 I took a risk with the terminology here. I guess I went a little bit over the top. This presentation is a little bit over the top. Okay, first thing I want to ask is how many people came to my presentation last year? Wow, so I got a repeat audience of about ten. Okay, last year we were over at the Alexis, and this year we were in the Riviera. I want to ask how many people think that this year is better than last year? Really, how many people think last year was better? Wow, I'm surprised. Last year we were out in the tent. I was out in the tent last year, and when I got to thinking about it, I mean, you got 400 hackers, they're out in the tent, we're out in the middle of the desert. I mean, doesn't that come pretty close to what the Fed's considered to be an Arab terrorist? An Arab terrorist, think about it. We've got 400 hackers, we're in a tent, we're out in the middle of a desert. Wow, I'm not going to get any smiles tonight. Okay, we're going to have a t-shirt giveaway. I'm going to ask some questions. If you're interested in getting t-shirts, you might want to move up to the front rows here. Nobody wants a t-shirt either, okay. Another thing I want to ask about, I heard a rumor that Defconn got kicked out of the Alexis. Is that true? Can anybody confirm that for me? What? Well, I kind of wanted to confirm it on one way or the other. Okay, okay. Well, who wants to pick up the t-shirt? Because I'm still not sure about a clear answer on this. This presentation is going to be about unconventional manipulation of information. Most of what is going to be in this revolves around two ideas. One is meme theory, which has been discarded and people think that it has no meaning anymore. I happen to disagree a little bit. And the other part of it is going to be a thing called shared context, which is basically what people believe, what the set of things that cultural beliefs that people have. What you can expect in this is that we're going to have a lot of nebulous examples. There's going to be a few outrageous ideas in here. There's going to be a little black magic, some weird science, and if you're lucky I'll talk about my sex life, because that is a joke. Okay, everything that you're going to see ultimately derives back to a thing called trams model. This is a communication model about how people talk and how they communicate. I've got Jack and I've got Jill. And ultimately, Jack wants to go up the hill and he wants to get a pail of water, right? So he's got some kind of idea in here that he formulates. He encodes it into some kind of common language. He transmits it to Jill. She's got to decode it, and then it's got to match her vocabulary here to be able to get something to happen. Ultimately, these two contexts need to be synchronized if we want to get anything done. Now, what we're going to do is we're going to go through here and we're going to see that there's two extremes. I can either have contexts that are completely synchronized if Jack and Jill were clones, or I can have contexts that have nothing to do with each other at all, in which case I can't get much done. If I happen to have fairly synchronized contexts, then Jill can communicate and I don't have too much problem in the way of conflict and transaction costs. But as I differentiate the information sets that each set of those people have, if Jack has a different information set than Jill, eventually what happens is I start getting transaction costs, trying to figure out what I'm really talking about between each person. And I get conflicts because people misunderstand each other. And it is a geometric function. As I include more and more people, I get more and more issues with contexts not lining up. This is a really simple concept. I did it like 15, 20 years ago in college. I did not appreciate the true value of it until the last four or five years once I started playing around with the data money. It is an extremely powerful concept that can guide you in certain manipulations of information. It's very powerful. One of the things that's different now, because of the internet, if we go back through hundreds of years, there's all kinds of people that are doing advertising. We've got politics going on. We've got armies that are conducting psychological warfare. We've got propaganda being put out by corporations. All these things have all been in existence for hundreds of years. But what's different now over the last 15 years is they now have a feedback path that's quantifiable and it has a history path. It completely changes the way, it completely changes the dynamics of the situation. I don't think the great majority of people understand that yet. This is probably where I went wrong, was the idiosphere. This is a memetic term. A term out of meme theory. What it refers to is the common set of all ideas of all people. Imagine that. Imagine a network stream with thousands, millions of channels. Each one of those channels represents an idea and it's shared between certain people. I may have a bunch of hackers that know about addresses and memory spaces. I may have a bunch of people that know about social politics and they know about lipstick. There's a whole set of ideas that are shared between different subsets of people. One of the more outrageous ideas that we're going to do right here is we're going to take the original OSI7 network layer and we're going to put another piece onto it to get an idea of what we can do with these ideas. How many people know about a thing called biodiversity? Biodiversity, if you apply some of these principles to information, we get back to the SRAMS model ultimately of conformity versus diversity. I have conformity where context is all identical or I have diversity, ultimate infinite diversity where nobody has any context where they can share any information. Where everything has no shared context, that's a bad space, I don't want to be there. If I have everybody knows all the exact same information, I don't want to be there either. Both of these are exploitable in certain ways. So ultimately the set of all things that we're going to do revolves around this place right here and what we care about really is which axis we're tending towards. Conformity produces susceptibility. If everybody believes the same thing, it becomes a problem. That's one of the reasons that China protects their internet so greatly. They have a much more conformist society than we do. Diversity, in biodiversity if you have ultimate diversity people don't share the same genes and you have infertility. Now Americans do love diversity but ultimately infinite diversity turns into a bad thing and at some point because of the information revolution we will reach a point where the amount of differentiation that's taking place among cultures, among subgroups in the United States will reach a point where it's no longer a value. I believe. Now DEF CON 13 submission. This thing right here always makes me think of other things. Last year when I did this presentation I did a thing called the meme miner which I build as the poor man's data miner and at that time I heard a lot of stuff from the audience that it wasn't really appropriate venue for DEF CON. I took and distributed that program to about 30 people in the audience and about 30 days later they called me and they said well this is not working anymore. So I traced it back and it turned out that Google had changed certain aspects of the days I knew that I was doing a call into and this is what you get. What? Computer virus or spyware application is sending us automated requests. So according to Google now I'm officially a hacker. Oh okay I'm a virus. Either way I belong here now. One of the things with meme miner what meme miner does is it gives you a time series graph of certain keywords out of Deja News. I use Deja News for certain reasons which we'll look at in just a second. But let's take a look at service oriented architecture. You can see that back at the dot com crash it took the wind out of a thing called service oriented architecture and it went dead. Now since about 2004 service oriented architecture has been gaining strength steadily for the last couple years and if you've been watching the news you notice that it's start SOA shows up in a lot of stuff now. If you had been tracking all this stuff you would have known right around in here that there was a comeback in SOA and you would have known to maybe get involved with it if you were interested in it or at least take a look at it which is what I did. One thing that's interesting to look at is things like movie stars. I run a graph on Travolta here and you'll see that the number one entry that he'll be remembered for is Battlefield Earth. I use Deja News because it's a centralized hierarchy. It has a fairly balanced taxonomy that's fairly controlled by a fairly centralized source. A lot of the RSS feeds now are being created ad hoc all over the world all over in different spaces so I'm not sure that an RSS version of a database would be able to produce the same results that Deja News produces right now. Deja News ultimately is a dead product but it still can give you valuable information. Now we'll take a look at the OSI-7 model. How many people are familiar with this? Everybody should be. Ultimately I have at the very bottom layer, I have a physical layer, I have a bunch of layers that tie together and at the top I have an application layer. What I'm going to propose to you is that we need one additional layer which is an abstraction on top of the application layer to be able to track memes, ideas as they go through the network around the world. One of the things I've done over the past years, I've tried to tie back the low level physical layer which ultimately is voltage current and resistance. Ultimately everything that's built, the IT world is built on one concept, the measurement of electrons through a wire. So if we could take these concepts and push them back up to a higher level, we might be able to do things at a higher level abstraction to be able to measure and manipulate information. So what I did was I went through and after you play around with this stuff for a while you start seeing patterns. There's latency, there's a thing called sentiment and there's also a thing called diffraction which have rough equivalents to the electrical values. Here's latency. This is the pope. When the pope died, it took roughly five days for that information to propagate around the world in peak. Five days. That's very fast. Most of the memes take considerably longer than this. There's always a latency through the network. We'll take a look at sentiment. This is basically a comparison, a diff of two graphs. Google Cool versus Google Evil. And you can see that when Google first came out, the net result was that they were considered cool. But when they issued that stock IPO, they instantly became evil. This is extremely simple. This is very crude, but it's very simple to run and there's already much more sophisticated versions of this being run. I don't know if anybody saw the Sundance Film Festival. They ran a much more sophisticated version of this to be able to predict the winners. Tom Cruise. This is another movie star we can take a look at. There was definitely a bias in his favor for a long time, six years. But once he started those Scientology Rants. Now, that's just the mem minor. This is basically a rehash of what I did last year. What we're going to talk about now is a little bit different. This is going to be more abstract. I think everybody is aware that the NSA is doing data mining. I think that's all coming out in the news in the last year. Nobody knows that? Well, what we're going to do is we're going to take a look at an extremely crude example of the kind of things that you can do with this type of information. This is stuff that I did just off Deja News with a program that took me like 10 days to write. If you believe in meme theory, then you believe each culture has a different fingerprint for how it reacts to events. If that's true, then I should be able to differentiate between cultures and I should be able to see that fingerprint if I look at two events. If I look at a single event at two different cultures, then I should be able to see differences in how those cultures react. What we're going to take a look at right now is off of Deja News. This is filtered for these countries. There are certain groups that you can filter off. If you look here, I've got England, Australia, and Canada. See how they've got a very similar reaction to the Iraqi war. Any time that Iraqi war showed up. Now we're going to take a look at Europe. Here's England again. If you look at Europe, if you look at France and Italy, they have very different fingerprints from what you see for the Anglo countries. And if you look at the Asian countries, you get an even different fingerprint. The Japanese have absolutely no reaction at all, and Israel has a very different, very subdued reaction. All these sets right here are all within two times of each other. The lowest data set was about 8,000 items. The highest data set was about 16,000. So that's the kind of information you can get off. You can do a fingerprint of a culture. If I can map something, then I can predict it. Then I can manipulate it. Now we're going to talk about, but probably drew some people in here. This is not something that people normally would want to talk about, but I'm going to go ahead and talk about it. I worked for a startup company in 2000. I worked for them for a couple years, and I built a team of people, about 20 people. And we went through a bunch of politics, and ultimately I left the company. I wasn't happy about the fact that I had built a team. I had put a lot of effort into building a team, and someone else had used a bunch of politics to be able to conscript that team and take it away from me. So I kind of had an idea in my mind if there was some way that I could use information to disrupt the company. This is a very dangerous area. So here we are, we're in the 21st century. We're in the USA. Does anybody have any idea of one of the easiest ways to disrupt an organization right now in the United States at this time? Pretty darn close. Are you ready for this one or are you ready? Here we go. Telling the truth. Just imagine if the government told you the truth about the Iraqi war, or they told you the truth about the deficit, or they told you the truth about, you know, the NSA data mining. Imagine if your realtor had told you the truth about that house you bought, and imagine if you had told your girlfriend the truth about various things about yourself. It would be completely disruptive. So anyway, the plan for sales logics. I was in the military, decimate. Does anybody know what decimate means? The true definition of decimate. One out of ten. In the olden days, if you could destroy one tenth of an organization, you could destroy its organizational effectiveness, and it would fall to pieces. So my goal when I left sales logics was to disrupt the company, disrupt the team that I created. I figured out of 20 people, if I could knock three or four people out, that it would basically reverse everything that I had done, set it back to a zero point, and it would be like nothing had happened. That was my goal. The three factors involved in the disruption was the fact that we had a lot of stock options and everybody read this Yahoo Stock Board every day. So I knew there was a central point of communication that everybody looked at every day. The guys that came in and took my team away had made a pretty serious error in the way that they had done things. They had fired people that didn't need to be fired, that shouldn't have been fired, and they created credibility problems for themselves. And at that time, because of the dot-com crash, the management was issuing things that people were already unsure about. So these three things combined together to create an environment that I was able to take advantage of. Yahoo Financial Chat Board was the conduit of the attack. The damaged team was the object of the attack, although several people thought I was trying to affect the stock price at the time. And then the method of attack was ridicule. Satire can be extremely effective, and I did not have a true appreciation of that until 2001. The CEO of the company, I'll give you an example of one of the things I posted. And it's fairly mean. The CEO has a resemblance to Yoda. Pat. So one of the things I posted was, you guys deserve a pat on the back. And then I posted this picture. So anyway, imagine a bulletin board that was filled with a lot of satire and also filled with a lot of attacks on information, financial information that was already in question. Anyway, this went on for several weeks, and I didn't really have a true idea of what had happened at the company. My goal was to knock out two or three people out of the team of about 20. At the end of about 90 days, I talked to someone and they told me what had happened. 90% of the team had quit. That was far beyond anything I intended to do. And now for the consequences of my actions. I went to go to an interview at another company three years later, and the CEO was the CEO for SalesLogix. So I walked into the interview room and he's sitting there and I walk in and he goes, oh, it's you. Has anybody here ever been accused of being themselves? It is an eerie feeling. And what can you say to that, right? What can you say? So I said, yep, it's me. And he said, you know, you caused a lot of trouble. And I said, well, I didn't know that until you told me just now. I wasn't sure. He says, oh yeah, you disrupted the company for weeks. Just simply posting on that chat board. You got everyone upset. People quit. There was commotion. And so I'm sitting there and I'm thinking, there's two sides of me, right? I'm trying to get this job. It's a startup company and I'm thinking, you know, I just want to get my job. I don't really want to get into this. But then there's another part of me that's thinking, yeah, I fucked that place up. And I was surprised. I didn't realize that it had caused as much trouble as I had. What I've been playing around with for the last couple years is various ways of propagating memes and forcing information to go in certain ways. The data mining is a passive, what? Yeah, I did get the job as a matter of fact. They hired me the next week. No, that's a good question. I'm glad you asked that. I really am. The company went out of business. I mean, it went under like a lesson a year later. I've been playing around with various ways of propagating information and a lot of this is very experimental. I've had good success with a couple things and I need to check the time here. What's the time? 20 after. Okay, wow, I went a lot faster than I expected. I cut a lot of this out because I thought I was going to run over. One of the most effective things that I've discovered, which comes up over and over again over the last couple years is a thing I called the mixed message pattern. This is a very abstract idea. Basically, I have a message that I'm going to propagate out to people which has two embedded meanings. You send it out and ultimately it's context-based on what the target is thinking, what the target believes or what the target has been exposed to. So this guy right here, I happen to know that we share this in common, X. So when I send this message out, he's going to see only X. He doesn't know anything about the Z part. And likewise with this woman right here, she knows about Z but she doesn't know about X. This is exactly how politicians manipulate you in campaigns. I don't know if people are aware of the, you know, there's all conspiracy theory and all this stuff, but I've seen enough stuff with George Bush with the Christian right. I don't know if anybody has seen those photographs and I probably should have brought one down. There's photographs that have very explicit religious meaning embedded in them that are released to the press, that have them in a halo or have a cross up on the podium or whatever. I never paid any attention to those things because I'm not religious. But I can see now after it was pointed out to me that they're basically using a mixed message pattern to send one message to the Christians and one message to people like me who are agnostic. It's a very powerful, this is a very powerful pattern and I played around with it a lot over the past year and I've been surprised at what it's been able to accomplish. It can be very disruptive. If you propagate one set of information to a team, one set of a team, and you propagate a different set of information to a team, you can create conflict and you can create all kinds of disruption. Now we're going to take a look at if you can detect if people are trying to manipulate information on you. If they're manipulation means. I do believe that there is a small set of people now that are deliberately manipulating the Internet in order to force ideas out and force things to happen. That was always true. That was always true. People always want to get their message out and they always want people to believe. But I think that there is a higher abstraction, a higher level of people that are doing basically what I'm doing. They're doing it from a much more technical and theoretical standpoint. Now we're going to look at a couple of examples here. One of the reasons I know about this is because I put my website together a couple of years ago and I tried to do this. I thought that if I pushed out information in certain places hard enough that I could force the meme into the general the mindset and make things happen. It doesn't work. The idiosphere, the sum of all people's beliefs and ideas, that has very definite resistance. It has quality of service and it has saturation. Those are things that I absolutely believe now after playing around with it for the last couple of years. It works just like a network and if you treat it like a network it changes your ideas about what you can accomplish. I don't try and push my website anymore. I just let it run. I let it run at whatever rate it's going to run because I've discovered that pushing it doesn't really do anything. It creates short-term effects but ultimately those fade out almost immediately. How many people here are EE's? Well, in the transistor world there's a thing called over-driving. Audio signals. If you push a transistor past its normal limits it'll cut the signal, it'll clip the signal and give you distortion. Saturation, exactly. If you do that with memes and with ideas you get a very significant pattern. You get an identifiable pattern. A normal meme would propagate through the network at an exponential rate and eventually saturate and it would look like an S-curve. It wouldn't always look like this but it would be close. An over-driven meme tends to look like this. It tends to spike as people drive it and they try and force something out into the mindset and then it dies off because people really weren't interested. So this is kind of a pattern I look for now because I did this two years ago. Here's Breitbart. Has anybody heard of Breitbart? There you go. This was going to be a hot thing a year ago and if you look at this there's virtually no knowledge. There is zero knowledge of this thing until July 2005. That's already a warning flag right there. There's zero knowledge of this and suddenly there is a huge spike up of people talking about it. Now I ran this graph almost a year ago. I went back a couple months ago and I re-ran the graph to take a look at what the results were going to be. Can anybody tell me what the results are going to be? Flatlined. Went nowhere after the initial push. They haven't been able to make this go anywhere. That tells me that the idiosphere, people's minds are saturated. They do have quality of service. They do have priorities that they set on the things that they believe and the things that they spend their time on. And it is very hard to massage those things because it has been a process of evolution that has got those ideas into people's minds to begin with. Now has anybody heard of Rocket Boom? Let's take a look at the Rocket Boom graph. It looks exactly the same. I have a lot of doubts, especially when I see financial incentives tied to ideas. I'm very skeptical about the coming wave of video that's going to sweep the internet. And I believe that the internet is skeptical about it too. Okay, vector web. One of the things that I've tried to do with my website over the past year is I have tried to basically create disease vectors. Does anybody know what patient zero is? That's right, the initial infection. What I've tried to do is I've tried to subclass domains of certain knowledge and understand how those things are tied together over the internet and determine injection points for certain ideas that would be more, you know, certain points would be more receptive to an idea than other points and then push those ideas out and see how they propagate over the internet. And so basically what I'm trying to do is I'm trying to optimize a disease vector and the viral marketing guys are trying to do the same thing. I've had limited success with it. My website is very successful considering how many people read it. I've got something like 12 or 13,000 links to it right now, although I only get about 70 or 80 readers a day. One of the things that the search engines could do, and I wouldn't be surprised if they're already doing it, is if they're basically collecting the same information. They already have the information they need. They know when a keyword was available on a website. They know when it was placed. They know where it is. All they'd have to do is basically add an entry point. They'd have to add a starting time. They'd have to go back and back work it to do a frequency check to be able to understand when it started ramping up. They'd have to go back to other sites if you really wanted to. I think you'd have to eventually categorize each of these memes into a taxonomy because there are channels. There are definitely channels that information flows with inside people's minds. They have to be categorized to be able to get any kind of reliable results. Right now, any of the search engines could put together some kind of basically disease vectoring capability with almost no work at all. Now we're going to talk about one more fairly outrageous idea. I wasn't sure I was going to reach this point. One of the things that's different about right now that has been different all through history is that every person in here is being exposed to a huge amount of information far beyond anything they ever saw before. If you believe that people's minds are finite, that they can hold finite amount of information, and you believe that what they're being fed over the internet is fairly random, then what that means is we have what amounts to a cultural diffusion. We have what used to be a very common culture driven by the three TV stations, ABC, NBC, CBS. That kept everybody into a very narrow, very focused mindset where everybody had a very common context, common things to talk about. Now we have the internet, which has millions of different ideas, and I personally believe that we're watching a fairly slow devolution into smaller and smaller groups. You can do a little graph to get an idea of it. In the olden days, I would have a mainstream culture that had a normal distribution that looked like this, but as we go along, and more and more people get on the internet and they abandon ABC or they abandon the news, all the controlled media sources, I believe that the fringe cultures are pushing out, they're becoming more popular, and the mainstream culture is shrinking. I can't prove this. I have samples that look like this. I have certain memes. I have a great deal of memes that I run that show this, but I cannot prove it conclusively yet. But I do believe that if this is happening, it's a big issue. And I don't think that the powers that be are aware of how much trouble it can cause. If we go back through time and we take a look at the way information used to be pushed, it used to be pushed out through ABC, NBC, CBS. These were very coordinated. It was always the same information. Everyone got exposed to the same things. Then we went along and we had the use net groups that came out in 1980, and that was another differentiation, wider range of information, and now we've got 10 million blogs, and that's it. That's my presentation. Very intriguing the analysis of the broadcast media in the last slide. Why did you choose broadcast media and then print readable in the next examples rather than the print media versus print internet? Was that intentional or accidental? It wasn't intentional, but the print media still, if you look at the majority of the print media 20 years ago, they're an oligopoly that was still fairly controlled, so they still fall pretty close to this model. I don't think there was 50,000 newspapers operating. Right, no, I agree. Okay, great, you've confirmed that. Thank you. I found what you did about cultural fingerprinting very interesting, but you did it based on subject to see how different cultures responded. I know it's a more daunting task, but have you looked at trying to focus on a specific culture and fingerprint the issues that are, kind of do it in reverse, so instead of picking it out. Find out what issues drive the culture rather than... Right, and I think that maybe if someone could actually pull that off, people may be surprised what their perceptions of what a culture is interested in and what the reality of what the culture is really interested in. I can't tell you because I don't work for the NSA and I don't work for the federal government, but I would be shocked if they weren't actively researching this to be able to manipulate. If you really look at the model here, I wasn't going to go into this because I didn't think I'd have time, but I do think it's an important issue and I think it will surprise a lot of people. Let's look at this model right here. I have a heterogeneous extreme. I have a homogenous extreme. Can you tell me what countries those represent? You're in China? United States and China. And if you're at either extreme, what did I tell you? You're disruptible either way. Okay, thank you. Just an idea. You mentioned the need for a taxonomy of vectors. I was wondering if you had noticed, or what categories of vectors you had noticed, obviously... What I did for my own self, I did a very crude categorization of technology, social, and a couple of other things. I only had three or four categories that I was working with. Oh, sorry, sure. So how are you... you were defining them by just technology, social, broad categories? Basically, what I did was I tried posting... What I've done is I've tried posting references. What I've done is I've run all kinds of graphs. I've run graphs on politics. I've run graphs on photography, on technology, all kinds of things. There's a certain set of things that it works well for. There's some things that doesn't work well for it all. And then by posting those references in certain news groups or in certain places, like on the finance boards, if I see something that relates to Yahoo or Google, like I posted the Google Evil on the Google stock board. That was very interesting. That's when Google stopped working, by the way. That's when MIM Miner quit working. I've tried posting to things like Free Republic, Democratic Underground, and the fingerprint that you get from these different sites is just... it's scary is what it is. The Free Republic and Democratic Underground, the political groups have virtually no interest in anything except arguing about politics. It's amazing. There's no curiosity at all. Have you played around with Google Analytics much at all? Is Google Analytics the trends? Yes. That's based on MIM Miner. The guy that wrote that, Bruno, he sent me an email about a year ago and he said, I really like your graphs. And he went around and he sent me the thing for trends about six months ago. I said, I want you to take a look at this and tell me what you think about it. And I was like, I think maybe you should pay me some money. No, I didn't say that. That means their information. Ultimately, it's their information. I'm pre-writing on their search engine. Have you looked at anything with the social networks like Delicious or StumbleUpon or Dig and how... I've tried posting into Dig. StumbleUpon was a good success of mine. StumbleUpon, a friend of mine sent me the link to it before it got as popular as it is right now. And I ran a meme on it and I posted it to StumbleUpon. And then StumbleUpon propagated out to their people. And then what I watched was my hit rate on my site Rise in tandem with StumbleUpon. It was pretty interesting. It lasted for several months. I was going to say I was just going to suggest that because I've had some success with doing that through StumbleUpon. I tried to do it with Dig and it didn't work at all. Dig threw me out in a second. There's too much. It's too big a community right now. I don't know if it's... I'm not sure if I buy that it's too big of a community. I'm not sure what I believe about the filtered mechanism about many, many people. I think if you threw the right item in... I just think I threw the wrong items in. Thanks. I had a question about your last graph. It kind of... I think it sort of displays a sort of unbalanced picture of the state of information in human society because your last graph examines really the last 50 years. But I think that what that's really showing is a state of renormalization after disruptive technology was introduced, namely radio wireless transmission. That's introduced. It's very expensive. As the cost goes down, you go back to a smaller mainstream, less of a mainstream. Same thing. Similar things are happening. If you look back in history, you have small kingdoms. You have villages, towns, not much communication between them. Then you get printing press. That is a disruptive technology. As that goes down, your mainstream fractures again. In view of that, what do you see currently in development that you think has the potential for being a new disruptive technology to regain that control of the mainstream? Exactly what I just showed you. The fact that if you understand this information, I think the internet is far more disruptive than most people realize right now. I think that the NSA, I think any government that really wanted to, or any entity, I think a corporation could do it, that wanted to spend a little bit of money on data mining and on disease vectoring could do a fair amount of manipulation. Now I do think, ultimately, one of the reasons I believe in this model right here, well, you know what, let's go back to this real quick. What's my time? 140, okay. One of the reasons that biodiversity works the way it is. If I have disease vectors, let's say that I start off with a context, everybody has the same genes, and I get a disease, right? It wipes everybody out. So there is a natural incentive to move towards some kind of level of diversity. If I'm being attacked by disease vectors, I have a natural tendency to move towards more diversity. If I get people on the internet that are manipulating it explicitly to be able to push out their company or make their money on it like this, what is the reaction to that? More diversity. So I think what you're seeing is a breakdown of the most common structures into smaller and smaller structures because that's disease resistant. I think it will become harder and harder to be able to manipulate things as more and more people try and do it, and eventually it will drive a model that looks very much like this, except it will be differentiated by much, much more. But ultimately what that leads us into in the United States, I mean, if you're old like me, I'm 50, I have a different view of the United States than people that are 30. I remember when it was a common culture and when it could move unilaterally and it could achieve things. And I think China still is that way. I think that if we have a great deal of diversity and a great deal of small groups that are very self-involved, it becomes very hard to motivate the culture to achieve large scale things. I question whether we could recreate Pearl Harbor right now. I question whether if we were truly attacked like 9-1-1, whether we would have the ability to respond like we did in Pearl Harbor because we're obviously not responding like that right now. Whether you believe in the war or not, I think there's a difference in the capability of the United States to be able to respond because of the level of diversity we have now. You mentioned something near the end about the finite brain. Do you think that young people who are growing up with a vast amount of information around them actually are thinking differently than people of my generation or your generation that their brains are becoming different? I do. I think that they have a different way that they think in a way that they approach certain things. But that may be true for every generation. Seeing some of the analysis that you've got here and it's technical nature, I'm reminded of some areas that may be a bit more philosophical and I was wondering if you're influenced by or have thought about anything with regard to say the work of either Marshall McLuhan or even getting a little further out there, William S. Burroughs? I don't know. I know Burroughs' name. I know McLuhan's stuff, the media, but I wouldn't say any of them have influenced me. The two people that have influenced me the most would be Kos, transaction costs. I can't remember his first name now, Richard or Robert. He did a lot of stuff about transaction costs in government and why certain structures exist in spite of a free market and how organizational structures minimize transaction costs. And the other one would be SRAM, the communication model. Like I said, almost everything in this ultimately devolves back to a SRAM model of context, shared context differentiation. Thank you. Which is why I put it at the very beginning. I was hoping everybody would walk away with SRAM. Hi. I am absolutely fascinated by your graphs and this may be a little bit out of your field but I studied metaphor analysis and I was wondering if you could offer perhaps a computational model for quantifying embedded metaphors, embedded memes. Let me give you a really relevant example to this con. In the past when we were talking about our thought processes we used a metaphor like, it was called the cows utter metaphor. We would digest information and that is how we as a culture thought about our thought. But now we process information. The brain is a computer now. It no longer digest things, it processes things. And before computers obviously nobody used this metaphor. So this is kind of like a meme that's buried underneath in our linguistic model. There's another one, all of this thought right now about cycles, time being a cycle rather than something that moves forward linearly that's a very eastern concept and as our cultures have blended that has moved into our speaking patterns. I was wondering if you could offer a model for mapping things that are embedded like that because that would be really awesome. There's other people working on that. The thing I had there with Tom Cruise is an extremely crude semantic map. I did that a year ago just off. I want to make something really clear here so that everybody understands this. This is extremely crude. This is extremely crude, very simple and it is somewhat untrustworthy. I do believe that you extract real information out but I've seen where it doesn't work right for various reasons. The graphs that I pulled out they're right maybe 70% of the time. So I want to make sure everybody understands that there is a huge amount of potential error here. There are people that are doing semantic mapping work like I said the Sundax Film Festival what they did was they put together sets of keywords that represented certain phrases and they were able to do a fairly good guess at what films were going to win the film festival just based on people's comments. But that's conscious stuff that people can think about consciously like their feelings about Tom Cruise or their feelings about the Iraqi war. I think anything in the unconscious is reflected. Ultimately we're dealing with language. And you have to hope that the subconscious is reflected in that because that's all we have to work with with the internet. Until we get video everywhere and you can build something that can analyze people's facial expressions. The more you talk about diversity the more that the few biology classes I've taken over the years are kind of rumbling around in my brain. In biology biological evolution is constrained by bottlenecks in which case there are only a few events that are available and then things will expand from that point. Already you've talked about some of the technological bottlenecks like transforming over to radio transforming over to television where things will expand. Have you noticed any event bottlenecks that occur for example the events of 9-11 things like that. I've got that on my website. You can clearly prove that there is a quality of service and when 9-11 comes along there is a huge spike up of allocation of bandwidth to 9-11 and terraces huge spike ups in those keywords and there is a concurrent fall in other keywords. There's things that people abandon and believe it or not one of those things is sex. I was shocked that at 9-11 the keywords for sex drop so there's definitely an allocation of bandwidth going on. I didn't talk about it much because I didn't think it was going to have time and it probably depends on the type of sex. You know I have a dangerous area. You must have missed the beginning of my presentation. There's a thing that I talk about that I call diffraction and what it is is it's the allocation and deallocation of allocation, deallocation of bandwidth in response to events and I think this is an important area. I haven't been able to follow it up. I have a finite amount of time but I'll show you a graph real quick if I can get on here. I wasn't planning to do this. Is it 11-43 right now? You have 15 minutes. Okay well in that case we'll bring this up. If you go to my website there's far more here than I went into. I cut a lot of this out because I thought I was going to run over time. Here's the finger printing. These are patterns that I've used in my website in order to manipulate information and force memes to come to me or push things out. Let's take a look at diffraction. This is very interesting to me. Once I started playing around with this I realized that the human mind really does work like a network and it does have quality of service, it has saturation and it has channels. This is the Pope dying. This is the pure keywords right here of the Pope's death. See that huge spike up? Everybody's talking about the Pope dying. What that does is it spawns off all kinds of additional conversations about who's going to replace the Pope, was the Pope good and what that does is it causes a concurrent decrease in bandwidth in other things. Because of this allocation of extra bandwidth there's now an opportunity here to force new memes in. There's been a disruption in the system. The system's responded by taking bandwidth which is marginally allocated to things that it doesn't really truly care about, just whatever the lowest level is. It takes and deallocates those, it allocates them to whatever the new entry point is. At this point you have the ability to change things. You have the ability to force additional memes, related memes in and capture that bandwidth if you can hold it. There's a thing I call the flea pattern. I use this in my website. My website writes on the back of the search engines right now. I explicitly use the search engines to drive traffic to my site. I do all kinds of things with search engines. I do searches on Google. I use Google in some of my graphs. A meme enters the bandwidth. It captures a certain level of bandwidth. At the point that it enters it actually gets allocated more bandwidth than it will eventually have, most of the time. If you can associate your meme, something that you don't want to have happen, with the meme that's entering and you can create some kind of binding between them, a shared context. A lot of times what you can do is you can capture that allocated bandwidth and hold it. This is a pattern, this is a very high level abstract pattern for being able to force information into people's minds. Scary, huh? Scary that I even thought of this. Right, exactly. That's why I think it's important that people understand that this is happening anyway. Just people don't realize it. I didn't realize it. This is happening right now. There are guys that are in politics. There are guys in the military. There are guys in advertising. There are guys in corporations that are explicitly trying to hack your mind. They're trying to push information into it. As long as you understand that, I don't think a lot of people realize that the dynamics of the situation are substantially different now than it used to be. They didn't use to have quantifiable feedback whether it was successful or not. I can measure when I'm successful. I can measure when I push a graph out. I can measure how many people read it. I can see what the reaction to it was. I can adjust. The internet has created a feedback loop into the real world, back into the virtual world, back into the real world. It works at a fairly small scale right now and I think it will continue to work at a small scale. You can do disruption of small scale organizations if you really want to. I know that for a fact now because I did it six years ago and I overdid it. I think you've already sort of answered this question, but I wanted to clarify. For the real sustained beams, the ones that follow the normal S-curve, do you think they are or can be influenced by individuals or are they all ground swellings from the 100 million blogs? That's a good question. I think that in general they're pretty much at random right now, but I do think that there is a rising effort to be able to explicitly manipulate things like I'm doing. I think the reaction will be to create more diversity and more diversion, more devolving into smaller groups so that the effect of any one person becomes less. I asked because I was thinking of, during your talk I was thinking of people like Tim O'Reilly or Cory Doctro who definitely I feel have influence, great influence on some of these memes, but it's possible that they are just the mouth pieces of the ground swelling. I want to point out that there's a difference between an idea and a difference of manipulating an idea. Generating an idea is one thing. Manipulating and forcing it out into the public is a different issue. There's two separate things there. Thanks. Time? Thank you.