 Thank you everybody. I'm just going to start a timer because I'm a diligent speaker. So I'm Martin from my society. You've probably come across this before. As soon as we organise this event but research events with great cakes aren't the only things we do. We're a small charity based in the UK and we build apps and websites to improve access to democracy around the world. And that is us. And around the world is all of the countries you can see there. And I'm sure Mark mentioned this morning how much of a global effort we're making. And the result of that is Alavatelli, our freedom of information platform runs in 28 countries. Fix My Street, which is our fault reporting tool, runs in 19 countries. We have parliamentary monitoring sites in places like Kenya, Nigeria, Zimbabwe and Iran. And we've had a lot of experience making our products and platforms work in these different countries and cultures. And it's not something that only a designer like myself should do. It's sort of something we've tried to introduce across the organisation. I remember in my interview the way it was put to me that everybody in my society is a designer. You're the only one with a job title there. So hopefully there's something in here you can take away to improve your own products and improve the way you work globally yourselves. You'll find that most of them aren't specific to civic tech. They are actually just sort of good design principles. And this being a practical takeaways session, there will hopefully be some. So working anywhere in the world, your number one responsibility is to understand local culture. And the most simple one I can put to you is colour. So colour is kind of a very subjective thing. And working in countries, we always get the typical don't use green, that's this party, the political party or don't use red, that means such and such in a culture. So that was when we were designed a candidate crowdsourcing site in Costa Rica. That was I think our basic colours were green. The message we were given was don't use green. So we didn't use green. People would assume it was related to this unpopular leading party. Colour is the easy one though. We often get much more subtle ones. So we did some work in the Dominican Republic with the police force. And this is a tool to report crimes. And we were working with a local design agency there. We don't always do the design. We sometimes use local partners. In fact, we've had much better results. And we were putting wireframes back and forth and we were finding that their wireframes would have so much text on them. They'd be laboured. And kind of the western way of designing is not much text, very clean and very clear. And it turned out that there was a really good reason for this. Dominican Republic is somewhere where corruption is quite common. And adding this kind of extra text added this legitimacy to it. So where people used to be and scammed and reduced to being suspicious of corrupt processes having this extra text, the local designers had found actually made the users trust it more. And something about trust itself. This is People's Assembly. So this is the parliamentary monastery site we run in South Africa. We work with a local partner there. Something that came up too is historical culture can be important to understand. One thing we did when we set this site up is we made sure that none of the servers, none of the data, nothing at all was hosted in the UK. We were very mindful of the kind of colonial past with South Africa. What we didn't want to do was end up with an ambitious journalist with a bit of a narrative to play, looking a little bit into the technical history and a little bit of the background of this site and being able to say that actually this data is hosted in the UK. It comes from colonial powers in the UK and we're trying to influence parliamentary process in South Africa rather than improve it. Part of local culture is going where the people are. So this is the next step, follow the users. Sometimes that's literally, this is part of our team working with people on the ground in Liberia doing research interviews. And sometimes it's a bit more figuratively. So working out with users what their existing behaviours are. This is a quote by an ex-colleague of mine who is here today and this Facebook is the internet, get over it. And this was kind of the western idea of Facebook as the internet is kind of, we push back at it a little bit. We kind of see it as something for our friends and mothers and things like that. But actually in a lot of countries, countries like Uganda and Myanmar, cheap phones, come with Facebook by default, you walk into a shop, you buy your phone. The shop actually sets up the phone for you, adds apps like Facebook customers don't even know they're in password or their email address is logged in, they lose their phone, they just go and get another phone and they get another Facebook account attached to it. Practically that meant when we've worked with partners in the past in those places where previously we might have built a prototype website or built a prototype app, actually sometimes just a Facebook page with a video worked even better than those things. And lots of countries Facebook tie-ins are pretty clear. So Ghana for example, Facebook access is free. So even people in very impoverished companies, in impoverished countries with access to poor mobile devices can still access Facebook. So in terms of cities and engagement it's quite a valuable place to look. And speaking of how people access your sites, designing for multiple devices is probably the best tip we can give you. This is a bit of a funny one. So designing for mobile in the west means Safari, iOS, Android, very high-powered devices with really decent cameras, lots of data. In countries like Kenya and Nigeria though it means very, very low-powered devices you use something called Opera Mini. If you've not come across Opera Mini before it's a web browser that actually does most of the hard work on a server. It renders the web page on a server and basically sends some pictures of the website. And it means that a lot of things that we're used to using in technical things like CSS and HTML just aren't possible to be used. So a lot of the times our apps that were fairly, you know, wizzy rich experiences on an Opera Mini browser just didn't work. So it's led us to find tools like BrowserStack, which is something that lets us emulate devices. We can make it basically simulate a very, very cheap Android device or a very old-fashioned feature phone. We test our devices in those and see how they work. And it's always surprising, it's always way slower than you think. It's always way clunky than you think. And we found that when you start to build these apps and when you start to design these new features, if you start here as the basic and then add the kind of iOS and the Android stuff on top later, it works really well. Sorry. Let's see how it's like. So all of this is for nothing if you don't know how to make a start. So this is your practical takeaway. So I'm assuming that most of the people who aren't technical people and don't directly edit code themselves but probably work with people that do. Something you could do right now, take back to your teams and put into your project plans for the future is to improve your front-end performance. Something we did at my society a couple of years ago. We chose a couple of our partner sites and we ran a front-end performance sprint. And a sprint is just a set amount of time to do a set of development and design work. It is fun, honest. And if it's not fun, it's satisfying because it's something that's a small amount of work, relatively iterative and very impactful. So this was for our South Africa partner, I think, first of all. Chose some low-hanging fruit. And this is front-end performance. It's things like making sure that when you load the page, it loads quickly. Making sure that when you click a link, you have to move quickly, making sure that when you do a search or choose an interaction, things move quickly. And there's a few ways to do that. And the easiest way is page weight. So that's basically the amount of data you need to download to your device. So one man, one day's work, one person, we reduce page weight by 40%. And it's not everything, there are other things to do. But it's like in terms of low-hanging fruit, it's probably the most effective thing we've done. And just a few days' work, you can make a huge difference. So putting it all together, understanding local culture, following the users, designing for multiple devices, and making a change right now. None of this should be anything new, but hopefully it's giving you something to take away and try when you go back to your day jobs. Thank you very much.