- Share your favorite videos with friends
- Comment on videos and join the conversation
- Get personalized recommendations
- Enjoy exclusive offers
Purchased a FORA.tv video on another website? Login here with the temporary account credentials included in your receipt.
Sign up today to receive our weekly newsletter and special announcements.
Good evening I am Chris Shipley I am the executive producer of the DEMO conferences and the co founder and chairman of the Guidewire group and I welcome you to this meeting of INFORUM a division of the common wealth club by and for young people who share interest and a passion for civic issues. Tonight I am joined by Jimmy Wales he is the founder of Wikipedia and also the founder of other Wiki related projects including the charitable organizations Wikimedia foundation and the for-profit company Wikia. Wales was ranked 12th in Forbes magazine, web celebs 25 and listed in 2006 in Time magazine's 100 influential people. He has changed the way millions of people access and influence information when he opened the collaborative authoring software the Wiki software to the world's internet users. Creating a website with boundless encyclopedic information that site is called Wikipedia. Recently Wales launched Wikia which is building the rest of the library with 1000s of wiki communities and it is intended to be a search engine to compete with Google and that emphasizes a greater transparency and promises less spam. Please welcome Jimmy Wales. Jimmy you had a have had a a rich career before you founded Wikipedia but I think that is certainly what you were so well known for and what so many people here tonight are interested in learning about so lets may be start in the middle of your carrier with Wikipedia and talk about sort of the origins of that project. So the the basic idea for Wikipedia came from watching the growth of the free software movement. So for many years free software, open source software had been steadily growing and becoming more and more important throughout the industry. And I was watching this happen and seeing programmers coming together to collaborate often times and clearly volunteer context and creating really fantastic software. If you remember the the early days of open source software people really tended to dismiss it as you know, well this is something fine a few hobbyists will play around with it. But overtime we were seeing you know, kind of Linux and apache and Perl, PHP all these stuff that really runs in the internet is the open source software. And so I realized that the free licensing model created a new type of social condition for people to be able to collaborate together. But for all the rest of us, if we want to collaborate to build something, you know, well what did you have available so if you wanted to co author a document with lots of people you could e-mail around word documents which is like a night mare. And you totally don't want to try to do that so I realize that hey there is a there is an opportunity here to have the same type of collaboration the same kind of social model but three things other than just software. And so Wikipedia was like the first thing there I mean the encyclopedia project I should say. So when I started I had PhD in philosophy and we had a project called Newpedia which was organized in a very top down very structured very academic format and was a complete and total failure. So when you say top down talk about what you mean by that? So what I mean by top down is that in order to participate in Newpedia you had to apply to become an editor, apply to write an article. People had to prove their credentials in fact for a period of time. That guy who was running it had people faxing in their credentials to prove they were you know, PhDs or were not. And the hurdles to participation was so enormously high that it just really failed but at the same time, what was good about it was that a large group of people had been talking for a couple of years, really good people about how to create an encyclopedia what it should be like. And we did all that before Wikipedia and so once we found out about the Wiki editing concept which had been on the web for since 1995 and decided to try that we already had a great community built up and so it was just able to get started right off bat. So the Newpedia community was able to transfer was that A fair number of them transferred yeah not everybody, but it's more that, a lot of the ideas and the spirit and that people knew we were working on this and that we have a lot of the I would say the ideology behind Wikipedia came from - the really strong commitment to free licensing for everything, the really strong commitment to neutrality to open this to transparency, all of those things came about as results of those discussions within the Newpedia community. So all the big ideas you might have tackled for a a giant collaboration, why the encyclopedia. It just seemed very obvious to me, and I remember when I had the idea, I was in a complete panic to get start because I thought somebody else would do it first. And then for you know for the two years of Newpedia which were failure in the first couple of years of Wikipedia, well pretty much nobody noticed right. But an encyclopedia is a really fantastic it's low hanging fruit for collaboration because it's a pretty easy thing to do. So if I say, you know encyclopedia article about the golden gate bridge right. Everybody in this room has more or less the same idea of what that supposed to be. You need to have the history of the bridge, you need to have all the facts that link and and so forth. You need a couple of pictures on there, may be one in the day, one at night. You kind of have a good idea what an encyclopedia article should tell you. We can fight and quibble about the details but we we all have a clear vision for what it is we are trying to accomplish. So for the encyclopedia, you know every article, you pretty much know what you are supposed to be doing and that's really important in a collaborative environment because if you say, you know every one in this room tonight we are going to get together and we are going to write a poem about rain. Well good luck figuring out what the hell that suppose to mean right, but because there is no structure and there is no simple obvious idea, so really encyclopedia is pretty easy. So we know that the traffic to Wikipedia is is quite tremendous. What is the collaboration community? How large did that community become? So it's kind of hard to measure, because you know we we tend to think of the community as being people who have a certain degree of involvement and commitment. If you come to Wikipedia as many many people do, you click on edit and you fix a spelling error or you make a sentence a little bit better, some little thing like that. Well you have helped in your own small way and that's wonderful but you are not really part of that community. I normally look at some statistics like the number of people who are making at least one hundred edits in a month and things like that. So you know if we want to talk about the really really active community we are talking about, may be around 3000 people in English and you have to keep in mind that English although it's the largest language in Wikipedia, is actually less than one third of the total work. So there is you know, equally large you know twice as many people who are not working in English, working in other languages. So may be 9000 people were wider like the core community And the number of entries today are approximately. The number of entries in English is 1.8 million some odd it's always on the front page of Wikipedia so I didn't look today so I can't tell you the exact number. And like I said that's less than one third of the total so it's over 6 million in in all the languages and so in terms of number of languages, we now have about 130 languages that have at least 1000 articles. So 1000 articles is not really much of an encyclopedia but that's the number that I always keep an eye on because, once I see there is a 1000 articles in a language I normally find there are five to ten really hard contributors, 20 to 30 others helping them. And that's when they really start to become a community, they have had a few fights, they have made up. They start to thinking about how they can get more people to come in, they started e-mailing academics, they started trying to make a press coverage in their local area. So I I consider the number of languages to be like a 130 or so right now. So when you you started your first experience with Newpedia, you - you were screening credentials of of users, what and what was your imagination about what what that credential user would look like and how does that today compare with the credentials if you will, the people with the profile of the members of these, most active members of your community. Yeah so, so this was a real struggle from the very beginning and I never thought it was a good idea to ask for credentials from people but was talked into it and there was some possible reasons for it. But what was really missing from that approach was, all the people who would just like to help out in their own way, whatever that way might be, so some people can contribute to Wikipedia, by being a subject area expert and some times those credentials or credentials that are very traditional and very ordinary. So for example I met a - an ornithologist at Cornell University who regularly checks up on the bird articles in Wikipedia - he doesn't write all that much but he checks that on it, and he has met a few people in the community. That's actually a - really convenient area to talk about this phenomenon because in ornithology they have had a long standing, very favorable relationship with the amateur community of bird watchers. So lot of the data that they have collected over the years comes from passionate people who are just really into birds and they like to go out and hunt for rare birds and report back to scientists on what they have seen. So they are very comfortable with this idea that they are passionate amateurs. And in some other fields, that's not been as common and so some times they are little less comfortable I'm guessing neuro-science wouldn't be favorable - Yeah yeah the, for some reason the brain surgeons really frown on the armatures and I just don't know why. There was a great there was a great April fool's article, I can't remember if it was in the Onion, or it was the CNet April fool's edition that, - said I was having WikiLasik so I was going to, you know, it may take the community a few tries but I am sure that they will get it right, anybody come by my house and yeah. So in terms of how does how does the community differ, another way that it differs, so some people are contributing as topic area experts but without traditional credentials. So the math professor is also World War II history buff and that sort of thing. We also have people who just like working in the communities, so they like doing things like spell checking or mediating conflict or just going through and rewriting, I mean I have met many journalists who say, you know as professional writers, you know this well I don't really contribute to Wikipedia, but some times I will see really tortured sentence and I cant take it, so I fix it, you know. And that's just different ways of contributing that may not be, and in fact in many in many cases you know, if you if you have a a physics professor, they be not be the best writer in the world, particularly in terms of explaining it to general audience and so you get these collaborations between very different kinds of people. So I think one of the the questions that, comes up around this issue of credential and the authority of the community really is to do with - what actually should be, in an encyclopedia and you left that largely to the community to determine what - what entries are appropriate for data and which which are not and how do those debates what's that process like? You know, so we have always aimed to be as compressive as we can be, and so there is several ideas that emerge for a long period of time, so one of the the old sayings in the community is wiki is not paper, so if somebody makes an argument that we shouldn't have something because its just why we are taking up this much of space, well its all on a computer disks, it does not really take up any space right, and we are not constrained by, needing to print out 30 volumes and so there is no reason to limit the content based on sheer size of the text. At least not in terms of the physical size - right. At the same time though we do know that there are limitations, that isn't to say there are absolute no limitations. So one of the classic limitations is verifiability, one of the things that can happen is, you know any person in this room who could could write an article about their grandmother, well someone here must have a famous grandmother, that will be fine, but most of us just have a regular old grandmother and there is not a lot that you can write about my grandmother, that anyone else could really confirm. I could write down some family stories, but you would have no idea if I was telling the truth or not, there is no way to verify it, so the community would say - that's not verifiable there is really no way for us to verify that. So that's one of the rules that that has come up over time, in terms of limiting this, another that's come up more recently as Wikipedia is gotten bigger and bigger and bigger in the English language especially is some consideration particularly with biographies of living people, that there are incentive problems, that some times if somebody is some body is very very famous, there is no problem, we need to have an article and there will be many different people with an interest, we can keep the article well referenced and neutral. But sometimes, if somebody is not very famous if they are only very famous for doing something bad, its really hard to maintain the article, it's really hard to write a balanced article, small town politicians - right who have some people who hate them, some people who like them, they come to Wikipedia and they fight and we don't like it, so get rid of them all. Those are the kinds of things where we realize that there are constraints; in the social model on how detailed we can be and still maintain quality and accuracy. So part of that social model I think and again it goes back to this credentialing idea, is really about transparency and trust and - and that's starts to lead the questions of of motive and of of incentive perspective, these sorts of things. Talk a little bit about the role of transparency and in understanding who the authors are and how they, what their motivations are for participating in the in the community So in in all Wiki's - part of what makes a Wiki work is the transparency. So when you make an edit at Wikipedia, Wikia or any other Wiki, it goes in to the history right, so everybody can go back and review all of the past versions that have ever been there. If you have a user id, a stable identity over time, people can, with one click see all of your contributions. So that you can build a reputation over time, people can come to know who you are, and decide whether or not they trust you, based on a long pattern of behavior which is exactly the way we do it in the real world. If you meet some one new at work, you don't know, are they good, or they are not good, and over a period of time you learn whether or not you can trust them, then you being to rely on them. And there are a lot of social pressures right to, continue to behave well. All of that is really important in Wiki's setting as well. When people imagine Wikis as being, you know, 10 million people each adding one sentence each, I always struggle against this idea of the wisdom of crowds and swarm intelligence in those kinds of things. So I don't think they are really accurate description of how this works. If you think of it as 10 million people each adding one sentence each, it seems a miracle that it isn't just rubbish. When you begin to realize that, you know the the articles about bridges right are [0:16:58] [by and large] written by people who are members of wiki project bridges which is a sub group in Wikipedia and these are just people who love bridges, and for what ever reason is their hobby know everything they used to know about bridges and they discuss how to make the articles better and they keep one on each other, they come to trust each other and its that small group that that can monitor each other that's really important and the transparency is what really makes that possible. When you get this this authoritative group and it's really got the world's most enthusiastic bridge experts, bridge hobbyist may be even, there comes a few question of sort of sourcing of information, the quality of the information and the ability for Wikipedia itself to be the authoritative encyclopedia. How do you see that how do you see Wikipedia gaining that authority as a reference, we know that lots of people do refer to it, but is this now, you know, is it moving to the point being a citable reference in a in a bibliography. Is it gaining that authority in the market? Right so in general no and in general I don't think it should. I always say you know, when I was in academics and when I was in the university, if I tried to sigh at encyclopedia Britannica, I would have been laughed out and appropriately so, that isn't the role of an encyclopedia in the research process and isn't something that we should aim at, as an example one of the one of the old rules, social rules of Wikipedia is - no original research. And there is a two fold reason for that as there are from, from many of our rules there is the the epistemic or the knowledge oriented reason and then the social reasons. So the knowledge oriented reason is, if someone's doing original research so original historical research or actually the rule came about in the early days because of of trying to do with physics crackpots of whom there are many online. So somebody comes in and they want to post their their new theory of magnetism which they have just dreamed up. Well we really aren't qualified to evaluate that, and open into community like this, its it's just not the sort of thing that we would be particularly good at, and so from a knowledge point of view we want to say, well there is other citable you know, sources, other references, other places we could go to confirm what you are saying, and if the answer is no, we say we are - have a really hard time doing with that because as a community its one thing we can check to see if the - the journal article you sighted from the Journal of Physics actually exists. And what it says, but we can't necessarily like start working through your equations or whatever. And then the social reason is, it it actually works out socially very well to have this rule of no original research, because when - a physics crackpot shows up instead of having to say you are a crackpot, please go away you can say thank you for your wonderful contribution, we just aren't qualified to evaluate it. It's - looks like highly original research, you should submit it to the Harvard Journal of Physics and I am sure they love that we say that, but so that's basically you know how we deal with that right and so because we consider ourselves tertiary source or we are not a primary source and that would be you know, raw historical document. We're not a secondary source we aren't a peer reviewed academic journal, we aren't a - a - something that's been published. We are a tertiary source, so we are a source that points people to other things so we are starting point for research, we we aimed to be a highly reliable and accurate starting point for research and that is a good thing. And when we fail at that, we are not happy about that. But to try to become a citable source is just not really what I think we should be trying to do So you talked about social rules and I thought it - just to hear you elaborate on that for a bit. And particularly, I think there is a Wikipedia etiquette of sorts about how one one edit someone else's work, how one submits there's certainly etiquette around, writing your own biography, these sorts of things. Can you can you just expand on that a bit so we can understand what those rules are? Yeah so, a lot of that is what you can characterize as kindergarten ethics. I mean a lot it should be fairly obvious. One of the oldest rules of Wikipedia in fact perhaps the oldest rule depending on how you count it is ignore our rules which is sounds a little scary, but the point of it is, the rules of Wikipedia should always be something that you don't have to consult and detail before knowing what you should do. Its all be fairly obvious that if there is something that is not obvious like - a stylistic tradition or something like that well if you violate it, you shouldn't get in trouble, somebody should just come along and help you and fix it. It's really important to to the project that we and its important for all, Wiki communities, in fact for all online communities, its important to have certain social standards that's, look there are certain ways that we want to try to interact with each other. One of the things is - assume good faith. This is something you will see Wikipedians say a lot. Somebody comes to an article, they start messing it up. You can assume that they are evil or you can assume that they are trying to do something useful right. Maybe its hard to see what it is that that they think they are doing, but if you assume that there - they probably have some good intention in there, some where, you can address them in a respectful way, you try to draw them out to say what is that that you are trying to accomplish here, I feel like the edits you made weren't that helpful but you know lets talk about it and so when we succeed at that we have much better dialogue, we have much better quality of work in the end than if we just had flamewars back and forth and this is really a big part of, the Wiki culture and what - what is I think the real change here from you know like a web form for discussing politics for this is a Wiki for discussing politics is there are the incentives built into the very mechanisms of the software to encourage people to be a little friendlier and be a little more reasonable hopefully. Then you have one of the the advances that Wikipedia is responsible for in Wiki software sort of separating but - but keeping next to one another, the articles that the document essentially and the conversation so that those things can happen in tandem. Yes, so every page has you know the article and then the - and then the discussion tab. And the discussion tab is itself is a Wiki which is actually very important for the - the social environment. Because, if somebody comes in and they start flaming being really rude or not just someone else would just delete their comments, everybody else doesn't have to see it and but that is all transparent, you can see who deleted what and why and somebody started deleting other people's comment, well they better have a good reason for it or they - they get in trouble and so on and so for so it's - it's a circular kind of incentive mechanism that keeps you honest. And yes, so separating those two out was very important, I mean it seems fairly obvious now but I remember in the very early days of Wikipedia when Tim Shell posted on the mailing list said, hey, you know we should really not be discussing you know the articles - in the articles which shall have separate space for that and that immediately got adopted as a convention later built into the software and yeah that's very helpful From - from Wikipedia grew the Wikimedia foundation. Tell us a little bit more about that So the Wikimedia foundation is a nonprofit organization that I have founded. It's a 501(c)(3) charity in the US. And the Wikimedia foundation owns Wikipedia, Wiktionary and the bunch of other sister projects that we can use and so forth. We had we spent about a $1 million last year and we'll spend around $2 million to $3 million, we think this year. The bulk of the money goes for hardware and bandwidth and we are supported primarily by small donations. So, the vast majority of the money that Wikipedia has ever received has been in the form of $50 to $100, ÃƒÆ’Ã‚Â¢ÃƒÂ¢Ã¢â€šÂ¬Ã…Â¡Ãƒâ€šÃ‚Â¬50 to ÃƒÆ’Ã‚Â¢ÃƒÂ¢Ã¢â€šÂ¬Ã…Â¡Ãƒâ€šÃ‚Â¬100 mostly from the U.S. and Europe, significant amount from Japan, Australia, and really all around the world last year we had donations from over 70 different countries in different amounts; some by small but small countries. And so, the foundation now has seven full time positions, seven full time employees and it's a number nine website in the world, and I bet nobody else can have a ratio like that. So the one of the things to understand, lots of people when they have just used Wikipedia, and probably this audience has - has used it more and knows a little bit more about it. But, lots of people are you know, I get interview questions and I realize after two or three questions, what they really want to know is where the Google blogs, where is the compound, where the campus with the 600 people administrators monitoring everything. That doesn't exist and in fact the the employees the office of the foundation has almost nothing to do with editing the actual content. There is one guy, Kerrey who takes phone calls in the office from people who are complaining and don't know how to fix something themselves. He liaisons with the community to let people know about the problems, sometimes he'll fix it himself. Other than that, one person you know calling the office with the complaint is probably not very helpful because they are they are quite swamped with it and try to keep servers running so. If it is it's a highly distributed community, not an organization and I think part of a trend that we are its clearly unfolding and I think very powerful this movement from really centralized organizations out into communities. And then Wikipedia may be well the strongest example of the power of a community to to inform, to shape and certainly that seems the work of the foundation as well. Yeah, exactly but I mean it's also important to recognize I think the the limitations of that, right? So, the Wikipedia community can do something really amazing which is why this great encyclopedia, maintain it, and monitor it, always be trying to improve and so forth. But the foundation and the community have not done such a great job, of actually pushing the software forward. We have very limited resources for doing that at the foundation and so actually over the years as its typical for free software and one of the common criticisms which is usually but not always valid is, it tends to be very geeky the user interface can be a little bit clunky. So if you go to edit an article in Wikipedia, its supposed to be easy and obvious what you are doing but in many cases it really isn't; it's gotten to be quite complex. So this is part of the reason why you know, I started Wikia is basically as a for-profit company and we have funding, we have a lot more programmers, a lot more employees, we have many people doing usability, study, design because we really do feel like this is just the first stage of this revolution. But in order for it to go further, we need to make editing of Wiki and collaborating in a community as easy, and intuitive and obvious for people as using a word Processor. So, you know if you remember the very early days of people using word Processors, you could always you know, you could see the paragraph codes and we didn't have WYSIWYGs for a long time. And then over the years, you know we got better and better and there are certain conventions that developed and I pretty learned more or less how a word processor works. We are just starting that process now in collaborative editing and so we are really trying to push that forward just to make it possible for a whole another circle of people to get involved in sharing their ideas and thoughts. Did you originally expect that the developer community would would advance the platform in the way that the editorial community advanced the document? Yeah. And I mean, I am not I am definitely not saying that they haven't advanced it - right. Because I mean media Wiki is a fabulous piece of software, it does what it does really really well. But it isn't completely user friendly, right. And so, it's amazingly successful in some ways and not successful in another ways. There is a - you know its its completely open source software, so there is developing community. There is an extension architecture that people are using to create extensions to the software to do different kinds of things but it's still quite difficult. It turns out that the learning curve to joining the media Wiki development community and really getting involved in making a difference it's much steeper than it should be, it takes people a long time to get involved and to really help us push it forward. So that's we are trying to fix that but it's you know it's a big job You described, Wikia is building the rest of the library? Yeah. So if you think about an encyclopedia, you know traditional library encyclopedia is normally - 30 35 books, its usually like this big or so. But then imagine everything else in the library, all the other possible books and works that people might collaborate on. And so the the idea is that we see that people can come together and collaborate on all kinds of things. Now there was a time when I thought - this is the only thing that makes this work, the only way you can really have a successful Wiki is neutrality the neutrality principle of Wikipedia is very helpful as a social mechanism for helping people to get along. But then I saw Uncyclopedia which is a parody of Wikipedia, which is absolutely hilarious, its certainly not neutral I am not sure what it is exactly but neutral is not the first adjective that pops in my mind. But it was is a very - what it is very successful Wiki and you have 10s of 1000s of articles. And so that was the first example where I really - the light ball sort of went on and I said you know it's not just the encyclopedia we actually can just do this lots of other things. So now we have over 3000 communities building everything from Uncyclopedia, which I thought was so funny that I bought it and from Uncyclopedia to help that because these sites we have the WoWWiki which is all about world of warcraft. So if anybody hears the fan of world of warcraft they would check out the Wiki. Although if you are a real fan of world of warcraft you are probably not here, you are probably playing the game. So but all these different kinds of things that people are building that are not the traditional encyclopedia, some of them are clearly advocacy there is sfhomeless Wikia sfhomeless.wikia.com where the local homeless advocacy community is coming together and documenting best practices, what are all the resources available in the area, talking about that, that's a really great one. And just really great people are using this tool. It's clearly not neutral right they are advocates and they are trying to push something forward and do something beneficial. And so it turns out hat what is needed to make a Wiki work is not neutrality but a shared vision. So when I talked earlier about the shared idea of you know you know of what an encyclopedia article, about the Golden Gate Bridge is supposed to be. It works equally well if you say if you say well this article about George W Bush is supposed to be a parody of an encyclopedia article. Okay everybody kind of knows what that is to. And in fact within that community they have certain house style and sort of inside jobs that they have built over times so that the whole thing has a certain theme to it and so forth. So it's really amazing that kind of things that are going to constitute the rest of the library and so that's what people are working on. Are you finding because these other Wikis the rest of library tend to have a a much more clear perspective they are not to your point neutral that the communities behavior interacting in any different way. You are learning anything about the social dynamic when there is a a clear perspective? Absolutely yeah. You will find things like you know the the the style of interacting can be very different. So we have a site ArmchairGM armchair general manager is the idea so armchairgm.com there it's a sports community so its all about sports and so they interact with each other very differently from Wikipedia, right there are bunch of sports nuts and they sort of tease each other taunt each other, they have you know they leave comments and they sort of have these fun little debates and is like a lot of little advocacy going on in there and we actually we had a lot of cool software for them very recently social software so they can friend each other and give each other gifts and things like this and make foe so now can be enemies with each other. But it's all good natured right. In Wikipedia I have to say around the lot of political articles I don't think it would be a good idea if we had to we had to mark people as foes and I think having this sort of spirit of togetherness is better but - in a sports community were its all just a bunch of people kind of joking around and having fun writing about sports, they write news article about what is going on its words, it's a totally different flavor and yet its something that the community likes to do and its offering the license and everybody, you know they are having fun with it. Now the other part of Wikia is search and search you claim that it will be better than Google and - and lesser spam. Talk about the vision that And not as many calories and you will look better while you are using it But - but will you get a date Yeah no so I - I actually am a little more cautious than all that right that, the I have been talking about search. So for me its - its like it's almost a political statement more than anything else, it's a political statement about what we as citizens of the world should expect from the search engines which are driving most of the traffic on the internet. We should expect a certain amount of transparency, openness, participation. All of those things are super important and this is really a statement that I think we can build this and I want the community to come together and try to build this. So, I view it that way and then Fast Company Magazine had me on the cover with this sort of cool picture and it said Google's worst nightmare and there I was - my picture, I don't - I don't know what about that - my mom bought 10 copies so - somebody liked it. But we really aren't comfortable sort of in that role of of like you know sort of the Google killer. But it's a great story so whatever. So, basically what we were doing we are bringing people together to say look we want to we want to cerate a search engine, all open source software, human participation, so we were right now, we are in the early design stages. I heard Jeremie Miller who is here tonight from Iowa. He was the founder of Jabber which is the open you know, instant messaging protocol and he comes from this long line of you know how instant messaging, its it's these close silos, if you are on one service you can't chat with somebody on another one. But it is very different from e-mail where it doesn't matter where you have to email, you can email somebody on a - it's a simple idea right? And so Jeremie invented the protocol so that we can begin to to break up those silos and monopolies and that's beginning to have a bigger and bigger impact all the time and you know Google talk supports Jabber and it's right default in Apple iChat and there is bunch of clients that support it. So that is what true Jeremie and I together with this shared idea that - this kind of stuff should be you know open, transparent and there is a huge number of different kinds of benefits that come with that. So the questtion I have in these is well I guess fundamentally is whether human powered search really can scale if you talk about and you think about the massive amount of contact even just in your own in your Wikipedia alone and then multiplied across the internet all of these documents. How do you scale human powered search? Is this algorithmic? Human assistance or Yeah - so I mean the way I look at this is that the - the philosophical question I think is pretty easy to answer which is the stuff that computers do well, they should do their part and the stuff that humans do well, they should do that part. Well, that doesn't answer all the question right? That's a sort of broad statement of how we should divide up the work. I can give you a really great example I was on my cell phone the other night and I wanted to go see Harry Potter and the order of the Phoenix which just came up. My daughter really wanted to see it. So, in my cell phone which is really hard to type on and so forth in the browser I typed in Google in the mobile search engine which is pretty smart. It knows I am in san Francisco, it always tells me that and so I typed in Harry Potter, phoenix, show times and I got all of the times, when Harry Potter is showing in phoenix Arizona. Well, I thought well that is a computer for you right? And that's not really criticism of Google because I don't know if anybody is doing that correctly right now. That's the kind of thing that you can always expect algorithms that are going to trip up on until computers are smart enough that you know that they can hold talks and conferences like this. I think we have to say look algorithmic searches has certain really strong limitations. That kind of mistake would never happen at Wikipedia right? You would never have that kind of weird ambiguous thing were somebody starts writing about Harry Potter and the order of the phoenix and starts talking about phoenix Arizona .That just would be crazy right? So that's an example where what I saw is being needed at that point and nobody is really designed how to do this share would be if you are - if you are there at that page and you realize that your the computer is doing something stupid and really needs to be told that this new movie came out its called the 'Harry Porter and the Order of Phoenix' and so people are searching for Harry Porter and Phoenix; they probably don't care about Arizona. Right. That's not really what they are looking for. That's a human judgment that could be fed back to the search engine and I think Google spends a lot of time trying to do that, Yahoo all the major search engines. They try to do as much as they can but this is were the scaling issue comes in. Any small number of employees at Britannica can't keep up with Wikipedia. And any small number of employees at a closed proprietary search engine can't keep with a properly empowered community. And so, as I talked about Wikipedia and the idea earlier that if you really want to collaborate on writing documents, emailing around word documents isn't the right way, we need tools to do it. If we really want to collaborate; we really want to bring together human minds to help us search, the current tools are what we need. Right. We have to work on thinking about how we can get information from people, how we cannot let people participate in a way that we can monitor their behavior, we can monitor each others behavior, reward that behavior, diminish bad behavior, all those kinds of principles that go into making a Wiki work are also really needed in this area. So just to clarify then, you are really looking at how humans tune algorithms; not how humans deliver better search results. Sure. Good. Something like that. We have a just a few more minutes for our conversation, but I would like to invite you to join the conversation as well and to do so just please go to the microphones in the aisle and you can begin to line up now. And as you do that, Jimmy - a question; you used to comment a moment ago, "Citizens of the world" and certainly Wikipedia and much of your work is finding its way around the world, you travel extensively. But Wikipedia and its social construct aren't accepted every where, when you talk about the fact that in China Wikipedia is not publicly open. Yeah. So we were completely blocked in China and we have been for quite some time now; about an year and a half I guess. And we don't like that. Chinese language Wikipedia is the eleventh largest, it is the large community it's growing perfectly fine - turns out, lots of Chinese people don't live in China. And so, it has a really active participation from Taiwan, Hong Kong, the U.S. and around the world. And actually quite a bit participation from China because the firewall is fairly porous and so there are some really brave people who edit Wikipedia despite its been blocked. I was in Hong Kong and meeting with some of the Beijing Wikipedians, we had a special meeting. With this point we have been blocked for like six months and you know I said I am worried that you know you guys are going to sort of go off into something else. And so frustrating you know Wikipedia and they said, "We are with you forever." So these people are very dedicated and so people within China are going to continue to edit Wikipedia and eventually they all realize that its foolish to block Wikipedia. But the I am going to be in China in just in September and I am trying to set up as high level meeting as I can. We have deliberately not really tried to pursue it at low level, bureaucratic levels; simply because our legal advisors tell us we don't even have standing to file a case, we don't exist in China and the only way to exist in China is to agree to censor and we are not going to do that ever. So it's really more of a at the political level. So I am trying to reach out as high level as possible to explain to them that well Wikipedia the neutrality policy is quite important, its not a haven for dissidents The community policies things quite well so at least in theory we are not that sort of site that should be blocked. I have no idea of that arguments going to be persuasive or not. But this is the only one I have to make. Might the community of Wikipedia participants in China though as a community make it a decision to accept concessions in order to see this open. So the external Wikipedia community and its rules and its social dynamic might that community make different rules, and is there room for that - for that in the Wikipedia world. Yeah. So having so, there is a couple of answers to that may be. So one answer is local language communities do make their own policies to a certain extent. There are certain core policies that I set on from the beginning. The neutrality policy is non negotiable I always said. Certain ideas about participation and openness in a vague sort of way, decisions about exactly who to ban and when and why and what kinds of behavior; you know, one of the rules is no personal attacks. But saying, no personal attacks is very culturally variable like what people would consider to be a personal attack or not can vary widely in different cultures. So, they make all those decisions for themselves. So some of the basic things; yes, they can make for themselves. Could they choose for themselves to you know ban editors from Taiwan who don't toe the Chinese Government, "No, that isn't going to happen." But its kind of an academic question you know, in the first place because having met with the mainland Wikipedians, they say basically you know they have no interest in that, they have not asked us to compromise at all. In fact they are taking great pride in. The mission for all of us is really about free access to the sum of all human knowledge for every single person on the planet; they mean it and we mean it and so there is been no desire from the community that we should compromise. If we if they wanted to do that, they can go to Baidupedia who - that's a - Baidu is the one of the big search portals and big portal sites in China. They copy all of our content and claim its all copyright of them which is not true, but and they let people edit and they censor and whatever and almost on one participates because that was just not that interesting. They want to take part in real movement so.