 This is an unusual time, after lunch. I know it's a bit more difficult than other times, so I appreciate your interest in user experience. My name is Marc Miquel. I come from Barcelona. I teach user experience in video games in a university of Barcelona. And I would like to share with you some reflections, some reflections I had during the past five years. So one of my main interests in Wikipedia is to improve its user experience. This is a desire, a wish, that makes me think in many different ways. Because user experience is something elusive. It's hard to define. There are many factors influencing user experience. User experience is how we use things, how we relate to things, how we remember them, how we actually think of them when we are not using them. So there are many different ways of analyzing user experience. And at the same time, there are many factors influencing it. And I would like to share some ideas. And when the presentation is over, to encourage you to discuss them or to tell me what you think. So first of all, user experience in Wikipedia is very important in order to make communities grow. Wikipedia has a huge potential that can expand to the entire education community. So we are far from being as many people as we really can be. And this is perhaps the most important reason that supports improving the Wikipedia user experience. If we grow in number, we can also grow in diversity. And therefore, this might improve also the different types of content that are going to be in Wikipedia. And how can we improve the user experience of any technological project or product? Well, there is a particular framework or a philosophical perspective which is called user-centered design. User-centered design is a framework created by Don Norman. Don Norman was one of the pioneers in user experience in the late 80s, beginning of 90s. And he proposed several ways that every product design can be better by taking them into consideration. First of them is to be iterative. So to continuously change the design according to how it is used. Second of them is simplifying the task, making things visible. This one is very important. Getting the right mappings. So it means that everything is consistent once we have something that it is explained in a particular way with a color, a button with a certain meaning, being consistent with that color and that meaning. Understanding the user needs was the main motivator and understanding their limitations, either they are cognitive, education, and so on. So if we apply a user-centered design process and we keep on redesigning, we'll eventually improve the user experience. So this is what we do in every technological project, either video games, computer software, and this is what would work as well in Wikipedia. So what are the factors that influence Wikipedia user experience? Well, there are so many, there are many, which is very good because it means that we can improve the user experience by improving very little pieces of the product, of the creation, starting by the content itself. We improve the user experience of the reader. The editor's behavior, that's quite controversial. The harassment, the way we talk to each other, we improve the user experience as well. The norms, the norms also shape the user experience, the usability of the tools, and so on. When these factors are not tailored to the users, there is friction and the user experience is bad. So we need to work in all of them in order to improve the user experience. So what do we know so far? Well, actually, in this talk, I want to bring a little bit of friction, a little bit of another point of view because from research, we know there are many things that are many problems that are not tackled yet. We know for several years that experienced editors, they tend to block newcomers, contributions at content and policy levels. We know that reverts to newcomers, they drive them away. And at the same time, we know that harassment is also another situation that drives new users away. And these problems are known for some years, but still they are not tackled properly. All the things we know from research is that there are some ideas that they work. They work because they have been tested, like trying to get participation from readers in order to engage them to become newcomers, allowed to send minimal feedback on the quality content in order to engage them as well to become newcomers or to assign tutors in the tea house or village pump in order to facilitate their entrance into the Wikimedia sphere. All these things, they have been approved by research, but somehow they remain unimplemented. They do not find its right implementation in order to have its most big impact as possible. And unfortunately, there had been some solutions which have been tested that would potentially improve the Wikimedia user experience, like the visual editor tool or the flow discussions, that remain unimplemented in many Wikis. One thing that something as a why do you see when you get editor that it's so common in many other platforms like WordPress should be by default for new editors. So I'm really wondering why this is not happening and this is a question that each of us should ask in order to improve new editor retention. So why this is happening? Why do we know there are problems? Why do we know there are solutions and why do we know there is software improving user experience, but still we are not getting as far as we would like? Well, I try to be analytical. I try to divide the problem into different steps. And I thought that the first possible step is that there is a low degree of knowledge or awareness of the problems by the involved actors. And these are foundation, communities, chapters. The second one, there is a lack of communication between these actors that they are not transmitting this knowledge. So perhaps this knowledge isn't the foundation, but it's not in the communities or it's already in the chapters, but not in the communities as well. And the third possible factor or the last step of this process is that the consensus decision-making mechanisms is somehow blocking these changes or it's possible implementations. I would like you to reflect on which one do you think is more possible. In my opinion, the second and the third are crucial. Okay, so Wikipedia has two unique factors, two unique characteristics that make it more challenging to transform it into a user-centered design process. The first one is the consensus decision-making. There is no central decision-making, so it's harder to implement changes, as I said before. And the second one is the outcome, the resulting product. The resulting product is a counter-centered totally. So it's not centered on the users, it's the user that they have to go around the product. These two characteristics, which are unique to Wikipedia and are essential to make it work, they present very specific challenges in order to implement the user-centered design process I said before. So I'll try to propose solutions for that. So why consensus decision-making is something unique but at the same time challenging? Because in a user-centered design process, we have new ideas coming from company managers that they have a strategy, they want the product to grow, to have more users, to make more revenue. They design, they evaluate with research, and they listen to the research very carefully in order to take decisions and implement them, and they continue iterating in this process. What happens in Wikipedia? Well, in Wikipedia, we have new ideas coming from the foundation, but very often from experienced editors, we have that the research is in the foundation and the decision-making is always in the communities. It's always in a group of users inside the communities. It's a different process. The actors are different. Within a company, the one who sets the strategy is the one who finally decides what to do with research in order to implement the changes or not in Wikipedia. The ones who set the strategy are the communities in the foundation, but the ones who are always deciding about what to implement are the communities. And this makes it more challenging. Why challenging? Because new ideas, they come from experienced editors most of the time. So they tend to create new ideas based on their needs, based on their own needs for the whole time they have been users. And less thinking about the needs of potential new users or newcomers. So what happens? It's much more difficult to set a strategy in long-term. It's much more difficult to propose some changes that not aim at the same users, but at new editors. So it's much more difficult to think to grow, but to think to make happy the same community. This is the consensus model. And it works, but it has this other side. And we might need to look for some ways to improve it. And besides the consensus model, there is also another, let's say, cultural factor, cultural values that I consider very important. Every now and then when I talk to Wikipedia and I listen to them, they really tell me, if I learn it, anyone can. That's very encouraging. I like very much the spirit, but that's not realistic at all. Because we are not the same. Users come with very different education path, very different sociological profiles. So assuming this is one of the big mistakes if we want to expand our community, if we want to extend it. We are full of cognitive bias. When I talk to game designers, I tell them, please do not do the play testing yourself because you know your game, you know it very well. Just give it to other players who don't know your game and you really know, you really see if it's usable for more players. Editors are not the ones more aware of what new editors need. So these values are useful for certain editors, but they are not useful at all for the entire community or the potential users. These values are often represented in controversies. A few days ago, actually in June, I found this post in Wikipedia weekly page in Facebook which presented the debate, open knowledge, mission versus controversial design features in Wikipedia. Whether it's more important the commitment or the features because they are controversial. Well, I think this is a really false debate. It is a really false debate because anything Wikipedia should not be a test of commitment, it should be easy. That's simple. If you have a high commitment, you'll be able to overcome difficulties. But if easier and you have commitment, the better. You're gonna do much more work and you're gonna engage more people. Or should we edit with terminals? There is actually the way to do it, but it makes no sense at all. So we really need to keep motivation, but we really need to make it simple. So this is why getting to consensus is problematic because there can be groups of users or the entire community that eventually might get into a close feedback loop. How some people are going to decide to the interest, to the best interest of these people that are not in the consensus. That was kind of the most difficult questions that I cannot answer. How can we represent newcomers if they are not even in the community in order to decide for technological features that would help them? It's really difficult. And I know I'm getting into the danger zone and this is controversial because taking power and privilege from people is always difficult. So our really community is able to take decisions to the benefit of every potential user. Should consensus be limited to only content? Should we let technology to another group of people, maybe larger? Should we change consensus for technology? I know this is conflictive, this is tension area, but we should really think about it. My proposal or one of the solutions I think that it might be useful in this direction is to say perfect a little bit more the consensus and the community because Wikipedia started as a collaborative project without users, without roles. And roles were created when the needs required these roles, content policies. So I think that we are getting to the point that we might need roles for users, roles for community, roles for implementing changes and roles for facilitating this information flow. And this is why I propose introducing a user experience flag inside the communities that would care for community health, that would care for user experience. This flag, as I imagine it, has three main characteristics or three main duties. The first one is to represent these users, these newcomers who are not in the community yet and therefore cannot debate whether flow is going to be useful for them or whether visual editor should be the default editor. Second of all, it should ensure communication among the foundation and the communities. It should provide the communities the information on what is going on, on what the time is being invested and therefore facilitate that they really accept these changes. With research, with evidence, experienced editors will eventually accept that it's important to do these changes to favor user experience. And third, follow community stats to know exactly what is going on, to know exactly whether there is retention, whether there is harassment and communicate it as well to other users. So we need to represent potentials and new users in the consensus. We need to put these roles at every step. And this way we're going to facilitate the consensus to take into account the right and needed changes in order to expand communities. At the same time, this role should be in the community. It should not be in the foundation. And this is really important because as I've seen there is a distance between communities and foundation and this user experience flag would breach this distance. And this is why it's really important that the information flow is facilitated by this role. And third, as I said, to follow community stats, harassment, new commerce retention, all this should be available to communities. So they take responsibility. There are amazing research papers about retention, about the effects of reverts, as I said before, but communities, they don't know about it. If this information is at hand to communities and may hand, I don't mean outside websites or not only foundation projects, but something integrated and put them as a goal, then communities would take charge of it. This is why this role would bring a sort of secondary goal to the project. We have the big goal of providing the sum of human knowledge into a place. This second goal would be to give the best knowledge experience. And I'm really happy that I saw that we are talking about knowledge as a service because we are right in this direction. So this user experience flag would comply with these activities and would really help in bridging communities and foundation. And with this, I'm getting to the second challenging characteristic in order to improve the user experience. And this second characteristic is the outcome, the other product, the content center architecture. So we know that content is the center of Wikipedia. This is why we create it, but this has consequences. Wikipedia has grown by accretion. There are tools, there are websites for pictures, websites for metadata, it's a universe and it all revolves around content, but it's really difficult to catch up with everything that is going on. And very often we realize there are tools that would have helped us if we knew that they exist when we started a particular task. Okay, complexity brings friction. This complexity is against the user experience. And actually with multiple creators can only create complexity. And you'll say, because I had this conversation before, okay, but this is the power of Wikimedia. We cannot stop that, actually not. We shouldn't stop that. This complexity will continue and will grow, but we need different ways in order to tackle complexity and to balance a content center architecture like Wikipedia it is, and transform it in a more user center architecture. And I'm gonna tell you some ways very fast that they are usual from a user experience perspective. And I'm happy that some of them they are already taken into account, but it's good to have them in mind. So there is a user experience tool which is called user journey, which implies dividing every part of the experience into different steps, starting in Wikipedia would be before registering, after registering, while contributing, while engaged and re-engaging. So first of all, it is really important to realize for new editors that it is possible to edit. And by putting in bold the edit label is not enough at all. There are still a big percentage of users that they do not know that it is possible to edit and there is new research supporting this fact. So we should look for ways to make this possible. We should make it more visible. Actually, one of the misconceptions or the ideas that I think that we should debunk is the division between editor and reader. This division makes it more difficult for readers to become editors because there are many different profiles who are contributors and they are not the typical superpower editor. There are researchers, there are photographers, there are event planners. And by saying editor-reader, we are making it a bit more difficult for this to be part of the community. So what I'm proposing, as I said, making things more visible. Few years ago, perhaps we were a bit afraid of showing that Wikipedia is a living thing but because of reputation, because of credibility, but I think that more and more we need to visualize all the things that are going on. The community events should be visible to newcomers because they might be engaged, not into editing itself, but into following what is going on and perhaps contributing in some other way. I'm thinking about the homepage as a page much more diverse than it is now. It could include calls to action for uploading pictures. It could include community events, topics that are missing in a language or translations. These are tasks that they would really call the very diverse motivations that users have because editors, they have very different motivations. So if editors, they have very different motivations, why should we all see the same page? And this is what I mean by making it more visible into the homepage. So after registering, once we have already created a nickname in Wikipedia, we need to teach users to create content to contribute somehow. In video games, we use something called onboarding plan which teaches players each of the steps that they need to play the game. But games are so complex and some of them, they have game plays for 30, for 80, for 100 hours, which is not very different from what many Wikipedians do for Wikipedia. And not all, they go along the same learning path. So what this onboarding plan proposes is to divide each mechanic into a step. So first you learn to move the character, then you learn to create an object, you learn to shoot if it's a game about shooting, you learn about creating tools. So imagine that for Wikipedia. Why does a photographer needs to learn about a particular policy? Why should all users go along the same learning path? Sometimes some of my students, when I tell them to contribute to Wikipedia, I say, wow, but this is like a jungle. There are so many things to learn and policies and they have the feeling that there are so many things I don't know where to start. They feel overwhelmed. Onboarding plan helps people not feel overwhelmed because they are carefully designed to set an order of learning for different profiles. And the most important, no time constraints. It must be an environment that is meaningful, so it's integrated in what they are doing, but at the same time, it should not put pressure. This is how a good onboarding plan works. So let's say that they already learn what they need to learn, whether they are photographers, typographers, translators, which have a role, and they want to start contributing. So by contributing, we can be motivated to contribute, but if the system itself facilitates us what we need, it's gonna be more usable and therefore the motivation will not need to be that high. And this is why I propose providing assistance or providing further actions in order to perfect what we are doing. If I'm a new editor, having a space in the screen telling me which policy might apply to this page or which other article might improve what I'm doing, would really facilitate my process of learning. If we make things visible, it's gonna be easier to use. Recognition, we often say, it's easier than to recall. And this is especially important during the first days because we haven't learned it so hard, so deep. So we need to recall it continuously, to recognize it. I mean, recognize it continuously on screen. These tools, they would show the policies, they would show the templates. Fortunately, in the foundation, they are already working in this area in order to improve quality features of articles. And I just remember one of them proposed in the hackathon of introducing pictures from the same article in other languages in order to extend the article and improve its quality. Further actions is really something helping user experience. So while we are engaged, and we already know how to move around Wikipedia, there is this big problem with complexity. And Wikipedia is good at finding content. It is easy to find articles, pieces of articles, but it's not easy to find collaborators. It's not easy to find available tasks based on your profile. It's not easy to find some particular parts of the communication, what is going on in the community. So we need more recommenders. And this is a similar idea to the one before. We need recommenders that they make this thing easier to find because we often need to go outside many other pages. And the last part, re-engaging. We might leave Wikipedia for a while, for months, for years. And our memory is valuable. We have a learning curve and a forgetting curve. So after several years, we tend to forget what we were doing, what we were using, who we were collaborating with. So knowing this, we should provide tools in order to catch up much more easily. One of the best tools, and proven in many other websites, is a dashboard, a user-centered space. This is not new, but it would really improve editor's life when they haven't been active for some period. It could include page creation stats, Wideslist, some research to follow activities from the communities. A dashboard would be something that would be useful for all different profiles in order to personalize based on their needs. With these measures, with these different ways, we could improve user experience. And I think that by creating a community experience role, we would really facilitate it because we would have someone that would represent newcomers, would represent users' best interest in the consensus decision-making, which is where decisions are really taken and where it matters. And they would ensure communication and follow stats and communicate them across the communities. So thank you very much. I would like to hear your ideas, to really understand what, how this could be implemented, who should I talk to, and really, how this should start. Thank you very much.