 Okay. So thank you very much for joining this session here today, which was put up by Raphael after we had a nice chat and we wanted to do this session here together. It's about open hardware. There are many people who run like open hardware projects or got started with open hardware projects in Debian and around in the community. So we were interested in inviting more people to tell us what's going on with open hardware around Debian and what is everyone doing. We already had some sessions here about electronics and details and so on. So this session is about an open and free hardware. We can still like discuss this definition as well. And yeah, so welcome to the session here, which is like for 45 minutes until quarter past five. And yeah, I hand over to Raphael who is from China and I'm Mario by the way from Germany. And so Raphael, you have some also some introductions here. Hi, I'm Raphael Lee from Shenzhen, China, which is a which is a city for we can find the electric electronic components very easily. So in with that kind of advantages, we've had made a lot of hardware related projects. We are in a community, which is called Shenzhen DIY, which is that which is derived from Shenzhen lag. So we are more Shenzhen lag is a pure software community and we want to make our hardware and we want to have our own place. So we make so we rent another place and set up the Shenzhen DIY community. Here we have some some mechanical devices, including CNC, a small, small desktop CNC machine and laser cutters and three different kinds of 3D printers. So maybe we make, we may make some hardware there. So where is the base in Shenzhen? I know there's the, let me move this microphone so I okay. And I know there is Hua Chengbei. There's there's this space for some troublemakers, quite funny name, but and they are make affairs and so on. Where is base in Shenzhen and how many people do you have already participating in your community? Our community have no more than 30 registered members. But for all the members who who is in the mailing list is around 500. So we are mailing the place we rent is in the central Shenzhen near Shenzhen University. It's quite easy to find. We moved to their last October. Before that we are in leaving a village in city. After that we're leaving a better place. And of course we want to know like about the tools and so on and what what projects are you doing. So maybe we start with the tools. So how much can you already do like with free and open source tools? How far do you get? Like how our customs are people to free tools? For free tools now for for electronics design, EDA tools and programming for embedded and even even software development PC we can do it always free software. For PCB design we use key cat and for embedded software we use ARM9, ABIGCC, ARM9, ABIGDB and other software we develop with GCC or Python or some other free software. So sounds like you're using a lot of different tools for different projects. So now we're very curious what kind of projects do you have already? What kind of projects are you using? So I can show you a little bit on the PC. We have a good lab. It's made a project which is called USB cable cracker. It's which is a device that can measure the quality of the different USB cables. It supports from type A type B mini USB USB 3.0 and micro USB and even USB type 3. So this project is all developed by free software. With that device you can when you have different kinds of USB cables you can measure all of them using this single device. So in our hacker community we have more than 100 different types of USB cables. It's very convenient to have standard instrument to measure all the qualities and cut the bad out. So we made this project. I hope this will benefit everyone who has a lot of USB cables. So we've sold this USB cable cracker for, yeah, we have sold 20 USB cable crackers in different kinds of make fair and offline meetings. It works pretty well. Even another hacker, Bunny Huang, who bought one for me. Bunny Huang is like the original hacker of the Microsoft Xbox. He's based in Singapore and also often in Shenzhen, right? So he bought one from you. So pretty cool. And you also like shot me a keyboard earlier today. So that's another cool project that you're doing. Yes, I built this keyboard for my daily use. Before that I used HHKB for typing but HHKB does not fit all my requirements. So I decided to make my own. Then we have this keyboard. There are a lot of customized functions in this keyboard. If you are interested, I can give you a short description. Yeah, please go ahead. It would be great. Well, this kind of keyboard, all the keys are aligned in rectangular ways. So when you put it in just in front of you instead of on the left corner, it fits your fingers more conveniently. And your left hand does not need to turn for maybe 30 degrees to fit that keyboard. So another, there are a lot of, there are eight-bit switch on the top of the PCB. So the user can select the different layout of this keyboard. It supports double rack and QWRT now. Maybe someday I will support Comac. It's not easy for, it's not hard for adding another layout into this keyboard. All the things you need to do is modify the firmware. So they are like different colors here. Is they different? Is it shiny or what's the reason for this? Some protectors or what's the reason for it? No, no. These are for just easier for identify the different keyboards, identify the different keys. There are the F key and J key are have some markers on it. So when you can type blindly without looking in the keyboard, other keycaps are modified for the same kind of reason. So the not often used keys are with keycap to tell the difference between different keys. So it's basically all these keycaps are modified for typing blindly. So how much is, how far is this keyboard open, for example, or when you do projects? How open can you do that? So we had like different discussions here at Depthcon as well that for example the chip is not open in many projects. The keyboard doesn't have a chip. But yeah, so how far can you go? Like for example, is anything laser colored or 3D printed or how do you build it like the different components? Well, the PCB is all designed using keycap which is a free software without problem. And the microcontroller on it is STM32. The chip itself is not free, but the two-chain we use is free. I use ARM9 EBI GCC for compiling and ARM9 EBI GTP for debugging. All the project is based on MacFail. And other parts, there are quite a lot of keys which I bought in from, I bought a second-hand keyboard and take the keys apart and then to use then build my own keyboard. Other parts are free, I think. Okay. And so now like when I talk to hardware people from China, like some people say to me, well actually like we know open source, we know free software, but like inside China it often feels as if it is open already. Because like hardware developers from different companies and so on, they exchange information, they have access to different toolkits and so on. So what is your reason for now going a step further and actually focusing on tools that are really free? Because we often hear from hardware developers in China, well for me it's no problem, I can get access to whatever I want. But what is your idea of making something and advancing the free software and free hardware, open hardware community? Well, Shenzhen DIY, we recommend to use different free software. But the freedom, the other side is that allow people to use whatever they want, even it's a priority software. So I think that is another kind of freedom. So I am using, I've replaced all my software development environment into free software. I'm encouraging my colleagues and my friends to do that. I'm recommending different softwares to them. Yeah. But do you have any idea, why are you doing this? Well, two years before I just installed my Linux on my computer and gradually I found it fits all my needs and all the functions I can implement in a very easy way. I can use different kind of scripts to automation to make my daily activities automation. And when I reinstall the system, I do not have to click my mouse here and there to set all my settings. All I want to do is synchronize my home folder into another machine's home folder and all my settings get back. So it saves me a lot of time. And for personal, for my daily use I wrote a lot of scripts to make some actions even faster. And I felt that is very enjoyable. When I found an answer of some problems, that answer won't change for several years. Not like Windows. In Windows, the different versions have so many differences. And when you get the settings here in this version, it won't work in another version. And it cannot stop from upgrading when you are doing something really important. And the computer stopped, started upgrading. You can do nothing with it, right? So this is great to hear. Because I remember in the past, people said, okay, free software doesn't always have the features required by users. It's not working for some people. And it seems like with the free software community, we advance much further than some proprietary applications or settings or operating systems already could. So great news, great to hear the story. And we have the first questions here. Okay, so I'm quite frustrated with the keyboard on the market, especially the laptop ones. And most recent trend makes all the keyboards on laptop suck. So therefore, I want to get a touch on the keyboard. Would you mind me to Okay, cool. Sorry, of course. Let's pass this around. Possibly pass around the audience. Okay, there are some functions you will not get familiar with when you when this USB keyboard is not really working. So there are a lot of function case implemented in a hardware way when you can, you can press the FN key and J for home and other for and another for delete. With this kind of layout, you do not need to move your hand around to assess different keys. So although there is up down page up and page down on my keyboard, but I can make I already make some other hawkies for for that. So so you cannot try this kind of functions to get a feeling of it. Okay, sure. Sure. So yeah, thank you. You find me in the noise lab. You can you can have a please have a look around then like pass the keyboards to others. Maybe others also want to try out and so so a request that I have here, for example, so you're not making the keys yourself yet. And probably production is quite challenging, but would be cool. Like if we get a keyboard that's completely open, like, let's say, schematic 3d, the 3d file, you know, for the keys and so on as well, maybe like it's something you could do in future as well. Yes, sure. I am going to share the reason that I did not put this project on the internet is that I did not write a good manual and good comments for the for the source code. That's the only reason. Yeah, but like now we have a lot of people here who would probably like love to help you and he's already very excited about it. So I hope to see that online soon. And we have the next question here. Just out of curiosity that I have you try other open sourced open hardware keyboard like GX G edge 60s. Yeah, I know that. Yeah, so how I'm curious that if you have you ever compared your your ones to there, like the budget issues or, or people layout costs, those kind of things. Because I remember I get a GX 64 about 100 US dollar. So how much is the cost to make yours? Well, the raw cost of my keyboard is really, really cheap. I got a lot of all the keys I use in my keyboard, that second hand. So it's for other manufacturers, the PCB, I have got five PCB for more for about 20 USB dollars. No, no, American dollars and others for the processor STM32, it costs me around $5. That's, that's most of the bill of materials. Then coach match. And for G edge 60, I remember that it, it is a AVR based controller. So with AVR base, the frequency is lower, the performance is lower. And there is not quite good to chain analytics to program it. Yeah, although there are some AVR dude. Yes, it's maybe I think that it's quite easy to write it from where as well. But you get a point because the frequency is low. So some, some cheater might, might happen if you type too fast. No, no, that won't be the problem. The scan, the human input frequency is much lower than the, than the possessors frequency. This won't be a problem. But for AVR, I've encountered some kind of cheater. I don't know why. Maybe the product production quality is not well as well. Okay. Okay. Thanks. Thanks a lot. So there are other open source free keyboards out there. And I would like to know, are there any other projects like because you said you have 30 people in the, in your space, working on that, could you name a few other projects? What are people working on at the moment to maybe give us a wrap up that would be very interesting. There are some leads on this Shenzhen DIY website. We have a list of projects here. We can, we can kind of, yes, the first round name is Artharnis, which is a press, which is a GNU official project. So the members then joined to the GNU because of this project. And there are some other reasons. So this, this project is started in Shenzhen DIY community. And it is based on web architecture. With simple command, you can set up your own web, web server and web pages. And these, the, the web server is, is aimed for high performance web servers. It, and there, the, the communication in this web server are all asynchronous. So, and a green thread is used. As we're focusing on hardware, are there any specific hardware projects, especially? Hmm. Okay. Yes. Sorry. There are some that is not well documented. Yeah. So documentation, documentation is always like a challenge. And but yeah, okay, so maybe, maybe you can just tell us like that. What are these projects doing? Well, we are, we have a lot of, a lot of members in our community can both design software and hardware. And we may come get some projects. So we have designed a climate station, which use 80266 processor for Wi-Fi SS. And we, with that kind of device, we can get the PM 2.5 and temperature and humidity and all of the data into our web server. This is, and we designed a laser cutter case for, for that PCB. Yeah. And a small screen. Okay. So you're doing many things. And, of course, like in your, in your hex space, how about like, when you have visitors, did you already get like international visitors to your space? Sure, a lot. If I come over, can I work like a week in your space and hang out with you guys? How does it work? There's one day, there are more than 30 visitors from different kind of the, different parts of the world visiting us. Yeah. So, so that's possible. And I just get in touch with you through the website. Yes, sure. There is email. Yeah. There is email there. Okay. Good. So, so this is great to know. And before we go like, to the audience, more in the middle, because I know that many people are doing their project. I would also like to show a project that the Foss Asia community is doing. And so open hardware, I don't know, some of you might have seen it already. So we also like producing open hardware in with a pocket science lab in like, in the community. And there is a main developer is the Patmal. He's in Sri Lanka. Other developers are in India and in Germany and in Singapore. So it's a very like, like, nice team like across Asia. And the idea of the pocket science lab is to have an instrument where you can measure different like electronics. For example, I can measure resistance with a capacitance measurement here. And also you can connect many sensors. So there are hundreds of sensors compatible. And this means you could use this device, for example, for like, let's say biology lessons or even chemistry for somebody could measure the CO2 in the air or you could measure even radioactivity. And it's like completely open hardware, except for the chip or what we mentioned already before. So anyone interested, please check out pslap.io. Sorry, developer focused website still. Yeah, what you can see here and you find a all the repositories there. There is an Android app. There's like different repositories are linked here. And we also have a desktop app where we look for interested package maintainers here in the Debian community who would like to package these different apps for the desktop. So that could be one place where we could collaborate more. And we are also using Keycat for this. If you're interested, please get in touch with us and check out the different applications. I will also hand this around. And I think now we would love to hear more from the audience, right? Raphael, like what are you guys doing? I know for example here our friend from Beijing was like yesterday already looking very interested to the hardware. So we know that there are many hardware developers here. So what projects are you doing here in the community? What's going on in the community? This is a buff, so we hope to get some more information from you guys. And I think the microphone is open. Please feel free to grab it and give us some information. Before we start the topic of what the audience make, I want to share my view of free software and open hardware. So since we are in DevCon, a lot of us are familiar, quite familiar with free software. The free software means you can distribute it freely and modify freely and get the source code and document freely. And we can use it in anywhere you want. But for hardware, the price is not free, itself is not free. And we are under the chip, the implementation of the chip is not free. But we try to make every other part free, including we use free software for hardware development including so the people can modify the design files freely instead of using some property software. So I think that is the main point of open hardware. A lot of companies use some property software for their kind of open hardware. But when the design document is released, we cannot motivate with what we can get. I think that is what I believe in open hardware. Absolutely. It seems like probably everyone would agree here in the room with that. I still personally have a challenge with, like I met Stormin and he said to me, call it free hardware. But I think free hardware, people actually do think it is for free. It is not the same as with software which you can share for free. You don't have to produce it. But I think we will figure this out. The important thing is actually that we do it. That we do share all the sources. I have a feeling that it would be very helpful in international relations if copyright issues could be much simpler. If there is a commitment to open free sharing of the source that could help a lot in the east-west, the China question of fake hardware. I don't have to tell people what that is. This is my hunch. What is your experience? Do you see that Chinese manufacturers or developers see this advantage? Can you get people to understand why it would be easier if the license was open? Can you convince people of the benefit of copy left? Do people understand? People will understand it gradually. We are telling people what free software is and tell them to try free software and free operating systems. But it takes time. I think since we are getting better and better, for our company, we are trying to make all of our development tools two-chain free. Our CTO is highly influenced by free software like 10 years ago. He convinced other colleagues to try free software and using them in development. We are getting better and better. But it takes time. Our company is quite rare in Shenzhen and in China. 10 years ago, even in community, you told people to use free software for development, for compiling. It's not possible. All of them use Windows compilers. Since we are getting better and better now. In my opinion, for free hardware, it is better to, how it is more reproducible by independent party. Currently, I am considering another free hardware, but my concern is the availability of the chip. You know that. The GD chip is only available in Chinese market. I think that I should not select the chip, perhaps. My opinion is that if it is free hardware design, but if it depends on some specific distribution channel, it's not good. That's my opinion. I think it's a very important point. I don't know if everyone knows about your project. Could you maybe also share? I'm from Gunno PG project and I am developing by free hardware design, the token for Gunno PG. It's called FS201. In this conference, I also distribute another free hardware design, a kind of toy of LED matrix. It's free as a beer. We have a booth there, FSIJ free software initiative in Japan. We are distributing a toy of LED matrix. Hardware design is available by KR CAD, tool chain, and firmware is also available as free software. Especially with your project, we had some links with another project, which is the NitroKey of Jansu in Germany. That was really great to see this collaboration and cooperation across projects, where actually in the NitroKey features of your project were implemented in that key. How did that work out? NitroKey just used my design of FS201. They depend on my development of Gunno PG SC demo. They are users of my work. And actually we have a very good relationship between FSIJ and Shenzhen DIY. Yes, we already met three times. Every year I visit. Yes, Nip-san is a very important developer in Gunno PG project. He designed a smart card like hardware to store your private key in. And it cannot be read out through software. So the work like this when you say public key is signed message into the Gunno key. And the Gunno key itself will decrypt the message and send it back to the computer. So the key is stored in the micro controller safely. And as far as I understand also the Linux foundation uses this project as the NitroKey is available by any kernel developer by the way who wants to get such a key can apply with the Linux foundation and get this delivered for free to their address to use it for kernel development. So it's actually like soon or maybe already it's the case that every kernel developer will somehow use your work as well, right? Yes, it will be possible. Okay, so this is really great and also like of course like the companies will follow in and we're getting like this idea of open free hardware out there. So thank you very much. And we have more questions here. Since that you have mentioned the chip is not open source but for nowadays I think that will be much more easier for people that tries to implement their own chips. Just like it's the last year November 29th at the RISC-5, the seventh workshop of the RISC-5 workshop. One of the kind of EDA tool. I don't know what should I code them but they are trying to set up service that includes a cloud based kind of very log or hardware description of the IDE on the cloud and they also have the whole open source tool chain for people try to verify their implementation, those kind of stuff. And finally if you have sort of enough money they can arrange shuttle for you to the Fab House to finally have your ASIC implementation. And the fee, how much money you should put to make your ASIC is there, they didn't mention it but I think the price will go lower and lower and lower. So maybe I think it's quite optimistic. Absolutely and so I see like you have a lot of comments on these topics so I'm personally very curious what are you working on actually and so what is your work because you have a lot of knowledge in these areas. Actually I'm just a graduate student and my work is not on the semiconductor. But I played around RISC-5 and open RISC back then when I was a year one of the college I bought a very, very shady, very shady IPG board and tried to burn the open RISC and round the index on it. So that's why I have interest in the open hardware. Yeah so this is great so but like you have a lot of knowledge in this area and if we have any questions here we can also come back to you for example like this is what this buff is about, like finding and connecting different people who are working in the area, right? Good thank you very much and another question here from our friend and from Beijing. Hello everyone and I want to address the greatest challenge for so-called open source hardware that is in hardware industry. The concept of free is not as adopted as well as adopted as such a concept in software industry. For example I can take Ubiqui, a famous security device as an example. Ubiqui is open to the community and provide open source software and management tools for us but however no one knows what the chip it is and even if you are willing to create a free Ubiqui, okay, a free Ubiqui and that's what he's doing. However as far as I am concerned that the security element chip is the document of that chip is not open, isn't it? Yes, okay, however I should say in hardware industry we should improve the adoption of the concept of free because nowadays many chips we cannot even find the documentation or the data sheet. That is a really, really bad thing for us designing hardware products. So absolutely and I've heard from many hardware developers that they don't want to sign NDAs for example like often you have to sign NDAs in order to get access to document. Because normally for individual developers we cannot sign NDA with companies. That is impossible. So we cannot get any design resources and we can only select those chips with open documents but those chips maybe those chips cannot meet our requirement. So we also need to talk to companies actually to release their documentation Yes, open document. Okay, open documentation. So that would be the next step that you recommend also what your student at Tsinghua University and so you want to see also more of this in order to actually be able to learn about these technologies here. What is your experience here? So I think RISC-5 is a great opportunity for open hardware. Quite optimistic in that area. That's my opinion. In our company we even made patches for KitKat. Maybe sometimes we we can put them into upstream to add more functions to KitKat. This is our contribution. Well, it will be our contribution to the free software community. Yes, I'm trying to learn more about Linux and trying to pack some packages for Derbian and other distributions. So this hardware actually costs money. I mean you have to produce it. It would be great to see more collaboration in the community to actually exchange also like experiences. How have different people like produced hardware and how did we how did we do it? Did you set up companies or did you set up foundations? How have you worked together with companies maybe in China or elsewhere? So I think this would probably be a topic for a whole new panel. Sharing experiences of real production of generating income. More open and free hardware can be produced. I hope we see more of it in the upcoming months and years. So I would like to thank everyone for their input here. Thank you Raphael. Thank you all the participants here who have given insights into their projects and asked questions and put in ideas. And we have more electronics and hardware topics in the upcoming days at Debian. Please check out the schedule and let's keep this work here up and running and really get to the next level of free and open hardware. Thank you very much for joining.