Slashdot Mirror
Cooperation Works if Majority Can Punish Freeloaders
plasmid writes: "Some Swiss economists ran an investment game... they found that if the majority could punish freeloaders, cooperation flourished. I think this has implications for cooperative peer-to-peer systems and, to a lesser extent, for open source development. I'm so inspired I plan to go out an punish someone right now, as a matter of fact." I had just read this article the other day (go memepool), so this Nature piece seems oddly apropos.
I know its funny, but this actually applies.
Think about it. If you have one large server that everyone tries to download from without others participating, several problems emerge. First of all, you saturate the bandwidth on that connection. Secondly, with limited resources, it takes much longer for everyone to obtain what they're wanting. And when the server is located by authorities and shut down, a major resource is lost.
Now, have everyone serve. Anyone looking for something can always find it, because its everywhere. They can always get it, because no one server is oversaturated, as the load is spread out. If one or even several servers get shut down, the effect is minimal. Everyone benefits when everyone cooperates and nobody is hit too harshly.
Now we have another form of potential punishment in this case, not from those that participate, but from law enforcement. Law enforcement, unlike the traders, is more likely to go after those who DO participate, and the freeloaders will get off scott free.
-Restil
Play with my webcams and lights here
Funny that Swiss Economists should come up with this conclusion.
Swiss - Sit back and watch the rest of the world fight tyranny and just rake the money in wherever and however it was attained
Economists - Earn money based on pseudo-science and predictions which are as reliable as those gained by examining chicken entrails.
Therefore should'nt we just punish Swiss Economists
Choose your allies carefully, it is highly unlikely you will be held accountable for the actions of your enemies
I seem to remember a statistic back in the old napster days that the majority of the people were freeloaders e.g. they just downloaded without offering anything themselves.
Now if we apply this swiss theory to p2p applications you know what will happen?
1) if the majority of the users are freeloaders then there is little chance that they are going to kick other freeloading users off the service
2) assuming that only contributors to the community get a vote then they will be faced with a massive task of getting rid of the freeloaders
3) once you lose all the freeloaders you are left with the people who adopted early and helped the service become massive, but you will have lost the majority of the userbase
4) once a service gets a bad reputation it sticks, and since these services gain popularity through word of mouth rather than regular channels you lose a lot of the potential users
5) lastly a particular p2p service may be good but there are a large number of services which are just as good and which wouldn't support this concept of co-operation.
Just my 2c
A similar set of ideals has been previously applied in psychological and darwinian non-zero sum games where there is a reduced personal gain but higher group gain from cooperation. These games challenge participants in finding an optimal outcome for both in cases where there are multiple iterations of choices to cooperate or 'defect' from cooperation - the website details only a new variant of these.
One model is that of the cold war. If both countries cooperate in an arms reduction treaty, they both win some, but for the individual country, a win can be made if their competitor cooperates and they 'defect' and build more arsenal.
This game has a matrix of possible points scored by each side depending on their individual choices.
. . . . . coop . . defectcoop . . . 3,3 . . 5,0
defect . . 0,5 . . 1,1
In the above situation, the two scores delimited by commas indicate the score for each country. If the countries both cooperate, each receives three points. However, if they disagree, one country will win, but the sum score is less. The interesting situation is if both defect - the value placed on these scores may also determine how the game is played through multiple iterations by two players.
Another variant is the prisoner's dilemma game. Two criminals are captured, and the DA will cut one of them a deal if they squeal on the other. Of course, if both squeal on each other, both loose big. If both are quiet, they will get a lesser charge. The dilemma is that the best group outcome is that they will both fare better if they are both quiet, but they don't know what the other will do.
The article listed is similar to this, but different that there is a cost involved in punishing the 'bad' player that doesn't pay into the investment pot. Here the game asks you to punish the uncooperative player with costs now, but the punishment might make them more likely to contribute in future rounds of the game. Interesting.
Well, spammers have the opposite situation, a small cost for them (sending 10000 copies of the same mail) costs much more to others (bandwidth, time reading and deleting junk).
The correct application of this work (although not to the letter) would be to 'punish' those who spam us with lawsuits such as is allowed in Washington State. Although it is a personal cost to call ISPs, file suits and such, if everyone were to make such small pains, we would all benefit greatly.
Then of course next there would be the freeloaders who do nothing to help but profit from our spam-eradicating work that need to be punished ....
This "game" sounds like a development of one which I read in The Economist a few years ago.
In that, the idea was a group of people had 10 beans , of which some were added to the pot and the rest were kept by the participant. At the end, the he pot was shared amongst all, and the goal was to maximise the indivuduals holding (with no concept of punishment).
This was carried out at a university (where else?) and it found that while students of most disciplines did the same thing, kept five and shared five, (only) students of economics kept 9 and shared 1. The summary of The Economist wondered whether this was cause or effect of studying economics.
I wonder if people like Linus Torvalds, Alan Cox, ESR et al would keep 1 and share 9, and whether Bill Gates and co would behave more like economists.
If the other sources somehow become slashdotted, NewScientist also has an article up on this.
It's up under the title "Anger plays key role in human cooperation".
--R
People will pay to punish - suggesting that their
notions of fairness outweigh selfish considerations.
This quote reminds me of an experiment. It runs something like this: A group of people is divided into two groups of equal size. Then each group is asked this simple question: We will either give both groups $2 per person or we'll give each of you three bucks and each of them four bucks. What would you prefer?
85% of the participants go for the two bucks.
Think of M$ - making billions just because they once written a not-so-good program! Isn't it freeloading.
Disclaimer: sorry for m$ reference, just couldn't resist :-)
This is an example of what most elites have nightmares about... the "masses" deciding for themselves what to do, through concensus and free exchange of information. This is the horrible, to-be-avoided-at-all-costs thing that many refer to as "too much democracy". The key is this: it only works if those with an interest/stake both get a place at the table and the ability to punish people who waste their time with lies and greed.
I'm convinced this kind of democratic, community-oriented "anarchy" could work at any scale. As long as everyone feels they are part of something meaningful, and that everyone else is taking it seriously, then you can actually get "competitors" to agree on strategies to maximize the Common Good.
A major stumbling block has been the desire to "punish" criminals by sending them into isolation (or rather, creating isolated COMMUNITIES of criminals), instead of focusing on a more "healing" punishment which would require the community to confront, shame, and supervise the trangressors's rehabilitation.
For example, look at the pyros in Australia. Doesn't it just sound right that they should walk through the destruction, meet their victims, and generally confronted the effects of their crimes? Is it really better to lock them away where they can learn how to hate society even more? How can they be accepted into society again if they aren't genuinely seeking to make reparations?
Just like laughter--a social sanction against rigid codes of behaviour--punishment should bring people together. As weird as that sounds, everyone has to share in the duties of rewarding and punishing members of society: the only way to find a common good is to have everyone agree on it. Don't let anyone tell you that you should leave it to the "smarter/better" people to make this decision for everone else. What is best for those with privilege and power is not necessarily best for all.
I've got a bad attitude and karma to burn. Go ahead. Mod me down.
I definately agree with you and have seen examples of this on IRC. A few weeks ago when the FBI busted all those warez servers, traffic died down quite a bit. I read somewhere that they found out the reason for this wasn't necessarily because people were afraid to serve, but because there were only a few people that basically served a lot that everyone else mooched off of, and those were the people that got busted.
On the other hand, though, I think there are other reason for not contributing. Case in point (switching gears somewhat to Morpheus), I have Morpheus running ALL THE TIME on a DSL connection. But when I want to surf the web or download something, all those downloads kill my bandwidth. For this reason, I've set the maximum number of users that can download off of me at one time to 1, and the maximum bandwidth allowed for that to about 3.5kb/s. Now, I'm not using my computer all the time and I don't have a problem with allowing more users and more bandwidth when I'm not, but it's such a hassle to change the settings every time I want to step away from my computer and every time I come back. So, it stays at 1 user at 3.5kb/s.
But, if there was a better program that would, say, allow 10 users at a time and my max bandwidth whenever the screensaver came on, and then brought it back down to 1 user for 3.5kb/s (Morpheus allows resuming, so the cancelled downloads could connect again later) when the screensaver went off (i.e. I'm using the computer), then I would have no problem letting people download to their heart's content.
So I guess what I'm trying to say is, sometimes contributing is beyond the convenient means of some people. Mayhaps a system (in general, not necessarily a warez system) where contributing was easier, or maybe even somewhat, if not mostly, transparent, would see fewer freeloaders and more people contributing.
No sig for you.
No. Punishment is a purely selfish strategy: spend money punishing someone, so they will invest more, so your profits increase. All this shows is that the people playing the game were able to come up with vaguely intelligent long term (selfish) strategies.
If they wanted to prove that people will 'pay to punish', they should have setup the system where the cost of punishing someone was so high that overall profits decreased - and seen how long people kept on punishing.
So next time I get moderation priveledges, I'm going to mod down people who haven't posted anything :)
I've been wrestling with the article's issue, on a game-theoretic level, for years. For example, many people simply do not understand what I say when I discuss the events and aftermath of
What Happened To The Censorware Project (censorware.org)
It's far deeper than ego or "personal", which are superficial reactions I get. In game-theory, the Prisoner's Dilemma teaches us that that individuals have an incentive to defect in terms of cooperative resources. Now, having said that, what then? What follows? How does one go about organizing a cooperative venture with this knowledge in mind?
To quote the article:
This is the exact argument I made passionately regarding the necessity of making there be some penalty for Michael Sims' actions in destroying censorware.org. It's the flip side of enlightened self-interest. Cooperation cannot be supported if someone can defect without penalty. But:
Indeed. It's not costless to create downsides. This makes it tempting to ignore their role in maintaining cooperation. They're unpleasant, to say the least.But what if it's nigh-impossible to have a penalty? This is an aspect where I think about "the power of journalism". As a programmer who has worked with journalists (many times unhappily), I'm acutely aware that as a general rule, journalists can harm me with manipulated coverage, much more than I can punish them via semi-futile protests about their actions. This is in fact my number-one publicity worry about anti-censorware work and how I'd ever get covered nowadays in Slashdot if I ever were to be sued like Dmitry Sklyarov.
So in the end, I don't have a solution. But the implications of this problem are NOT abstract, in fact are very immediate.
A lighthouse is for the common good, but can't exist without being charged for. However, due to its nature (it just emits light), you cannot deny service to those who don't pay - they'll see light regardless of whether they've contributed.
The dilemma is - as a ship owner, you have no incentive to pay for upkeep as the service is delivered to you anyway. This works right up until the moment the lighthouse has to close, at which point it becomes in your best interest to ensure everyone pays. Note that - everyone, not just you. If only you pay, you're still at a disadvantage.
Can't remember the exact terminology they used - I think it's a form of 'free good', but I'm prepared to be corrected on that. Why these researchers felt the need to reprove a very old and established theory is beyond me.
Cheers,
Ian
PS: 'A'-Levels - the exams in the UK taken when you're about 18.
http://www-personal.umich.edu/~axe/research/Evolvi ng.pdf
Yes, well, they play on somewhat of a different level from most of us. Lask I looked they had bandwidth in excess of 150mb/sec. Most broadband consumers have an upstream between 128-512k. Just a small difference.
Also there is an interesting phenomena, the number of users a given amount of bandwidth can support exhibits better than linear scaling. For example, a 56k modem is pretty much only practical for a single user. Put a second person on, and it gets really slow (even slower than normal). However when you get faster lines, you can get more than equivalant number of 56k users. For example a full DS-3 line is roughly equivilant to 800 56k modems. However you can easily have 1,500 or more people using one, with no ill effects. Not only that, even with 1,500 people, each person will still find it much faster than being on 56k. Why? Simple, because not everyone is using it at the same time. Even if all 1,500 people are present and logged in, they aren't all always making requests of the line.
To give you a real world example, I work at the University of Arizona in Tucson. We have two internet connections, totaling about 115mb/s of commited rate. However we somewhere in the realm of 20,000 computers on campus (I don't have exact numbers). Doing the math you find that the bandwidth is woth only around 2,000 56k modems, yet internet access on campus is still nice and fast, even during busy times.
Along these lines, the more bandwidth you have, the less it matters to loose a percentage to P2P filesharing. For example if you have a 10mb connection, and half of it is being used to outgoing file traffic, I highly doubt you'll notice. 5mb is still plenty to do whatever you want very quickly. However if you have a 512k DSL line and half of that is being used for outgoing traffic, you are going to find the reduced bandwidth fairly noticable.
Interesting that this post should come up at the same time that "Beautiful Mind" is in the theaters
The finding of the Swiss Economists is close to the very premise of pure democracy and why forms of it have by-and-large overcome monarchic states. Combined with the assumption that game theory and John Nash's work is based on(see Beautiful Mind -- or better, read his research) "that equilibrium can be predicted when you take into account that each player acts in his/her own self interest", you have good theoretical evidence supporting the findings of this research.[Actually both Game-theory and Nash tend to start with the presumption that people will act in their own self-interest first and foremost]
In order for the majority to have the power to punish freeloaders, they must first have power to begin with. With majority vote and regular turnover, the opportunity to enact this is provided for. If everyone acts in their own self interest and they have the power to vote, then freeloaders MUST be punished.
If the majority are freeloaders, then those that contribute least will be punished. (Napster is shut down, but everyone who knows how to contribute still has access by some means). If this "freeloader" society is self-sufficient, it will eventually turn itself around if it is interested in self-survival. In the case of government, democracies turn themselves around because the cost of non-cooperation is death. Napster and p2p are bad examples becase the cost of community-death is not as dire as individual-death.
The summary of this rant: community works if either 1) the act of cooperation is equivalent to the act of acting in the majority's self-interest and/or 2) acting in the majority's self-interest does not lead to the destruction of the community. True democracy allows for consistent societal change in both of these directions.
Punishment is a purely selfish strategy: spend money punishing someone, so they will invest more, so your profits increase.
Incorrect - In the actual experiment, if you chose to punish a `freeloader' then you paid out of your own profits, and no one else's. The games were not iterated (played repeatedly with the same cast of players), so any consequent change in the freeloader's behavior would not be to your benefit. Perhaps on the next time around, the freeloader would have a change of heart, but even if s/he did this was not likely to be to your own benefit.
Thus in the context of the game, choosing to punish was a very counter-selfish act - not selfish at all, but quite the opposite. That's what makes the research so interesting.
-Renard
(including the words "Five words")
./configure --with-vxworks; make ) were not compelled to release their changes by the XFree license. Result - a less than stellar server, that locks me into a buggy and feature-lacking OS (Don't say it - as soon as I have the manpower my project will be converted to Linux.)
I didn't see anybody at +3 making the analogy to the GPL vs. the BSD-like licenses.
In a sense, the GPL "punishes" freeloaders by denying them resources - "If you don't share with us, then we won't let you have a share of the pot." If you won't contribute to the shared codebase, you cannot take from the shared codebase.
Compare and contrast that to the BSD-like licenses that don't have the "Release the source" requirement - a freeloader (certainly Microsoft, possibly Transgaming, possibly Lindows) can take from the public pool, not give back, and incur no "punishment".
I used to think that RMS was a crazy, extremist bastard. Then something happened to cause me to revisit that thinking. I work professionally with a product called RtX, which is an X Windowing System server for the embedded operating system VxWorks. RtX is derived from XFree86. I've had several problems with RtX - it won't recognize certain graphics chips, it doesn't support font server use, it won't do anything but 256 pseudocolor, I cannot easily add key bindings or LEDs to the keyboard routines, and (most importantly) it won't work under the newer versions of VxWorks. None of these would be insurmountable problems if I had the source, but the folks that did the conversion of XFree into RtX (and it isn't a trivial conversion, not just
I know I just enraged the "GPL is tyranny, BSD is freedom" crowd. But please, think about this for a moment. If you wish to continue to use the BSD license for your code, wonderful. However, any code I do off-hours will be released under the GPL, for the reasons stated above.
www.eFax.com are spammers
Just as I'll claim if you try to use traditional economic arguments to justify "ownership" of software, or whatever, analogies between physical property (money) and property that can be duplicated (software, information) just don't hold up. The fact that you can share software or information with a friend without losing it yourself makes a HUGE difference in any kind of economic game. (There is some cost, for instance bandwidth in a peer-to-peer system, but I think it is mostly negligible.)
However, I would expect that this result does in fact hold for IP-less software economies as well. I am just saying that making direct comparisons is always trouble.
There is another method for ensuring cooperation and fair behavior in peer networks. And it works the same was as the method described.
It is called social discovery, and it works by having each and every peer create a view of the network that suits their interests and needs. In such an environment, the freeloading peers will not be viewed as valuable peers and will be dropped from your peer group(s); no longer used, and no longer using your resources.
On the flip side, there is a strong incentive to become a better, more reliable peer yourself, as the quality of peers you can associate with is directly related to how they perceive *your* quality to them.
If you want to be able to tap better, higher quality peers, then you should keep your node available longer, more often, and also share more resources (whatever they may be).
The project I am working on that implements this social discovery mechanism is called the ALPINE Network and there is also another social discovery based project called NeuroGrid.
I am biased towards this kind of approach, but I think it provides the best long term solution to resource discovery / searching in large peer networks.
This is why copyright is supposed to be limited to a short period. If if still were, the system would work as intended. Much the same with patents (which originally had a longer term, because of they're supposed to represent a physical device, with the additional associated costs). The flaw is not in the basic concept, but in it's perversion by our government.
There's a long article (which unfortunately is not available online) discussing experiments like these and coming to the conclusion that the vast majority of people value "fairness" over material success in this particular case. I was particularly fascinated by the experiment known as the Ultimatium Game; the article says that only 4% of people, IIRC, choose what the mathematically most beneficial solution. (In other words, in 96% of cases people would choose the "fair" outcome over one that was objectively better for BOTH participants.) Worth checking out if you are interested in this kind of thing.
"Biped! Good cranial development. Evidently considerable human ancestry."
Art. 13 Right to Privacy
e ga ff/swilaw/fconst.html
1 All persons have the right to receive respect for their private and family life, home, and secrecy
of the mails and telecommunications.
2 All persons have the right to be protected against the abuse of personal data.
http://www.eda.admin.ch/washington_emb/e/home/l
-s
It's not available online, but the January 2002 issue of Scientific American has a very relevant article titled "The Economics of Fair Play". It discusses the nonrational dynamics of how groups of human expect and enforce fairness. Definitely worth a read for open-source economic theorists and fans of intriguing behavioral-psych experiments.
When all you have is a hammer, everything looks like a skull.
Let's see... freeloaders use up the service without contributing. And you think it's a bad thing to get rid of them because if you do, then "there goes your userbase." But if you kick these people off you GAIN: more bandwidth, more room for people that contribute to the service. On top of that, a certain portion of potential freeloaders will be more likely to contribute if they find out they are in danger of being kicked off the service, thereby increasing the value of your service even more.
As long as it is made very easy to contribute to a service, you should not lose any meaningful users.
"And like that
I'll use a Napster-like system as an example, but when I refer to "songs", you can easily substitute "movies," or "naked pictures of Natalie Portman," or just "files."
1. It costs one Point to download a song from another user.
2. Users have unlimited Points for a certain trial period (some people will try to re-register every day to get around this, but that problem may or may not significant enough to affect the service).
3. Users get a certain (small) number of Points each day.
4. Each time a song is downloaded from a user, that user gets two Points. This will be the primary means of gaining Points.
5. Note that the person the song was downloaded from received two Points for the transfer, but the person who downloaded it only paid one Point. This means that the total number of Points in the universe will increase by one for each song that's transferred. This if fine -- it keeps the system from being too strict. You can take up to twice what you give, which should be generous enough for most people's tastes.
6. People who have a lot of songs to share will have many more Points than they could possibly spend. This is fine. If you're even moderately generous, you shouldn't have to worry too much about what you take.
7. Most people who generously offer the songs they have will wind up with more than enough Points. Those who DON'T offer what they have will find themselves frequently running short, and will be encouraged to start offering what they have.
8. To further motivate people to accumulate a lot of unused Points, have a "Hall of Fame" listing top Point-holders, top new Point-holders, fastest-rising Point-holders, etc. People love stats; witness the people who'll install the D.net or SETI client on 5000 computers primarily to increase their rank in the stats.
9. For further motivation, offer additional prizes for accumulating Points. Maybe a person who reaches 100,000 Points gets a T-shirt, or a person can exchange 10,000 Points for a coffee mug.
10. Who pays for the T-shirts, and the service? Users with low bandwidth who otherwise would have a hard time earning Points can earn them by an alternative method of contributing to the service: financially. Whether you make songs available to users of the service, or help the service meet its financial needs, you have to contribute to the servicein SOME way to get a significant share of songs from the service.
11. Another way to encourage people to earn large numbers of Points would be to give preferential download treatment to higher-ranked users. For example, if the person hosting a song has configured his client to only allow other users to only download 200kbit/sec from his machine, and five users try to download from him at once, the 100,000-Point user might get to download 100kbit/sec from him, the three 10,000-Point users might get 33kbit/sec each, and the 2-Point user might be forced to download from a slower host.
12. This'll not only encourage users to offer more of their songs more generously so that they can download from faster hosts than those who don't, it'll also ensure that people with slow connections will get some people downloading from them (and thus the people with the slow connections will get to earn some points too), rather than every single user swamping the fastest hosts, bogging them down until they're slower than the slowest hosts.
Ideas? Suggestions? Flaws? Discuss.