 Thanks, thanks. So I guess before I get started I'd like to get to know where the audience, how familiar they are with the topic and the talk. How many of you have heard of a company called ChibiTronics? Okay, actually a lot of people have. And how many of you saw or heard about the talk I gave last December at 33C3? Okay, good, not too many people. So otherwise you'll be seeing a lot of the same stuff again. So I want to talk about making technology more inclusive. And the first thing I want to do is try to motivate a bit. And the first question I have, particularly amongst an audience like this, is like how inclusive are we about technology in society today? So this is a graph actually from a survey that was done recently that shows how much of society is included in technology. So that tiny little green line on the top, that top 5%, those are the people in this room. They actually were more like even beyond that. These are the people who can do specialized tools and search queries and discard what are called distractors and false results. The people in the purple bar are people who can do kind of searches using multiple sites and applications. And then everyone beneath that, which is over 50%, can basically either do like a basic search query like what is open source? They could look that up but they couldn't use multiple apps to try and figure out what it is. A number of people can't even delete an email and there's a large group of people who can't check email as well in the world today. And so living in this bubble, it's very easy to think that we all use technology, we're all included in technology. And then while it's important that these people are included or not, well the problem is that society as a whole has decided that voting is a good idea to decide things and so then let's go ahead and try to vote on things like surveillance or net neutrality, DRM, your digital identity, your right to own and repair, open source or even the selection of emojis. Trying to explain these concepts to people who have trouble doing a search query is not going to get very far. And the problem is at the end of the day is that democracy is very hard. Things like in or out or like should I vote for a liar who endorses sexual harassment of women is a hard decision for people to make these days and you can't just rely upon society to just make sort of without trying to educate people on the issues. And so for an example more relevant to this audience here, open source. Everyone here is, I mean open source is the name of the conference, it's Foss Asia. We all believe that open source is a good thing. Is it a good thing though, right? I'm an open source activist, I believe that open source and liberty movement is about empowerment. But if we just give people the source code and say hey, you should figure out yourself, have we really truly empowered people? And there's one example I like to give where there's a screen saver called X screen saver which last year popped up a time bomb saying hey, this thing's terribly out of date, tell your package maintainer to go update it. And a lot of people are really mad because when they log into their machine, this time bomb comes up and they're like what the hack have I been hacked or whatever it is. And I Google around and stuff like that, you can see this flamor that goes on between the maintainer and the users and this guy has this wonderful little clip where it says like when they even bother to tell me what version they're running, I say that version is three years old. And I say but this is the latest version of my distro SIP. And then they say your distro sucks but then the other people say I don't know how to compile from source. And then, you know, when he says herp derp, I eat pace and everyone goes away unhappy. And so the point is that even though if you give people source code, they don't know how to use it, both the people who make the source code and the people use the source code aren't happy because there's an expectation on both sides of who's like doing the maintenance and who's actually running the code. So at the end of the day, openness is powerless without inclusiveness. If you share your source code, you're only empowering a tiny fraction of society. But consider that we need, you know, the government's help to enforce our licenses and we need society's help to sponsor and patronize our efforts. And so if you believe that open source is a moral or social imperative, then you need to empower more people to preserve and sustain the practice. But teaching people to code one by one is not efficient. It's not JWZ's responsibility to go and tell people how to compile a screensaver, right? This should be done by a group of people who want to make sure that this happens. And so when you say, hey, Bunny, there's a problem with my power regulator and my battery life is too short, what do I do? I would look for big wins, really easy wins. Things are obviously wrong with the systems. If you have a linear regulator and you're running off of battery power, I say you should probably turn that into switching regulator, you're going to save a lot of power, right? And so if you look around at a typical conference like this one, you can see the gender ratio is pretty predominantly male. And so that already shows that there's a large part of society that's not being included even into the 1% of technology. And so I feel like by trying to focus in on trying to get more women involved in technology, we can easily increase the numbers of people who are included in open source. And oftentimes, when you have bugs that are this pervasive, it's oftentimes emblematic of larger systemic problems. If you manage to fix it, a lot of times you get even a bigger win than 2x at the end of the day. So just to give you an idea of how bad the bias is in terms of including women into computer science, this is a recent survey that was done of bachelor's degrees awarded by gender and ethnicity. And about 15 to 10% of women are getting degrees in the undergrad range. And then it's not in this lower graph here compares how computer science is doing relative to mathematics and physics. And the dark shaded part at the back end is how many women are involved. And even compared to mathematics and physics, women are highly underrepresented in computer science. So it's actually a particularly acute problem in computer science that women are not included. And the problem is not just those those stats were often the United States. The problem is not just the United States. There's a kind of global survey that shows the, you know, sort of the degree trend across many countries. And oftentimes degree rates are as low as like 6 or 7% to like, you know, 20% or something like that. So it's like, it's a pervasive problem that's happening everywhere. And I find that the thing is doesn't quite happen exactly everywhere. Like I think the bias might be cultural. I spend a lot of time in China doing work there. And I find actually a lot of the staff I work with in the technical fields tend to be much more gender balanced. I actually asked my contract manufacturer to send me a picture of one of the recent gatherings. And I counted through the photo and it's about 62% female. And this is not, these are not, you know, just, you know, desk workers and administrators. These are engineers and design, manufacturing, project management, logistics, all this sorts of stuff. And so it's not that there's like some inherent reason that women are not involved. I think there might be a more of a cultural reason. And so I started looking around and found two interesting experiments and cultural change. There are two well document studies from Carnegie Mellon University and Harvey Mudd College talking about some efforts to try and change this. And when these universities engaged in these exchange, these, these experiments, you would see, for example, in Carnegie Mellon, they went from below 10% graduation rate of females to about 45% rate in about, in about five years. And similarly, Harvey Mudd had a similar result where we went to go from like below 15% to about 50% in about five years. And the key insight from the studies is that computer science is unique in assuming that you already know it before you study it. When you go to college, you expect, you start learning in a C-like language. And the professor doesn't go over syntax, doesn't go over the basics, doesn't go over the rules or anything like that. And so when you didn't already play with computers before you went to college, you're going to walk in a classroom and be like, this isn't for me. It's as if you went to, like, a medical program and they say, they expect that you've done surgery at home. Or, you know, a law school already expects you've already done some trials or something like that. And maybe you messed up, but whatever it is, you've played around with it before you got to school. And so, or, you know, it's like the experience that, you know, some people have, you walk into a Chinese class and all you see is a bunch of Chinese people trying to get an easy A. If it's great on a curve, there's no way you're going to get an A. You just walk right out. And, and so what HMC and CMU did is they redesigned their introductory CS courses to have a greater diversity of courses, including ones that assume almost no prior experience. And they also try to contextualize computer science. They try to do more practical and applied programming in interdisciplinary studies. In other words, you don't learn programming by implementing GCD algorithms and whatever, you know, Fibonacci is your typical C examples. They actually say, oh, here's something a little more relevant to your life. Here's, here's why you might want to learn programming. But the problem is that we can't count on changing every undergraduate computer science program. If you look even deeper, you find that the bias, the general bias starts at extremely young age. Here's a graph of the number of engineering and construction kits that are targeted or bought up and given to women versus men. And you can see that even at a young age, girls only form a very small minority of people who are being given engineering construction kits. And there's another study that was, that was happened. This is actually recently published in science, where at the age of five, when shown a pair of people, like a man and a woman, you say, which one of these are really, really smart. They're more, they're equally like they say man or woman. But at the age of six, somehow, the bias shifts. Even the girls are more likely to pick the boys as being really, really smart when they're shown a pair of people. And so, you know, whether this, you know, what causes a shift is still not known, but this is something that has been shown to exist in society. So, I've recently teamed up with some people to try and experiment and change where we're looking at doing paper electronics. So, if we change the underlying ingredients of how electronics are made, can we contextualize it? Can we make it more accessible to more people, just in an intuitive fashion? So, you're not trying to actually push it along, but you actually have people pulling to try and learn this sort of stuff, because it seems more relevant to them. So, instead of starting with a breadboard and a resistor and a diode, you start with a piece of paper and some copper tape and some stickers. And so, this started out as a research project with a lady named G.C. at the MIT Media Lab, and eventually this ended up turning to a company, which is called ChibiTronics. And the initial results are very interesting. They're very positive. So, when we first started, we launched on a crowdfunding website called CrowdSupply.com. The green area up top is the percentage of male buyers, and the pink area in the bottom is the number of female buyers. That was in 2013. And relatively recently, if you look at our numbers off of our direct sales site at Amazon.com, you can see it's flipped around. We have almost like 60 to 70 percent female buyers versus 30 percent male buyers. And it's not like we've tried to sort of, we don't pink out our kit. Like, we don't just make it like a kit that's necessarily targeted at women. What we've done is we've made the core materials more accessible to a greater group of people. And I think, you know, the high representation of women getting into it is just, I think, an example of, there's a lot of demand there on getting into it, but they just hasn't been met by people who develop technology. And so, we want to get beyond simple circuits. You know, ChibiTronics is just about doing like, you know, sort of, you know, LED circuits and greeting cards and that sort of thing. And so, we've designed a new system called Love to Code, which is a system aimed at increasing diversity, particularly at young ages in computer science. And we've kind of decided to focus on three different things, which we think will help increase the diversity. One is balance. So, we want to make engineering as equal design. Familiarity, which is using materials that are universal, so they don't require a lot of explanation to get it to. And finally, simplicity. We want to reduce the barriers of getting started and not make people think that they're stupid from the beginning for not knowing how to install their USB drivers. So, on the topic of balance, everyone is now familiar with the idea that design can revolutionize technology. It's not just about bare circuit boards. You know, when we had the MP3 players, the iPod came out and that revolutionized MP3 players. And then, when the iPhone came out, the revolutionized smartphones, and we all have come to sort of expect a certain amount of design. But these examples highlight how much sort of an inclusive type of thinking can change a technology field. You have the same code bases running on both of these, but they just, one of them is more intuitive and more appealing, and therefore a bigger group of people wanted to adopt it. So, the problem is at the end of the day is that blueprints alone are in inscrutable. If I show you, like a circuit diagram and a bit of code, you don't know, it doesn't have any particular inherent meaning to you. You have to read it and understand it what it means. But if I go ahead and I take that exact same circuit and I fold it in with, you know, a piece of artwork or something that tells a story, you can immediately relate to it. You know what the purpose of this code was. And so this is an example here where this code and this circuit here is transformed into a set of circuits on a piece of paper and then on top of it is drawn a field of dandelions. And it's an interactive artwork where you can go up and you can blow on the dandelion and like the white sort of puffs like fly away. And so immediately technology takes on the life of its own without having to explain to people why the technology is interesting. And so when we design a curriculum, we want to feature engineering and design as equal partners. On one page, you're immediately challenged with an exercise to create something like a circuit or a pressure sensor. And then when you flip it over, you're given a mostly blank page with a little bit of an icon asking you, well, why did you design this? What story are you telling? So at the end of the day, the result can be more compelling if you have a stronger inclination towards like the art and design side. And what we find at the end of the day is that diverse thinking is inclusive. So when you go ahead and you try to measure achievement on one axis, say engineering ability or artistic ability, these are sort of reductive measures of goodness. And so you find that on the engineering axis, if you just measure on engineering, all you have are really good engineers. If you go on the artist axis, all you have are artists. But if you combine the two together, you find that you don't have to perform really well at both of them to create something that's actually very compelling and very interesting. And so your ability to achieve in these exercises is not just how well one can code. If you just run in and all you know is how to write code and you don't actually complete the exercise on the other side, you're going to see a bunch of kids like looking at someone else's beautiful thing that's actually doing something perhaps even simpler in terms of code, but it's more compelling. If you only know how to draw, then of course your lights don't turn on and then that's also not as compelling as someone who managed to get their little heart to flash. And so at the end of the day, by combining the two together, what we do is we create a much more larger space for achievement. And this is, I think, what is sort of the core idea of when people talk about STEAM, S-T-E-A-M, Science, Technology, Engineering, Art and Math versus STEM, where they drop the art. When you talk about STEAM is about creating this inclusive space so that's not just about measuring your ability on one axis and making people feel if they aren't strong at science or technology, that science and technology is not for them. Angry fans. And so familiarity, another thing that we focus on is we want to use familiar things are naturally comfortable. If you start with a breadboard, it takes a bit of explanation. You want to, this is actually from the Adafruit website. You actually tear off the back of the breadboard and show you how things are wired on the inside so you understand what the holes mean and why they're connected to each other. Whereas if you give someone a piece of tape and you say start playing with it, they're going to lay down pieces of tape and you can say, well, wherever you see the tape is where the electricity goes. That's much more comfortable and much more natural paradigm. Easier to explain. The cool thing is paper is also an expressive material. So although we start with paper and copper tape, you can immediately go and create things that have very, very wide range of expressive capability using paper. And paper is also an engineering material. This is actually an example of a circuit that Zabs, who's in the audience here, made up to buy us a transformer. And this is all done on a piece of cardboard with resistors and so forth. And as you can see, one of the cool things is that it's self-documenting. Like, you know, he's drawing it out the circuit on the cardboard and then putting the copper tape and the components on top of it. So you don't even really need to look at a schematic to debug this. It's all up there. So in paper versus breadboard, you know, you know, you can see that actually paper is compatible with SMT and Thruho, who was breadboard only does Thruho. Paper natively supports comments. Breadboards don't. Paper is thin, flat, and foldable into 3D shape. So when you're done, you can go ahead and you can fold your circuit over. You can hide into things where as in a breadboard it's kind of bulky and difficult to mount. The problems with, and in paper also the substrates basically free, right? It's a piece of paper. Whereas a breadboard they're a little more expensive. The problem with paper is that they're not reusable and may require some soldering and the components themselves when you put them down are typically not reusable. So this has trouble finding adoption in classrooms where they have a very limited budget. So we've gone ahead and we've adapted the idea of a microcontroller to put it into the form of a paper clip. You can go ahead and clip onto a piece of paper. This is done by combining rigid and flexible PCB technologies together. And so the idea is that you can go ahead and craft your circuit on a piece of paper and then you slide into your clipboard and you let it down and then the circuit can go ahead and turn on. That way you can combine sort of the best of both worlds of both electronics and paper craft. And so this is what kind of the final the final evolution is looking like we started from like a clip prototype on the left. And this is actually very close to our production clip prototype that you can use. The final thing that we try to really focus on in fostering diversity is simplicity. So now you have a piece of code on the left is your little Arduino loop that flashes a light. How do you get it into your device? And a lot of us here take it as kind of, you know, motherhood and apple pie that USB drivers can be installed. And we know we know where to go in the in the in the control panel or whatever is on the command line to figure out what the problem is. But it's actually extremely difficult for people who don't know and some people don't even have the right to be able to do this. Like, you know, and particularly teachers in schools, they lock lock them out of their computers. And also people who who have nothing but a smartphone don't have the ability to install USB drivers. And the problem is is that kids these days, parents are not buying them laptops. They're buying them tablets and smartphones. And so if you say you have to go home and do your homework, you need a laptop. You know, kids say, hey, mom, can you buy your laptop? Mom says, this is for work. You don't touch it. And then the kid can't do their homework. And so we've come up with a solution where we use sound to to upload cloud compiled code. So you go ahead. You log into a browser. You write your code and we transmit it through an audio port into the device. So this is what the user experience looks like. We have a little sort of Arduino like code environment on the left. And if you hit upload, it modulates it into a sound which if you unplug your headphone port, you can hear it. It sounds like static. And then when you plug it into the device, it all goes. Actually, if you have questions about it, you can talk to Sean over there in the audience. He he wrote all of that. And so at the end of the day, you can use this technology to allow LTCTB program with almost any device. From smartphones to laptops. If you can play music on it from a browser, probably you can use our our system. You could even, in theory, use like a record player or a cassette tape to go ahead and program these devices because it's just sound at the end of the day. They would have to have very good frequency references, but you should be able to do it. And at the end of the day, the reason why we're doing this is that open sources built on inclusiveness. It's not just about commit and forget. Like if you just commit something to the cloud and no one pulls it, did you write any code at all? Sort of like the tree falls in the forest question. Actually, at the end of the day, the whole point of the reason why people come together at conferences like this is about pulling, emerging and forking. We want to accept new ideas and we want to improve our code base. We want to share our code and create a community around it to sort of drive these things forward. So at the end of the day, I think it's important to empower more society to understand a value-open source because it's risky to vote on things that you don't understand. So that's my talk about making technology more inclusive. The left-to-code platform will be in open beta real soon now, hopefully in the next month or so. If you want to find out more about it, follow at chibi-tronics or visit chibi-tronics.com for updates. Thanks. All right. Thank you, Bunny. Any questions from the audience? Any questions from the audience? Any questions? No? All right. Maybe a question for me. Okay, sure. Let me think for a moment. So what have you been doing with chibi-tronics, you know, this left-to-code platform? What's next in the future? Yeah. What's your plans? Right, right. So we're looking to get this microcontroller launched and so we're going into beta in the next kind of month or so. And the concern that I personally have is that we hit actually all the compatibility promises that we make, that we actually build something that works with all the browsers and all the different configurations out there. So we're going to go in beta for about a couple months and kind of collect feedback. And then after that, we're going to have the full product launch which will hopefully include things like more educational material, you know, a binder that works with the clip and some power sources and sensors and a kit of parts and that sort of thing. So people can get started on their own and not just have to figure it out from the raw. All right. Any questions from the audience? No? Cool. All right. OK, so hi, Bunny. So it's great to see all this work that you are doing. It's so much different work. And I'm sorry, I also have another question a bit in another direction. Sure. You have been very active last year and there was this big news with your cooperation with Snowden, of course. Oh, OK. It was Snowden. So I mean, like, technically, maybe you could like tell us a bit because I know that you can you help like to develop a case for the iPhone. Yeah. Yeah. Or you did that. I mean, maybe you could like talk a bit about it. Yeah, that's actually a whole separate talk. Yes, of course, like a few insights. And the whole project is still in development. So that's part of the reason like I'm not talking about it at this point in time, like in a public setting. Yeah, I know this topic is a little bit different for most open source conferences, but I feel like it's something that even if it's not as exciting as artificial intelligence and self-driving cars, it's very important. And I'm spending a lot of my life trying to work on this and address this issue because I think it's going to be a big sort of social problem for the world at the end of the day. But also the problem of making sure that reporters can report safely and that we have something to monitor the surveillance is also important as well. And I mean, if you want to find out more about the status of that, maybe come by and chat with me privately. I can maybe show you a couple of things. I think like in regards to open source, I think it's exactly the same thing that we're doing because open source for me means that we are in charge ourselves. We're able ourselves to control what we do. And I understand that's also what you are trying to do with this project. But on a side note, of course, we are all very curious, how could you get in touch with Snowden and how did that cooperation actually take off? Sure, in a nutshell, he was working with a reporter who also knew me and then they were talking, they were doing a work, a thing on surveillance and my name came up and he was like, hey, you want to just like, we can ask him these questions and then it kind of moved on from there, so. Okay, so and have you met him? Have you been to Moscow? No comment. Okay, no comment, okay, just very curious questions. Thanks a lot. All right, thank you very much, Bunny.