approach is particularly suited to the internet based collaboration tools, in particular wikis. I also have some insights to offer a cryptographic approach to the club system, which might allow P2P distribution of collaborative documents, while preserving group privileges.
Edges of Cryptographic Security -- The SSL cryptographic protocol offers a choice of a number of security properties: integrity, confidentiality, encryption, one-party authentication, and two-party authentication. But there are a number of security properties that very few deployed cryptographic protocols offer. These include perfect forward secrecy, undeniability, deniability, authorization, delegation, multi-party authentication, shared secrets, etc. What are these security properties and how are they useful? Why have they not been successfully broadly deployed?
The SSL Story -- When SSL was first proposed it was broken within an hour. Even when Netscape fixed those problems, it wasn't clear that SSL was going to win the battle of security protocols. SSL was competing against SHTTP which had backing of RSA and an industry consortium. The credit card companies merged their standards and were backing SET. The internet community was moving toward SSH. Microsoft was doing its own embrace and extend protocol PCT. So how did SSL win to become the broadest deployed cryptographic security protocol? The answers may surprise you.
I welcome any comments or suggestions for links on these topics, or any new topics that you feel are closely related.
Posted on April 4, 2005 at 01:08 AM in Business, Games, Science, Security, Social Software, User Interface, Web/Tech, Weblogs, Wiki | Permalink | Comments (2) | TrackBack
Dunbar, Altruistic Punishment, and Meta-Moderation
In my post about the Dunbar Number I offered some evidence on the levels of satisfaction of various group sizes based on some empirical data from online games. There I was able to show that even though the Dunbar Number might predict a mean group size of 150 for humans, that in fact for non-survival oriented groups the mean was significantly less, probably between 60 to 90.
I also offered a second hypothesis, that there is a dip in satisfaction level of groups at around the size of 15. Unfortunately, I could only offer anecdotal evidence that this threshold existed. My personal belief was that this dip was caused by not enough "attention" being given to everyone and that group gatherings of this size risk becoming too noisy, too boring, too long, or some combination thereof. Yet groups of this size are not large enough to allow for different perspectives (i.e. insufficient requisite variety) or for other group processes to come into play.
I was reading through the current (12 March) issue of New Scientist, and found an interesting table in the article Charity Begins at Homo Sapiens. It stood out to me as it showed a dip approaching zero average cooperation for group sizes of 16 -- almost precisely same place that my hypothesis predicted.
Digging further, I found the original source for this table was published in Nature back in October 2003, in The Nature of Human Altruism written by Ernst Fehr and Urs Fischbacher.
What the chart shows is actually not empirical data from human experiments, but instead data from a game theory simulation of the prisoner's dilemma. This is the classic zero-sum game where if you cooperate, the joint payoff of the players is higher. However, in zero-sum games there is also incentive for the players to cheat by defecting and thus taking less risk. Robert Axelrod showed in The Evolution of Cooperation back in 1985 that in spite of the statistical best individual strategy being that of defection, cooperation inevitably evolves.
Fehr and Fischbacher, in their Nature article, took this idea a bit further by creating 100 independent simulations with group sizes ranging from 2 to 512, and then executing each simulation 1,000 to 2,000 times. Each generation of the "players" was allowed to evolve different strategies of cooperation vs defection, the classic successful strategy being Tit for Tat. They would then evaluate the percentage of players who had cooperative strategies.
If punishment of defections was ruled out, they discovered that over the 1,000+ generations of the simulation that the rate of cooperation quickly crashes, such that at the group size of 8 a little over 50% cooperation evolved, and for groups that are larger than 16 none cooperate.
Next they added to the simulation "Altruistic Punishment". This is the ability for players to punish those who did not cooperate -- however, such punishment is at some cost to the punisher. Earlier game theory research on altruistic punishment has shown that cooperation flourishes if there is some price for punishing defectors -- if you allow punishment at no cost then cheating strategies emerge. This earlier research determined that allowing individuals to punish at some cost that yields them no material gain will paradoxically result in an average gain for everyone.
Adding altruistic punishment to the simulation increased the amount of cooperation that evolved, such that groups with the size of 32 would have 50% cooperation. But even this had limits; at the group size of 128 no cooperation would evolve.
Finally, if they added to the simulation the ability to punish those who did not participate in punishing (i.e. didn't pay the cost to punish defectors), then the percentage of cooperation that evolved was never less then 60%, and in fact got better as groups got larger.
This is a very interesting result. To explain it in different terms, if you have a system that depends on sharing some commons and there are no process or trust metrics, a group as small as 16 may find themselves not cooperating very effectively.
The idea of commons can be as simple as how much speaking time participants in a meeting share. The time that each participant uses during the meeting can be considered the shared "commons". If there are no enforced rules, with a group size of 16 there will inevitably be someone who will abuse the time and speak more than their share.
With some simple rules (some type of process to partition time more fairly), or through some trust metrics (punish those who abuse the commons), larger groups can gain value from cooperating, but even these groups have limits. As long as there is some effort required to punish those who abuse the process, eventually the price for that effort becomes too high, and no one is willing to punish any longer. You see this in moderated discussion groups, where the guardians of the common good have to spend too much time moderating, given the number of people that wish to participate. Moderated newsgroups and early versions of SlashDot moderation had this problem.
The most interesting observation from this simulation is that for larger groups, you needed to have a system to punish those who did not participate in punishing -- or to put it another way, get everyone involved with the process, not just a few. I think that this explains something that I've always wondered about SlashDot's meta-moderation system -- with this feature you grade those who have moderated, and if someone abuses it they will no longer be able to moderate. What I suspect, given these results, is that if you don't participate in the meta-moderation system, you have less a chance of being able to moderate; thus SlashDot has a system that punishes those who are not involved in participating.
I do have some questions based on this simulation -- I would like to see results with more granularity then powers of 2. I'd also like to know the percentage of players participating in punishment -- I suspect that you'll find a very small number of "moderators" are required in the middle graph, even when cooperation evolves successfully. I'm curious as to how many moderators are required to participate in punishing the non-punishers for cooperation to evolve in the third scenario. Also, in meta-moderation this simulation emphasizes punishing the non-punishers -- is that as successful as rewarding the punishers? I'm also interested in investigating some of the further works where Fehr and Fischbacher are among the co-authors, such as this study using MRI (Magnetic Resonance Imaging) that shows that our brain's pleasure systems are activated when we altruistically punish.
I'm sure that the findings of this simulation are not the only reason why there might be a dip in group satisfaction at group sizes around 15, such as I observed. However, it does offer some interesting insights into group size thresholds, both for the threshold at 15 and for threshold of non-survival groups which I've identified as being significantly less then the Dunbar mean of 150. It also shows that for large groups we need to offer not only moderation, but meta-moderation capabilities that involve all the participants in the process.
In summary this research offers me another widget for my social software toolbox: in any group process look for the commons, allow participants to participate in identifing defectors; determine what the costs are for such identification (which may be as simple as requiring some attention or charging for such punishment); and encourage participation in the common good by punishing those who do not participate in seeking out defectors.
Some other posts about the Dunbar Number and group size issues:
Posted on March 17, 2005 at 11:38 AM in Science, Social Software, Web/Tech | Permalink | Comments (5) | TrackBack
