Embedded - Bonus: From Solid Con

Episode Date: June 3, 2014

Elecia attended O'Reilly's Solid Conference, recording a few of the people she met there. Note: this episode is recorded in a noisy location. Erin Mulcahy at littleBits (@littleBits), magnetic electronics modules Jack Mudd at Onewheel (@RideOnewheel), powered sort-of skateboard with only one wheel and auto-balancing Laurie Yoler from Qualcomm, she spoke on Intelligent Connectivity. It’s What’s Next Taylor Stein from AutoDesk, a Fusion 360 evangelist (free for hobbyists!) Ahmed Daoud from Playtabase, makers of Reemo Terrence McKenna of Panoptes ("No drone is safe until it is a Panoptes UAV.") Michael Holdmann at Coversant, on XMPP communications protocol Also, thank you to O'Reilly for giving away copies of my book. 

Hello, this week we will not be having our usual good studio sound. This is going to be a segment show in a pretty noisy environment. I know Chris White is going to do the best he can to make it sound good. But, well, here, let me explain a little better. This is Embedded, the show for people who love gadgets. I am at the SOLID conference. It's the Software Hardware Everywhere conference put on by O'Reilly. This is the first year, so it's a little strange being here and talking to all of these people about embedded systems. What's really strange is that some of them don't know that they're talking about embedded systems. What's really strange is that some of them don't know that they're talking about embedded systems. Even as we talk about connectivity and hardware

and all these things, it's like all these people don't know. But then there are a lot of people who do know, and they're the ones I'm going to be talking to. The first person I spoke with was Aaron Mulcahy of Little Bits, which was a pretty fascinating little booth. But let me let her explain who she is and what they do. I am a community and program manager for Little Bits. So I do a lot of education initiatives, partnerships, content, and also community programs. So engaging people who are using LittleBits to extend it to the community, doing projects, and also events.

So what is LittleBits? LittleBits is a system of modular electronics. The best part is that everything is magnetic, so you can't snap the pieces the wrong way. So there's no need to solder or to breadboard. You can simply snap the color-coded pieces together and you can quickly have a great tool for prototyping, learning, or if you just are interested in electronics. And what's the age range? We say 8 and up just because of the magnets, but we see people of all ages using little ones.

How long has it been around? We started shipping product in 2011, and we've been very focused on our mission of enabling everyone to use electronics. And now this year we are very focused on the word power, so we've been releasing a lot of new modules, wireless, logic. We just announced we have the Arduino module and we have the cloud module, so you'll actually be able to use little bits and connect to the Internet. Wow, that's sort of the Internet for things for eight-year-olds. Yes, it's very much about democratizing the Internet of things and being able to build things yourself. What is the most interesting project you've seen built?

We have a really amazing jukebox on display today, which is combining the synth kit, which is a modular synthesizer. It's a music synthesizer. Music, exactly. And what we've done is use the Arduino to actually code in different songs and there's cards that have a light filter and so when you put the card in the jukebox the light sensor picks up which card it is and actually will play different tunes. So it's using all of our components, the Arduino and music, so it's a really powerful instrument.

Why did you come to Solicon? I came to Solicon. Joey Ito has always been a long-time supporter of Little Bits, and I have a dear founder who's actually out at MIT Media Lab. But for us, it's a great way to showcase where we've come as a company with the Arduino module. We have a lot of really exciting products

for this year. Well, I hope you have a good show. Thank you for speaking with me. Thank you so much. Aaron took me back to the Little Bits booth to listen to that jukebox. And you know, the Star Wars theme in 8-bit never really gets old. The next person I spoke with was Jack Mudd about the neat contraption that had people braving the wind and then zipping around outside. Well, we're here with OneWheel.

OneWheel is a self-balancing electric skateboard. It looks kind of like there was an old game, BC, where it's like one wheel and then you stand on either side of it. It's like a skateboard that is weird. Yeah, yeah. You know what, it's tough to describe because there's not really anything out there like it. People associate it with sort of a Segway skateboard crossover. But it's tiny. I mean, Segways are big. Right, yeah. Segways weighed over 100 pounds. This is much, much smaller than that. It weighs between 20 and 25. But yeah, it's basically a whole new way to ride

and a whole new riding experience. And that's what's really special about it. So it's got electronics in it. This is not just a mechanical solution. What do the electronics do? Yes, it's a digital vehicle. So the entire riding experience is controlled by software. Basically it has gyro and

accelerometers in it that take the angle of the board thousands of times per second. They feed that into a computer controller chip which digests that information and tells the motor how to run. So by adjusting the algorithms that the computer controller chip uses, you can completely change how the board runs, which is actually what we're doing right now. We're in the midst of fine-tuning our firmware, which is a really fun process because Kyle, our CEO, and Daniel, our control chip expert,

plug the board into a computer, type some code, unplug it, and then they give it to me and they say, go ride it around, and then I go shred a little bit in the backyard and come back and say, well, it needs this or that or this, and then the process begins again,

so it's a lot of fun. Who's the primary audience? Yeah, we did a Kickstarter campaign in January, and we raised $630,000 in three weeks, over 1,000 backers. So from that demographic, it looks like people are interested in commuting to work on it. A lot of 35-year-old engineers who want to get off the train and go to their office in the city. You know, they can bring this on the train. It's a lot smaller than a bike. Yeah, it's a lot smaller than a bike. You don't have to be in the bike car. You can bring it into

a coffee shop. I do all my short errands on it. It goes four to six miles. And how fast? 12 miles an hour. So it's really an efficient way of getting around. And it's also an enjoyable way. We call it recreational transport. I saw you giving a demo earlier, and it did look like one of those 30-something engineers trying it out. And he got it pretty quickly. How fast does it usually take for people to go from

wow there's no way I can balance on this one wheel to oh I could totally zip around? Yeah it's exceedingly easy to learn. It's self-balancing so it's really only as difficult as you make it. If you relax and stand up pretty straight it does all the work for you. So, you know, we find that most people hop on and probably about half of them zoom off right away and about half of them take a few minutes before they're zooming around. So, pretty simple. What are you doing at Solid? Just showing it off or you have other plans? Yeah, you know, it's fun to get the Onewheel out there, introduce it to the world, which is sort of what we're doing this summer.

And Solid is a great community of these people who understand the importance of hardware and software coming together, which is exactly what the Onewheel is. You know, it's sort of the embodiment of hardware and software coming together and changing how we live. Well, Jack Mudd, thank you for talking to me about OneWheel. Absolutely, thank you. For my next interview, I perhaps accosted a woman who was just standing peacefully near the coffee urn.

I struck up a conversation. Well, have you ever gotten part way through a conversation and then realized the person you're talking to is sort of famous and you probably should have known that? Yeah way to go me. So I talked to Lori Yoler of Qualcomm. She's the president of Qualcomm Labs so I kind of wish I'd had the business card first. And then I decided to record outside, in the wind, and pretty much everything that could go wrong went wrong. I am so lucky that Lori had a sense of humor about things.

Okay. Very high-tech setup. Yes. Okay. only the best in podcasting exactly after a few more minutes of me fumbling around i asked lori what she was speaking on i am speaking at on intelligent connectivity and why that's going to be so important for all these fantastic new devices and new applications that are being built around the name, the Internet of Things, which means a lot of things to a lot of people. But it does mean that 25 to 50 billion new devices are expected to come online in the next 10 years or so. And all those devices are going to need connectivity.

And that has a lot of interesting implications for what we're going to need in the industry and how we ensure that all those devices are easy to use by consumers and are connected to all the other things we want them to be connected to. And you work for Qualcomm. How do they play into this? We play in a lot of different ways. We think that many of the same technologies that we've built for smartphones and for tablets can be used in all kinds of new things. And that's when the wind got just too bad and you

don't want to listen to it. Clearly, Lori and I have a lot of things we could talk about, many points we'd agree on, and probably many things I would learn. So I'm going to try to have Lori on the show for a whole hour of asking what Qualcomm Labs is up to and what she can tell us about fixing all these problems with wireless that I've been whining about. My next conversation was unexpected. I wasn't really sure that I would talk to Autodesk because, well, because CAD tools are pretty far from what I do.

And yet, I really enjoyed talking to Taylor Stein about Fusion 360, and now maybe I have one more thing on my list of things to do. My name is Taylor Stein. I'm the Fusion 360 product evangelist for Autodesk here in San Francisco. And I have a background in mechanical engineering, and what I do now at Autodesk is go around the Bay Area working with designers, artists, engineers, helping them use some of our newest software, specifically Fusion 360. It's a new tool that really crosses industrial design and mechanical design. It really lets you take something that's an idea in your head to a real product in your hand. So I came over to your booth more because I'm interested in the 3D printers and I know I can print things directly, but I'm more likely to want to fiddle with them and those

CAD programs cost a fortune. But you told me something different. Yeah. So for Fusion 360, if you're using it for non-profit, just hobby, something around your house, just for fun, it is free with an enthusiast license. So when you download Fusion 360, you can check the box as an enthusiast, and you'll get a one-year license as an enthusiast. So if you're just doing 3D printing around the house for fun, anything like that, you can use it for free.

I cannot win. We can do that again. Okay, so you told me it didn't cost a fortune. Yeah, so Fusion 360, we do have an enthusiast pricing option. So if you're using it for non-commercial reasons, just as a hobby or something around the house, it's absolutely free. So when you download Fusion 360, you can check the box for enthusiast and you'll get a one-year license. And at the end of the year, if you're still using it for non-commercial reasons,

you can just re-up for another enthusiast license. So how non-commercial are we talking? I mean, if I print a few things for Etsy, does that count as enthusiast? Or is that, when do I cross the line and how much does it cost when I do cross the line? So there is a bit of a gray area. It's a little tricky.

What I would say is when you first download the software, just get your enthusiast license if you're not using it for commercial reasons. I believe we capped it at 100 grand of revenue to then start having to pay so if you're using it for small etsy projects probably not getting around 100k in revenue um you can still use the enthusiast license um but later on if you are making money off of it we'd like you know to return the favor and pay again so if you do pay for the commercial license, it's $25 a month with an annual license of $300, or month to month, whenever you want, at $40 a month.

These aren't easy to learn, though. I know that from just kind of watching over other people's shoulders. And I'm embedded software, so software background, a little bit of hardware. So schematic capture is not foreign, but mechanical design is. What do I do next? Where do I start? So for Fusion 360 specifically, we do have a lot of learning tutorials.

If you go to fusion360.auditus.com, we have a learning center where you have a bunch of tools that you can look and tutorials to walk you through the process to get started and kind of help with that onboarding process. We have a lot of also videos on our YouTube channel to check out that show different tips and tricks and different projects we're working on, but Fusion 360 itself is so intuitive that you really don't need an engineering background to do it. You can start with a primitive shape

and start pressing and pulling on it and really tweaking the design as you want without having to do much defining of points and curves and splines and all that nasty stuff you can really just get straight to making organic shapes or really dimension shapes whichever you want that is neat thank you for telling me about it i think i don't have a new toy to play with um so 3d printing i think is what i'm most interested in. Do you have any suggestions at which 3D printer? So, I don't really have too many suggestions. I would just say I have a MakerBot myself.

I really like it. It's really reliable. I have the Replicator, the fourth generation that I love. It's just a cheap FDM printer, similar. There's a bunch out there now. They're becoming more and more popular.

But I would just, you know, take a look at what you're wanting to make, take a look at the build sizes, like the build plates, and really just look at reviews and pick what you want. I think at this point the technology is kind of getting to a point where all the printers are kind of plateauing at that price point and are really similar, and I don't think you can really go wrong. Great. Thank you for talking to me.

Thank you. After Taylor, I met up with Chris Gamble of The Amp Hour, and we hung out for a while. He had a better recording setup than I did. I liked my iPad with microphone, but his H2 recorder, well, I think I'm envious now that I've listened to some of these back. I also went to a technical talk about MQTT, a protocol for sending messages from things like sensors to things like servers. It was pretty in-depth and deep.

I also went to Chris Gamble's talk on open hardware, which was just as interesting, but not as technical. Solid was an odd mix of theory and tech and funding and engineering coolness. I'm not quite sure what it wants to be when it grows up, but I'm not sure it knows either. And this was the first year after all. The next person I spoke with was the CTO of Playtabase. They had a demo of their product Remo, but I'd better let him explain. I'm Ahmed Dawood, and I'm the CTO of Playtabase. We're a wearable and gesture-controlled tech

company out of Minnesota. We really exist to solve problems that we see that people have accepted as part of the status quo, and we try to solve those in fun creative and innovative ways through technology with the product we're creating right now is Remo and it's a wrist-worn gesture controller that allows you to interface with technology around you and the internet of things that exist now in the world and I saw one of your engineers using it he it looked like a large-ish sports watch sort of device, and then whenever he would move in a certain way, it would turn on or off the nearby light. Correct.

And so that's Bluetooth connected to inertial sensors. Well, the inertial sensors are on the bracelet, and they're measuring the user's or Pat, my engineer. They're measuring his motion, and based on his motion, they are identifying if that motion fits one of the predetermined gestures and then responds accordingly, sends over a message that says, this gesture has been detected. What it's communicating with is a receiver that the lamp is plugged into. And the version you saw is fairly large, and that is our current prototype. Our next prototyping iteration is going to bring the bracelet down much smaller, closer

to the size of a Fitbit or one of the fitness wearables that you see around. The only difference is this is not a tracker, it's a controller for the world around you. It really exists to extend the human body's ability to control things around you without requiring you to interface with a smartphone or taking you away from the world that you are in into a digital or virtual world to control some lights around you. You're able to directly,

as we refer to it as the mouse for the physical world, you can point and click, quote-unquote, to control things around you. So are these for sale? Are you in production and development? We will be releasing to market in Q2 of 2015. So, optimistically, oh, 2015, so you've got a year. Yes, yes. We want to give ourselves some wiggle room because it is a challenging space. We're very happy with the prototypes and the development that we have right now,

but we wanted to give people a realistic timeline to expect this on the market. So far, there have been no wearable products that have actually met their release to market date. We are trying to be the first in that area. And are you going to do Kickstarter pre-orders? That's a thought that we're flirting with right now. The reason for that being twofold. With a crowdfunding campaign, you're generating funding, but more importantly, you're generating the interest.

Yes, you're creating the buzz. You're bringing attention to the product that you're making and the thing that you're doing, and you're able to gauge how excited people are about it. So far from what we've seen in industry and professionals in the field they're very excited about what we're doing and what we're seeing are is that the applications for Remo are coming at us faster than we're hypothesizing them. We're seeing people from from almost every single market segment or industry

saying I can use this in some way or another to improve the industry that I'm in. Oh, I thought that too. I wanted to use it as a mouse immediately or as a controller for all sorts of different... Yes, I can see that. Exactly, exactly. It's awfully early for you to be out talking about a product here at a conference like Solid. Well, we don't necessarily think so because we're fortunately creating a lot of business-to-business partnerships for some of the applications for Remo before its market launch.

So we're in late-stage negotiations with a lot of companies that want to kind of white-label Remo or use it in different ways for their specific applications or their specific products. And so that's why you're at Solid, to meet more businesses. Yes. The other aspect of it, we're from Minnesota, and in the Midwest, if you're a hardware startup, it's a lot harder for you to raise capital. Doesn't 3M sponsor everybody out there?

No. Of course not. No, 3M has a history of trying to acquire everyone out there. We are very proud and happy with Remo. We love where it's going and we want to maintain control of that. We don't want it to be acquired by a large organization all of a sudden and for it to change its mission. We're really a mission-driven company, as cliche as that sounds, but we really are and we pride ourselves on being that way. I left a career as an engineer to lead technical development in Play-to-Base because I believe

that this is a product that can restore autonomy and dignity to people. And Play-to-Base is the company that makes Remo? I'm sorry? Play-to-Base is the company that's making the Remo device? Correct. Okay, just making sure. Yep. So this product for me, the first application I saw is still the one I'm most excited about. It may not be the most sexy application, but it's a healthcare application. I said a lot of people have lost mobility due to health conditions or due to accidents and that really is something that is corrosive to one's dignity to have to be reliant

on others on a consistent basis. For me to be able to contribute towards bringing autonomy back to a lot of those people, giving them the ability to control their environment, not need someone else's help to do that because the interface that requires you know using a large remote or interfacing over an iPhone or an iPad to control your environment may not be accessible to them due to lack of dexterity due to lack of motion in certain parts of their body so to give them a single wrist worn device and a controller for the world around them to allow them to regain and restore their dignity and their autonomy is something that i can really stand behind and something that i truly

believe in that is great well i should let you get back to your booth we're about to have another flood of people thank you so much for talking to me and good luck thank you very much thank you moving from products that won't be shipping for a year to things that are taking orders now but seem futuristic and strange, I went by one booth that was showing a UAV, well, showing a quadcopter, really, and picked up a card and it said, no drone is safe until it is a Panoptis UAV.

That was pretty neat, so let's hear what they're saying. Well, I am Terrence McKenna. I am the CEO and founder of Panoptis UAV. And what we do is collision avoidance sensors for small UAVs. So we really think that all of these things that you see, like on the 60 Minutes special and... The Amazon Droid? Amazon drone, all that stuff.

You have drones delivering dry cleaning and tacos and all those great ideas. Any of those things to actually become real reality needs a collision avoidance system on there to make them safe and robust. So how do you do collision avoidance? I mean, you're not building a network of all of them. That's true. That is true.

We actually incorporate echolocation sensors and vision-based sensors into the shell of existing platforms. And so it's a vehicle agnostic sensor that you just plug into any quad rotor, and all of a sudden you have collision avoidance capabilities. Any quad rotor? That doesn't, I mean, any? Really? Any? Yeah, so we're actually basing a lot of it on research that we've done based on bio-inspired methods. So we have, it basically hears like a bat does and sees like a bug does to avoid those obstacles in urban environments.

But we take advantage of a lot of some standard interfaces that quadrotors have to make it vehicle agnostic. So in your booth, I heard that it doesn't affect the software on the quadcopter if it's already got an autopilot. How in the world can that work? Well, that's exactly it. So we actually don't want to interface or we don't want to change the software on the existing system because typically they're very well tuned. They fly very well as they are. All those loops are tuned.

We don't change those dynamics at all. We just interface and intercept the inputs and the outputs of the autopilot. So you basically just unplug their autopilot, plug our system in, and then plug the autopilot back in, and you're good to go. And so their autopilot maintains all the stability that it normally does, and we just work at that last 10 feet. When you're about to hit the wall, we intercept at that moment, in the last 10 feet there. Why are you here at SolidCon? Well, we're from MIT, some back there, and our office is in Kendall Square in Boston. So we know the guys at the Media Lab pretty well and did some research with them. And it's just a great, great environment out here. Really great way to see all the different applications and use cases for quadrotors that we hadn't even thought about.

Are you selling your product at this point? Yeah, so we're taking pre-orders. We just launched our company about three days ago, so very new and very excited. The website panoptisuav.com is live now, and we're actually accepting pre-orders. So we'll be shipping those by summer, and that'll be working primarily with the DJI Phantom, which is one of the most popular quads that are out there now.

And do you worry about the legal aspects of this, or are you still on the this-is-for-amusement-purposes-only bandwagon? Legal in terms of the FAA and things like that? Yeah. Yeah, yeah, yeah. So we're actually working with them. There's a bunch of committees and workshops that are going on right now to figure out how people will be integrating drones into the national airspace.

And so we're a big part of that community. A member of my team, the CTO, is actually a PhD at MIT on collision avoidance for large-scale aircraft. So there's a lot of great ideas and technology that are much more mature than just what you see there on the shelf. That's just our first product. We plan to release a lot of more and larger products for larger-scale aircraft as well. Are you going to do more than obstacle avoidance? Yes, we plan to do collision avoidance, to sense and avoid using ADS-B and TCAS technologies as well

to actually, eventually our big dream is to have large scale UAVs flying in the national airspace just as large scale aircraft fly now. And so you have manned aviation and unmanned aviation hand in hand flying all across the country. So you have unmanned FedEx deliveries and unmanned aviation hand-in-hand flying all across the country. So you have unmanned FedEx deliveries and unmanned DHS deliveries, all these things. And so, you know, the first step right here are small quadrotors because we feel the FAA will let those fly sooner, and then we'll work together to get those large-scale aircraft flying in the near term. That sounds great. Thank you for talking

to me. Yeah, my pleasure. Thank you. The last person I spoke with on microphone was Michael Holtman of Conversant. Conversant is the world's first Internet of Things service bus that is actually certified and mandated by the U.S. Department of Defense because of our interoperability and our security features. And you were telling me about how it is based on one of the old protocols that I knew from years back that Jabra was based on. What is it? What is the acronym?

Yeah, absolutely. So 1998 XMPP. And XMPP, it stands for Extensible Messaging and Presence Protocol, was developed as a communication and federation protocol. Specifically, a good example would be SMTP with email. The ability to, no matter what server or what type of system you're on for email, SMTP actually is able to federate. That's the same thing that XMPP does.

It allows you to be on any type of system, any type of platform, and XMPP federates and allows a communication between those different protocols, different platforms, and different systems. And so if I had an embedded device and I was making firmware for, I don't know, say an internet-enabled pedometer. How would XMPP help me? So that's a good question because there's a couple of different ways. One of our clients is a large HVAC manufacturer and they have a network of 200,000 buildings with between 125 and 1500 devices in each building that they do remote device management for. One of the things when they went out to do this network is they couldn't go into the building owner

and say change out all of the systems that you have in place and put new ones in that are with smart systems or capable of able to communicate over the internet. So in order to, you have manufacturers like Honeywell and Train and others that are out there, Carrier and such, and each have their own flavor of protocol that they like, whether it be a data protocol like BACnet or Lawn Works or Modbus. What we did in that system, instead of actually going into the embedded device, we say go ahead and leave the embedded device, leave the control systems in, in the building that are working fine and are very efficient,

and take that controller and put on a client and turn that controller into a gateway. And then what we do is we have applications built, and now we have a partner with a company called Kleister as well as KTC, and they're both companies out of Sweden. And the capability to then take those different protocols that I mentioned or protocols such as MQTT, which is another message queuing protocol, or AMQP and such, and Kleister and KTC built a bridge software that actually will take those, translate, transform them over to XMPP,

and then the delivery of that from that gateway up to our system, and then to send controls back through the secure DOD-mandated and certified XMPP protocol, where when it gets back to the gateway with the controls, it'll unwrap that protocol, change it back into the, convert it back over to the, whichever protocol the data or the communications protocol is working within the local area, and send that data back to the device. Michael and I spoke for a little while longer, but maybe you heard it in that recording. They started to reconfigure the booth out from under us. So next year, I think I'd park a little closer, lure people to my car and record there where it'd be nice and quiet, at least compared to the show floor, which was not a good place to have a technical conversation. And I don't know what's

going to happen next year. I don't know if they're going to keep it at Fort Mason. I don't know if they're going to try to zoom in on one of the particular areas or keep it eclectic. Overall, I had a good time, and I look forward to seeing where Solid goes. I also look forward to getting back into the studio where the conversations are much easier to understand. In the meantime, thank you for listening. The show notes will contain links to everybody's

company that I've talked to. And thank you to all of the people who spoke with me. It was really interesting. I learned a lot. I hope you did too.