 Alright, so the meeting is being recorded. Why is it a perfect segue? Because one of the really bold things about Wikipedia is its universal character and its universal aspirations. And it's been holding that aspiration since its inception for the last 20 years. And we see that both in the content, it's an encyclopedia, which the very word encyclopedia means complete knowledge. So there is this spirit of universality there. And also in terms of the user base, Wikipedia tries to be universal, is free for all, and it tries to be universally accessible. And the only thing that sometimes hinders that is censorship. But the aspiration is one of universality. And it's also obvious in the vision of the Wikimedia Foundation, which is to empower everyone to share in the sum of all human knowledge. But there are important hurdles towards achieving this universality, both within language versions and across language versions. Within language versions, we see a big excuse with respect to important biographical variables of the contributors. For instance, I don't know what the latest number is, but at some point, 90% of editors identified as men. Even in readership, rather than editorship, we see a majority of men. Across language versions, we see huge disparities with respect to the size of Wikipedias, with the number of contributors, also with respect to the quality of content. We have some versions of Wikipedia that have been hijacked by people with a certain political agenda. So that creates disparities. Lowering these hurdles and decreasing these disparities that hinder the universality of Wikipedia is a really important problem for the community. In fact, it's at the very top of the agenda for the Wikimedia movement. In the Wikimedia movement's 2030 strategic direction, knowledge equity is featured as one of the two main goals. The other one being to enable knowledge as a service. I'm quickly quoting here from the movement's direction. It says, as a social movement, we will focus our efforts on the knowledge and communities that have been left out by structures of power and privilege. We will welcome people from every background to build strong and diverse communities. We will break down the social, political and technical barriers, preventing people from accessing and contributing to free knowledge. The importance of this mission is what inspired the conversation that we're about to have. That conversation will be under the motto towards a worldwide Wikipedia. The snippet from the manifesto that I just quoted is very broad and it's inspiring, but there's still a long way to go. And our goal today with this conversation is not to discuss further Wikipedia's universal mission at an abstract level, but rather to dive into the matter at a tangible level from the point of view of concrete real-world scenarios. That's why I added a subtitle to the motto, which in its entirety is towards a worldwide Wikipedia one step at a time. And this is also why we invited the two guests whom I'm about to introduce to you. They're both not only visionaries when it comes to connecting people all over the world to knowledge. They're also people who have gotten their hands dirty on the job. Before I introduce these guests though, I would like to spend just a minute thinking about what we want to achieve in this conversation. What is the criterion that we should meet in order to be able to say that the conversation was a success? And what I propose is to pick one concrete place on earth, let's say Kenya. And to imagine two concrete people in Kenya and think about the problems that they're facing today linked to Wikipedia and how we could overcome those problems one step at a time. The two groups of people that I propose to think about are supposed to represent Wikipedia's two main user groups, which are readers on the one hand and editors on the other hand. So let's think of Richard the reader and Esther the editor, two Wikipedia users from Kenya. By the way, I'm not proposing Kenya because it plays a particular role from the Wikipedia perspective, but in the spirit of really keeping us grounded and concrete and also because our first guest is from Kenya. Her name is Catherine Adea and she is the director of research at the World Wide Web Foundation, where she is responsible for coordinating a research team dedicated to understanding and removing the most important barriers to achieving the vision of a web that is safe and empowering for everyone. She began her career as a research fellow at the United Nations University's Institute for New Technology in Maastricht in the Netherlands. She then moved back to Africa in order to expand opportunities and to drive development there using the web and digital technologies. And she has pursued this mission in a very broad sense within the public, private sectors and within the civil society sector, as well as with international organizations. So welcome Catherine and greetings to Nairobi from where you are joining us today. Our other guest is Danny Vrandicic. He is head of special projects at the Wikimedia Foundation, where he leads the work on abstract Wikipedia and on wiki functions. These are two new Wikimedia projects that Danny will tell us about himself in a bit, I hope. Previously, he was with Google where he worked on the knowledge on the Google Knowledge Graph. And before that, he received the PhD from the Karlsruhe Institute of Technology in Germany. Many of you may know Danny as a founder of Wikidata, the knowledge base that today backs important aspects of Wikipedia. And in the keynote, Yolanda also gave a brief shout out to Danny already. So we are all primed. Also Danny was the founder of the Croatian version of Wikipedia. So welcome Danny and hi to Berkeley, California from where you are joining us today. So to open up the discussion, I just want to kind of throw some quick speedballs at the two of you. Catherine, one of your missions with the World Wide Web Foundation is to bring the web closer to people all over the world. What role does Wikipedia play in your work? Can I at least just even say hello and invite? And the fact that I've never really had a lot to do with Wikipedia and I think I've been, I feel like I've been thrown under a bus, but I'm willing to be under that bus for this particular meeting. With a lot of people in it too. So you know, when I think of this question, the first thing I ask myself just very briefly is would Wikimedia exist without the World Wide Web? So when Satim invented the World Wide Web, he did it out of love to make the web for good. And I think I read some in one of the Wired magazines that Wikipedia was also built on love. For that alone, Wikipedia has a role to play as we aim for a web for good. And Wikipedia's goal is really to provide the sum of human knowledge. Again, I've read this somewhere available to everyone for free. And at the moment, it is already one of the most successful collaborative efforts in the internet, but with many challenges that I think we'll discuss as we go along. So just a quick one, I'll just give you high level. Our interest in the World Wide Web access and affordability. Women's rights online. Remember, women are very unlikely to be online compared to men. We have data. Data rights. First is critical. People are to fully embrace the web or even to fully embrace Wikipedia. So with those, Wikipedia has similar concerns, but in line with content and contributions. So we need Wikipedia to assist us, whether at the World Wide Web or us, as even people where I come from, to manage the issues above. For example, without meaningful connectivity, which I'll mention later what it is, basically access and affordability, without more women being online, without more people trusting data on the web, then Wikipedia's work will be in vain. Thank you. Great. Thanks a lot. I'm looking forward to unpacking these issues in what is about to come. Danny, Wikimedia runs through your CV like a red thread. Nearly 10 years ago, you worked with Wikimedia Deutschland, then you moved to Google, and now you're back with the Wikimedia Foundation. What caused your move? The same reason I moved to Wikimedia Deutschland back in for Wikidata from academia. Basically, I believe that the Wikimedia movement is the place to do this kind of projects. It shouldn't be done by a for-profit entity, and big changes to the Wikimedia project should really come from the Wikimedia movement itself and be done by the Wikimedia organizations. Great. Thank you. That's great to hear. It's great to hear that there is such a big role to be played by this community. Let's jump in towards those questions that I mentioned in my introduction. Let's go to a concrete place. Let's go to Kenya. The Kiswahili Wikipedia is one of the biggest Wikipedias written in an African language. It has about 61,000 articles, and it's the second most popular Wikipedia version in Kenya and in Tanzania, after English, in each of those countries. Catherine, I'm wondering what role does Wikipedia play in life in Kenya, on the ground in every day? In your experience, is Wikipedia used in similar or in radically different ways in Kenya compared to other places, let's say Europe or North America? When I listen to your question, I'm trying to debate how honest should I be, how should I balance, but I'm behind the screen, so let me just be as honest as possible. And first of all, by the way, thank you for getting it correct. Kiswahili is the language. It's not Swahili. And the funny thing is this is such a common mistake that is now so widely accepted. I even get confused myself, and it's not your fault because many people now use them interchangeably, but please do remember Kiswahili is the language. And then when I think about our education system, our education system in Kenya is English first. I speak Kiswahili, then we have very many local languages, I speak Luh. Most Kenyans on average are going to speak four languages. And then you're going to find that in many Wikipedia pieces that I've looked at, and I'll be very frank, they're written first in English, then translated into the other languages. So for example, they translate into Kiswahili. And the content is not always relevant. So some people, we may search, but let me tell you, it's not necessarily for many of us in research in this country. Our first go to place, you'll go there and you'll want to verify something else. And I'll give an example, I need to think, like for example, if I think in, I'm Luh, my mother tongue, I first spoke in Luh when I was born. So I always think in Luh, then I had to translate it into English. Then I'm speaking in Kiswahili. So this is where the university's linguistics and languages department could really help. And I wanted to give you an example, Bob and the rest if you don't mind. I always remember, I had a professor in my university and a graduate who spoke Luh first. So he was Luh. Then he spoke Kiswahili. Then he did his masters in English and did his PhD in Russia. So one day in class, I was asking him a question and he said something like, he meant to say, Catherine, clarify what you're asking. But this is what he said, I've never forgotten. Catherine, put me clearly. You know, that was the direct translation from the language. So if you don't get it right and you mix, you need to get the experts to be able to help you think deeper into what really you need in that language and what the context is. And you've even asked me about Europe. I lived in Europe. I've lived in Kenya. Used in Kenya more in Europe than I did in Kenya. And maybe it's because the content that I was finding in Europe was even more relevant, more factual, more correct. In Kenya, there is so much content is wrong. That's why I said I feel like I've been thrown under a bus because I was one person who had parked on the Wikipedia site. When there's content of very key people, for example, that are very relevant, that is so wrong. I'll give an example. One of the great Pan-Africanists in Kenya, Pierre Lolo Momba, is still written that he is, I don't know what he's teaching at the law school where he left three years ago. And he's doing so many things in the Pan-African movement. I would like to talk about my father who was a famous treasurer, that's a whole different story. So I think my point here is that there's so much content, there's so much that is available from legitimate sources. So do we go to Wikipedia first? The jury's out. What I may advise is that we must invest in research. We must invest. Voluntaryism, is there a word like that? To volunteer, to volunteerism, it has its role. But proper research cannot be compromised, especially for some content, to give it credibility and reliability. And especially a lot of stuff from the developing countries, I'll be frank with you. Like from Kenya, it's quite poorly researched. I'm not criticizing, there's a lot of good material. But when you balance, so when you go to a few and you get poorly written material, you give up on the whole thing. And the second thing is it's very hard to get a lot of good people to volunteer. There must be a balance. And I'll talk about the meaningful connectivity that I was going to talk about, I mean, whenever you want at whatever point, because you've got things like access, affordability, you've got women, there has to be a balance. How do I get these people to contribute? And then finally, you have to focus on issues that matter to that country or to that context. Where is the pain point? Where is the pain point? So sometimes I look at the content, or I look at some of the people who are the key people on Wikipedia in our country. It's all the, what we call them in Kenya, the celebs. So I don't know what you call them elsewhere. So they may just be people who are always out there. And doesn't mean that they're the people who really everybody wants to hear about. So maybe let me leave it there because you could ask me more, because I could talk forever. And I said, you throw me under the bus and we need to be there. You raised a great point. You need to get, you said you need to get people to contribute. And here I would like to turn to Denny because he has this experience from co-founding or founding one of the Wikipedia versions. It was the Croatian one in your case, Denny. So what does it take to build up a strong community of contributors from the ground up? And do you think it's a process that can be repeated all over the world? What would it take to build a luo Wikipedia? It might already be there, but what would it take to flesh it out? That's a great question. And I really don't know the answer to that one. It's not that I went to the Croatian Wikipedia with the intent to found it or whatever. It's something that happened. It was just one of the first ones there and started doing things and became the first administrator and so on. But I didn't have a plan of going there and thinking, oh, this is what we need to do and stuff like this. Just like, okay, let's start. This is an Wikipedia. I want to write those in Croatian. I want to make it available in Croatian and so on. So I really don't know what it takes. What it obviously takes is volunteers. What it takes is for people who feel that they can do this, that they want to do this, that they have the time to do this and to actually go in and do it. And plenty of those. It's not enough to one, two or three. You need a community there to build that. And knowledge creation in the time of Wikipedia is really a community effort. So what you would need for Lua is basically to have people who are comfortable writing content in Lua and who have the time, who have the ability and who have access to the web in a strong way that they can actually go in and write. That's basically what we need. I mean, with the project that I'm working on now with Epsychic Wikipedia, I hope to actually free up more of those resources to make it available. We can talk about this then when we get to it. Great. Yes, we definitely get to that. But maybe harking back to that immediately. Catherine, what do you think, given this scenario that Danny just described, having a good strong internet access, being comfortable writing in a local language, you say Kenya is English first, so that might already be problematic, I guess. How do you see the constellation? Are we set up for success or are there huge hurdles to be overcome? Frankly, go to write is not a problem. Even I, as I'm talking to you, I can write in Lua, I can write in English, I can write in Kisohini. Time may be a different factor, especially to volunteer. So it's getting the balance and getting people to be able to do it consistently. And maybe it requires some strategic thinking on how it can be done. It's not impossible. But then now you bring in an issue that if you allow me, maybe then I'll bring in something from the Web Foundation, which is so critical in countries like, especially in the global south or countries like Kenya. And that is also people who have meaningful connectivity. You know, people who, we talk about meaningful connectivity. Many people may not understand that meaningful connectivity is a tool to raise the bar for internet access. And this is a work that's done at the Alliance for Affordable Internet, which is under the Web Foundation. And we look at it under four dimensions. Regular internet use, minimum threshold, daily use. How many people in the global south, for example, are able to use the internet daily? An appropriate device. Enough data. That's the problem. Like I was very worried. I was really praying that I'm going to be able to have this meeting and I'll not be embarrassed by dropped connections and things like that. And a fast connection, minimum threshold, 4G mobile connectivity. So, you know, you need people to be able to have this. And this is the work we're doing in the Web Foundation to make you understand that, hey, to get access, to get more people to be more included, you need this. And so when I look even at Wikipedia, you want more people to volunteer. You know, what Danny has said is excellent. But if it was Catherine, I would have found it easier to do it when I lived in Europe than when I lived in Kenya. If that answers the question. But I would do it. I find this very interesting angle because it's an angle that we rarely think about. I think I'm speaking for myself. But also, Danny's approach is more at the, I'm thinking of this as a high, as a stack of, you know, you need the very basics. You even need to be able to go online. And then there's a whole layering. And at the very top, we can maybe have software tools that support people in creating content and so on. And Danny has been doing a lot of work in that direction. So the buzzword here is abstract Wikipedia, which is a project that you already introduced two years ago at the Wiki Workshop in San Francisco. So it makes us very proud that we kind of got the sneak preview two years ago. And that back then you were still with Google now, you're with Wikimedia. And you're working with full force on abstract Wikipedia. So maybe, Danny, you can say a few words to or in order to introduce everyone to abstract Wikipedia. How does it work? What purpose does it serve? And how does it connect to this problematic that we're talking about today? The idea of abstract Wikipedia is basically an extension of what we already did in Wikidata to have one common knowledge base that everyone can contribute to no matter what the language is. And then to have this content be translated to the individual languages and be able to be read by any of the contributing languages. So this means that someone can contribute, for example, in Bengali. And it can be read in Lore or a song can contribute in house sign, it can be read in Igbo. And it don't have to think, oh, should I contribute not to the English Wikipedia and have more readers? So should I contribute to my local Wikipedia and write about it? So you're not splitting the community effort in this direction anymore in order to create a kind of baseline Wikipedia. So I don't expect, because this is obviously a difficult problem, so actually the language about this kind of talk, so don't expect us to replace and know the core articles of in each Wikipedia and each culture. But this is like a baseline of knowledge that should be available to all Wikipedias. So that, for example, one of the things that I experience when starting on the creation Wikipedia is I really wanted to write about the island where my mom was from. I wanted to write about a little village she is from. I wanted to write about a little town where my dad is from. But you know, how could I justify having an Wikipedia that has an article about a village of 160 people, but doesn't have yet an article about Nigeria? So it was like, it felt like I first have to describe the whole world before I can get to the things that I actually want to write about. And what knowledge do I have about Nigeria? What knowledge do I have? I don't have any reasonable knowledge to actually write the articles about it. I really could look it up and so could anyone else. I would like to contribute about the things that I was actually knowledgeable about. And this is true for many people in many cultures. They should be writing about their cultures, should be writing what they know about, should be writing what they experience, to write their voices to it. So having a background, a baseline of knowledge available frees up the communities not to write about all these topics. You don't have to write about species in continents far away. You don't have to write about towns in Croatia. You don't have to write about historical people in Russia and so on. But they can actually focus on their own people, on their own histories, on their own towns and villages. So the idea is that everyone can contribute to this common knowledge base because it's written in an abstract way. This is for the name of the abstracting from the specific natural languages and then being translated into the natural languages for reading. So it looks like the normal Wikipedia. But at the same time, it is more comprehensive because everyone is contributing to it. It is more up to date. It's more current because everyone is working on a single date set and set off. We have 300 different languages and who knows how up to date each of them are. And it's hopefully more correct because we have more eyeballs seeing it. Following line is loaded. Given enough eyeballs, all bugs are shallow. So if we have more people working on the common knowledge base, we will have more correct, more current, more comprehensive knowledge available to all of the languages. And then the local communities can still focus on the topics that they really care about and write even more about and stuff that they can't express in the probably limited expressivity that abstract Wikipedia will allow to them. So abstract Wikipedia is really a goal to enable the communities to create a baseline of knowledge and then to add the knowledge, which means there will also be, you will actually find about a lot of more topics in your language than information. You don't have to switch over to another language just because it's the language of education in your country or the language of a higher language that has more prestige or whatever. But you can actually stay in your native language where you're most comfortable with and read the content and access the knowledge. So this is the goal of abstract Wikipedia. So this would be a solution to the problem that Catherine mentioned earlier about the Africanist whose name I would never call it a solution. I would always call it the contribution because we're not going to solve anything with technology. We're just providing tools that help find solutions. And so maybe we can go back to our concrete scenario that I sketched in the beginning. We have Esther, the editor who wants to contribute to the Ki Swahili to Wikipedia in general. And now she, Catherine told us that Kenya is English first in terms of education. So it would probably be unrealistic to think that she only speaks Ki Swahili. But let's assume that there is someone who only speaks and writes Ki Swahili and now they want to contribute to abstract Wikipedia. Maybe we can play this through very concretely how this, and let's assume there is good internet access so the lowest level is solved. But what would be the steps? It's clear that it's not there yet because you're still building this, but in your vision, what would the workflow be? Esther would basically come to a website which is completely translated in Ki Swahili, all the content is already in Ki Swahili. And she could decide on which topics she would work and find, either find gaps in the knowledge base like oh, there are things that I've nothing written about or there are things that I'm missing about a specific topic or find things that are incorrect for some reason. And then she could, using a form-based interface likely, edit the content in abstract Wikipedia and contribute for it. So this would be like the strongest kind of contribution. People who are actually willing to edit the abstract content in Wikipedia and create more in the common knowledge base. The editing is again, completely in Ki Swahili. It's not in English or whatever, but it's just like Perfiki data. You can actually edit everything in Ki Swahili and contribute in it without having to know any of the other languages. But there are many other modes of contribution. Esther could also, for example, if she's not that confident with the exit interface, for example, just contribute a geographical knowledge about Ki Swahili. For example, what are the words that represent a specific concept? What are the forms of the verbs that we already have in intellectual graphic extension of Wikipedia and so on, so that we can use this to actually generate the natural language. The interface would guide her and tell her, oh, if you want to contribute this kind of information, here's what we're interested in. If you want to contribute this kind of information, here's where you can work on it and so on. So it would have all kinds of flows where we hope that we can gather as many as possible contributors and volunteers to help us build the common knowledge base. It will make more and more knowledge available. If everything goes well, she will also always see the impact of her work, like how many people are actually reading the things, how much more knowledge did I unlock with the system that I did with my contributions and so on. So this would be roughly the interface she would be working in. Katherine, what do you think? I think I could jump in in very many different ways. And first of all, I honestly congratulate any form. I read through what you're doing with abstract Wikimedia and I think it's brilliant. I think there's a real opportunity there. I don't intend to take away from it. I think one of the things I appreciate that you've even said is that it is more complex than simply translating the context in an abstract way to the languages. And so my issue would be more of the relevance of the content and the source. But to be a bit more candid and maybe it's good to maybe contextualize a little bit, Kiswahili is not just spoken in Kenya. Let's agree, Kiswahili has a wide reach. And right now I know even South Africa is planning to introduce Kiswahili into the education system. But for those who know, Kiswahili is spoken in the wider East Africa, Congo, Rwanda, we are many. And to contextualize further, when I was saying about my education system, I come from the Generation X. So Generation X were taught Queen's English. Then Kiswahili was an optional subject with French and German. So if you are thinking global, we are thinking about the future, we are thinking where we will be, of course we are going to choose French and German. But now the current, the generation that came after Generation X, Kiswahili is now a compulsory language. And trust me, they are very who prefer to read in Kiswahili. It's just that is there content for them. So I may be very eloquent in English and all that, but there are very many others. In fact, the Kiswahili newspapers, I look at, there's a man who drives me. He only rushes for the Kiswahili paper. And when you look at their content, their content is sometimes a little different to suit what they're interested in. They're not interested in all these great definitions of abstract media or what they want to know is what is this, you know. So I think what I'm basically trying to say is the example that Denny has given is very relevant. But when I think of robots, I don't know how much time I have, but I like the way you thought of Esther and Robert. And I wondered how real these people would be. And maybe how much the Wikipedia has consulted and edited people, Robert and Richard and Esther. The reader and the editor. I think, sorry, I think Richard, Richard is actually more real than Esther, in my context. But Richard gets disappointed often, depending on his region, country or locality. His disappointment is sometimes with Esther. I found myself writing that. She tries hard, but sometimes her good is not good enough. And there are many reasons for that, which we've raised. I'm seeing people raising them in the chat. There are many we're aware about them. But Richard may want to read more, but meaningful connectivity may be a challenge, which I've said and other issues as well. However, it is more of a challenge actually to Esther, who wants to be able to provide quality but has other challenges as well. There may not easily be addressed by the current Wikipedia model. And that's why I actually like this abstract, the abstract Wikipedia that he's talking about. So I think it's time for Wikimedia as it implements its excellent study of knowledge gaps. And even having just from what we have right now is a real opportunity to bridge the gap between Richard and Esther. And I like the fact that you use Richard and Esther. They could even court each other more closely and maybe you could have a marriage very soon. Thank you. That's a, I think it's a great hint because you're alluding to these, to the human aspects of knowledge creation, right? There is the bare facts, but then there's also the people that you need to have in order to get the work done. And so I'm wondering to what extent abstract Wikipedia can help with that. Then do you think that will it also help in creating communities of humans who co-create knowledge? Or will it mostly be a tool for making the humans that already do it more productive? I really hope that we will actually grow the number of Wikimedians working on the different projects. We started with Wikidata where like more than half of the people contributing to Wikidata actually have now been contributing to Wikimedia projects before. So we did have an influx of completely new contributors. And I hope to achieve that with abstract Wikipedia and Wikifunctions as well. One of the reasons is actually that I believe that we can maybe achieve that is that currently if you look at underrepresented languages, if you look at Lua, which as far as I understand doesn't have to be a media language addition, you know, why doesn't it have one? There are plenty of speakers of it, but for many of them, and it's the same thing with creation when I started with it, it has changed since then. You don't necessarily believe that you're doing anything substantial here because how could you write a whole Wikipedia? You can't, right? And if you don't see enough other people contributing, you don't think this will never go anywhere. It's like an open source project where you're completely trying to stand by yourself and it's a big project. You can't. You need more contributors. You need the confidence that it can actually work out, that you can have a comprehensive Wikipedia in your language and so on. Otherwise, you're not actually willing to contribute and you might think it's actually much more effective if I contribute in English or in Kiswahili or in any other language, which is already there and bigger and has a much bigger chance of succeeding. In fact, more than half of our language editions, about 150 Wikipedia's, have fewer than 10 active contributors and you know, 10 people in their free time. Can they really build a complete and Wikipedia? That's challenging and the belief of that might just keep more people away from it. With absolute Wikipedia, I hope that we can, you know, create the perception and the actual substance too, but also form a perception that we will have a comprehensive Wikipedia in your language, in your native language. You don't have to switch to another language. You don't have to give up and just, you know, go to Facebook and check with your friends instead, but you can actually contribute to a project which has a fair chance of succeeding, which kind of gives you more energy because you're a bullseach because you say, yes, we can do this and you can see the places where the knowledge is missing, you can go in and contribute to it. So the hope is that we can actually change, you know, the whole incentive mechanism around the Wikipedia, particular around the smaller Wikipedia projects in order to make it more believable that your contributions are actually substantial, that your contributions can actually make a big difference. And once we have more content, we will probably show up in more searches and people will actually be confident about, you know, going to Wikipedia and search for the knowledge in their language. I hear so often stories of people who are, who are speaking a smaller native language and then they say, well, I don't even look up the knowledge in my language, because I can read English anyway, I just go to the English Wikipedia, why should I do it? And, you know, the Israeli Wikipedia, the Croatian Wikipedia or whatever, you know, the same thing is happening in many other languages. There's a map, if you can find it and show it, I have trouble with the spelling stuff, which shows which language Wikipedia's are the ones who are most accessed in each, in a given country. And, you know, the list of those languages which I access around the world is pretty small. So in many, many countries, people access a different language addition than what is actually even the standard language of their country. So here it is. So you can see even in places like my native Croatia, English is the most accessed to Wikipedia. If you go to Central Asia, you see Uzbekistan, Turkmenistan, Kazakhstan, the other countries, all using the Russian Wikipedia instead of their own native Wikipedia, which exists, which has times a thousand of articles. In Africa, you can see the history the Europeans imposed on Africa very clearly and which countries are using English and there's almost, there's no language that uses Kiswahili, for example. There's no country that uses the Kiswahili Wikipedia as the main Wikipedia. And I really hope that abstract Wikipedia will eventually make this map much more colorful, much more diverse. And in this case, actually much more gray because gray is the one switch. And that we will have access to knowledge in many more languages that we currently have. It's a good point. I even have the stats for Kiswahili. Even in Kenya and Tanzania, it's 96% usage of English and only about 4% is Kiswahili, which is the second most viewed language version in those countries. I think we could discuss for a long time, but at this point I want to open it up to the audience because I saw by like with one quarter of an eye monitoring the chat a bit, I think there are a lot of questions. So Isaac, do you want to pick a question and throw it at our panel? Sure. I think I'll just go with the first question, which goes back to the very start of the panel and was sparked by a comment Denny made about the distinction between industry and media. And the question I think goes to both of you though, which is what roles do you envision for university labs and shaping Wikipedia and the web in general? Yes. University labs have, based on my experience, have the great opportunity. You can actually sit down and think. You can go and analyze problems. And I see university labs always having a multitude of responsibilities. One responsibility is to raise problems, to find and describe them and to actually tell us, look, this thing is not going well. One responsibility of university labs is to try out crazy ideas to see how they can work and ideally plug them into existing systems and see how they would actually change and benefit from that. But also to think beyond the short term cycles in industry, for example, to think with an eye on what changes that are really profound could be made if we change certain basic ideas and foundational systems. But also on the other side, for example, then systems like Wikipedia or wiki functions in the future will have a lot of interfaces where you can plug in results from the research labs and see how does this work out. And I don't expect research labs to provide production-ready code and production-ready systems, but I would hope to find ideas, inspirations that can be taken up by systems like Wikipedia and implemented there again. It's also don't think that Wikipedia is all production-ready code. We actually have a vast ecosystem around us, which has come from volunteers, contributors, and so on. And the code quality is very uneven in this space, and we're still somehow managing to pass by. So there's a lot of opportunity for research labs to have major impact and to try out ideas. The lovely thing about the Wikipedia project is that we provide so much data and make it available, which can be analyzed. It's not just the content in Wikipedia. We provide every single edit, every single history. All of these things can really be used to look into it, to see how the incentives work in the world. What are the correlations between contributions from a specific region and, for example, the demographic and economic numbers in these regions? How do specific changes in the software affect specific demographics? All of these things, I would love to have research working on. And depending on your background, there's so much more you can do. Can I comment briefly on that, Robert? Yes, absolutely. Because I've got a background in academia, and I've jumped from academia, come back outside, but I now see, not I now see, the role for academia is clear, but it's just that sometimes, especially in some of our constituencies, it's not very well appreciated. And the question is asked, I'm thinking more, my mind just went to an African port. You know, an African port is the place where people used to convene around as somebody's stirring and cooking things. That's why I'm seeing them. And because they've got, they've got the convening power. These are many of the universities, as he's saying, can convene those people in the grassroots, can convene these editors and the volunteers, can even convene their students. They've got the language departments. I just think that universities have such a huge opportunity, especially in the global south, that is underutilized. We're always looking for somebody else first. And I'm not being critical, but we've got some fantastic universities that could actually assist. So I think it's more of a collaborative way of doing it, but I see the port. My mind just keeps going back to the port. That's great. Isaac, do we have more questions? Yeah, we've got two, I think, kind of about abstract Wikipedia. So maybe more specifically for Danny, but I think Catherine had already made some points about this too. So I would encourage her to chime in a few thoughts. But the first one is how to kind of handle the tension that abstract Wikipedia adds this additional layer of abstraction and therefore would seem to raise the threshold to contribute from people from traditionally represented Wikipedia and just thinking about that tension and comments on that. We're thinking about this problem a lot. So we will really throw major UX resources at hoping to resolve that and figure out how to make editing Wikipedia as easy as possible, even with the abstract layer in between. And editing Wikipedia currently is already not exactly trivial. So it's not that we're competing with the easiest editing interface in the world. But still, you're completely right. We will have to work on a UX that is as inclusive as possible. Now, given that the benefits are so much higher, because you're contributing to all of the language additions immediately, we still think that even if the interface doesn't manage to become as easy as editing in a natural text, we hope that this will still be worth read because you're basically editing in several hundred languages at once. So we hope that this angle can help with creating incentives for overcoming the hurdle. And finally, we're not taking away the possibility to actually write a natural language because we're just augmenting the Wikipedia, so we're not replacing them. So you can still contribute in your actual natural language and just write the articles and layer that on top of what you would be doing in abstract Wikipedia. And this will become hopefully more interesting and again, there will be more motivation to do it because you're contributing already to a pretty well-established Wikipedia, not creating a fresh from the start one. So I hope that we will have for different people and the different skills and motivations to have more channels and we have today to actually contribute to knowledge. One thing, for example, is that we are thinking currently very hard to, so the Wikipedia Foundation launched a relatively recently a Wikipedia app on KyOS. KyOS is a low resource operating system, which is as far as I understand it's widely used in Africa, which doesn't require as many resources and as a powerful smartphone as others. It basically works in the feature form. And we are thinking like, can we actually integrate a contribution interface into KyOS, the KyOS Wikipedia app to make it possible for people to contribute even in, you know, with this kind of technological setting and not to have the barrier of you requiring a pretty good smartphone, if you're requiring maybe a laptop even or something like that in order to contribute. So there are many different options where we can have contributions flowing into the system and make it and make more knowledge accessible to everyone. I don't mind jumping in, maybe just also be frank that I used Wikipedia right now as I was sitting here because I wanted to tie in a paper I read from Denny on wiki functions and abstract wiki media. So at least Wikipedia was able to tell me that what wiki functions are and why it's closely related to abstract wiki media. And then from your paper I began to see great opportunities in abstract wiki media, depending on how it's done. Because I think one of the things that I appreciate especially is the wiki functions which has a secondary goal, it's providing a comprehensive library of functions. And so what I really like there is that it's going to enable people without programming background to compute answers to many questions either through wiki functions or through third party sites. I think that's really attractive and accessing the function. What I also found very attractive was that wiki functions and abstract wiki media are expected to drive a number of research directions, knowledge representation. So there is a real opportunity here. But I think what you also appreciate Denny which I so I'm not saying that it's now a criticism from me, it's it was in your own words. You said a major risk is that contributing to abstract wiki media and wiki functions becomes too difficult. So you see the difficulty element there is some of the things we need to demystify and maybe be able to bring it to the level that it could help everybody get to the same goal. And because all of us whether it's a worldwide web, whether it's wiki media, we are all trying to get to one goal and the goal is to bring more inclusivity in the digital, more digital inclusivity. So we try to access, we try to work in collaboratively. So this is really an opportunity. But as you say there's some difficulties in it. I was wearing my academic card and I was seeing opportunities. But I love that you actually criticized yourself. So I didn't have to do it. Thank you. Yes, Kevin, I totally agree. I mean, this is exactly the thing that really, really worries me to not make it too complicated. And I would love to have insights that we already are planning to have UX research and UX design specifically worked with communities and people in Africa, in India and so on and to understand exactly what are the barriers to contributions and what might be the barriers to contributing to our UX and how can we actually overcome those. But I would love to hear from someone with your background as well and have you maybe as a kind of advisor or something for the project to so that we can come to you and say, do you think this can work? And what I actually really love is that you brought up the wiki functions projects, which are having a very terrible hard time to sell. I mean, the absolute Wikipedia project is like, yeah, more knowledge to everyone. But I also think more functions to everyone. This is awesome. You know, this can help so many people. But I'm having a terrible time to actually explaining what our functions, why are they useful to people and to find a way to tell this story, to narrate this thing. And I'm really looking forward to help with how to tell this part of the story. And I'm super glad that you're excited about that part. No, I mentioned I'm under the bus. So if you want to join me under there, I might demystify what I'm thinking about in terms of wiki functions. If you don't mind, Robert, I'm seeing somebody asking a question on the abstract wiki media and is there a plan to monitor the content bias? We know that every language edition has its own biases. Is there a risk that the largest community's vision on a topic can spread in all languages? Correct. The risk does exist. The risk definitely exists. And I even see it even in my own country where there's certain languages that just or certain categories that might overtake and slowly push away the smaller ones. Then you're forced to think in a certain box. I think this is an excellent question from Tizano, Picardi, and what you've written there is correct. I do agree with it. It's something that has to be considered. Thank you. So I think the energy is rising, but unfortunately it's the time to wrap up already. And let me do this by throwing another ball at each of you, which is kind of to reflect a bit on... We set ourselves this motto toward a worldwide Wikipedia one step at a time. So now one hour later, maybe we can take... It can't be expected that in an hour we solve a lot. But I would like to ask if there are any new perspectives that the two of you might be taking away from this today, perspectives that you hadn't seen before? Because I think if that's the case then we would already have achieved something good. Then you want to go first. I think one of the things I'm definitely taking away is digging a little more even in what you're doing in abstract media. And in being able to be in this panel, I've come to appreciate a lot more because I found myself doing a lot more research, which is very interesting because it was such as like new work in this area don't know as much. And I think about this discussion in Wikipedia, especially in the global south, it means there's some disconnect that we need to bridge. So I know one of the objectives that we keep media has is to scale areas that haven't been well represented, but to do that you need resources. Not all volunteers are researchers or editors. I'm just being frank, but I appreciate the effort they have made. I mean I've looked... I've written, I've even looked at Swahili, Wikipedia. There's some fantastic stuff there, but I'm sure they're stretched. There's a lot more that could be done. So resources need to be available. And so I really do appreciate. I'm not criticizing, but I'm just saying not all of them. And we need to get to the next level. And to get to the next level, we have to be brave about it. What is not working right? What do we need to do? Do you just want to move to the areas that have not been... We've not been to before because we just want to be included or we just want to say all languages matter or all people matter. Let's do it. Let's be serious about why we want to do it. And I'm sure we have. That's why we're having this conversation. And finally, find out, especially these areas that we want to scale to, the countries and all that, what do they care about? And then we work to deliver what they care about, in whatever language. And remember, those with cheaper and easier access can volunteer more. I'm done. Thank you, Catherine. Denny, your last words. Thank you. I named it a lot from Catherine today. And I will actually go back to the recording in order to unpack, particularly your opening statement, in order to really understand the situation of potential contributors in the region and to see. I also really much appreciated her shouting out to Wiki Functions, which I was planning to just not introduce today at all, because I thought this is a little complex and so on. And I'm taking away that I really should work more on this narrative and figure out how to tell the story, but also to keep a really mind on, you know, and this... I'm saying Wikipedia won't be a solution, as I said earlier. And there's so, so many layers of this, the access to the internet, the perception of the different languages by the individual speakers and the understanding of the potential contributors and readers of what an encyclopedia actually is. So all this and how they expect information to be packaged. There's a lot of data unpacking. I saw the one I following a little bit in the text. There have been a lot of great comments in the chat and I want to take another look at those and see what's going on here. And also to be reminded to leave enough interfaces to work with academia in particular and research approaches in academia. So I have to really make sure that there are opportunities are visible and opened so that people doing research can think, how could my work actually fit into that and make the most impact of the project? So there's a plenty of stuff that I'm taking out here to work on. Thank you very much both. It's been a great pleasure and honor to share this panel and to kind of moderate this conversation. It's been, I've learned a lot and I'm sure that it's the same for a lot of people in the audience. So thanks a lot for a very long day and Danny for getting up at the inhumane time of 9 a.m. So thanks a lot and I hope to be