 He has given up on that. He just salivates me. He's a sharpening face. He's a fool. He's a good guy. He should come on. He's a outright developer. He has bought Red Lighter all my life. In the now? That was last year, right? Before he gave us a go. What? I don't know. I mean, it's like this. You don't know the way to die. Okay, what about members? I think people have to list you. He would like me to go to the final. So why not? Because I don't think it boosts my morale. He doesn't know how to deal with it. I would be nothing if just... I don't know how to deal with it. Okay, that's fine. Now, can we say it's not fair for them? We're going to run up and down. We don't know how to deal with it. Yeah. What's your question? I don't know. What's your question? I don't know. Can we say it's not here? Why can't we say it's not in the office? I don't know. Thank you. So Stefan is out of town as well? Yes, he's travelling somewhere. That's why he asked me to go to town. This is a node red priority. Um, it's blue mix size. That's more than a node red. Okay. A bit more things in it. So what are you going to do? A spark pool? Yeah, this is a spark pool. That's a new one, right? Yeah. Yeah. And this one is, it's based on E-Speed. So it's basically a demo device. I assembled it from the ground. And it's just basically to show different levels of existence. It's a D20, no, it's a mini temperature sheet. It has a PLD and it has a button over here. It's not that, it's quite nice to use. Yeah, but I like the idea of it just like this. So what did you get? Adafruit or, is it from Adafruit? No, no, it was kind of like a unit for... It's a kind of internal project that we had with IBM. So they made these units? Yeah, it's actually one guy in IBM decided to make this kind of thing. Right. So this is some more IRT stuff. And she basically published the way how to assemble it. And I just ordered the parts. Oh, you did it yourself? I did it myself. So what are you going to do with the case? Oh, that's from this... Aero Pass. They have a lot of these project cases. Yeah. You put the OLED in there, isn't it? An OLED? Yes, it's on the screen. You touched it yourself? Was it part of the case? No, no, no, I just touched it myself. So I had to adjust the case a little bit. Cut things out. Unfortunately, it doesn't have a battery. So that's something that I can still work on. Yeah. Do I get another one? But there is no space there anymore because I was like... I had to... I mean, it's a little bit of a high-class thing. In addition to the battery, it's sort of like a child battery. So it's a little space like a shield. Yeah, it should be something like a shield. Is that your favourite Beatles album? Is that your favourite Beatles album? Have you heard? That's my favourite T-shirt. That's a pretty good album. I can probably go more forward with myself. Actually, I think I've listened to one of them. Did you get some food? No, no, I didn't. Do you have a pizza? So yeah, this one is pretty nice. And that's actually a replica of the Yahi. The whole thing? Yeah, the whole thing. So you use the Yahi project as a monster? As a base. That's the western scene. Is that the Xinyi? What's the name of that company that makes them? It's a Chinese company. I think... I bought it as a group. But that's basically something that's smart for me. Have you done any correlation between the readings for that PM2.5 report? I think the roll-ons came off as some kind of proposal. I asked him the same question with the panel last week about being careful with consent. It's related to numbers, but it's about hundreds of accuracy. Which is quite a big error. Actually, that's one of the things that I was trying to address. Because I'm collecting data from these things and these things. And from these as well. You're publishing it somewhere, or not? Right now it's a pilot, so I have a UI for it. But the data may be not up to date. And you're putting it all into emails? Have you got a dashboard? Yeah, I'm going to show it as well. Yeah, I like that too. So I have one too. I think I have seen a lot of YouTube videos. Yeah, they have. I think you can also go ahead and take a look. I think you can go ahead and take a look. Thank you. You know, it depends on the... So I'm just going to close this up here. And then I'm going to make the... This is... That's the other problem is that you don't have to write anything. I think I'll just leave it here. I'm just going to leave it here. I'm just going to leave it here. I'm just going to leave it here. I'm just going to leave it here. I'm just going to leave it here. I'm just going to leave it here. I'm just going to leave it here. I'm just going to leave it here. Okay. I'm going to do my French. Oh my goodness. Oh my goodness. Oh my goodness. Do you want to teach me? Oh my goodness. Do you want to teach me how to shop? How to shop? How to shop now? Yeah, she has a... Actually, I want to talk to each other. She has a printing business. Yeah, yeah, yeah. I have seen him... Yeah, so he has a printing business, and he has... Yeah. Catching... Catching a... You will. I will know what he is doing. I will know what he is doing. No muy no. No. Interesting. Oh, because blog last year, I am. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Yeah. Okay. Okay. also have the color coding similar to what NYACI has. So it will change the color of this particular measurement here. Now, when I did this, I started thinking, OK, so I've got one sensor, I've got the UI, and I've got the whole infrastructure in the back end. So why don't I extend it a little bit? So now the data is coming into the IoT foundation. So first, I think I will show you the architecture here. So the data is coming into the IoT foundation. It's one of the services on the Linux. And then it goes to the Node-RED. Node-RED is the tool that allows you to create nice transformation flows for any kind of data that is coming into it. It can be IoT foundation, but that can be other data sources, and I will talk about that a bit later. So after I'm getting this into Node-RED, I'm actually persisting this into the database that is called CloudNet. It's JSON-based database, but through the SMS. So here, I'm getting the data. I'm transforming this into the GeoJC, persisting to the database. Then at the same time, it goes here and checking, in this case, it's checking the temperature, because I also connected the temperature and the sensor to it. And then sending over here SMS to a top Twilio service into my mobile phone. So that actually can be some kind of extension of the Yaki project in the future, so people can subscribe for the notifications for some particular area in the city. Oh, OK. I'll try to speak louder. So next thing was to tap into the data from Yaki, and it doesn't work right now, but the idea was that I can subscribe to the Spark cloud, the cloud that is used by the particle photon and all the devices from the particle to send the information. It's similar to our IoT foundation, but it uses its own protocol. And then I also can use it to basically this one node over here that does it for me. So I can specify here the topic that I want to subscribe to, the token, and things like this. And also, transforming to the format that I support and then put into my database. And then I tap into the data of the NEA. So NEA has the API that is published, and they also publish the PSI index, but that's only five sensors comparing to 20 sensors of Yaki. And I created a small node for NEA API. So here I'm just putting the ID and then choosing the data set, and I've got three of them right now. One of them is a PSI update. So that also denotes how it transforms the dating to the appropriate format and also puts, again, into my database over here. So on the Bluemix side, that comes into the two parts. One is everything that comes to managing the hardware. And that's the Internet of Things foundations. And then the Node-RED application that is running on the Node.js, basically. It's a Node.js application right now. And then my cloud and database, where I'm persisting Node.js. And that's another thing that I was playing around with a little bit, but it doesn't work right now. It's called the real-time insights. So it can show me the statistics if I will have like hundreds of devices. I can see how they perform and how the date is coming from them in a kind of nice dashboard kind of thing. And then the second part is everything that comes to UI. And that's also quite simple. It's Node.js app. And the same cloud and database. And in Bluemix, the nice thing is that I can share the services that I have across my applications. So here, I'm using the same database, basically. And on the Internet of Things, it looks like this. So I have this Yahoo device connected. And there are some. Basically, when I'm registering the device, I'm getting the security password. So my device will use the password to connect to the code. So it will be kind of a first thing in security. And second, I can use the SSL to encrypt the traffic. But I'm not using it right now here. And from this UI, I also can see lots of statistics on what was published and what was the simple data that was the last message that was coming from the device. And then what was the connection log and things like that. So I can do some kind of analysis on the device and how it's performing and how it's connecting. OK, seems like this one doesn't work well. And the real-time insights, that's the thing that I didn't manage to make work yet. But the whole idea is that, as long as I'm getting the data from the IoT Foundation, I can generate real-time analytics on what kind of data is coming, what are the sensors, how they're performing, are there any alerts on the devices and things like that. And then I also can create a simple rules. They are a bit simpler than a log graph, but they can do things like sending SMSs or emails and trigger some kind of actions in our asset management software that's called Maximal. And actually, there are lots of other things that we have on the data analytics. Like the one that I want to explore, and I didn't have a chance yet, is geospatial analysis that allows to create some kind of geofencing applications where I'm online, I'm getting the data through MQTT and then I can trigger some actions if my device is crossing some kind of virtual border on the map and things like that. So this is the complete, let's say, project, but I have another one that is ongoing. I mean, complete in terms of, at least I can show something working, the both hardware and software. But I also have one more thing to show. And that's another nice hardware, I think. So first of all, I probably will show this one. So this is my hardware for Yahoo. As you can see here, I've got the particle photon, the dust sensor over here, the dust sensor from Groove. And DHT went into the humidity and temperature sensor. So they're all connected to particle photon. And this is another project that I've got. It's actually based on some internal project going on inside IBM. It's more like a demo device, the IoT Box, as we call it. And it has a very nice packaging like this. And it's based on ESP8266. And that's basically the core thing. And it has a small LED screen over here. And the same DHT22 sensors and a small button. So this thing, basically, also can connect to the IoT Foundation. And you can see here it's saying MQTT connected. So it's currently connected to the cloud. And then the thing that I'm still working on is, like, trigger some kind of actions and do some kind of data analysis on the cloud. Basically, push the button, something should happen or trigger some action based on the temperature and humidity. So yeah, I spent basically the whole weekend assembling this thing. And if you want to create the same thing for whatever kind of purpose, just for fun, it's all open. So you can do the same. I can just share with you some links and you will pick up the, pretty quickly, I think. Because for me, like with the basic soldering skills, it took like two days. For you, I think it's going to be like half a day, maybe. So this is it. This is all I have to show. Do you have any kind of questions? Any kind of sensors? Yeah, I used DHT-22 because I like that it's humidity and temperature two things, not one. But originally, it was connected to another temperature sensor. If you know the one with the long wire and the temperature sensor at the end, which you can put anywhere, basically, it's more like industrial kind of thing. So ESP is very cheap. I bought, like for $15, I bought five of them. And then the most expensive is the screen and the box. The screen is about $10 and the box is about $12 or so. I don't remember. The rest is very cheap. So the total cost is around $40, maybe. And what else is inside the box there? So I have ESP8266. I've got the small LED that I can control from the ESP. So I can basically publish the event on some topic in the MQTT. And then it will trigger LED to change the color. Then just the temperature sensor, the button over here. And then how it's called, the step down transformer from five volts to three volts and the micro-USB connector. And that's all basically it. Yes? My question is, actually, first, I forgot my role. Let's actually find any agents at the December last year. 24-month impressions. Mm-hmm. OK, I was counting. I was counting here. Oh, well, no, so they don't publish the location. OK. But it's, yeah, in fact, the power supply analysis in the network has integration with the G-end ones. Question on just on a little bit, you hadn't yet done the coming for the sample that I'm doing. Do you have a little bit of a question? Just don't get there yet. So my problem is that I need coordinates and the actual data together. And in your case, you have. Ah, and I don't mind that there is a blue. Yeah. They are there. OK. I can fix that. Yeah, so that's basically done, I think. Then once I will get both of this, then I can do it. That's fine. I'll just add the zero numbers to the micro-USB. I think it should have some connection somewhere. Yeah? Yes? Yeah. Yeah, there is an open source library for that. And if I will switch back, I can show you the UI. Yeah, that's some of the things that I probably missed. So thank you for this question. This is my, oops, this is the UI for this IoT box. So it has a small, like, HTTP server running there. And I can control the configuration through the Wi-Fi and through the browser. So I've got a number of things here. The Wi-Fi settings, the MQTT settings, and password to protect it. I can change, like, the options of two sensors that I can connect. LED control page, and that's basically, I also can control the LED through this UI. And check the temperature and humidity measurements. You'll see it's 28 degrees and 47% humidity. The data may be not very accurate, because I think I overheated the sensor when I was sold running it in. And then the nice thing is, another nice thing is this over-the-year update. That's the thing. I think it was published just not that long ago for the ESP8266, so it can do over-the-year updates for it. It's a very nice thing. Yeah, and this all based on the open source. So just this guy in UK who assembled this all together into the one nice box, which is very cool. The sources for this project are going to be published. He's going to publish them soon, just need more time for this. But the firmware itself is available. So if you assemble the same thing with supported sensors, then you can just uphold his finger and play around with it. But it is also based on some open source, so you can just take the sources that he used to create. Any other questions? I wanted to make motivation to actually take the blueprints and try to feed with the load of data. For example, for the hands, I mean it's maybe a few months before it's going to be finished. And then afterwards, you want to be able to do the files. The hands is usually like three more years. So I mean, I have also myself two sensors, et cetera, which are, you know, you could put together at least one time ago. But it's just like you want to take the scalability of the system, or what is the true motivation of actually setting up that from hardware and software? My first is to learn, because that's basically for me to play around with things. Second is to have a nice demo for blueprints, because eventually that's something that we are selling as a company. Third, I think it has a potential with this subscription thing. Remember I showed this subscription to send the SMS, right? So what you can do with blueprints and the geospatial API, you can create a UI for the citizens to subscribe to the notifications for some particular area. Let's say the place where they live or the place where they work. And then when the measurements from Yahi together with NEA will reach some threshold, they will get the SMS. And that can be paid subscription, for example. That's just an example. Then the data itself, like we at IBM were saying that the data is the natural resource, the new natural resource. So when you get the data, you eventually can figure out what to do with this. And the environment data is a kind of interesting thing. Let's say if you want to share it with property group, for example, that can somehow affect maybe the ratings for the property in Singapore, why not? And feed back to NEA at some point if they will finally decide that they can rely on this sensor network. So that can be plenty of options. And also, the NEA data is actually a 12-hour prediction. Yeah, 12-hour prediction. That's for. This is done by then. Yeah, that's done by then. But that's not for PSI. It's for the weather, which I know. PSI is just one of those. OK. Thank you very much. Thank you. I can pass around this one. Which, oh, this one. I need to disconnect it. I don't have it. I can pass around this one. So if anybody wants to have a look. Am I on? No, you're right. I feel like I'm the snake oil man coming on next to get things ready. No commercial talks. No commercial talks, ah. Next. Like so, like so, like so, like so, like so. Not like very, very, very, like so. I need, I need to. OK. What do you think, everyone? My name is Willem. Am I on a mic? Am I on a mic? You are live. I'm live. Oh, dear. It's live streaming. Hi, everyone. Hi, internet. This is Willem. And I'm here at Blocks Everyone live at Hackware. At Hackware. My name is Willem. And I am now trying to set up stuff. I'm going to do a couple of things because according to the schedule, I'm supposed to speak on three things. And I only have 20 minutes. And I don't know how I'm going to cope. So what I'm going to do is the best is we do a demo. So the good thing is I have tested the codes and I've done a trial. It won't go wrong. It won't go wrong. All right. OK, first of all, just a bit of myself. Some of you, you know that I've been doing this maker thing for the last couple of years. There's a new thing which I'm doing now, which is the IoT. About six months back, I had a chance to chance upon the existence of this really interesting chip set called the ESP8266. By chance, I happened to meet the guy who actually made the chip. I happened to the guy who owned the company that made the chip, Srinan. He was in Singapore and he gave a talk. And since then, I was really enamored by what he can do. And I really believe that this thing is quite awesome. I know there are also other chips around, but nobody came to visit me, only Srinan with his ESP8266. So TI, Broadcom, if you are listening, come and visit me. All right, so today, I'm going to talk about things which are really quite different. Many of you, when you give a talk, you always like to talk about the best product. Am I right? Today, I'm going to show you the worst product I ever come across, which is an ESP201. How many of you heard of this before? ESP201 is one of those banggood stuff that you see on Facebook. This is actually the ESP201 is actually a breakout board for the ESP8266. And I believe it costs something like $3 US. There's this white piece. And the most interesting thing is that it is so useless that it needs to have a piggyback board, which actually has all the components that you actually would require for doing any IoT education or doing some sensor. It has a relay. It has a DHT. I think it's 11. It has the LED RGB buzzer. Sounds remember when we do our Arduino lessons? We have all the components that we ever use to practice Arduino. Imagine all those things can be packed inside one board. This is it. Guess what? This is only about $14, $16. This is really cheap. I tell you, whoever is going to come into the ESP8266 business is going to lose money. Because how can you compete with these kind of stuff? But we are going to try anyway. One of the new projects that I'm working on is called Maker Asia. And Maker Asia is basically think of it as what if there is a way to help makers be involved in IoT hardware or creating IoT solutions, products, and services. What if there is a way? So together with a friend, a business partner from the Chiang Mai Maker space, his name is Jimmy, we actually came together to form this new setup called a Maker Asia. And one of the things that we wanted to do is to create, first of all, a hardware, a universal hardware that it is easy to use. And in order to create the best hardware, we have to look at the worst hardware possible. And then we learn from it. So today, I'm going to share with you how to use the ESP201. So I have three with me. So what I'm going to do is I'm going to just show you very quickly how to set up a very simple personal network. Ingredients. I need my own Wi-Fi router, gateway, slash MQTT broker, which is all within this little Raspberry Pi. Hope it works. If you actually scan, this is actually on at Tespa. Tespa is a name of a project which recently tried to go on Kickstarter. They tried to kickstart this hardware called the Hawk, which is meant to be the Lexus of the 8266 hardware. Unfortunately, it did get funded because a lot of work needs to be done in order to market it. But nevertheless, we are going to go ahead to cannibalize the Hawk and add a bit of Rojak from Singapore, add a bit of stuff from Shanghai, mix it up with some Seat Studio stuff in San Chen, and then voila, we're going to give you a new hardware called the Expresso. Maybe I'll just very quickly talk about this first. How do I make it expand? Isn't it? All right, IoT is important. Everyone should be involved. Everyone should be involved because it's the next wave. If you do not know why it's IoT, you should not be here. This is geek hackers, right? Current stuff, we need the cloud. The cloud controls the device. What is so special about Expresso, which is the new hardware that we are trying to develop, is that it has a couple of cool features. Number one is OTA. OTA has been around for a while. It's one of the features that opens up that A266 actually allows you to do. So this is very good, particularly for people who let's say, you know, supposing I want to put this and embed A266 into thousands of consumer devices. Imagine I have a security fault. If I have to call back, I'll be in big trouble. So for a situation where I'm going to embed it into thousands, the ability to do OTA, firmware, OTA, it's really, really important. And this is really something which I hope that the new project will allow us to develop something that is going to be very user-friendly for people who are going to embed this into numerous consumer products. It has rumors, has it that it is one of the first few Wi-Fi protocol that allows what we call mesh network. It is true. In fact, I'm on camera, but let me tell you this. Every A266 chip is actually over-designed. There is a hidden feature, which is a mesh network that actually is embedded even in the earlier version of A266. So it is something that is overly designed and only very recently the SDK or the software caught up with the hardware, which makes this a very awesome product. Imagine if I embed Beacons, Bluetooth Beacons, and the Beacons actually talk to one another. Currently, it doesn't talk to one another. It just broadcasts and ping. What if it talks to one another? Imagine the possibility of the kind of solution and services that we can build. It's really amazing. I wish I can show you a smart config demo, which I wanted to, but I guess I forgot to download the app. It's actually an app which allows all my devices to be configured to a particular Wi-Fi gateway or whatever instantaneously, simultaneously. So introducing ExpressO. I know this month, but this is just a concept. At this point in time, we are going revision four. There's a revision four. So by mid-October, we should be able to release the final prototype. Then maybe it's in time for the next headway. So think of ExpressO as a small form factor, a small little device, a small board with Bluetooth and OLED display built in. And if you don't want the OLED display and the Bluetooth to strip it off, it is actually an ExpressO light. The ExpressO itself, it's only made more powerful by adding on an expansion board, which is a piggyback board called a mocha. For coffee lovers, mocha is the port that is used to make ExpressO. So we believe that mocha will be something that allows us to squeeze more juice from the ExpressO, as well as mocha will become a universal development board for people who are already using things like Node-MCU. And hopefully in future, or even 201. And hopefully in future, Edison. Intel, are you listening? Hopefully with Edison, with even particle chips. So we hope this is actually a project which we are currently working on, just started, only about two, three months back. But we hope to come up with the idea with the actual prototype very, very soon. So yada, yada, why? Because we wanted to approach the IoT so-called space not just the hardware, but it includes an app, software, as well as a cloud level. Now currently, there are three communication methods, Wi-Fi, Bluetooth, GSM. In future, we believe that we are going to move on to having more application involving GSM. I believe the next speaker is probably going to talk about it. But if you look at the applications, Wi-Fi is still the mainstay, especially when you talk about fixed infrastructure. Bluetooth will be very useful if you have those kind of things that you take on the go, or personal objects. Where GSM is, imagine you want to track your dog, you want to track your car, where does it go? GSM is actually one of the better communication methods. And we believe that something like this would allow more people to actually build applications around these use cases. So currently, I'm also wearing another hat, which is I'm with the one-maker group, which where we operate the prototyping lab. And one of the things that we really want to encourage is more people to build stuff, more people to prototype stuff, and hope so that we can create really, really exciting and vibrant ecosystem in Singapore. So we wanted to play a part in this first by building hardware that is quite upstream. And hopefully, these are the possible spin-off, which will be so-called something that people could monetize, not just us, but people around us, people like 12 gigs, 11 awesome people, or something like that. So basically, these are the market that we wanted to reach. I am a pure user. I'm also pure by the bloodline in the sense that the guy who taught me Arduino is the founder of Arduino himself, all right? So the rest of you are bastards. Oops, bad joke, bad joke. So because of that, later on, I will just show you how to upload very simple so-called stuff from the Arduino. Thank God to whoever wrote the Arduino library, we can now talk to tricky things such as A266, right? So there is a huge market for open source hardware. And I must profess that the first thing that we're going to do is, once we develop this, we're going to release even the schematics how we actually build it. So if you want to build it yourself, carry on. So there is no at an IP or whatever. So this is something which we want to write on the existing ecosystem that we are familiar with. So ultimately, we hope to add value to the whole value chain where we can create something for distributors. We can create something for startups. We can create something for maker space to onboard people into the IoT space. And we can hopefully, through this model, get even better ideas to improve this product. So some of you are familiar with this. This is just where do we position in terms of how much is this, still working on it. But we understand there are already players out there. So yeah, so people ask, since there are such big players out there, why bother? So the question is, why? So the answer is, we can say why not? Or we can say, look, I think the market will continue to grow. And the only way to grow is, you are the one growing it. So we want to be the one to seed. And hopefully, we want to be the one who reap as well. So we are trying to kickstart something very soon. So this is just some manufacturing schedule. So expect more news on that. This is just myself, Jimmy, and Svian. So last week, I managed to grab Pao Si to be one of our advisor. So thank you very much. And that's all. So right now, I'm going to quickly show you something which is going to be what we call it. Remember, I said earlier, to build the best product, you've got to work with the crappiest product. And the crappiest product we came up with was the 201. I just need someone with power source. So that this is someone with power. Because I only have two USB. All right, let's do a quick demo. Can you? I only have three more minutes. No, no, it's OK. No, no, you just need power. So USB. So for some of you who have not tried this before, the guy who designed this is very strange. The guy who designed this is very strange. It has two so-called USB ports. But one of them is not for data. One of them is for power. The mini-USB, the micro-USB, is actually just for power. Because in China, they make a lot of this power bank. So you can actually power this for power bank. So there's no data transfer. So I just need power only. So is it? And the interesting thing about this is that it comes with a switch. Awesome, right? If you don't want to use, you just switch it off. But the switch doesn't work because they use the cheapest switch on Earth. It sucks most of the time. So if you want to see, you've got to, you know. All right, so this is the awesome thing about this device. I just want to do just a quick test because this board, earlier on, I made some disembarrassing. First of all, let's see what we can do. So currently I've designated this board. This board comes with a couple of things. One of the things is what the earlier speaker had earlier on is actually a humidity and temperature sensor. Now, if you want to program it, we actually have a library of stuff. All this code is kind of provided in the example. And you just need to kind of like tweak it and upload it. Now, the trick about uploading this is that, about this hardware is that when you want to upload a new sketch, you've got to do a silly thing. The silly thing is you've got to switch it off and turn it on again. Don't ask why. It has this weird so-called feature that it doesn't refresh itself. And you've got to do it in this manner. I'm not sure whether I have problems, but most of the time it will, because remember, this is actually the crappiest product that I have. Now, I have an error message. So when you have an error message, what do you do? Try again. Switch on, off, and then try again. You see the entire demo? I just show you how crappy it is. It's uploading, and I get error message. You know Murphy's law? When I did it at home, you don't get this. When I do it here, bad demo. OK, moving on, moving on. But you're supposed to work it out. No, it's supposed to work, it's supposed to work. Now, let's try this. Now, although it's crappy, but one of the good things about ESP8266 is that every chip or every board has a feature which turns the ESP8266 into an access point. So let's hope this thing works again. I'm going to hold it. Oh, dear. Why is this happening to me? You know if you hold it long enough, it will turn into an access point. So I'm going to hold it. This is bad, this is bad. It's not like nothing is working. If you have a right board, it's the other one. Good question. Good. It's the wrong board. Oh, yeah. I think so. I think so, I think so. Don't. Too soon. No, no. They pull out one so that you only have one board. Yeah, no, yeah, no. It was even right the first time round. I know I'm on camera, but do you know that actually with this right, technically I do not need to buy anymore? Do you know boards? Ever. Again. Shucks. This definitely looks like a better option than the other one. Sorry? This looks like a much better option than the other one. Yeah. OK, I tell you what. Is it OK if I pop out everything here right now so that to avoid less embarrassment? But we will do it afterwards, OK, afterwards. Come over to the booth at the back. Shall we move on to the back? Yeah, yeah, let's move on. OK. Yes, yes, yes. There's always network in. Yeah. All right. Thank you very much. Thanks. I believe when Steve Jobs did the apple, he was... I have kids, so I brought them along. Later, I'll pass them around. So I mean nothing beats looking at it personally, right? So I brought them along as well. Here's the clicker. So let me just set up those. So it seems like the topic for today is like IoT, right? Everyone's talking about IoT, right? Yeah, I think... Oh, just let me set up. So I don't know about IoT. It's like an over-high password for me, but there's another debate for another time, so... Let's not go into it. OK, so everyone's good? So let's get started. So some of you might not know. So you are new. So my name is Kenneth. I'll be talking about the two topics I mentioned earlier. So the first one I'll go for is the clicker. So everything is shared on my GitHub repo. So if you need more information, we can chat offline or you can find more info there. OK, so this is a picture of the clicker. I have it here today. I didn't pray to the demo god, so it's not working now. So yeah, I'll just click using my keyboard for now. OK, so this is the clicker. So this is sold by a company called Punchthrough. And the reason why I chose it for this project is because I already have it available. So I bought four in a pack the other time when they first launched. So I have them around. And they are running on the Arduino platform that I'm familiar with. So that's the reason why I chose it, and it's pretty small. So as you can see, it's just so big. So without, I can pass around now since it's not working. OK, you can take a look. So yeah, because it's small, so I chose it as well. And I've been thinking of making this for a very, very long time. I always wanted to buy a presenter tool for presentation, but I never wanted to do that. It's like $30. So I'm like, OK, I'll borrow from someone. And so we had a gig camp last Saturday, just last Saturday. So I thought it would be nice to make something for the presentation. So that's the motivation. OK, so let's look at how it works with some code. So it's pretty straightforward. So this is for the left button. So basically what you see here, it reads the button for input. So if it reads, then it sends. So you'll see over here that it does it for 10 times if it doesn't receive a response from the computer. So initially when I did that, when I clicked, then out of 10 clicks I get one response from the computer. So I thought I could do better than that. So my idea was to make sure the computer was sending something back to me. And once I get something in response, I got that response and I would just exit from this code. So as you can notice, it's very crude for people who do a lot of programming. There's no debousing or whatever they're doing here. It's like, oh, if I detect a change in a signal, then just send, send, send, and get out. So yeah, it's pretty crude. But I did it in a day also because there was not enough time. So that's for the Arduino side. So it's pretty straightforward. And for visual feedback, I just toggle the LED on and off. If you pull it out, you can see the red LED blinking. So yeah, that's just for visual feedback to know that the device is running. And this is running on the computer. So pretty straightforward as well. So once the serial port gets the data, it will just send back. So you just read whether it's L or R. And using this package that I found online, it's called RobotJS. So it's already a package that allows you to control the mouse and the keyboard of your computer without doing anything. You just have to call the command and it will work as it is. So it saves me a lot of time. So the whole thing took me around almost a day to complete. And I had a working clicker that's not working right now. So once we eat that more time, but there's no demo. Let me see if it's working. So if you look at this one. So this is the app that's running. So oh yeah, it's working. So someone just click right? Yeah, try clicking right? Yeah, so it's working now. So you can see it's live. So the guy's clicking and yeah. So it's pretty straightforward. So you can do a lot more cool stuff. The cool thing is you can make it super flexible. You can prank a friend or you know. Like that's what people used to do right? They make a USB device and plug it onto someone's computer and the mouse starts running all over the place. Yeah, so this is something that you guys can do. But just don't tell them I suggested to you. So yeah, this is a clicker. I thought it's a fun little project that you can do. Okay, I have to quit this. So yep. What is the link for video? All right, so that's one part of the presentation. I thought this might be more interesting for you guys. So it's pretty cool like the first speaker doing the Polygo stuff. So those of you who don't know Polygo is an IoT company. So they sell deaf kids that focus on IoT products. So yep. So the one I'm going to show you today is called the electron. So it looks something like this. Let me pass them around. There's two, like one in the black casing and one in white. I'll explain to you why. So you guys can take a look. Just don't fiddle with the wires because it's still pretty better and you might sort out something. So just look at it. After the presentation, I can unpower it and I can flip the underside and show you what's going on. So just to give you guys a quick introduction is a cellular deaf kid with a global data plan. That's the first very interesting point and they are successfully funded in April. So in Singapore itself, the data is US 399 for one megabyte and thereafter is 299. So you might be wondering like it's kind of pricey and one megabyte. What can you do? So in the next slide I'll tell you more. So the interesting thing is also that you are cheap and there's no contract involved. So at any point in time, you can just suspend the plan. So that's what makes it so attractive. So this is from their Kickstarter website. What can you do with one megabyte? So they calculate that you can roughly send probably 20,000 messages a month based on the protocol they are running. So these are some interesting things that they can do. So you guys might want to consider that. So here I'll go into the details why there's two models. So the one we have one is the U270 and one is the G350. So if you can see over here like G350 is for the 2G network. So it's running on 2G. And we have a 270 model in Singapore using 3G. So the people in North America is running on the 260. So a reason why there's two variants is because for example in Singapore in 2017, they're going to stop 2G service. So the minimum you're going for is 3G then 4G and hopefully 5G when we are in 2017. So this product is good but two years time it's going to be phased out in Singapore and some other parts of the world. So it's important that it came like a 3G variant. So that's the reason why we have two as well. So like I mentioned they are offering three kinds of package over here. So you can see that this one is for 2G. It covers South America and the Asia region as well. And 260 covers the North America for 3G. And in Singapore we're using the 270 which covers the Europe and some countries. But there's something weird with the Japan network and stuff like that. So they are excluded from the def key. Where did the numbers come from? Sorry? It's someone else's wish to know that it's not available. So this is from the cellular module manufacturer called U-Blocks. So do you know they have a cellular module? So the one that the first because using the other one is Wi-Fi module. So obviously they swap out the Wi-Fi module and add in the cellular. So you might be wondering like why haven't people done this before right? So cellular modules are complex piece. It needs a weird footage like 3.8 and 3.3. So to make it happen it's super weird. And when it starts to send messages the current can spike up to like 1.5 amps in a spike. So power management, voltages and stuff gets really tricky. So that's the reason why they are def key so popular. Because they swap everything in one package in an Arduino familiar platform. So this is the delivery timeline. So we are going to hit October 1st pretty soon. They are going to finalize on the hardware and somewhere around January they are going to start shipping. So this is the semi-uplitted timeline. So I think they are probably delayed. I think some of the people here in Singapore ordered. So this is the beta unit that you are seeing right now. So probably just a few in Singapore right now. So you guys are getting like a rare preview. Okay so I am going to tell you what I do behind the scenes for the beta program. So there is nothing that I can't tell you because everything is open source. So yeah. So this is the unit when I first got it and the sim card. So they are using this provider called Telefonica. One of the world largest mobile plan provider. So what do we do for beta basically? Like a few things where we get it. We try to run through the documentation that they provide us. We see whether what works what doesn't before we reach the general public. And we do a lot of firmware testing and feedback. So they give us a new binary and we test tell them what is working what is not. The likes are not in sync with what they describe it to be. Stuff like that. And I do some random probing. I think that there might be some issue here and I start to probe and see if there's an issue. So one issue I went into is the charging issue that I will share with you more. And yeah you will have to expect the unexpected for beta. So last Friday suddenly all the beta units worldwide went offline. Yeah. So Telefonica had a wrong setting in their account and they were in the test mode and it expired and all the device went offline. So for you guys who know like the SLA agreement right it's a pretty big thing. So yeah they're trying to resolve that now. So I mean it's lucky that we are in beta right now and we found out that this might happen. So they are in talks with them to resolve that. So yeah that's one interesting thing as a beta test. Suddenly your device just goes like offline. Yeah and everyone in the world in the communication chat will start to talk to each other. Yeah then you know something wrong is going on. It's really more of like more tinkering. You keep playing and playing and playing with the device and get familiar. So I'll talk a bit about the charging issue that I noticed. So I had two units that were... So I placed them under reset mode. So basically it's not running and I charged them and they were fully charged. So the lithium batteries were fully charged and I started running them with the USB cables on. And after a day or so I noticed that both of the units went dead. I was like maybe there's something with my power supply and I just power cycle everything and it got working again. And I experienced it for the second time. So something's weird there right. So I set up some testing stuff and do some data logging. So if you notice over here this... I think two days ago. So you notice that after I did some testing I could get it to charge itself every time the power drop. So what this graph is showing is the state of charge of the lithium battery. So you see here it's around 98%. Once it starts drooping to 0.93 the automatic charging kicks in again. So I was trying to test and make sure this was working. So it was charging and discharging and it kept working pretty well. So before that was a bit of a weird setting. So there's the input voltage threshold was like 4.82 volts on the chip and my USB was like 4.80. And it took me a day to realize that. There's some settings issue and some power issue that they didn't tell us because it came like this in the box and the settings were not initialized on every unit so there's some weirdness. So that's one thing that I managed to resolve after doing some beta and some testing. So here's the setup that I have running on my table. So one 2G, one 3G unit and another one is their first generation Wi-Fi DevKit, the core. So what it does is these two units will publish their state of charge every five minutes to the cloud. Then it goes to the cloud then it goes to the device running the Nokia display there and it will start to blink rapidly to tell me that there's a new data coming in. Then you must be wondering why you have to go through such a long process. It's because the TCP application on the Electron is not fully baked and I tried to compile and it failed so I had to do some extreme ways to get the data available to me. In the past I had to plug in 2 USB port on the serial terminal and look at the number all day and try to figure out is there a change. So I had this Simplify setup. So it makes it easier for me as a beta person. So that's all I have to share. I use this plotting service called thingspeed.com It's an open source data plotting and monitoring tool that you guys can find as well. It might be a useful resource that you want to try out. It's very easy to set up for a particle depth kit. I can log in and show you guys but I'm done with sharing and if you guys have any questions, happy to answer them. Why is one white and one black? We've been the issue. No, no, no. So I did it on purpose, right? So I keep... That's another joke that I need to tell you guys. Only yesterday night when I was preparing the slides there I realised that I've been labelling the 2 devices wrongly. I was labelling the 3G device as a 2G with a 2G name. So it's like electron 2G device, a 3G device. So only last night I realised, I went to ask them, oh, I better not tell you guys the wrong thing. So I updated everything as of today's slide. So I made a black and a white to tell myself which device am I staring at now so that I don't do the wrong thing on the wrong one. So that's one reason. Any other stuff? So I can probably pass around this one as well just to show you that it works. So there's no lock-in to their SIM card plan but just that they provide a lot of advantage to you but you can just plug into any SIM card that you want. But I think the minimum plan in Singapore is, I don't know, 3G or 1G, something like that. And I don't know, it's like 10 bucks a month. And there's an interesting story that I thought you might like. So the founder went to Africa for his vacation and he brought his electron along. And his electron connected to the cloud better than his phone connecting to the network. So that's one interesting fact. And he was moving around countries during his transit and it seemed to work pretty smoothly. I don't know how they're going to figure out the roaming charges and stuff like that but it seems like you can just roam and the charges wouldn't change much. So that's a question that I want to ask them tonight. But how is it going to work? I think it's really interesting if you could just move your device around the world and your pricing plan doesn't fluctuate that much. That would be an interesting advantage that they have over getting a local SIM card. Anything else? So telephonica, right? I think in Singapore the carrier that is supporting them is Starhub. I think so. Just they will tell you it's like telephonic card but they don't care who's below. Because I'm holding an international number if I'm not wrong. I have no idea where that number is taken to also. But it just works. So the interesting thing about the cellular DAF kit is it allows you to deploy in places where Wi-Fi is not available. So people are talking about IoT and IoT and stuff, right? So I don't think the real thing is about IoT. It's about you being able to remotely control your device or get useful data out of it or do some application that you could never once thought of doing. So the battery that is shipped with us is a 2000mAh. I unplugged it from the USB at 5pm before I left. Right now it's something like 80% for the 2G and somehow 40% for the 3G. So I don't even figure out why so much difference. But most of the time the kind of project that you would want to be working with is long sleep and you get away and send data. So for like a 2000mAh probably can last you at least a month. I would safely say at least a month. But right now I'm doing testing so I'm like 24-7 running and running. Is there a Wi-Fi ready on board as well? Just say we know. There's no space for anything else. I can show you the underside of the body. You guys have more questions about the clicker? I feel that this one might be more interesting for you guys. Use case for this one? So my friend is trying to work with this company doing a green recycling bin. So they have this compactor inside there. So they need this kind of application where cellular is useful. Because there are places where they deploy that they won't allow you to tap on their Wi-Fi. Or there's simply no Wi-Fi. So that's my application. I would really love to place it on my bicycle or something. With GPS and stuff like that. So that's something I'm looking into as well. It's really interesting that you can track your stuff. And the other one that I want to do is a hiking device. So I'll go hiking and get a lot of data. I don't want it to... I want to enjoy my hiking activity. I don't want to worry about data uploading and stuff. So it will just do at the back end and upload wherever I'm traveling to hiking. So these are some applications that I think of that might be useful. Anything else? If you're tracking a CEO, then maybe that's... They did that. So just a disclaimer. I'm not working for Particle. I have to say this at every presentation I do for Particle. I'm just a more active community member. And I get involved with beta. So I don't work for them. Can you get the discount? No discount, but free hardware? Yeah. Anything else that I'll hand over to... Okay, thank you very much. So we've come to the end of the talk. Are there any... Anybody else want to say something? So we have got a Facebook group where we communicate with someone and share information and stuff. It's probably our main communication channel. And then if you don't use Facebook, there's one person there, then sign up for meetup.com. Let me tell you where we're going to meet up. There's a hackathon happening on 10th of October. It's actually a Meteor JavaScript hackathon. There's an IoT category. So if you're interested in doing that, doing some developers or if you don't meet up already, then... Anybody wants to hire? Yeah. Do they need a job? Anybody needs a job? Yeah. Do you need an announcement? Do you want to make about events or... Meetup or whatever? Yeah, they did that at the back. Yes, yes. So regarding the connection to the data network you can start up IW's fascinating thing. And one thing that I'm aware of that you might be interested in, in case everyone of you, is the internet.org initiative by Facebook. It's providing internet to the two thirds of the world that does not exist currently. So in terms of providing opportunity to gain access to technology like this and also information that all of us are sharing and which are not available to everyone. It's normally not available to all of us. All of us being one-shot in many years possible. But I thought there was potential in this that there was a chance that this would be put somewhere in Africa. Just a while. This is my watch. But it raised possibilities in my head and I just want to encourage you and congratulate you on these things. Cool. I'll pass the message along to the team because I didn't develop the hardware. They did the hard part and I did the testing. But thanks for the encouragement. Just in case you guys don't know, ESPA to 6x will be supported by Particle as well. Not directly by them but another partnership company that is making a dev kit. It's called OAK. So you will be able to connect the ESP to the Particle Cloud in probably a few months time. So if you have a lot of ESP in your box, soon enough you will be able to connect to the cloud. Is it? Yeah, no. Yeah, I'm not sure. I didn't bank them. Thanks, Kringley. Today I'm speaking on behalf of Hacks Hackers, a relatively new group that's still trying to get our act together. One of the acts we're trying to get together is a hackathon for Hays data that's going to be on 13th October. It's a weekday, unfortunately. But if you happen to be free or interested or can take that as a training day or from work, whatever. The point is we're looking for people who want to build civic solutions like visualization apps or forecast apps basically with the NEA data and other sources of data that are comparable. And you can find out more information about this on khack.evendbrite.sg or khack.evendbrite.sg You're talking mostly about the Facebook group or the VLT group? Yeah, thank you. Any minute now? We'll make a vlog. Make a vlog. Alright, in November, it's about a much much time from the 7th November to the 29th of November, we're going to organize the Makers' Vlog. Makers' Vlog is actually, think of it as a three week maker event that is happening at the Pavilion at Far East Square. So expect to see things like flea market, workshops. On the weekends and at least on Fridays, there will be conferences on people working with 3D printing, IoT, drones. So these are being planned as we announced this. So it will be in the month of November. So we're looking for people who, for example, you like to present something or you like to showcase something or you like to just be part of the festivities. We are actually trying to put up a call. Next week we will put up a call for people to gather, to explain further. So just watch out for the call. Thank you. So for the FPGA Singapore Community, so we tried to look into your calendar to check one Saturday towards the end of October, which is coming soon. Try to do a full day at Hackerspace where we're going to present an introduction to FPGA. Bring some calls. If you're interested to start to learn about FPGA programming the hardware, we try to make a full session on a Saturday in Hackerspace. So we will announce this and then try to send some, maybe preparation type of work, what it is, et cetera. So my point is to really make it a whole Saturday in Hackerspace where we have like a full day and people can try to interpret things to an FPGA, start to learn from the code and really just get started with FPGA as a topic, basically like an Arduino or Raspberry Pi FPGA technology and software and hardware. Any other announcements? Jimmy, are you in talk with your friends? Yes. We can just sell them the website. Yeah, so in November also there is a big DevFest week of developer events happening and one of them is going to be a Notebot competition and Notebot workshop so it's going to be a day-long workshop. This is around the JSCon. So Notebot is basically a way for people who have not sort of like, basically introduction to robotics and getting into robotics using NodeJS and usually like an electron or an Arduino or something similar. So all the details are on DevFest at Asia. I'm not running it but a friend of mine is running. I did it last year helping some new blood to start running that stuff. So it's really interesting if you are sort of new into robotics, it's a really great place to start learning and building smaller robots and moving up. So just check out DevFest at Asia for the details. There's also a bunch of other sort of most software events happening but it's one hardware event. Other than that, is that all? Any other? So I haven't known yet about 28th of October we will have a VR meetup. So this time we will be having OSVR from the reserve to showcase their headset. So this is virtually ready to meet up? Yes. Just search for Asia VR Singapore Meetup. Please post all your links in the group after they just go and post it so that everybody can go and just check it. Otherwise, we do have to wear the last Wednesday of every month. So the next month will also be the last Wednesday and don't post exactly. We will again announce at Facebook and meetup.com and the regular places where you found out about this, you should be able to get to know. And as always, we are looking for both sponsors for venue as well as seekers. So if you know of a space that would love to hold us about this kind of people so you can let us know either to Facebook meetup or just talk to us and then we will go around and explore different places. Last month we were when they posted us in their awesome place it was very interesting to see what they are doing so we'd love to move around and see what people are doing. And at the same time if you have your projects to share take this as an opportunity to clean up your project, make it demoable and I have all the news and come in this way here because sometimes you just need that last little kick just to say, you know, I have to fix it before the next hacker so pick it up for the challenge and tell us that you are going to demo it and that will ensure that within one month you will have your project ready. So just contact any of us, Rick, Sani, Fuzzi and me, we have organized this meetup. Thanks for coming and we'll see you next month. So you're just wondering like what's the under