 Hello, thanks for the warm welcome. My name is Stefan Lennart. I'm from the University of Ringsburg in Germany. And the reason why we did something more special on the search is we wanted to pull out the complexity of full searches where the students and the teachers called us and said, oh, I don't find my course. I don't find my teacher. And I'm new here. I don't know what to do. The reason was make something more intuitive, right? And that was the reason why we adapted a little bit, Google the instant search, and make it more convenient for the people. So the integration itself, this is the start page of our eLearning system. And the integration itself is done in the header in the navbar. And it's basically two functions only. It's the quick search where you enter and you get instant results. And if you want to do more research on it, which many people do, but mostly they use the quick search, they can pull the full search button and they receive more insight in the courses. Let's go first to the quick search. It's functionalities. Like you see here, if you enter like a teacher's name, you get the course category. I, myself, have a course category. And although I'm a teacher, so I could click there. And you come to the, how to say, to the built-in Moodle page where you get the list of all the courses which is teach to have. Like I said, it's based on the Apache Slucin search engine. And we did the indexing over courses, respectively the long names, the short names, and the ID. So teachers and students can search for long name, short name, and ID. The trainers, it's first and last name. And the categories. Also, we are a long name, short name, and ID. That's for the quick search. The next thing is the full search if you take the class. This is how it looks like. I've made it green. I made a mistake on it. I wrote biology with three O's, meaning spelling error. But it's not a problem for a Lucine because it can guess and has phonetic search. If you would enter it in German, you would get an English result also. So its search characteristics are the phonetic search. You have search term highlighting, like you see in German biology, Unterricht in green. The weighting of search, which is more higher rated or lower rated. And a detailed listing of where is it, what is it, and when was it done. Is it an old course? Is it relevant? Is it not relevant? It helps a lot, I think. The key benefits, especially also on Lucine, because it's very powerful, you have more flexibility and reliability on the search. It's quite straightforward integration. The special features, like phonetic search, is on board already. And that was the main reason why we used Lucine, because first we started with basic database queries, and it's not performant. I would not recommend database queries on the fly. It pulls down the server, it's not good. So Lucine is very good on that. And because the Lucine index is always done one day per night, so if a course is created by day, and on the same time, the search would not find it. We did it hybrid, so everything's older than one day, goes over Lucine Apache indexing, and the rest is going via the basic database queries. It is like a compromise. We called it hybrid search. Somewhere it's about user satisfaction. Since we implemented it, we did not get any call about where is my course, where is my teacher. They found it directly. They use it almost every time. Just some small words on the implementation and the structure. On the back end, we have the Lucine indexing, directly in the root of Moodle data. We create, like I already said, a nightly cron job to rebuild the indexes. And search result preparation is done via JSON and the delivery via HX and PHP. The front end itself is encapsulated. All JavaScripts are encapsulated in ES6 modules within the structure of the applied theme, like in our case, Boost Union. And for the calls itself and the integration in a nav bar, we had to do some small code snippets. And the JS call itself in a nav bar. It's not a lot, but it has to be integrated like Hacked. Just some words on the outlook. If there is time, I would like to do a full integration as a full plug and play module to say, OK, download it. You get everything packaged, click a button, and you have it on your page. And what also would be great is further investigations on the Lucine's functionalities to make the search even smarter. It has more power than we use, maybe. OK, that's it. I just want to speak a little bit about a few ways that you can work to create an honest assessment environment in Moodle, and this is specifically with Moodle quiz. First, I want to mention the fraud triangle, which is a concept from financial fraud, but it can also apply to academic dishonesty. And that is that you have the greatest probability and occurrence of fraud when you have the most rationalization, motivation, and opportunity. That's the triangle in the middle. And motivation, you can't really address with Moodle quiz. You can use other aspects of course design to look at that. But with quiz, you can address rationalization a little bit and opportunity. So those are the two that you want to shrink. And those are the ones that we'll look at. To reduce rationalization, you can use an honor statement. Three question types you can use are description, true, false, and short answer. Excuse me. Research has shown that if you include an honor statement at the beginning of your quiz, which could be something like, I agree not to use outside materials. Maybe you explicitly say what they are allowed to use if it's open book or open note. Then students are less likely to cheat on that quiz. So for the rest of this, we'll be about reducing opportunity, which you can do with quiz settings, question types, and safe exam browser. Some of the quiz settings you can use are navigation options during quiz attempts. That includes shuffling. You can shuffle the order of the questions in the quiz and the order of the answer options within a question. You can also choose whether to use free or sequential navigation. This allows you to decide whether students can skip questions and return to them later, which is free navigation or sequential, which forces students to go through the quiz in order. And if they don't complete a question, they cannot return to it. Timing includes the open and closed dates of the quiz, which gives the window during which students can decide to take it, which you may want to make larger or smaller. And the time allowed, which is the time that they have once they begin the quiz to finish it. You may want to put some limitations on that just to reduce opportunities to share with other students. Review options can also be used. There are a couple of time frames, which are immediately after the quiz. And then after that, later, you may want to consider whether or not you want students to be able to still see the attempt, which is the quiz questions themselves. And whether you want them to be able to see what the right answers are. If those things are visible, it's easier for them to share with other students once they've completed the quiz and the other students may not have started yet. A couple of question types that are useful are random questions and calculated questions. Random questions can be used for any subject. The instructor creates a pool of questions in the question bank. And then in the quiz, you build the quiz by selecting the number of questions you want to grab from the pool, which is a category or subcategory in the question bank. And then when the student begins the quiz, Moodle will randomly select for that attempt which question they will get from that pool. Calculated questions are more suited to math and science subjects with numeric values. And here you create a question and provide parameters that Moodle uses to create numeric values for variables. So for an example, if you had a problem about the perimeter of a rectangle, you might have variables for length and width. And so you would use those variables in your question and then give Moodle some parameters for the kind of numbers that you would want to use. And Moodle will create a set of numbers. And you can specify how many. It could be 10 numbers. It could be 100. Then once those are created, when the student takes the quiz, Moodle will randomly select some numbers from the set that it has generated for the student when they take it so that you don't have each student getting the same numbers as every other student is drawing from a set of whatever size that you chose when you set it up. Safe Exam Browser is available in the Moodle settings. And it prevents students from opening other websites and applications on that device. This means it's suitable for proctored environments or a classroom or computer lab setting where you don't want to maybe have to watch every screen every moment to make sure that they're not accessing something else. But you can kind of see if they're pulling out a phone or another laptop. If the students are taking this elsewhere in their room, wherever you can't monitor that so they can easily access whatever they need on another device. There are three options for configuring Safe Exam Browser and your settings. The first two are within the quiz settings in the Moodle quiz. The ones to configure manually have a list of settings and you can choose yes or no or how you want to use those. You can use an existing template if one has been set up by your admin, which basically they have pre-selected some options in the settings and you can choose whatever template applies to your case. The other option is to upload my own config file which would create outside of Moodle. That's what the Safe Exam Browser config tool program that you have to use on a computer. That gives you some more options but it's also much more complicated and complex. Then that would have to be uploaded to Moodle. Whichever of these is available also depends on what your admin has decided so these may not all be available on all Moodle instances. So hi everyone, good afternoon. My name is Maxwell Fundy, co-founder and director of a company called EduTab Africa. EduTab Africa is a youth led company in Kenya with a small group of multidisciplinary people who work towards transforming education through use of technology. So I actually had prepared for 15 minutes but it's gonna be very very quick so I'm going to rush through the presentations. So where we work, we work mostly in Africa in different African countries but for this presentation I'm going to focus on two areas where we work. One is in the eastern side of Kenya in the refugee camp called EduTab which is near Somalia and on the western side of Kenya where our main office is in a town called Kitale. We work mostly with Moodle in these offline settings which are low resourced which do not have like super good computers, laptops, desktop and things like that and we find solutions that work well in those offline contexts. And that relates to the question I asked about simulations from Open University because part of what I'm going to talk about is how we use simulations in offline contexts for primary school kids. So some of the technology we use, we focus mostly on using portable single board computers like Raspberry Pi that we installed Moodle on them. Some of them have been inspired really a lot by Moodle Box for example. Moodle Box has really made this work very easy but initially we used to install these devices ourselves manually step by step, step by step to make this possible. But now with Moodle Box this is really, really possible and we also look at these devices are also cost effective because when you think about Raspberry Pi then it's really affordable. And then the other thing is trying to improve Raspberry Pi through some of our partners like inclusive education. For example, the box that's on there is a Raspberry Pi that has enhanced memory, a bit of Wi-Fi capabilities as well as a cooling module because it can get really, really hot. So first case study is Dadaab, a refugee camp in Eastern Kenya Somalia where we were running a pilot project with Save the Children. And this was looking at how can we use Moodle in offline contexts using the Moodle app. In this case we used Moodle app together with the Raspberry Pi that we set up together where we had to download content onto tablets, on the Moodle app on the tablets and this was used completely offline. And this was to be done in one school which reached about 320 students. And these contain two learning areas. One was mathematics and the other one was social emotional learning. These were very short five minutes videos that were localized to Somali language because the children at the refugee camp only speak Somali. They are still learning English and Swahili. So these were highly localized videos that were very short in social emotional learning and mathematics. Beyond the students we also had materials for teacher professional development and this included playful learning materials that introduced teachers on how to use tablets in groups and also in corporate playful learning. Second case study is near where our office is in western Kenya where we were using some device but using the government provided tablets. So in Kenya in the year 2016 they are about, the government rolled out what they call the digital literacy program where they were equipping primary schools or public primary schools with windows tablets. And now these are probably six, seven year old tablets and there wasn't enough capacity building for teachers to effectively use these devices. So edu-tub came in and worked with teachers to enable them use these tablets and part of what we did was including simulations from FET which are made by Colorado which are downloadable and you can put them onto the Raspberry Pi and then what happens is students can now access those simulations and other materials using the tablets at school. This was of course happening in groups because again enrollments are really really high so we have to put students into groups and things like that. So I already talked about teacher profession development and some of the things that we were talking about and some of the materials that we put in, videos, we had simulations, we also had podcasts that edu-tub produces on African children's storybooks which enables literacy education. So we had different learner experiences where some of them who are really passive became much more spoken due to the conference they get over time and we had lots of feedback based on what we did with teachers and students as well, seeing about improvement and how happy they were and things like that. So one of the things that we looked at also was how do we now quantify how these things were used and we had two problems. Number one, when using the Moodle offline app, it was really difficult to have actual data collected for how things are used but when using browsers we had this kind of data possible to be collected and yep, that was the main challenge and yeah, lots of lessons from a work. I think adaptations of content to local context is super important and also capacity building for teachers is really important. So moving forward, we are looking at how do we now use these offline technologies more and report better in ways that informs, for example, the entire Moodle community on how better can we develop solutions for offline context. So I leave it at that. Thank you so much for listening to me and I'm looking forward to questions and having conversations around our work in Africa. Thank you. All right, now we have time for questions so if we could call all of our lightening presenters back up to the stage. I know you just came down but come back up to answer questions, we're gonna have, yeah. So Anna has the microphone. Please raise your hand for questions for any of the three presentations we've just seen. I have a question from Sarah regarding assessment. We haven't used a safe exam browser so far because we had some challenges. What do you recommend as a best practices for safe exam browser? So in terms of the settings that you might use for it, I would say one thing to consider is if they have URL filtering, so if there are websites that an instructor might want students to be able to access, we have some that want students to be able to use a calculator. So we found a website that has online calculators and we found that you can allow that and if they put the link in either the quiz question or a description question at the beginning of the quiz then they can access that link and use a calculator that way. So think about something like that. A lot of them are ones that may or may not be useful but there are things like show wifi, audio enabled or disabled and I think those are useful. Like if you were doing something where students would need to listen, you would need audio but otherwise I don't think there would be a problem having enabled if they don't need audio because there's not gonna be anything for them to listen to and I think if you show them more things like the wifi control or whatever, they can tell if they're having a problem with wifi and then let you know that okay, that's why I'm having a problem with this quiz. I can see that it's a problem with the wifi. So I would recommend deciding what they do or don't need and then just maybe taking away whatever you want to be restricted instead of just allowing only certain things so that they've got the most visibility of what they can see as well. And I'd also say a practice quiz so that they can download the browser ahead of the actual quiz and they can see kind of how it works, what they'll need to do with their computer. Depending on how it's set up they might have to disconnect their laptop from any external monitors. Just so they can get familiar with that in a low stress environment. Thank you. One follow-up questions regarding downloading the browser for their students out. Have you had any issues from the student side that they're not able to connect or download that application? Thank you. Right. We've had a single student so far that has had an issue and that sounds like something with the install got corrupted on their computer. And that's apparently a very rare case that happens but all the other students either, they downloaded or if they ran into a little issue they just tried downloading it again and then got it running. Hi, I've also got a question for Sarah. First, thank you for a very good talk. It was a good rundown of the available options. And I thought introducing the fraud triangle was a good point of view. I want to say something that's probably going to sound very naive but I think it does is worth saying. When you're thinking about the fraud triangle and motivation I think we shouldn't overlook that when you're designing your assessment if the students think the assessment is a worthwhile use of their time that it's helping them learn that it's not just a box ticking exercise they're much more likely to engage with it honestly. So good assessment design is I think quite an important way to reduce students motivation to cheat if you can find ways to do it. It's certainly easier said than done. Sorry, that was a statement, not a question but you can respond. Right. I'll just say yeah, I think it's more of an instructional aspect of how do you create your quiz from a design perspective rather than what can I do with settings and everything, definitely. I have a question for Stefan. So you said that you would be scanning the whole Moodle data folder if I understand correctly. So I don't know how much space this takes and how long it takes to scan the whole thing into the search index. Could you give us some numbers? No, we did not make the indexing is not over the full content what Moodle has. We index over the categories, course categories, courses and the teachers. Everything together is 60 megabytes and it takes about half an hour in the night to create the index. So it's not over the full. I have a question for Stefan as well. Stefan, do you, how would one implement the search that you've built? Is there a repository for it or is it just something that you have? Yeah, that's the problem. Yeah, first you need the source code of Lucene itself. We put it on a root folder and from there we made on the Moodle source code also a root folder where everything happens like kicking off the cron job. That's quite, how to say, off site of the Moodle core. It's like, okay, you take the folder and put it there and it works. The only thing which has to be hacked with the Moodle core stuff is integration in the theme. So it's not a lot where you have to change this on Moodle core source or this and this. We created for Lucene and the indexing, one folder, only our stuff and a small snippet in the theme itself where you have to hack it in this way, right? So, but to make it more convenient, it would be better to package it completely and have it like a normal plug-in or module to say, okay, download, click install and you have it. Man, that is an outlook, right? Like many things. Hello, I have a question for Maxwell here. Yeah. Thank you. I'd like to congratulate for your amazing work. It's amazing how to see the data, how you can reach those communities. So I have two questions. The first one is if you have a sense or an estimation of how long does it take to implement a whole training framework or lessening framework in a new community. And the second, if you could share with us your lessons on how to overcome digital literacy in those communities. I'm talking from the UNESCO side, so it would be great to listen. Wow, those are two very good questions and quite deep as well. So in terms of training, I think this varies from context to context in terms of what's needed and level at which different people we are working in in the communities are, for example. In Kitale, the second case study, these are people we've worked with for at least three years, so we know them for quite a number of years. And therefore, when we started this, it was harder and took more time. But now it's much easier because they also recognize how helpful these technologies are. And also in Kenya, for example, there has been a very big drive even by the government to sort of encourage people to get more digital literacy skills from either training by the government or local partners and things like that. And teachers are not left out because even the teacher service commission has put a lot of their services online, so at least even the ones that were really technology-phobic are now trying. And so I would say for different communities, it takes different times based on the context. Now on the second question, please if you can ask the second question again. It's about digital literacy. You covered a little bit, but if you could explore more. Yeah, I think I sort of covered it a bit more broadly, but then there are some materials we have also developed. We also have our own eCampus, where Edutab Africa eCampus, where we also try to build some teachers who are able to get onto the platform. But sometimes we also have our team who do face-to-face training. But remember, we also have a small team, so we can only do so much, but then we also leverage a lot on open source OERs that we can also share with people to upscale themselves. Yeah, this is for Max, it's just a comment. I'd just like to appreciate the work of Edutab and your in Africa. And I think the work you do in Africa, what we do in India is highly underrepresented in model because we are working in completely different environments where quality education is not accessible. In our case, we even have data. It's not that we've also used the SPCs and so on. But I think that we need to see a lot more of the use of Moodle for the genuinely sort of deserving communities. And I'm really happy to have attended at least one lightning session on that, so great work. Thank you so much. I think that keeps our team going. And thank you so much. And we'll be looking forward to reporting more on part of the work we're doing because I think this is my second time in the Mood. I came last year and this year, and definitely there will be something next year because we're also growing. So thank you for that.