 And this is the first time I've had that kind of silhouette. So I noticed that the population really reduces when it comes down to identity systems, data improvements, and people who are interested in this thing. So my purpose here today is also to talk about our journey as a nation in Singapore, going through this whole idea of data privacy and promise from a government perspective. And from there, basically, at the end of this, what you walk away with is the lesson about when you are trying to launch a nationwide solution for identity systems and how you preserve data privacy and how you even get this across to the general public. Because we felt that there is really a huge gap between what the developer has in mind and what is really on the ground. The kind of thing that we speak about in conference, you can't just bring it out of the public and expect people to understand it. We felt that there's a huge gap. And there's a lot of lesson learned from building such solution. And basically, we are also aware that right now, I don't think there's a decentralized way of representing our visual images on these centrally rendering things. And we have a solution for that. And basically, we are going to share that for those of you who are a bit more technically inclined. So it would begin as a story. And the story starts with, you know, the previous talk was about blockchain under block technology. And then with the hype, it helps driving the political, there's the political drive to do it. And then all the questions starts with like, hey, can we blockchain this? So of course, maybe I think it would help if I show you what exactly this open administration opens it before we jump into it so that you get a better understanding of where the perspective comes from. Since there is internet, I'm going to go right to end. This is basically our PowerPoint. It's really simple. So you have a JSON document as the data for itself. People saw it in the hot drive, email, whatever. And basically, at the end of it, it's rendered in a way that's familiar to people, right? You don't have to use a stash at the door. You don't need a robot, although you could have one. And basically, what your document is structured in a way that is, it looks good and looks familiar to you. If you get it from your university, it's going to look like your university certificate. And of course, we do support things like, right, wanting to get a set. And it's just really cool when you show people that you can have videos on the set. And really, just upon that, we do enable data privacy. And this is something really cool that we want to talk about later, where I can obfuscate something, but I couldn't change any data often. So by doing this, I can actually remove data at a moment. But it doesn't affect the hash of the document so that I can later on verify this document again without the issuer reissuing this claim, right? So I'm trying to keep the technical aspect of this well so that if you're interested in the technical aspect of this, feel free to ask me about it at the end, right? So that's pretty much what we have. We want it to be really simple for citizens and people in general to use. So the journey began like that. The three things, you know, you start with, like, can we block changes? And really, you just want to revalid it again just to make sure that it's really alignment on the business as well as the technical. That's our version one. And then beyond that, that's where we realized we messed up in time, right? I'll tell you about that. And yes, when we launched version two, two actually patched up certain things, right? On scaling it beyond just Singapore alone. So the first part is really just the proof concept. And it started off with our polytechnic who just came to us and said, hey, can we have our blockchain on, can you have our certificate on the blockchain? Right, the other way around. So the key issues here is that what is in a certificate? What is a certificate, right? Where do you want the data to be stored? Definitely not on the blockchain. And how do you issue them so that the scalability of the blockchain does not affect you, you know? You have seven, you have a few transactions a second and what's not going to help you scale with $50,000 in a batch, right? So our solution was really simple. Or we put data into a file, we hash it, we get a hash. We take the file, we put it into a web render that does the rendering as well as the checking of the promise. It does so by just rehashing through the document and then checking it against the smart contract on the Ethereum. And basically there's two lists, right? One for issue status and one for revoked status. Anyone can do that, right? So in terms of data, it looks like that, okay? So there were, we prescribed a very strict schema because we think we knew what a certificate data looks like, right? So we say, hey, all your data must fit into this schema and it goes there. And then basically we use a level three to actually batch on the issuance of, you know, tens of thousands of certificates at one go. Yeah, then we managed to do like tens of thousands in a couple of minutes, no problem there, right? And the only information stored online was actually just the Muggle Route 64 character hash, okay? So there's no information linked to Opsize, right? And basically at the point in time, this is how our product looked like and we told ourselves, hey, we were ready for the market. Let's watch this to the citizen, looks like that. So there were a few key assumptions right here at the point in time, right? We assume that we knew what goes into a certificate. It's untrue, right? Later we found out that certificates are not just aggregation credentials, right? And different people think of different things where you talk about aggregation credentials, right? And schools are definitely not fine with their default look. They want their logo, they want it to look like the way they do and then they want it structured into tables in very weird way, you know, group by semester set. That wasn't quite what our data structure supports, right? So that comes open since version one where they form a consulting. So this is really cool, like a lot of schools got together informally and then say, hey, we are gonna form a consulting and you are gonna build us a product and we say, okay, sure, why not, right? And at that point in time, the challenge is this, how do we go about sanitizing data but get a lot of customization amongst the different school? How much data goes into the document itself? Do I add just the name? Do I add the ID specs in there? Do I have a whole year in there? Do I, you know, all sorts of things. How do we customize the looks of it? This is the first time the visual thing came about and how can we ensure people who issue open sets basically find the identity of the issuer to a real world identity, right? Because we work with students, right? So at this point, basically we had this thing where we have a template key, basically it's just a key and we basically allow schools to do a GitHub commit to our repository to change the certificate style and then basically we use the key to toggle the different styles, right? And then from there, we introduce the user control data privacy, which is the one that I'll show you that we can remove a single line or a part of these certificates without affecting the integrity of the search, right? And also we allow for additional data such as like signatures, such as like images and all sorts of things to go within the data file so that people can customize it, right? And we don't prescribe, we have a more flexible data schema in general, right? And basically how we deal with identity is really simple at a point in time, right? Basically we have a file, right? It's called registry.json and basically it results to this the student name is that person this student name is that person and this student name, I don't know who is that I'm sure a warning banner tells me that I could not figure out who is that go figure it out yourself and that is the, you know, they took correct address, figure it, right? And it wasn't adequate, right? At this point, there were a few assumptions we assumed, right, that, you know, it's good it's very good to have one or two templates and you can put them all in one place, yeah. And it's really easy to say who can issue such a document and who couldn't, right? Later we figure, even at a government level you couldn't do that because how about private training providers? Those are really tricky, tricky spaces, right? So that comes version two where we look into decentralization in general, right? At this point in time, the medias or high top schools are all high top they would say like, hey, launch this and then they put it onto the newspaper, right? Just like all blockchain projects, right? And this is really scary. So we log up to get out and then we see, oh my God, this school, this is from one school, by the way. So one school decided to put, you know, have different template style and this is just from one school. So part one, part two, part three, I don't even know how to review this. We spent a lot of time reviewing this and basically it's a lot on this. We get questions like, hey, how do we link? Seriously? Can you turn off Solarheel? Yeah, sure, why not? And people say, I need to install a custom form, right? So if you install a custom form, every single school in the world is going to get custom forms. Yeah, sure. And then when we give recommendations, there's really just different standards in technical abilities and then there's this like, is this just recommendation or could I prove this truth? And this is the best, right? Can you approve my proof request because my user is doing me ready to eat tomorrow? Yeah, sure. So at this point in time, scaling became a big issue for us. So the next question that we want to tackle was, how can we scale beyond just Singapore? Do we scale beyond just a few schools? How can we not manage those school requests? I don't want to see any more of those school requests. And basically we just want the template changes without the change, right? How can you just specify how you like to do your styling without coming to the government and knock on our door and say, hey, can we review our code? We don't want to do that. And also automatically verify identities of issuers who are not in the registry because there's only as much people we can fit into one registry without going like, hey, everyone's going to get in, right? Because you really need to verify each of them and it's a lot of work in there, right? So that comes the central lies rendering, which is one of the two key components of our version two, which is to use, first of all, iframes, right? To then render a document from another site that is not from our site, right? Basically within your certificate data, once you specify, hey, render me the view of it from a specific URL, conforming to a certain rendering specifications. And then basically outside, basically just go over to that site, pull out a blank page in the iframe and basically pass data through post message into that frame and have that frame rendered out to be shows, right? Basically, this is how you look like in the data. We have a template, remember just now it was just a template P. Right now it has a URL and says that it's a renderable type. Basically, it instructs the browser to go to the URL, throw in the entire object itself. It should be able to resolve, right? If it's following this standard. So this is how it looks like, right? So this is an example of the document that you saw just now. And basically, really just in the code, you can see it's a iframe and then it goes to the specific source demo renderer .winsets.aro and passing the entire certificate object there itself. And that party is then responsible for rendering up to view. So this way, we would not have any more pull requests on visuals because that is the list of concepts for us, right? And the next question is then, now we have to figure out the visuals, how then do we do identity? So for us, we were thinking like, hey, we are already using such mechanism, it's called DNS, right? Can we then use DNS to then prove the identity of the issuer? So it was really simple. It's just like you tie up the, you have a protocol and then you say that I'm on a train, maybe make that, that one and then your smart contract address that I'm issuing from is that address. And basically just have to match the one within the document. And it was that easy. And basically with that, you can really just forget about the list or the long list of like people that you know, and then basically just have this thing like issued by a cycle domain, right? Or really just any domain itself. So the moment we roll out this, we realized that there were schools around the world who were kind of watching us. In fact, there was one school that after we launched this, they kind of made it and they posted the news of today, the news and then we later found out from the news that hey, they are using our technology and it was really nice for us at the point in time knowing that hey, people are using it without telling us, which is great. Right, so at this point, we're just going to share with you what's next. So the key technology underpinning our open source is actually open devastation, which is the document format and the whole ideology behind the rendering, the identity, the document structuring, and all those things. So we are open sourcing that full schemes and we are right now working on the documentations. The source is all open. It's just the documentation is a little all over the place. We are trying to tidy things up there. Open to tell is fully open source with documentation ready. And the other cool project that the Singapore government is also working on is Portrait Trust and we are basically bringing this technology into the trade ecosystems. And basically it's just working on trade documents and on top of that, we are also looking into how we can tokenize deeds in general. So in the trade sector, there's this thing called the deal of the leading. Basically it represents ownership of cargos, right? And basically we are looking into how to tokenize that thing and make it recognized at an international level. So for that project, we are also working, we are working with international bodies like the UNCFACT. In fact, we were the one who host UNCFACT a couple of weeks ago, a meeting with Australia and China, right? So if you're interested in any of this kind of project in terms of documents, notarizable documents or either disease system, feel free to talk to us, right? And basically that's all for us. And this is the documentation. You can have a go at the demo. The slides are the web at this URL or the QR code. So we are hiring. We are free to explore use cases. And if you have suggestions or feedback, please go ahead and contact me, right? Thank you. Okay, we have a lot of time for questions. So if anyone has questions, please go ahead. So we're going to start with the hash function. And the hash function. We can show it to my six right now. Yeah, but then again, we can't show how the hash function at any time. No problem with that. We just use whatever that's useful for us, right? If that's the case, then I'll be right around. So I'm going to start with the hash function. I'm going to start with the hash function. I'm going to start with the hash function. I'll be right around. And you got my email so anytime, right? Thank you.