 Okay, then let's get started. This is Kai Hamacher. He was yesterday already talking about Bitcoin. Today he's talking about resilience towards leaking and why Julian Assange might be wrong. So let's not lose time and get started real quick. One round of applause for Kai, please. Yeah, thank you and good morning from me too. So this talk will basically be concerned with the WikiLeaks philosophy, I would like to call it and kind of an historical account and some work and this is ongoing as you see there is a version on it. I uploaded a paper which basically describes the research questions that I think need to be addressed somewhat to evaluate or validate, falsify, whatever, some of the theories that Julian Assange wrote some years ago actually. In the pentabuff, when you click on it, you have to rename the file somehow, the PDF extension was lost and I already got complaints about that. Well, anyway. Okay, first let me start with a disclaimer. So this talk will not be about whistleblowing per se. Okay, so whistleblowing as a personal decision of someone who might be an ethical dilemma or something like that. So it's absolutely and only about the WikiLeaks philosophy. It's therefore also not about any particular leaks. I mean, there were a lot of them in the press and all the resulting stuff we saw from boycotts and then the reaction of some pseudo-anonymous people that somehow feel that they are entitled to do whatever. It will not be about personal problems and between the main actors, which actually prompted me to ask where's this non-emotional responding dude explanation here that got lost, I think, between the major actors. It's also not about legal analysis of what's going on and, you know, between all the different states and anyway. And it's also not about all the stuff that has happened or might have happened or hasn't happened in some European country involving romance and all that stuff. So let me start with an intro. And I'm more or less giving now an historical account as I perceive it. And this is probably the only or these are the only texts or written evidence of what was driving the establishment of WikiLeaks. This is what I learn from what happened. It might be different and it might have evolved to something completely different these days. But in start, at the start, so in 2006, 2007, I think these two texts with where the later one is basically just the version two of this thing and you can get them from Krypton is basically like the founding document. And I think in some block it was already called the Anarchist Manifesto 2.0, these combined texts that are the base of what I would think was the general idea, so to speak, behind WikiLeaks or the establishment of WikiLeaks. Whatever it morphed into. So there are some key passages, I think. So this is now a quote or these are all quotes and the one says we must understand the key generative structure of bad governance and it's always about bad governance and you know people doing something wrong and the generative structure is then identified by or in that quote we will use connected graphs as a way to apply our special reasoning abilities to political relationships. So there are some key terms, it's a generative structure, it's a connected graph and probably political relationships. And that is a graph and you have some people here, there's some leader, there might be a right hand, they interact closely and there's some lobbyists who talks to a politician who ultimately has access to all our money and probably that's all about that this guy wants to get hold of and then there are some people just not really involved right, just working there or whatever, so and that is then a connected graph as described by Julian in these papers and well, yeah, so people talk to each other, communicate and it's really a communication theory that is established in these papers and it goes on. What does a conspiracy compute? It computes the next action of the conspiracy. So the understanding is that there are people interacting, communicating and they decide what to do next to achieve whatever goal and that is not specifies what the goal actually is and now comes I think one of the key passages, it's rather long but sorry, so the most secretive and unjust, whatever unjust here is, an organization, the mall leaks induce fear and paranoia in its leadership and planning coterie. This must result in minimization of efficient internal communication mechanism and consequent system-wide cognitive decline resulting in decreased ability to hold on to power on so later and so on, sorry. Okay, so meaning so if I somewhat, you know, provide for leaking or for non-quality communication so to speak, so reducing the reliability of the communication especially when it comes to secrecy then this induces fear, right? There is a tax as it's written in the text so you increase the cost. It's not as that efficient anymore, that conspiracy and at the same time there is also included the notion that some other organizations which are just might also be hidden hit by some damage but that is acceptable because that damage to that organization is not that large or that bad as to the unjust system or graph or network So there is some notion of collateral damage here also involved, right? You can really hurt someone or an organization but it's not that bad because that guy will survive because he's not unjust. Okay, if you look into these these these underlying mechanisms that is a showcase of biologism and what does that because of what well there is this notion of decreased ability which would be called in biology a fitness, right? Because this this whole picture is that over time this organization evolves from something to something because of this next thing, next in action thinking here it has some survival value, right? A fitness if it's not that good in surviving and computing its next action then well it will cease to exist and that is actually what what what what is really biological thinking here and its fitness and then this next action that is now for the mathematical orientated people that's actually kind of a Markov chain here so meaning that the status in the new so tomorrow is just depending on the state today and not on some ancient history so it's really something that is going on and going on and going on and has to prove its value from day to day to day the problem is that with all of this biologism is most likely it's not true because it's most likely a falsified or well a shortened or well the two simplified model to just say oh we have a fitness and then the fitter guy will survive and the unfit guy will not survive and that's it it's basically well it's it's not that bad, but it's kind of an of a comic picture of biological complexity that you see here that might also be put forward by some creationists, right? They always misunderstand what's going on and that might be true here just to again emphasize what the picture is so you have a conspiracy black, right? You have some normal people yellow and then you have a leak that I probably cannot see it. I try to symbolize that with a decrease in the coloring here so with every leak the connections within the conspiracy get weaker and with every leak therefore you get basically transparency, but at the same time you decrease the fitness or the value of that network so now using this and really taking this for granted and you know evaluating it and saying well probably it's true. What would be the effects? So and now comes a thing that might be hard for somebody here or for some people here, but you cannot avoid it those fans of the Alternative Flows podcast might have already heard about that approach although not about the details but that is modeling complex systems with ordinary differential equations and while this is numerics and this is really about numbers don't take the values themselves for for truth it's only about tendency so something goes up something goes down from now on So and what is a leak? Well a leak might be some kind of a function, right? There's an information transmission so all the time nothing happens and then suddenly there is a document or whatever appearing and that might be the time course of a leak right Okay, so what happens now? There is a conspiracy. There is a number of people involved in that conspiracy. So journey of science is always talking writing about a conspiracy It might be an unjust system whatever just or unjust is so now take x of t the size of that conspiracy or it's you know Well, you growth whatever at this particular time and then there is a growth, right? So whenever it's it's commanding some resources it can grow to even acquire more say of the wealth of the society that is immersed in Or it's just the gross is zero then it's a static conspiracy, but that would be included So what is now the dynamics? Well, this is now the ordinary differential equation And that is this is an equation that describes the change of the size of the conspiracy With respect to time. So per time unit, whether it's a week or a day or whatever There is a there's a change that is just the variable parameter r and that describes whether the conspiracy is growing declining or just static over time And then we have to include somewhat Somehow the the leak right and that is reducing the ability in julien assence Thinking right so it must be negative here. There's a negative sign And well, it must be somewhat proportional to the current size of the conspiracy, right because The more the leak induces fear that is obviously somewhat related to the size. So the number of people participating in the conspiracy Okay, and these are You cannot see that the plot is Okay, so these this is time and this is the size of the conspiracy and there's a so red and Black are just two different starting sizes. So two different conspiracies and you just see without leaks They are just growing linearly, right? It's just getting bigger and bigger and bigger And what you probably can only see in the front rows is here is basically the leak happening And I will just go with the pointer where the line is so it's decreasing But then when the leak vanishes and is over, it's just increasing more and more and more and so on So basically what we see here is these lines here just go parallel to those lines Meaning that that leak in this very simple picture is just kind of a delay But it's not really, you know, shutting the conspiracy down reducing tremendously the number of people participating in it And it's even getting worse When you say that the accumulation of wealth or of influence of capability to conspire Grows right because then we have actually the gross is then not only just some number But it's then proportional to the current size. So it's growing and growing faster Which would be called exponential gross in biology And what you see here is well, these are the two different conspiracies They just grow over time and then here's the leak and you probably cannot even distinguish the lines So in this very simple picture of biology or of a community The leak is more or less completely irrelevant Okay The problem here is that this is a too simple picture because there is something probably called feedback A leak might not just induce fear But think about that the more you hear about a perceived Group of people who can influence all the politicians and the law The more likely you are to overestimate the size So that might induce a feedback in society on the perceive On perceiving what's going on right at the same time a A group of people who's frequently appearing in the media even in movies might appear very a big big threat And you conform to their perceived values although they are just rather small And probably in this community the Illuminati Thing is probably something that goes in that direction, right? So it's perceived to be but eventually it's probably not that big even vanished Okay, so to illustrate this feedback I just saw that cartoon recently To make the things short. So the girl is telling the guy how I love this new technology. This is so great I would like to experience more technology in the future. So I go And get frozen for 30 years Then I will wake up in 30 years and the technology will just you know be much much much better Turns out She wakes up 30 years later and there is no new technology because all the engineers thought the same and were also frozen it so And this feedback right because everybody is perceiving something or believing something and eventually that leads to that What you perceive or predict predict doesn't appear at all, right? That is a feedback loop somewhat In a cartoon picture like Okay, what would be the feedback in in in the system where leaking is present or where people conspire? well First of all this conspiracy can only be a tiny fraction. This is the framework i'm not talking in so the conspiracy Could only be a tiny fraction of the whole society because if it's large enough, right? At least in the democratic state and i'm leaving out all the funny tyrannies and Tatorships in the world because they are out of question anyway, but just referring to a democracy whenever the group is large enough Then it doesn't pay to be a conspiracy, right? Found your own party Sorry There was a question Well, I mean the yeah in in mathematical. Well that depends on your perception, right? Is it 51 percent? Then you are right then you have the majority, but probably it's nine percent already there I heard that especially in berlin some conspiracy so to speak um Got a lot of votes so that it obviously pays more to work officially than behind the curtains, right? okay So tiny fraction meaning one out of ten thousand or something probably right that's an estimate But then you cannot really rely on the society to enforce your rules, right? Because if you are really working against or parallel or orthogonal to the society Then you cannot go to the judge and say hey this guy in our conspiracy didn't confirm to our internal code because you are secret That means you have a huge coordination problem. You have to enforce the rules and You know adherence to the internal code by some other means than the official ones and this coordination problem is even worse because from again biology so in that neuro biology and and human biology we know that people roughly can you know Holds or Keep intact keep working relations to roughly 150 people That limits the size of the conspiracy in itself, right? Because either you are a hierarchy, but then it's hard to attack Then it's easy to attack that as a hierarchy in the conspiracy But when it's more loosely connected then this number already You know sets an upper limit to the number of people participating because then When you go over that number you don't have a social relationship and then you definitely need to have some law enforcement by whatever means And that means that is that alone the resource constrictions. So the The cognitive restrictions the coordination problem the problem that We are not talking about conspiracy anymore when it's 51 of society is already a feedback Although not a a complete feedback in an active sense, but rather an implicit feedback So that is actually well known in biology That is called caring capacity. So how many individuals can survive in an environment? This is the same thing here, right? How many People participating in the conspiracy can survive in society When There is not a number k of people who could participate. There is again our growth rate So and this term here is just then the feedback So the something this this term in parent thesis That is the feedback saying oh, okay. This is now growing to too large too fast It's somewhat declining because it doesn't work out. You are all buff What you could coordinate Okay, what does it lead to well Our two different conspiracies just converge To the same size one from below because it started smaller and one from above because it started bigger And it's always going to the same You know value of the same size. That is what's happening. This is already feedback and Well needs to be taken into account Okay, now all leaking leaking again should reduce the growth, right? So it's negative here I just use the same example. So some leak happens Okay, what happens then? There is another delay here even it drops below what the the final value would be And then it's growing again Well, it doesn't really matter again, right as in the even more simple picture. So this feedback Is stable against whatever perturbations And this is a perturbation Okay, now you could use these models to probably Develop a leaking strategy because if you have an amount of information say a number of documents However, you measure the amount of information Is it good to publish it in small pieces or in one large thing right to to hit with a big hammer or With the several small hammers. What is better? Well, so This is the sum of all the leaks that you do over the full time period is just a constant Right and it could be something like that. So you publish two-thirds somewhere in between and one-third at the end Or you publish frequently periodically all the time Always a small amount. Does it really change anything? Well, so the question is if I have two different leaking strategies with the same documents was the same Uh, some or the same overall amount of information that I can leak. Does it really change anything? Well, the button line is no There is a distinction for very small numbers of leaks between odd and even but in the end it's also well Kind of stable against that and it doesn't really depend on how you start your conspiracy at what size This is now again the same picture but for much much larger systems and some thousands and 10,000 systems There is no correlation between whatever you choose for your leaking strategy and the overall effect. So These are correlation measures that give you whether there is a connection between how you leak and what the final effect on the conspiracy might be It doesn't really depend on the size. It doesn't really depend on The time points when you leak. It's just random noise, right? So The system when it has this feedback is stable against leaking at that point at least so For a short summary in between So we had the starting idea that there are conspiracies That are graphs, but he doesn't really address so Julian doesn't address the fact whether it's a static graph or an evolving graph It's definitely time-dependent. So the static picture couldn't be true He claims that's a selection, right? So the fitness value the survival value of conspiracy of just systems is better Because they don't rely that heavily at least on secrecy So they are not hit that hard by a leak than an unjust system who needs to For this coordination problem, for example, who needs to rely on secrecy And then the damage to to a just system is kind of a collateral damage And then I have to ask the question, but who is The person to decide what is acceptable and not, right? He doesn't talk about that And then there are the resource limits that are described, right from cognitive staff from money probably whatever There are bonds and that provide for a carrying capacity We learned already that Choosing different strategies in this model and this very simple model does have no effect whatsoever You can publish immediately or frequently small amounts. It doesn't really change In the average Yeah, and and the effect up to now is just, you know, kind of a Temporary delay, but it's not really giving big Or has no big impact But the same time the simulations or numerical solutions Are kind of a framework to evaluate what the security of the system or its Stability and robustness is Okay, now going beyond that simple picture There's even more feedback, right? So What people always get Frequently up to always get wrong is the picture of evolution Uh, and that's why genetic algorithms don't work. Um, because Evolution is not about a static fitness function, but this whole picture up to now Claimed more or less that the fitness function is kind of stable over time But there is co-evolution in nature and probably there's also co-evolution here, right? Because a conspiracy might learn Where is the the whistleblower? So they get rid of him. That is already co-evolution. There is a feedback in that sense, there are fluctuations um, we we investigated that in some other system already there is a So a process called fixation that is closely related to the fluctuations It might happen just by chance actually, but it might happen that a small fluctuation Leads to a so-called fixation event where one population takes over the other So the conspiracy gets really hold of society although it has in the average a small fitness value But just because out of you know pure luck it could happen And then there is probably and we see that I learned these days Counter-strategy when there is leaking you could probably exploit leaking to get rid of another conspiracy that is, you know, working for get Wanting the same resources that you want or you could even exploit it to get rid of some Person that is undesirable and these days I learned I read some some story that the DARPA is now funding some research how to do leaking correctly so to speak To well as a kind of a counter-strategy And probably that is what we already See these days Anyway, okay, so and then there's another thing The network size itself is probably not the only feedback value because if you are about To join a conspiracy or not Then it's not the number of people that is attractive, but it's rather the value that you get from that Conspiracy or the value that you lose when you do not join the conspiracy, right? so you must somehow Transfers the the the graph the conspiracy into a value for every participant or would be participant Okay, so in the following I will More or less model all these effects, but not this one because that is much more complicated Okay, because feedback and co-evolution is one of the most important concept in theoretical biology And that is because say a prey Is just running faster or well the children of that prey learn to run faster by some evolutionary events of A muscle protein in the muscle, so it's running faster, but that changes The fitness of the predata, right? So the the antelope is running faster So the children of the lion have to learn to run faster too But if they run faster too, then it pays for the next so for the grandchildren to run even more faster, right? So there is the feedback and in time you see this going up and down of fitness And the fitness function is a function of time and not a static Objective function as people frequently do in with genetic algorithm optimization Okay, what would be the effect in our modeling? So again, we have the change of now the conspiracy I there are several now because for co-evolution you need at least two, right? There is again our gross then we have our carrying capacity that is this for example, this Dunbar number of 150 That you can only coordinate and then here now is connection of one Conspiracy to the one we are modeling here. So if they talk to each other, they could Collaborate they could compete or they're probably neutral, right? Whatever And you just have to sum up all the influences and that is that is a model that is good to get into Co-evolution and these are now coupled Differential equations and the gross rate Is actually a function of the value of all the conspiracies, right? Is it valuable for me to join that conspiracy? Okay, it's gets even more worse because we need this value and there are several ideas how to value or evaluate Quantify the value of a network There is something that goes like a polynomial in the size So n is the number of participants in the network So the value is then roughly n square, right? So it's getting larger and larger The larger the network is so the conspiracy here Then there is something that tries to assign value to the to each individual connection And that grows then even more. So that is now exponential With some additional terms in the number of participants in the conspiracy And I will not talk about this much more involved model because there are so many free parameters That are just not known But basically this guy is or this is a special case of this guy anyway Okay, so it's getting more complicated, right? Just to tell you that it's really bad here because now we include the leaking and what how might leaking effect The time evolution of our conspiracies now Well, it might reduce a gross rate, right? So the economic value Say Some leak about a product you don't get that patent because it's already leaked So you cannot grow as a conspiring company say in In the music industry or whatever and then there is another kind of leak that somehow affects the The interaction between all the different conspiracies and the current one that I'm modeling, right? That is a second term. So these are the two places how leaks Might affect the time evolution of the size Change of these conspiracy And well one leak might actually lead to two to the two terms at the same time It's just different places where leaks might induce effects And that also is here the counter strategy, right? So every counter strategy That is exploited by xk. So the kth conspiracy in society to reduce The effectiveness of conspiracy i is then included here too. Okay Even more impressive But i'm this actually a reduction I'm not talking only about two conspiracies because it's the smallest amount that I can have And that compete with each other that leak Disfavoring each other that are immersed in a society, right? So there is the size X of conspiracy one the size y of conspiracy two and then there is the overall society z but that is fixed because Everybody adds up to say 80 million in germany and that means that this actually three differential equations are just two because by that Conservation property. Okay. So we now have to integrate these two differential equations But we have so many unknown parameters So and I do not know what the parameters are and I told you it's although it's Supposed to be quantitative the only thing that you can learn from these models is if you change this up Does this quantity go down? Does it go up or is it, you know, not a function of it? That's everything I can do so Therefore the only approach to learn something is to just take every parameter possible and look what's happening When I do not know anything about the system, you know, then I have to take Everything into account that might happen. So now what I'm doing is I sample all the parameters in this rather complicated formula Especially about the betas which describe the interaction either competition or a collaboration and the different ours and well the different The the amount or the the impact leak has And I'm distinguishing between this reeds law and the Metcalif law on how to evaluate or value a network But these are many parameters now. This is basically a seven dimensional space So you have seven dimensions as you can easily depict, right? And visualize and that is a problem because what I want to show you is that there is a Distinguished feature a signal, but I cannot show you that in seven dimensions I'm very sorry, but I tried hard But there is a mathematical thing to to do at least an approximation to that And that is called pca principle component analysis and you don't have to understand the details But the overall idea is you have something in a high dimensional space say three dimension And by that method you can map it into A small dimensional space. So we would have our seven dimensional space and we map it into two dimensions That's a basic idea I can you can look that method up in Wikipedia and it's probably It's implemented in basically every numerical package Mathematica octave are whatever so This is our seven dimensional space and we reduce it now just to two dimensions for visualization purposes And the idea is to rotate this huge body That you look at the surface that is the most important one and what is important That is figured out by that method Okay, if I simulate now the two conspiracies in that big society What's happening? well that so the distinction here between the black and the red dots is After this mapping into two dimensions of this huge data set right there were four millions four million simulations for each plot here Is the black dots are the one where one conspiracy got extinct. So it vanished the other was much more Fit to survive and acquired more resources And the other was extinct in the sense that it number of participants went down to zero There is no conspiracy more anymore, but there is another one, right? It's only one left And the red dots are the ones that are no extinct extinction So there are two conspiracies life and kick in here And you see that these two things are at least particularly separated So in this there this tells you that in either method how to evaluate The network value You get a differentiation So there are parameters. There are strategies of leaking or impacts of leaking That provide for an extinction and some other that don't provide for extinction That is a message of that slide. So you can choose somehow a scenario Where you get rid of at least one conspiracy Okay The problem is already here in that numbers The the relative growth of the surviving If if if it's only one surviving conspiracy Is a factor of 12 or factor 14 Indicates that probably the one surviving conspiracy Is the one surviving. Hey good news for those guys, but it's even better They were just much much bigger as they would be if the other conspiracy would also survive and compete for the same resources And now you can use the pca thing. I would not go into details, but you can get really What are the important parameters and you could see that here? These are the parameters And this is the the importance of that parameter for the distinction between the black and the red Group of systems of scenarios and whenever in absolute value a number here is large that tells you that that parameter was Important to distinguish and separate these two scenarios or the two results of extinction and no extinction And it depends slightly on what method you choose to evaluate the the value of a network But it's always the interaction between The two conspiracies So whenever two conspiracies Collaborate they do not got they will not got extinct. Okay, not that surprising, but it's the most important effect For extinction or no extinction So whenever they get into kind of collaboration, there's combined survival fitness is much higher. That is the message here Here that is the interaction between one conspiracy and the society here the picture is not really clear It depends really on which model you choose Here there is no value at all And here it might be a small impact And that would for example model the social norms is whistleblowing accepted. Is it, you know punished or Actually put people whistleblowers to be heroes. That is kind of this value here depends What doesn't have an effect at all is leaking Um In the sense that the the bigger the conspiracy the more leaks there are which you would assume there are more documents There's more information that could be leaked. There are more potential whistleblowers But this is not really, you know, the important thing what is important. However To distinguish extinction and non extinction is the second parameter And that is The more powerful one of the conspiracies is The more it could use blackmailing and counter-strategy. So using leaking as a strategy to hurt the other conspiracy And that is important to get rid of the other guy And it's actually the same of the same model of magnitude of the collaboration or competitor Come competition between the two right So this is this important This guy is probably not important. What's not important at all is really the starting size of the conspiracies So they will just grow to whatever carrying capacity Okay, bottom line the interaction between the conspiracies is the most important effect Society is not that important The leaking is only important when they do it either in collaboration or as a competition But that's basically it Okay I already told you that the average gross is much higher and you see that here So again, black is where you have extinction of one conspiracy And red is where you have no extinction with the two models of network evaluation And this is in our histogram. So this is a probability How frequently you find a relative size increase in the two scenarios And you see that while the relative size increase is rather small when the two conspiracies survive Small But when one conspiracy got extinct, then we already know from the previous slide that the average size increases 12 or something And yeah, and this is actually the logarithm here So the logarithm of the relative growth and it can get really really big 10 to the fifth So really big up to becoming some party or whatever in society So you pay a price You might now think that it's good to have conspiracies that somehow compete and leak and do whatever But then the one surviving is much bigger and your problem might be much worse, right? Because you got rid of that smaller conspiracy or that not that fit conspiracy But then the surviving one is a much much bigger threat because it's bigger and therefore it has more value more resources more blah blah blah Okay, that brings me to the interpretation Um, so for extinction We we learned For the extinction of one conspiracy that the competition between them is most important one of the most important effects Are one of the most important feedbacks Leaking could be a strategic weapon for one of them to get rid of the other because they ultimately Compete for the same resources, right money and when we when I get rid of my competitor There might be more money for me What doesn't really have an effect is In in that framework at least the social norms does society, you know Emphasize and and encourage motivate and value the importance of whistleblowers The psychological effect. So if I perceive conspiracy as to be very powerful Has no effect because or not that big an effect I do not necessarily want to join that The size of the conspiracy. So the good news is if you today conspire to do to achieve something Just start with two people in the end. You might grow to something The drawbacks of this whole leaking scenario, however, is that you might get rid of one conspiracy That's fine But you will then pay the price that you're facing a much larger opponent because that guy grew On the resources that were freed by the extinction of the other group Okay, um, how much time is left? Oh, okay, because This was now really that that you have more or less a static society was, you know, fixed norms What's not include of fixed laws. What's not included is what happens if the rules of the game change So if you change the leaking scenario or the leaking capabilities of people what changes when You know, the the the legal norms in particular change how people could collaborate or Compete and that brings me to one of my Beloved subjects that is intellectual property because that is somewhat related also to that stuff because it has a huge impact on leaking for several reasons And the importance is probably best described in that small quote By margetti. That's actually the guy from this picture company The most of the pictures these days come from intellectual property is the oil of the 21st century so everybody is Will at least fight over it and that is the key productive resources or resource in the end And that that is that the the current trend is not that good. We could have already known In 1949 I cannot emphasize that enough one guy well, basically the The saint of what you probably would call neoliberalism, but what I would call market economy So the hero so to speak already wrote That when it comes to copyright and patents blah blah blah a slavish application of the concept of property And that drastic reforms may be required So that was the guy who was proposing market economy to the extreme was already writing in 1949 Against basically the extension of intellectual property and that is Not just an ancient idea, but also a recent thinking as you can learn in all those nice books. I particular recommend the master switch by Tim Wu, which is explicitly Discussing the stuff When it comes to information Societies or information industries And you can learn that this information thing Intellectual property thing is really a big problem So what is the impact of intellectual property in in our scenario here in our framework? So again, I have the growth of our conspiracy and now Think about a conspiracy In quotation marks as say the music industry and the film industry and the blah blah blah industry Right, and then you might leak something that they do Most of you are aware of actor and other other things that they do whatever So you might reduce the growth either by economic stuff by leaking internal documents. I don't know So what is the effect on those two terms here? You have now competition between different industrial sectors So there might be an an internet company that is not favoring some restriction But there might be the music industry which is favoring that so you have competition here Okay, so first of all several groups if you would like to regard them as conspiracies Share common goals right to introduction of all these well horror things That lead that several that leads to the fact that several of these Collaboration or competition scenarios are actually larger than zero. That is that is we have Collaboration here and if you look closely that means they compete for the same resources But the problem is we come after that To that point that the resources are not that constrained. So the carrying capacity Is is not really existing at that point So the collaboration exists There is positive feedback now And that is a problem. So we don't have any checks and balances in the system anymore The leaking might be even reduced or hindered at all because I can claim Oh, this document is my intellectual property. So you cannot publish and enforce using these things then To shut down the leaking at all. So probably the terms even vanish, right? And then we see that they basically do that in plain sight, right? It's not really a conspiracy and secrecy. I don't think that they are really hurt by any of the leaks because they just do it And they don't give anything about it. So The feedback is switched off There is no feedback in the sense that they try to avoid bad publicity because they control the publicity, right? And then there is something missing in this whole model and that is important for the carrying capacity We see an extension of intellectual property all the time and that means you accumulate value up to now I had only included in the modeling and that is Now basically your research proposal for the future. I have only included The value at present time So I'm going around count the numbers of people and say, oh, everybody is contributing something to my conspiracy And that is a value But now I have a stock of patterns of, you know, intellectual property, whatever And that accumulates over time and it's extended even more and more and more So it's growing and growing and growing and growing in comparison to everything else Which is just probably static or even declining So that means in mathematical terms that we cannot longer model that as a differential equation But then this is an integral differential equation and in particular it will go into the carrying capacity Because if I extend everything every protection, I can just grow and grow and grow Not in the number of people, but in the impact on society So IP is really kind of a meta-strategy to counteract everything Which is a control feedback to, you know, have a competition to have some feedback that reduces the growth to an absolute control of everything Okay To summarize the talk So I started with Julien Assant's starting idea, which is a very simple picture of biology that conspiracies are graphs And they can be attacked, for example, by leaking And the unjust ones are more relying on secrecy. So they are rooted out more easily or More frequently Yeah in the average that might be true The problem is there is feedback. The full dynamics is much more complicated As you have seen from the feedback groups that induce some stabilization There was against in this very simple picture. There was robustness, resilience towards such leaking scenarios If you want to fight conspiracies Then there are costs If you get rid of one conspiracy the other is typically much more Viable and will grow much more. So the old latin approach of divide and Divide at impera Might be, you know, a better approach. You have two conspiracies Yeah, but they are much smaller and the combined effect Is probably not that bad as long as they fight each other and don't find a framework as for example Our common interest like say the IP stuff to share goals and then to collaborate as long as they compete That's actually good Okay, the outlook is that such a framework extended probably and fit to some parameters Is good to evaluate strategies or counter strategies in leaking and The effects of whatever we see in society And it could be Good to develop some counter strategies To cope with the effects on the just systems whatever just from your point of view is Okay, that is basically the summary and the outlook of what in the future I would like to address too And that brings me to my last slide in which on which I wish you a merry crisis and a happy new fear Okay, we have a few minutes left for questions I would ask you to please remain seated until the talk is Completely over so not a disturbing question with answer session If you have questions, please raise your hand and the microphone will come to you Hi, thank you for your talk You started with a motivation or you motivated your talk with this Graph theory and then throughout your talk you you actually considered the conspiracy as a homogeneous group if I followed your arguments, right so that Especially under attack actually it's known that groups tend to be or like graphs Networks will be more or will become more heavy-tailed So for example the the enron data set the email data set if you're aware of that So that under the attack from the legislation the the communication was more scale-free than before So I was wondering if you if you search about what will change in your in your line of arguments Considering scale-free or heavy-tailed degree distributions Thanks a lot for that question. It will change the results But it will not change the criticism I have because that's even a much more involved feedback loop, right? So we have leaking if I Understood you correctly. So you get a denser connection people Get more or rely much more on each other that would be included in that model probably So you have the inhomogeneous connections here and if you make them time dependent as you see it here So you de-value or you devalue those connections that you previously had But you devalue those more that were more in the past because you cannot rely on that people that Strictly so that would be in in such a model to value here the The the network or the sub network of the conspiracy itself the problem is that is then not that's just now pragmatic approach that would include much more free parameters and I wouldn't have a sampling space of seven dimensions as I had in this More simple modeling approach, but I would have I don't know millions because so how many people do you include every guy gets, you know Two parameters for every connection he has and if the if the conspiracy has n people Then I don't know probably there are n square or of order n square of order n Many connections and every connection brings me a new parameter. So I would have I would have end up at least with some thousand parameters So I use the most simple thing to illustrate that there is There are different effects of leaking and there are different feedbacks your feedback is definitely worth In which to be investigated, but it's much more complicated and more time consuming Well, we now have a question from someone who's watching the stream at home Here is a question in from the IRC How good how good is your? Mathematical mathematical model because from here it looks like To talk about the physicists that try to bet on a horse race by using a model with spherical horse in a wagon Sorry, I didn't get the last part of the question Which one the last part of the question or something with horse race, okay, okay To bet on a horse races by using a model with spherical horse in wagon What Spherical at that point. I'm lost. Sorry. I didn't get the question I didn't get the question Well, I think they are just thinking you're trying to approach a very realistic problem with the very abstract abstract model It was a metaphor. Yeah Guys, it's in the middle of the night. It's 12 or something Sorry, okay. Now I got it. Um again Yes and no so the thing is It's not good for saying oh this conspiracy will Grow to the size of 50 or 52 people no 23 people right But rather it's good for if I You know increase say the social norm that whistleblowers will be you know accepted more Frequently than the size of the average conspiracy goes down or something like that and Obviously, there is an abstraction here but there is This is the approach that is well the currently available approach To evaluate feedback basically This is all about the feedback mechanisms and what their effects Uh, what the effects of feedback are on the overall system dynamics You I'm not claiming that this goes, you know, always You know with the time period of two days or whatever I'm just saying in the average this goes up then this might go down or if you change this then this grows Grows in this vanish or whatever. So um It's not perfect But it's better than just arguing on the level of this basic two simple arguments That you find in the so-called founding documents of wiki leaks. Um, so it's A more detailed investigation. It's not the over truth. That's right Well time is over. Um, I'm sorry, but Well, yes, I don't know. Maybe you'll stick around and outside the door and you can ask questions there