 team. We're a relatively new team, and I wanted to share some of some information on the sort of direction that we're going in with our team and the work that we have been doing and thinking about doing and kind of just have a bit of a discussion about your thoughts on that, your feedback, your input, try and help guide this work from your experiences. So this is going to be a fairly interactive session. So if you're in Feedloop or YouTube, please join through to Zoom on Feedloop if you want to participate. I'll try and also look at Feedloop chat and like the etherpad and stuff, but I think if you're in, if you're in Zoom, that's going to be ideal. So the title of the talk was about improving mobile web for experience editors. The sort of more specific lens that we're working through is talking about content moderators. And so let me just quickly tell you what I mean by that. So for us. Oh, thanks, Andrew. Yes, I'll, I'll try to speak slow. So for us, content moderation means processes which review, report, or make changes to the contributions of other editors. So that might mean processes like patrolling recent changes, reporting bad articles, using admin tools, undoing edits. This isn't just limited to administrators, if I say moderators. I do also just mean active editors that are kind of doing things beyond just writing article content. I want to make it clear that to some extent, all editors are moderators, you know, most Wikimedia contributions are building on the work of others. And so to some extent, when I'm talking about more moderators or helping moderators, I'm also just talking about helping editors as a whole. So earlier in the year, we published some user research, we spoke to a few dozen editors across a bunch of different Wikimedia projects. And we published this report called content moderation in medium sized Wikimedia projects. So we hear a lot, I think, from the biggest Wikimedia communities about their problems in terms of content moderation. And we wanted to make sure that we were also thinking about how these sort of slightly smaller communities that are still needing to do a lot of content moderation work, and try and figure out how they were experiencing that and what problems they had, so that we as a product team could sort of figure out some ways of helping there. We focused that research in the Tamil Wikipedia and Ukrainian Wikipedia communities. But as I say, we did speak to editors from a wide variety of places, as well as reading a bunch of previous research and requests and that sort of thing. And the main thing that we learned was that mobile web was pretty bad for content moderation. So we have this project page on MediaWiki that you could go and check out. We basically decided that we observed that as we were speaking to all of these content moderators and active editors, they were either saying, I don't want to use mobile web because it kind of sucks or I can't do the things that I want to do on it. Or if I do use my mobile phone, I sort of switched to desktop mode and I pinch and zoom through the desktop skin. And I'm also an active editor on English Wikipedia, so that didn't come as a surprise to me. I always exclusively edit on desktop. But I think it speaks to an area that we're really sort of missing in terms of our focus because something like half of our page views to Wikimedia projects are on mobile. And there are a pretty wide range of communities, whereas many as 40, 50, 60% of editors are primarily editing for a mobile device. And if they're getting frustrated, if they're not being able to do the things that they want to do, then we're missing something in the sort of experience for them that we can hopefully improve. So I'm going to get to a point in a minute where we can sort of chat about this a little bit, but I just want to quickly explain some of the work we've already done over the past three to six months or so. The first project we picked was a fairly small one. There's this sort of overflow menu on mobile web when you're on a page. There's obviously an edit button, a history button, the watch list button, but there's some other functionality that sort of gets hidden away behind this drop down menu. It includes things like moving the page, page information, linking to the Wikidata item. It also includes administrator tools like protecting the page. And while that was already implemented, the only way you could get to it was by turning on the advanced mobile contribution setting. And so we thought that that maybe wasn't ideal and there might be a way to bring that to all editors so that they have access to those sorts of features like moving a page without having to go and find some setting that isn't hugely intuitive. And so we made some improvements to that menu. We added the block user button if you're an administrator and you're on a user page. And we also did some research with newer editors and found that menu wasn't very confusing to them. It was kind of understandable. And so we decided to roll that out to all editors. So now if you're a logged in editor and you have a user account, then you should see that drop down menu on every page now. The second thing we've been working on is preferences. At the moment on mobile web, there is no link to user preferences, which kind of surprised me when I realized this. You can open the search bar and type special preferences and then go through to the page. But that's obviously very unintuitive if you don't know that page already exists. And that's obviously a super critical page for editors. It includes functionality like changing your password, changing your email address, updating your notification settings, turning off banners. And so we thought that even though maybe it isn't specifically a moderation feature, that seemed pretty high priority to us. And so we're currently working on a new design for that page. And then we're going to make sure that it's linked in the interface so that users can kind of access that page and take a look at it. So I've seen at least one or two questions in chat already. And I just want to kind of open up for maybe quick 10 minute discussion on mobile web in general. I don't know if you want to put this in Zoom chat or in feed loop chat, but maybe raise your hand or something if you try to use mobile devices to edit on a regular basis. And then I'd love to hear what your experiences are. What problems do you face? You can feel free to either put that in chat or unmute if you're on Zoom. I do see there's one question already about whether we include blocking users in the scope of our team. And I would say not specifically. There are other product teams at the foundation like anti-harassment tools and trust and safety tools that kind of spend more time thinking about blocking users. We did end up adding block user to that dropdown menu because it was missing. But I think any further than that we're unlikely to sort of make changes to how users get blocked. So that aside, feel free to unmute or stick in chat if you have thoughts about your experiences of using mobile web at all. So Cintra Abdul says, I use mobile devices to edit. I'm curious what kind of edits do you make? Are they sort of big edits? Are they small edits? Are there typical activities you do on mobile that you don't do on desktop? Randall, go ahead. Hi, Sam. Nice to hear from you. Yeah, generally, as a more experienced sort of user, of course, there are sometimes when you have sort of don't have access to your computer and edit from mobile phone is sort of very necessary. But yeah, it's generally nice to see sometimes, for example, now a Discord server on Ukrainian Wikipedia, we have this user that only edits from mobile phone. He's kind of a new user. So all the time we hear some like sort of arguments why is this there, not there, something like that. So yeah, it's kind of it's important to get sort of this perspective of new user maybe. And yeah, it's it may not be very obvious sometimes. But yeah, yeah, generally, we need to make any improvements, I think, in mobile devices. Yeah, it's nice, nice work. Yeah, thanks. Yeah, absolutely. Thanks for sharing. Abdul Rafi, do you want to speak? Hi, Butch, go ahead. Okay, I posted my question on the chat. Actually, this is an ongoing issue that I've gone through. This graph chart, I've been already made about 1000 articles using this template and I cannot I am not comfortable undoing this. I'm using this graph chart and it's visible on the mobile browser, but on the on this Wikipedia app, it's it's not visible. So charts can be seen there. Gotcha. That's good to know. Yeah. And so this is an important point is that we're the scope of our team is to think about mobile web. So through a browser, the apps are kind of distinct and there are other foundation teams that work on those. But I'll make sure to take that to them and see if there's not already a fabricated ticket or something that they could look at. I'll make sure they know about that. I'm just scanning through there's a whole bunch of comments on on zoom. Yeah, Michael, you mentioned that you don't enjoy editing from mobile because it is kind of limited in terms of functionality in the interface. I wonder what kind of functionality do you find is missing? What is it you're trying to do that that you aren't able to do? For example, I handbooks, but when I try to switch into the editing mode, I seem to only have access to only the sand, the info box plus the intro. But the general article is not visible on the device, even when I try to pretend the orientation of the screen. Still, I seem not to really have the content, entire content of the article displayed. And as well, the font screen really is the interface is really small if you are really, if you really want to do like reasonable editing and any other projects or data. I believe it's something because font screens are always small in size. Yeah. Yeah, absolutely. I think I've seen discussion on that as well that on mobile you can only edit sort of section by section. So if you hit that main edit pencil, it actually tries to get you to edit that like top section and you can't edit the page as a whole. I think that's something we could definitely look into because yeah, I've also had that frustration when I tried to edit on mobile. Renvoi. Yeah, I just wanted to sort of find some maybe at least one example of things that should be maybe looked into. Yeah, that this user was sort of reporting to us. So yeah, it's interesting, but I don't really know how because I didn't really do it myself on mobile. But in fact, it is quite hard to find this sort of move from Wikipedia to Wikidata from mobile phone. Yeah, there was some mistake in sort of template infobox on the right side of the like, yeah, if you're on mobile, not maybe from the right side, but yeah, so user would try to sort of change that and yeah, moving to Wikidata wasn't as easy as maybe should have been. Maybe, maybe not. So yeah, that's something to maybe think for. Yeah, thanks for sharing. Jung Jin? Yes. Hi. Can you hear me? Can you hear me? Yeah, we hear you fine. Oh yeah. So I'm Jung Jin from Korea Wikipedia by the way. So I have several experiences that share I'm not sure it's gonna be maybe duplicate because I joined the session a little bit late. But yeah, so what I was feeling about personally about mobile web is like, I can, as an administrator in Korea Wikipedia, I cannot able to use my word of control the administrative access to the mobile. I know the editor side, I'm pretty sure other peers explained many concerns, but like the administration side, I know it is like the most major problem because like for example, if I want to protect or block, it is kind of like very messier interface so that I cannot able to using it and I cannot able to using like somehow the desktop mode. So like the how I using the Wikipedia mobile is like the I just like to manually convert it to the desktop mode and using it that way because of the administrative action. I know Wikidata is another problem, but I know previous person already spoken, so I'm not gonna do repeat again and yes. So the biggest problem also is the mobile interfaces basically just look like just like the reader readers interface and I don't feeling like that does not encouraging any people to edit or like that's not giving the throwing the message that you cannot you can able to editing or those those things. So I think that's kind of important concept in order to do in order to get abstract the mobile especially because nowadays many people are using just only the mobile not the desktop interface to just like the wrapping and the contribute over the project. Yeah, I think those are great points and it's got it segues on great to what I'm about to speak to. I just want to speak to one question that Amir put in chat about the strategy or planning to have a responsive interface that is you know not a separate website you know at the moment we have this dot m and there's this very hard line between sort of mobile and desktop. I think that's a discussion that we should have and where our team is talking to the reading web team who are doing the desktop refresh work at the moment. We've been speaking to them quite a lot about this work and I think that's something we're going to keep talking with them about. I think they'll be in the best place to work on this rather than our team we have quite a small team. But yes I would love to see that as well I think it would make considerably more sense than this sort of hard line desktop versus mobile paradigm that we have at the moment. But okay thanks for sharing your input I'm going to move on a little bit here and actually build a bit on what Jungjin just shared which is the mobile interface is very much a reader focused interface. I do personally agree with you there and I think that's because that's what a lot of the work has been focused on in the past on mobile web. We've historically at the foundation I mean been thinking primarily about readers when we've been making changes to the mobile site. I will say the reading web team did some great work on advanced mobile contributions a few years ago but we kind of want to build on that and see if we can make this better for editors as well. So one of the big questions that we have then is like how can we increase the number of mobile first content moderators and if reading is good on mobile and you know I think making your first edits is also pretty good on mobile we have the growth teams structured tasks that you can kind of engage with and finding the edit pencil I don't think is too difficult. But past that you know how do you then become a member of the community and an active editor and a moderator. And that question is partly spurred on by some data we were looking at. So we pulled some data on desktop first editors and mobile first editors so by that I mean editors who make more than 50% of their edits on desktop or mobile. First of all you can see that there is a lot more active editors on desktop than mobile which maybe is a concern in itself you know if we have half of our page views being mobile readers then I might personally expect that number to be higher for mobile. But even off those active mobile editors they are much less likely to participate on talk pages much less likely to participate in project namespace pages much less likely to have an advanced user group like patroller or rollbacker and almost no administrators primarily edit from a mobile device. And yes these are numbers from all projects combined. So of the $3,200 administrators 99% of them primarily edit from a desktop device and that number really stood out to me. It really feels to me like if you're a mobile first editor you should have you know something approaching an equal opportunity to become an administrator and be successful in your project and be able to help out in that way. But at the moment almost zero editors almost zero administrators primarily edit from a mobile device and if you look at mobile only contributors that last column so that's 100% of edits on mobile web then that's sort of even even lower across the board. And so this is kind of one of the questions we have is like we want to make mobile better for active editors but we also want to think about can we bring more editors through that pipeline from making their first edits to being an experienced active editor patroller administrator in a way that can then support the community more broadly. So as I said mobile first editors are much less likely to be moderators and administrators and again I want to spend five or ten minutes just having a chat. What do you think the barriers are for new editors in terms of using those moderation tools on mobile? And when I say moderation tools I mean things like reporting bad content that they see or undoing edits patrolling. What do you think the biggest barriers are to those editors? And how can we help mobile users discover editing tools and discover moderation tools? Perhaps thinking about this in a way that isn't going to increase the amount of vandalism for example you know not making it too easy to just go through and make very rapid edits for example. I'd have to hear your thoughts on that. Young June I see you have a hand up but I don't was that from before? Do you have a something to share this? But yes but the only thing I can add for a point is like some like a user maker gadgets are not like comfortable or an accessible via mobile like that's the point I can pop popping up. Yeah thanks. Yeah thanks. Yeah there's a comment in chat from Vera about it being hard to find out about tools when you're only familiar with the mobile experience and I think that's a great point. I think sometimes when we as experience editors when we think about it we we already know what tools are available and so we might go looking for those tools that we're familiar with but if you're a new editor and all you've done is you know make an edit and you haven't spent much time on the desktop site you don't even know what tools are available to you and so we definitely want to think about how to make that mobile experience inviting and sort of explanatory you know and set it up in a way that teaches you a bit about how all these tools work and there's a comment about how many mobile first editors are very active i 100 plus edits a month I guess that's a group likely to become admins. Yeah I think you're I think you're right and we haven't dealt into that data any further than just looking at those active editors but I would really like to speak to those very active mobile editors who who aren't moderators or administrators and see what they think a comment about having a twinkle like interface on mobile would go a long way I think that's a great idea that's something that we did look at one way we were doing our user research and I think even for desktop some of the twinkle functionality really is like very very useful and very important and I know that that's broadly an English Wikipedia only tool I know that other communities have imported some of it or tried to import it but yes I would love to see that and I think it would be especially useful for mobile okay well I'm going to move on from this slide there's going to be some more opportunities to to share your thoughts because I want to share with you sort of a theory that we have and this is a fairly new theory so I would love to please please be critical of it and and tell me what you think but we have a theory about how new editors become active members of their Wikimedia projects and it relates to four key pages and these key pages I've been referring to as kind of revision navigation they're the pages that help you navigate the edits that other users are making because we think that after an editor has made some of their first edits you know after they have copy edited an article they are interested in or they've done a structured task from the newcomer homepage at the moment I think they don't get then guidance on on where to go from there and we were sort of thinking about this and we think that there are sort of four key pages where a new editor learns some really fundamental things about how Wikimedia projects work and so these are the page history which lets you see every edit that is made to a page if you're a brand new editor that might be a novel interesting thing to you that oh I can look at the edits other people have made specifically they can go through to user contributions and see every edit that an individual contributor has made they can then from those pages go through to a diff and actually look at those individual edits and then of course they can use their watch list to track edits to those pages and so our theory is that those four pages are sort of core to graduating a user from they've made you know their first edits to they understand sort of how the wiki works a bit more broadly and how they can navigate those pages and see what other editors are doing and then maybe engage in content moderation if they see the undo button or they find the recent changes for you that sort of thing and so we've been thinking then and I'll show you some some things in this in a minute but we've been thinking then about okay if these four pages are important then what are they like on mobile at the moment and how could we improve them but first I just want to give a couple of minutes do you agree with this idea that these four pages are sort of critical to to becoming an active editor are there other pages or workflows that you think might be missing here I'd love to hear your thoughts so there's a comment in chat saying that when I give talks I usually show recent changes first and then show how page history and user contributions are filtered I think that's a great point we we haven't included recent changes here but I can absolutely see how that is a how that's a key element when you say that you show recent changes first is that is that brand new editors people who haven't edited at all or is that kind of to people who have started editing and Darcy says these aren't the only that scroll these aren't the only lists used for content moderation recent changes cleanup backlogs yeah I completely agree this definitely isn't isn't you know everything that an editor needs to do content moderation and we're trying to focus really on the first steps that they can take after making their first edits and making sure that there was sort of key core components are good yeah that's a great point and the recent changes is important but it doesn't scale well for very large workies I completely agree with that that's something we found in our user research when we spoke to contributors from the larger workies they said I need filters for recent changes I need auras I need you know tools like huggle for filtering edits and then we went to small workies they said recent changes is fine I can see every edit that has happened to my workie if I filter for anonymous edits I can see the last week all on one page and so yeah I absolutely see that Darcy asks the newcomer portal on the user page seems to work really well is that on mobile yes it is also mobile actually think it might have been developed as kind of a mobile first experience and then brought to desktop even and so mobile editors can get that that new home page and I think we've been speaking to the the team that develops that tool I think there are opportunities to for editors who have done some of those recommended edits sort of take them then through to more content moderation or reviewing patrolling sort of contributions because at the moment they're mostly sort of building on existing content I'd love to think about what those sorts of contributions might be like Nick comments that too much space is wasted on mobile so it gets sold destroying to browse through more than a handful of recent edits that's interesting yeah a lot of white space is something we've noticed as well and I think it's always a balance between and I think if any of you saw the talk earlier from the reading web team about the new vector desktop interface it's it's sort of a similar issue we we want information density for experience editors but also we want it to be navigatable and sort of clear for people who aren't so experienced and so it's it's constantly a trade off cool well thanks for all your input on this and I'm going to move on to the next topic which is assuming that these four pages are pretty key where do we start as a as a team in terms of making improvements and so the first step that we think is most important is improving the diff page really all of these pages and all these workflows come back to looking at an individual edit and seeing what what changed and in our view the current diff page on mobile doesn't really support a new editor in understanding what is happening the page hasn't changed for a very long time it was designed a long time ago as far as I understand it and at the moment it's missing some pretty key features and links so the one that the communities have raised a lot is the undo is missing you can thank a user for that edit but you can't undo it I know that some communities have a gadget that allows you to undo but otherwise that button is just not there you also can't get through to the page history you can't get to a user's contributions very easily your watch list is a little bit hidden here and so this really stood out to us as something that needs improvement yeah michael points out that the diff also doesn't work with red green color blindness absolutely not only is that diff kind of a mess to pass that was just someone copy editing a paragraph if you're red green colorblind then that's actually really hard to to figure out on desktop of course we have that two column view and it's yellow and blue and the obviously the position helps left and right but on mobile that green and red is the only way that you can understand the difference between added and removed text so just to be clear this is the current state this isn't this isn't our changes this is what it looks like right now and what we think needs improving um Darcy asks is the diff algorithm the same as used on desktop um it is different because on desktop you have a two column view and so um the algorithm only needs to to show you what was removed over here and what was added over here on mobile the algorithm needs to put those things together and as you can see from that example there show you what was removed and also what was added right next to each other um and I think it could be improved a lot um I've seen other inline diff engines there's a tool called wiki diff I think which has oh sorry wiki ed diff also uses inline and it looks a lot better than this one because it breaks up less of the words so you get more of a removed sentence and then more of an added sentence rather than being like every other word so I think we can make a lot of improvements to that that diff engine so that that's clearer to to look at and so uh having looked at this and and sort of decided that this isn't great and we could make some improvements to this we've started making some very early sketches this is you know absolutely version 0.001 um of of kind of our explorations here this is very early days but we made some very simple wireframes of what a new diff could look like on mobile so um some of the key things we were thinking about here is obviously getting rid of that red and blue uh sorry red and green diff change and making it probably yellow and blue given that that's how it is on desktop but we could also introduce some more elements there like crossing out the old text so that even if um color blindness is an issue you can still tell what was added and what was removed compacting some of the top of the diff so that um you can see the diff more easily um straight away making sure that there are links to the page history to user contributions and of course making sure that there's an undo button um on the undo button also we were talking about the fact that if you undo an edit and this is true on desktop and on mobile it opens the full editor so that you can make a partial change you know if you wanted to tweak the edit as part of undoing it um but we looked at some data and found that in the vast majority of cases when someone tries to undo an edit they undo the whole edit um making a partial edit is pretty rare and so we were thinking especially on mobile why not have it so that when you click undo you can just edit uh edit your edit summary and then confirm the undo right you don't have to open the editor at any point um though you can if you want to um so this is just kind of an early exploration we're not um we haven't tested this with anyone yet I think you're more or less the first community members who are seeing this um and so I'd love to get your feedback on you know what do you like or dislike about those sketches um how do you feel about the inline diffs and what do you think about that undo um process what what comes to mind for that I see that Amir has an immediate thought that this is dangerous do you want to expand on that if if I'm allowed to speak um there's a reason um revert and undo uh uh separate on desktop uh revert was much earlier revert was there when I joined in 2004 undo appeared around 2007 or so and it may or may not be the reason but uh 2007 is also the time um if you if you check by data uh when it became harder for new editors to join and stick maybe it's related maybe not but it happened more or less at the same time that the undo button appeared so there is a theory uh that uh that was one of the reasons it uh became harder to join uh Wikipedia so uh this will essentially give the revert permission to everybody uh so that's uh there should be some very careful thought dedicated to this yeah thanks for showing that I completely agree um I'm always wary about making it easier to do these sorts of things because on the one hand you have you know as an experience editor you might think yeah you know I wanted to be able to do this as quick as possible um but making it too easy is also dangerous because we know that the process of having your edit reverted is not a good one um when you get a templated message when you get that notification that says your edit was reverted you're much less likely to continue contributing and so I completely agree if we were to do something like this we would want to both be doing a lot of user testing and also a lot of data um monitoring to make sure that as we gradually tested it and rolled it out and ab tested it that we weren't seeing like a hugely negative effect there on uh on retention uh Mike how do you add a custom edit summary in that since the box is all filled with the automatic text uh yeah that is a question that we'll have to answer um I mean I think this would probably work potentially in the same way that it does at the moment where I think maybe if you click into that box it could the text could be left aligned right so you can add your comment at the end like on desktop um that's something we'd have to explore um like I say these are very early initial explorations and so that those kinds of questions I think we'll get into if we if we move forward with this there's a link in chat to a good gadget for partial reverts that's really helpful um I will I'll take a look at that um I think like other product teams we're often looking at gadgets and user scripts for inspiration um you know if if editors are already um fixing a problem themselves then that that's a good signal that we should consider that approach in the software by default um and so I want to just then uh open up a bit more broadly um given that we've got maybe five or ten minutes left um is there anything else you want to share with us about content moderation on mobile um are there any tasks that sort of stand out to you as maybe well suited to mobile um or maybe things that are just too difficult to do on a phone that that we almost shouldn't bother trying to trying to make fit into a mobile screen size um yeah anything else that you think our team should should be aware of fixing coordinates on nearby articles is a is a fun a fun micro task yeah oh and young yeah how this looks in a tablet um is a great question because um I think I think we often forget about tablets um we think about mobile screen size we think about desktop let me forget the tablet uses the mobile skin but obviously blown up a bit so um those will definitely as we start making you know better designs and um more detailed designs definitely tablet screen sizes will be a part of that um and yeah nick making the the link to switch from mobile to desktop more visible that's an interesting one because on the one hand I agree uh that that could be beneficial on the other hand I think ideally mobile would be good enough that you don't need to switch to desktop um we're actually currently implementing some some data logging so that we can monitor what uh where and when those clicks the desktop happen because I think that'll help us figure out which mobile pages aren't working um or don't have the functionality they should do um and so we will um we'll be thinking a bit more about that but and yeah michael I've heard that that link sometimes doesn't all work um so that's something we'll uh we'll also think about um okay so we have a few minutes left um there's just one final thing I wanted to say then um quick comment yes Darcy uh notes that even if mobile is good enough to never need to switch to desktop users should always have preference I agree I'm not saying that we should get rid of that button at all um certainly you should always be able to switch if you need to um so if you want to stay up to date with the work we're doing and have more input in this um we have a page on media wiki it's a moderator tools there you can find links to the project pages for the work I talked about that we've already done but also that we're planning to do um that work on diffs and revision navigation isn't quite up to date um today but I'm hoping to update it in the next couple of weeks to sort of have more of those wireframes and the things we're thinking about there um you can contact us directly probably the easiest way is to email me um but also you can find me around the wikis um we also have a newsletter which is brand new we haven't sent any issues of that out yet um but you can find that on media wiki at moderator tools slash newsletter please go there and sign up either with your own user talk page or with a community venue that you think would be interested and we're planning to send that out maybe once every few months um just with sort of updates on what we have done what we're working on and sort of where you can give us more input and give us more thoughts and feedback on on what we're working on so um that's all I have um I appreciate everyone's input I think this was a really interesting discussion I'm going to go and copy the entire chat into a document and make sure that I have that available to me to to read through again later um this was really helpful um to our work so thank you everyone for coming