#369 – An Interview with Jason Huggins

Chris Gammell: This is The Amp Hour Podcast. Released November 26th, 2017. Episode 369. An interview with Jason Huggins. Welcome to The Amp Hour. I'm Chris Gammell of Contextual Electronics. That's me. That's my cue. That's you, man. Can we get a robot in here? We need a robot? Can someone get a robot to push the button for Jason?

Jason Huggins: Can we automate this? Let me get my tape player here. This is Jason. I'm Jason Huggins, the founder of Tapster Robotics.

Chris Gammell: Welcome, Jason. How are you doing?

Jason Huggins: I'm good.

Chris Gammell: Coming to you live from all the way across Chicago, Chicago. Jason and I do live in the same city, but not really. I mean, Chicagoland's pretty big, so we're not in the same spot. Nope, we're not. I'm over all the way on the other side of town. The other side of the blue line, really. Exactly. Yeah, I live close to the airport. He lives close to the other end of the blue line. People who have been here will know what that is. And I should also mention, for people that are listening up front, we will be having a meetup, a 3H, which is the local hardware meetup. Hardware Happy Hour is on Tuesday, I think starting about 7, downtown. So there's a link in the show. And Jason might even be there. Someday. Someday I'll get there. Right.

Jason Huggins: Show and tell was my favorite subject in school, and that's pretty much what that is. And so it's embarrassing I haven't been there yet.

Chris Gammell: Right. Less beer in school, though. True. Yeah. From what I can remember. That sounds like something you'd be interested in. I remember you and I hung out in a German restaurant, and you just had a bunch of robots. It was me and you and Kat, and you were just like, there was beer and sausage all around and robots on the table, you know?

Jason Huggins: And this was in San Francisco, I believe. Yeah, that's right. Yep. Right. Right. Those things happen more often at random German sausage restaurants in San Francisco than it does in Chicago.

Chris Gammell: Well, we only have a couple German restaurants here, so that's fine. Yeah. Yeah. So tell us, how did you get here? Why are we talking right now? Also, I should mention that your Twitter handle is what I sometimes refer to you as, so it's Hugs on Twitter. Yes. Yes. H-U-G-S on Twitter. Right. So if people hear me say that, I sometimes get confusing looks.

Jason Huggins: Right. And I would like to say it makes sense because my last name is Huggins. Yes. It's Hugs. But hopefully it's not awkward that it's just a random guy who goes by Hugs. I'm more of a fist bump guy, actually, so if you see me, just, you know, you don't have to give me a hug. You still might get one.

Speaker ?: Right.

Jason Huggins: I do have a funny story with the president involved Hugs and Fist Bumps, but that's a story for another day. No, I definitely want to hear that too.

Chris Gammell: Yeah. But yeah. Maybe you can work that in as you're like your chronological working your way up to now. So you come from the software world. I do.

Jason Huggins: So I think our paths, I think, kind of overlap in that sense. If you were, I think I've seen you describe this with contextual electronics in the sense of maybe the biggest nerve that you struck, I suppose, as far as people finding your content is software people trying to get into electronics. I think if I'm characterizing things correctly.

Chris Gammell: At least if I haven't struck that nerve, that's what I'm going for. Right, right.

Jason Huggins: And I am one of those people. That's a software person desperately trying to get into electronics. I think I've been slowly, rapidly, but rapidly making progress since around the turn of the century, which is now a phrase that seems like actually a year. That was a long time ago. It is. Some people who are not listening to this were born in the turn of the century.

Chris Gammell: that's a child going to college next year. You're like...

Jason Huggins: Yeah, yeah, yeah. But ever since... So ever since around 2000, I've got a lot of ideas that needed motors and LEDs and all kinds of stuff. And so I've been trying to figure out how to make those things work. But so in that sense, we have similar interests. But I can't remember exactly how the first kind of conversation happened other than me just like heckling you about how long your YouTube series was for getting to Blinky. I don't know if I just reached out to you and said, hey... That has continued, yeah. I don't know if I just reached out to you and said, hey, dude, make this shorter or if there was some other setup there. It was probably,

Chris Gammell: let's just say generically Twitter.

Jason Huggins: That's probably the easy way to say it. It might have been that. It might have started from a Twitter heckle and then, yeah, went from there.

Chris Gammell: Yes. So prior to hardware though, so you were JavaScript? Is that right? Other stuff? I remember front end something.

Jason Huggins: So, I mean, my favorite programming languages are Python and JavaScript. These days, all my stuff is mostly JavaScript. I do like... I can go down a rabbit hole just talking about all the different stuff. I actually try to keep the JavaScript I do like browser-based. Keep that pretty like vanilla and simple. I use... And also I'm using... I'm making like user interfaces for controlling my little robotic arms. So I'm not using whatever the kids are into these days, like Angular or React. All these other kind of things for making web apps. Yes. I'm using mostly 3.js for 3D stuff and I'm blanking out of the name of the SVG, but the 2D graphics, I'm using SVG under the hood. And then that talks to JavaScript running. Well, I can go down the whole stack, but I'm a big fan of Johnny5 is the library I use. Oh, yeah. That controls the Arduino and it's all JavaScript. But my background before that was... Did a lot of Python. If I'm on startup number two with Tapster, my previous company, Sauce Labs, still going on, still going strong. It's awesome. It was buzzword compliant for 10 years ago. It was cloud computing.

Chris Gammell: Oh, yeah.

Jason Huggins: Check. Whatever. No, it's an amazing, awesome thing and it was all written in Python. And before that, I was at Google doing a similar thing, doing web testing infrastructure, things like that. But I mean, purely software job. Yeah. And I've always had dreams of getting into hardware and electronics. And so now I am.

Chris Gammell: Right. So and people who are front end compliant, they would know like Selenium. Is it Selenium or Selenium? I don't know how you say it. Right.

Jason Huggins: So yeah, I started the Selenium project in 2004 or so. I've been told I pronounce it incorrectly, especially people with a British accent. I think the way we kind of thread the needle here is the element is Selenium, but the software testing tool for browser automation is called Selenium. That's the way. Yeah.

Chris Gammell: Anyway. Nice to differentiate, right? Right. Yeah.

Jason Huggins: But I think that's the thing I'm internet famous for. Or the way I like to joke, you know, if it was Andy Warhol, has that quote of, in the future, everyone will be famous for 15 minutes. Mm-hmm. I'm famous to 15 people. All right. Yeah. So I'm a nano celebrity.

Chris Gammell: Knowing you by name. Yeah. No, that sounds familiar. When all of our wonderful listeners come up to me at meetups or something like that, they're like, oh, the amp hour. I'm like, yay, one person. Awesome. Right. Right. I love it. Yeah. So that's great, man. Yeah. And that probably leads into the fist bump with the president. That was not a, that was an offer. You actually worked with the president. Well,

Jason Huggins: yeah. Yeah. So, so he had a website that didn't launch so well. Right. A couple of years ago. And if you read the congressional oversight committee report, because, you know, as you do, one, one of the, you know, what do you call it? What do you call it? The, the reports that needs to get written after some horrible thing. One of the chapters in that was like testing or lack thereof, or all the different things. Right.

Chris Gammell: And this is healthcare.gov for people who didn't put that together. And that was Obama, not the current president. Someone who listening to this a hundred years from now,

Jason Huggins: what what's a website? Right. But, but yeah, yeah. So for two months of my life, all of November and December, 2013, I always had to clarify, I was started helping after it launched. Right. Right. Before. Right. And then, and then it's actually plugged in again, six months later for when they're working on version two. I think like the whole, the whole experience there was, I felt like I was a fire alarm installer in the middle of a fight. Of a firefight. At least the first time around. Right. I was like, man, you know, I would have been way more effective, more useful six months earlier. Right. And so plugging in six months later, it helped to have that context of the firefight. So kind of like, no, Hey, you know, flames go. Everything's on fire. Right. Like, you know, do this to prevent that horrible thing from happening. Right. But I, I did feel like it was, it felt, I felt like I got more, not done. I mean, there were some awesome things that we did that first time around, but I felt the contribution was more substantial, whatever you want to call it that, that second time. I mean, if nothing else, I kind of figured like I wanted to stick around to make sure that they did like the second launch in the next year. Right. Like, you're like, like actually, are you actually doing this?

Chris Gammell: Right, right, right. I didn't give up my, my holiday time just for, for funsies. Right. Right.

Jason Huggins: So it was kind of like enlightened self-interest to see that it, like it survived launch year number two. Anyway. Yeah. So, so I, I have this, you know, software background. Um, and I, yeah, I do hope that my particular skillset is not, uh, a required, uh, you know, you know, of national importance like that ever again. I hope it's just, you know, stays kind of a boring, useful skill. Um, well,

Chris Gammell: one thing I really liked is the, uh, I remember one of the first times you and I met, you had told me about like, I don't, I think it was your analogy, but if not, then it was someone else sitting with us, but it was like, testing is kind of like, like garbage collection, right? Like in the real, like being a garbage man, like it's not super glorious, you know, it's not like the, the top job, but damn, you notice when it doesn't happen. Right. It's like, right. You know, one of those things.

Jason Huggins: I'm not sure if I've used that. I'm not sure if I use garbage collection.

Chris Gammell: I'm trying to say you're a garbage man, Jason. I'm just,

Jason Huggins: I have no, I have nothing against the sanitation, uh, you know, engineers of the world. Um, uh, but, but it is, it is something kind of like the utility company or something like that. You don't notice it until it's not there.

Chris Gammell: And people always pay for it too. That's the other thing is that like when, when people figure out that like the, the issues that causes when you don't have it, that there's actually a business model behind it. That's, and that's the real thing that I,

Speaker ?: I,

Jason Huggins: I mean, so there is a certain level. I, we did not go down this path, uh, with the previous startup, but there is a certain level of like, it's, you're kind of like selling, uh, insurance or it's kind of like, if you watch like an ADT, uh, like, uh, home burglary prevention, right, right. You're selling FUD. They're selling fear, right? It's like they're, you know, do buy this thing, uh, or else this bad thing will happen. And testing is kind of like that. It's, it's, you're trying to prevent this bad thing from happening, meaning like you put something out there in the world and it's broken and you're losing revenue by the minute or use it. There's a sort of reputation, all kinds of stuff like that. Um, and the problem is when you talk to two kids in a garage, uh, working on their startup, they haven't had that massive failure yet. Um, right. And so they're like testing, whatever.

Chris Gammell: Especially at scale, right? Like,

Jason Huggins: right, right, right. Um, but like, you know, big companies where they've had that embarrassing outage, uh, then yeah, two things happen after that, you know, 3am outage, a massive budget opens up for testing and for monitoring. Uh, right. Yep. And if it's not there, it's like, okay, well, I'll, you know, we'll follow up in a year or two. And yeah, anyway.

Chris Gammell: Yeah. No, I mean, I, I remember that from manufacturing places I worked at. It's like, Oh, well we can't get another test stand. And it's like, and then like a couple of products go out that just didn't get tested or got, you know, kind of shuffled through. And it's like, Oh, what do you know? Got a new test stand.

Jason Huggins: So there is interest. It is interesting though. Um, coming from the software world, going into electronics and hardware that, um, it, it does feel like the electronics world does take it more seriously. Like, I mean, everything coming off an assembly line, like it, I mean, you would be laughed out of the room. If you said like quality control, like what's that? Like, Oh, we don't, we don't test our boards. Um, I mean, so there's a certain level of like, you don't, you don't just pay it. Let's service, I guess. Sure.

Chris Gammell: But I guess that depends on the industry too. Right. So like, uh, uh, you know, large scale consumer product might have like a built in self test, but you know, it's not going to get, it's not doing full calibration. Like, like when I was at test equipment company, right, that was huge. A huge piece of it was the calibration setup, you know?

Jason Huggins: Right. Right. I mean, it's all relative and very specific to the particular case, but there's, yeah. Um, but it is interesting. Just, you know, how do I say this? Um, I, I used to just always wanted to do electronics. Like I didn't know how or where to get into it, but I knew I wanted to do something there. And meanwhile, I had this software career that became kind of, you know, I had the specialty and testing tools and test automation. Um, and I, you know, almost like I was on this path toward like a midlife crisis. Like I'm going down this one direction. I want to go in a completely 180 over here. And what I didn't realize though, is that, uh, one, like hardware is becoming more software. I like, or at least the practices are becoming, um, you know, software practices are becoming more, more common in hardware. But the other thing is like testing is a, is a activity that happens in both of those. So it's nice that I've been able to make that switch by being in this testing world, uh, and it ended up not being a, uh, diversion. It ended up being kind of like a, a cool prerequisite of sorts to, uh, get my foot in the door. Right.

Chris Gammell: Right. Well, and I think too, the, uh, the timing was really apt with the, uh, the rise of touchscreens and everything like that too. It's just that there were some solutions, but you know, the ultimate thing is that you needed a finger to, you know, basically enact a bunch of apps. Right.

Jason Huggins: I imagine anyone who's listening to this and knows no idea who I am. Um, right. They probably like, what, what, what is maybe? It's time to say what Tapster is. Yeah. Right. Right. Right. Uh, so it's, it's, uh, Mark 15 minutes, right? Yeah. There you go. Yeah. Like what the hell? Um, I make a little robot. Um, and it's like a little robotic finger that taps buttons on, on phones. It tests, you can use it for testing touchscreens. So touchscreens are like everywhere now. So, uh, they're in cars, they're on phones, they're in all kinds of places. And so I make a little, uh, robot for, for testing that.

Chris Gammell: And so that has changed a bit too. Um, so what rev are you on? I actually, I don't even know. I haven't seen you in a while. I don't know. What, what rev of robot are you on?

Jason Huggins: Oh, I, I should, I should probably, you know, I need to go back and watch my, um, Iron Man movies and just like, you know, have like follow that naming convention, Mark 15, Mark 16. I don't know.

Chris Gammell: Are you at the 15,

Jason Huggins: I don't know at what point he achieved flight with his, his boots, but, um, yeah. I, I mean, my first robots were made out of basswood, laser cut basswood. Uh, I moved away from those. That was a long time ago. Um, yes, you can start a robotics company, making your robots out of wood, just kind of putting that out there. Um, uh, so the, the, the latest thing is called sidekick 1.5 and that's actually oddly the third version, I think of sidekick. Um, I have, I have, I would have to count the generations of it, but I've been working on it. I think my first, my first prototype was 2011. Um, and I've been kind of working on it since then.

Chris Gammell: Okay. Yeah. And I, I remember seeing, I think it was the first one, but I remember seeing it on Tindy to start with. Yes.

Jason Huggins: The very first product name. Like I, I didn't believe people when they said they would buy it, they wanted to buy it. Yeah. So, uh, I, I put it up on Tindy. It's actually, there's a funny story there. Um, uh, I, I guess I created my project, um, maybe six months before Tindy launched. And so the way Emil, uh, solved that kind of marketplace chicken or egg problem of how do you get customers and, uh, you know, vendors all in the same spot, he, he solicited people who made projects first. And then when he launched for people to come in and buy stuff, he had kind of a fully stocked e-commerce store, if you will. Right. So I remember like maybe like may of, was it 2012? Maybe he, he was soliciting projects. So I put mine up there. I think my official project listing is the robot that plays angry birds because I hadn't implemented any other functionality. So it was literally, that's all the robot does. Right. It just plays angry birds, nothing else. Um, and then he emailed me three months later, totally forgot about it and said, Hey, you know, there's three people on the waiting list. Uh, you haven't supplied like a PayPal ID. Like as soon as you do that, you will get three orders.

Chris Gammell: Remember like what?

Jason Huggins: And this is, this is also ridiculous because again, this is at the time I was making these out of, um, uh, laser cut basswood and I was flying so much back and forth between Chicago and San Francisco. I had a membership at tech shop San Francisco and that's how I made all of the parts for my, for the kits. So I, I would have to fly to San Francisco to fulfill these orders. Um, luckily I went there once a month or so. So it all just kind of made sense, but it was, uh, it was a very surreal moment. I guess this is like summer or late summer of, of 2012 when I got three orders for literally the product is the robot that plays anger birds. And so like every moment, uh, as I'm laser cutting these kits and putting them in the mail, it's like, I can't believe this has happened. Why?

Chris Gammell: Why are they doing it? Right. Did they want it for that? Did they figure it? Was there any tie to the actual testing piece at that point? Right. Right.

Jason Huggins: So a bunch of people, uh, in various places, like I like to call them, you know, the people who work in the secret robot labs of the world and they're everywhere, by the way, everything that is made, um, above a certain dollar amount, I guess, um, has, they have a secret robot lab and it's full of robots and they're, they're just clicking buttons. And so there was one company, they bought three. This was my first, uh, repeat order. They bought like two or three of them. And I'm like, okay, I definitely have to find out. They're either really into angry birds. They're the people who make it or what's going on.

Chris Gammell: Like get alive folks. Come on.

Jason Huggins: Right. And they're a company there. They're, um, they sit between the phone makers and the telecom networks and they certify devices on networks and they certify networks for phones or something like that. They're kind of like service level agreements.

Chris Gammell: I actually deal with that now. Yeah. I did.

Jason Huggins: I like a service level agreement enforcer, if you will, like if they've got whatever. Um, uh, I guess anyway. And so, uh, when I was talking to them, I was like, wow, I didn't even know that that thing existed. Um, anyways, so I had a lot of, uh, I had a cool conversation with her founder and we both got to, you know, that dream of sorts of, of the, the future of, of robotic testing is, uh, you know, I guess this is slightly dystopian, but it's kind of like, you know, that scene in the matrix where Neo wakes up and it's just pods everywhere. It's like, that would be the dream. But like, instead of humans and pods, it's like just robots and, you know, uh, phones tapping, tapping on screens. Right. Right. Right. Right. Right. Right.

Chris Gammell: So, so like, if it was like, uh, like, uh, like that scene in Silicon Valley too, where they, they actually have people in like Bangladesh clicking on stuff, but now you could just do it with robots. You're saying,

Jason Huggins: um, I haven't, I think I've watched up to, uh, I definitely watched the first two seasons. I'm not sure if that's in season three or the last one. Anyway,

Chris Gammell: they're, they're buying users. That's the idea. So,

Jason Huggins: I mean, it, it, so every article that's ever that's written about like, uh, jobs and automation and robots and the future and, uh, the future being horrible. I do read that and try to, you know, I mean, there's a certain level of like, I'm trying to figure out like, how do I, uh, uh, do a whole bunch of robots and automation stuff and, and do it ethically, um, and not, you know, be the downfall of human society as we know it. Uh, okay. Right. Yeah. That's good. That's a good goal to shoot for. Yeah. Um, so, so, and, and with that said, I, I have been, I have been in that conversation with a particular customer and they were like, you know, if it's a dollar more than this amount, it's cheaper for me to just hire hundreds of people in the Philippines. I think it was. And so it was really weird sitting there in this couch at the, um, at, I think automate Chicago was the event that, um, we were just sitting on a couch, uh, talking about, uh, robots versus humans and hanging in the balance for a whole bunch of, uh, jobs in the Philippines that I, I didn't even know them until that sentence until he, he mentioned it, but, uh, it was kind of weird. Uh, it's like, okay, so what do you do about those people who now,

Chris Gammell: did you raise the dollar, the price by a dollar or no?

Jason Huggins: Well, it gets more complicated because they were, they were, uh, they were, I didn't, I didn't have to worry about it because that, that, or a particular order didn't happen. They were, I think they were bidding on some other contract with one of the big three car makers.

Chris Gammell: And right. I think at that point, like those, those kind of like, like that level of ethics where it's like someone else, is really making that decision. It's like, look, keep that economical. I think. And then I,

Jason Huggins: I mean, what I would like to say is, I mean, my, my most optimistic view of this is that, um, uh, if that job is, is staring at a phone 12 hours a day, uh, tapping the big green button and it could easily be automated with my little, you know, shoebox size robot. Um, I would, would love, uh, for that person to then, you know, move on to learning how to program. JavaScript and Python. And then they can take your job program. My robots, right. Right. Right. Right. They can kind of move up the tech stack and not have to, uh, click lines. And also the other thing is like, at least with some of those things, uh, we had a different conversation with a different car company and they had a particular test scenario where they wanted it to run, uh, 12 hours, like overnight. And they wanted to sit, maybe trigger, uh, like a buffer overflow or something like that. They're just going to like go backwards and forwards on one of like the screens or something like that, like go into some panel and then come back out. And they're going to see if like, maybe just walking around the app for 12 hours straight, if that would trigger some kind of thing. Uh, and that's something that actually you couldn't have a human do because, you know, if nothing else, uh, it's illegal. It's illegal. It's, it would be illegal and immoral to have a person, uh, clicking for 12 hours straight without taking a break. Right. So there's certain things where it actually is more ethical and moral to have a machine do it than to have some particular test case scenario, you know, have someone, um, do it anyway. Um, yeah, but it's, it's, uh, it's interesting how this will all played out. I think, you know, I, um, when I was a kid, I, I think, uh, there was a, there was an existential crisis, I think about the robots coming in at the time. It was, uh, Japan was coming in with all their robots and, you know, they're going to destroy all the automobile, uh, manufacturing jobs. Yep. And it's weird. Cause that kind of happened. It's kind of true. Like, you know, like I haven't been inside a, an assembly line, but like most car doors are probably welded.

Chris Gammell: I was going to say, I've been in a Tesla assembly line. Uh, and, uh, yeah, there, there's some robots. Uh, yeah.

Jason Huggins: Yeah. I mean, so, so it's like, I'm sure there are some people welding a couple of things here and there, but maybe, um, right.

Chris Gammell: But quality goes up and, yeah, I think it's, it's a general shift to, this is, this is going to happen no matter whether or not you're, you make the robot as everyone else makes a robot. I think that's the main thing. And that's what I've said in the show before too. It's like, right. Look, if you want to, you know, if you're worried about that, become the robot maker, basically.

Jason Huggins: Right. Right. I mean, so yeah, if the robots are going to like wipe out humanity, be the maker of the robots and, you know, make sure that you get your friends and family on the white list that, you know, well, okay. Good human, right. The problem with good human versus bad human, that's like just a binary Boolean field of stored in the database somewhere. So like if you're, if you're off by, you know,

Chris Gammell: there's no slider there. Yeah.

Jason Huggins: Yeah. Yeah. You have to be really careful that the good human, bad human, you know, that is, that feature is, is tested. Got it. Yep. I'm kidding.

Chris Gammell: Anyone who clicked off at that point. Yeah. So, well, let's, let's talk about, as you make these robots are going to kill everybody. What, what, how, how has the process been of, of making these job killing robots, Jason Huggins?

Jason Huggins: So every, every step of the way, I mean, I mean, since, so, um, I have to frame this kind of in my, in my theory of, you know, the future. Um, my favorite, one of my authors,

Chris Gammell: I love talking to Jason, by the way, because it never is just like, Hey, how is that circuit board? The circuit board worked out fine. It's so, no, no, no. Well, let's get, let's get deep, man. First,

Jason Huggins: let's talk about the future of humanity. And then I can tell you about this Arduino, you know, uh, shield.

Chris Gammell: I'm making an apple pie. First, you must invent the universe.

Jason Huggins: Yeah. So, so Kevin Kelly is my favorite author. Uh, he's a, uh, futurist.

Chris Gammell: We met him in a week of his thousand true fans theory.

Jason Huggins: And, and I don't know if this is his theory or if he's just articulated it. And I remember bringing it this way, but, uh, he's got the theory that basically, uh, over the long haul, um, everything is basically, you know, everything even now, but like everything is like 51% good and 49% bad. So, so basically like you can spend your entire day, literally almost spending, you know, you could just consume a ton of news and 49% of it's going to be horrible. And it's going to be really, really bad stuff. Like people dying and all kinds of things. Right. Uh, but the thing is like over time that, that 2% difference differential between 49 and 51, that means like, um, penicillin is discovered and like, you know, uh, polio is cured and, um, you know, all of these things that actually ended up making life on, on earth better. Uh, it's, it's, it's hard to notice that when 49% of everything at any given time is. So, so there is a certain level where like jobs getting displaced, uh, because robots, uh, like that's kind of in the category, 49%. Um, my, my view on the 51% is that, uh, and I, I know this will come across as kind of naive, but this is kind of where I'm at. Like I, I got into this stuff because like, I want to make more art. And I think that like, that's, you know, top of that self actualization kind of pyramid. Like I think, uh, making a cool thing, sitting back and kind of enjoying it as a work of art, like, and, and all the skills that went into it, I think that's like a good use of time. Uh, it's kind of a cool thing. So if there was more people who could be freed up to make cool art projects, I think that's cool. Um, and if they could avoid, you know, trudging along, clicking buttons on screens for 12 hours, that'd be good. Anyway, so I, I want to create more art in the world. Uh, and that's why, and I'm doing it in this particular way. Anyway, now I've totally forgotten.

Chris Gammell: Rising, rising tide lifts all, uh, button clickers out of the, out of the, the quagmoyer of button clicking.

Jason Huggins: So what I'm trying to do with tapster, like, it's not like this evil ploy to just like get everyone unemployed. Like really, this is, I don't actually think that. Uh, for me,

Chris Gammell: this kind of playing on that, you know, right.

Jason Huggins: For, for me, um, I'm trying to, I've got, uh, very particular art projects that I want to get done. And, uh, I haven't, you know, uh, no one's come to me and just given me a pile of money saying, I want you to make that art project. I imagine some people do have that job. Uh, you get like an art commission and, and there you go. Um, I haven't stumbled into that, but what I have stumbled into it is like, um, I, I made this robot and, and all the people in the secret robot labs who actually have a bunch of buttons that need to get clicked. Um, they want my robot. So, uh, and the cool thing is that the skills that I know, need, you know, with a 3d printing and CNC milling and soldering and all kinds of stuff that I need to make my art projects are the same skills I need to make these, uh, products for companies. So, um, so anyway, what I'm working on is basically kind of like this little economic engine, a business model, if you will, like, you know, to, uh, become my own art benefactor. Um, yeah,

Chris Gammell: I like that anyway.

Jason Huggins: So it's kind of like, you know, um, I don't know. I haven't been a burning man, but, uh, there's a certain level of like, you know, the people who just kind of work, work on these huge, uh, massive art projects that really have a market of one, like, you know, the person who thought of it and wanted it to exist. Um, you know, I've, I've got things that I want to exist. I imagine someone who's listening to this, who's, uh, it's probably really frustrated with me right now. Cause I mean, there are so many, there are so many problems in the world and I'm not, uh, uh, uh, ignorance of that or, you know, immune to kind of hearing all the bad news. I mean, there's a lot of bad stuff happening all the time, but there's a certain level of like, you know, you could, you could just wallow in that 49% of horribleness. Um, and you can go to a really dark place or like maybe start, you know, change your compass heading towards thinking about like, Oh, maybe things will get better in the future. And, you know, art is, is one way to kind of, you know, uh, it's kind of a soothing kind of a thing that, um, I like that.

Chris Gammell: So, well, let's talk about the, so let's talk about the skills that you've built up because you said you've, uh, you've built those from scratch effect. I mean, I guess 3d printing, not like anyone was born with that, uh, in, you know, the times that we were born in. Uh, so, so how did you get started with that first robot? I mean, the, again, you said it's a, uh, a bat, you said Basswood, uh, it was started with Basswood. Right. But I also remember you had this, uh, this kind of, uh, modular screw down.

Jason Huggins: Yes. Right. Um, I could go. Yeah.

Chris Gammell: So maybe explain, could you just kind of like make an audio picture of what the robot is starting out and today?

Jason Huggins: Right. Um, bleh. So, I remember a Delta robot. That's why I'm saying. Right. Right. Right. So there's a certain level of like, I know all the jargon and I have to figure out how to like translate that back to the non-jargon word. Um, it's a, it's a robot army clicky, uh, articulated thing, right? Right.

Chris Gammell: It's a finger floating in 3d space. My,

Jason Huggins: my first robot was car. It was a Cartesian gantry style. Uh, that would be,

Chris Gammell: I didn't know that actually.

Jason Huggins: It kind of looked, it almost looked like, you know, the old school, uh, uh, Mendel rep rap. It had that kind of like, if you look at it from the side, kind of triangle, like the, exactly. It looked, it looked like that kind of house, I suppose. Right. It looked like an, uh, like a generation one or two rep rap. Basically. That's what it looked like. Um, someone saw that. Um, and I moved to Deltas and, uh, because it was playing a video game. They're like, you know, it should be fast. Uh, and so Deltas are fast. And so saw the YouTube videos. If you haven't ever seen a YouTube video of a Delta robot packing or, uh, stacking pancakes, you should go. I was going to say,

Chris Gammell: uh, yeah, that's my former employer. ABB has one where they, they're sorting, sorting sausages as well. That's another one that recently came out. It keeps coming back. Great. Um,

Jason Huggins: sausages.

Chris Gammell: Oh yeah. Right. Maybe it was jerky. I don't know, but either way. Yes.

Jason Huggins: Uh, and then, uh, since then I've actually moved to, uh, and I don't know how to pronounce it, whether it's scara or scara, but I've moved to a parallel scara. Um, it kind of looks like if you put your hands to your arm, your two hands in front of you and you're like, you're imagine like you're stirring a big, you know, pop that of spaghetti or something like that. It kind of looks like that. Um, uh, there's, I think someone's made, they were kind of internet famous of sorts, something called the plot clock. If you've seen it, it's, it's a similar kind of design. I haven't seen it. Whatever. Um, but, but the, um, but to your question, the material that I used to make it, um, the actually, again, oddly, but maybe not oddly at this point, the tapster and the robot, like that was really a spinoff, uh, from this art project that I'm pursuing since 2000. And, and, and, ah, man, there's like a five hour version of this and like, I'll try to make the pithy version of it. But, um, the particular art project I'm trying to get done is, uh, basically motorized pin art or, uh, or people don't know the, the name of it. Usually it's pin art or pin screen. It's like that array of metal pins. You can kind of put your face in.

Chris Gammell: You always got that at like the, the science museum and you put your hand in there and then, right. Right. Right. So,

Jason Huggins: so the idea that I had all these years ago was like, what if you could motorize ever soon? One of those pins. And then it would be like a real 3d display. And so I, at the time I thought that was interesting. I was doing 3d simulation on my 2d computer screen and I had that little pin art, uh, display. And I was like, ah, it'd be so cool if I could get the 3d simulation that's on my screen on this little pin screen. And it'd be awesome.

Chris Gammell: So have you seen that, uh, Coca-Cola billboard that does that?

Jason Huggins: Yeah. I mean, it's, it's come up a bunch of times. Uh, I've talked about it. So I've been so annoying about this, talking with my friends. Occasionally when things pop up like this, uh, my friends will, uh, uh, send a link to me. I think the most frustrating in that moment when someone's like, Oh yeah, it's like that thing. While I was in DC working on healthcare.gov, uh, the MIT inform project launched. Uh, and it's basically that exact project.

Chris Gammell: It's like when you're actuators that move the ball around. And so here, so here I am the, the, um, what? Michio. What's his name? Uh,

Jason Huggins: I forgot the name of the, of the principles on the project, but, um, uh, but, but it's, it's, it's these motorized, you know, it's, it's, it's the idea. Um, anyway, um, but it's, it's been this project that I'm pursuing for a long time. I think the key thing is like every single one of those pins, cause I need to have so many of them, like tens of thousands of them to make it like a super epic thing. I need to drive the cost down of every, uh, pin. And actually I'm using the wrong word, I guess like the formal jargony technical word would be linear actuator.

Chris Gammell: Yeah.

Jason Huggins: I need thousands and thousands of linear actuators and I could go buy some off the shelf.

Chris Gammell: That's I think what the MIT people are like industrial and, it's super pricey, right? They, they did it with money. They fixed it with money. So I did. I, yeah.

Jason Huggins: Well, MIT solved for, for glitz and panache. Uh, you know, they didn't solve for cost. Uh, and so, um, right. As,

Chris Gammell: as Nadia peak says, uh, MIT is really good at making a paper, uh, and then stopping. Right. Right.

Jason Huggins: I mean, so I've got like a, uh, a massive crush, jealousy, love, hate relationship with the MIT media lab. Like I love everything that they come up with. And, and, and other times it's like, Oh, darn. I wish I had their PR department. Right. Exactly. That's,

Chris Gammell: they are great at that. Yeah. But anyway, anyway,

Jason Huggins: they're not making lots of robots to send to people either. So, right. So the key thing for this art project is, is, uh, I had to drive down the cost, which means like as much as I can make myself, um, then, and I don't have to, you know, buy a bunch of parts, whatever, then I can drive down the cost. And my, my goal is like a dollar or two per linear actuator. And then I can, and then at that point it's like in the ballpark of, uh, you know, I could do this. Um, anyways, so the thing I've learned out of making, I think robots and anything mechanical electronic, um, I mean, it's kind of like a triathlon of sorts, right? You've got the software, the electronics and the mechanical. Um, and so I had the software background, uh, and I did some, um, some mockups, uh, uh, and I'm also, I'm not so great at web design. So there is a, there is a software demo for this. Um, but it's, there's literally nothing on the site that tells you anything, any context about it. So pin thing.com. Um, I can never figure out what to call my project. So I kept calling it pin thing. So it's pin thing.com. There's nothing about it. There's no even like secret message in like the view source. It's like literally that's it. But I said, I had the software demo, uh, web demo for a while. Uh, thanks to the maker movement and open hardware, open electronics and Arduino and Adaprute and spark fun and all of the chocolatey goodness of the last 10 years of, you know, the electronic space. Um, that was, that was easier to get done. So I had the software side, the electronic side was getting easier. Like, you know, all these tutorials and how to drive a simple motor from an Arduino, all that stuff was easy. But, uh, what, what was the huge stumbling block was the mechanical side of things. Um, at least, you know, a couple of years ago. Um, and I had, I finally got to a place where I was actually prototyping with Lego Technic, uh, pieces. I don't really, whatever. Anyway, so, so, uh, I got to a point though where the particular prototype.

Chris Gammell: explain what that is though. The Lego Technic is the fixed hole spacing. It's like the, the yellow, I always think the yellow and the gray ones, right? Is that right? Or is it different colors?

Jason Huggins: Oh, it's different. Uh, anyway, uh, Lego Mindstorms also uses Lego Technic pieces. Yeah. Um, and, uh, I, I, I've never gone to like a Lego convention or anything like that, but I do, uh, you know, there are people who are, you know, uh, super into Lego. Um, I've used it as like a prototyping tool for whatever contraption I'm trying to make. Anyway, uh, no offense to people who go to Lego conventions, but there's a certain other, there's like, I'm using it in a particular way. When I, when I realized again, from this particular point where I need to make 10,000 of these things, I actually priced out like, you know, the, how much does Lego cost per kilogram or whatever. And it's really expensive, right? Yeah. It's, um, and then I thought it was interesting. Like it's out of, made out of ABS plastic and all these 3d consumer 3d printers are coming out and they're also used ABS. So I started pricing out before I even got a 3d printer. Um, you know, uh, you know, could I make these materials? And I think it came out to like, just with filaments, uh, it's like 10 X, seven X cheaper than Lego per pound or something like that. And if I moved someday to like making my own filament out of like pellets or something like that, it's like another 10 X improvement in the cost, whatever.

Chris Gammell: Or if you move to injection molding, which we'll also talk about.

Jason Huggins: Right. But the key thing is I, I had this design finally for one of these linoleum actuators made in Lego technic. And, uh, and I was like, okay, I'll, I'll be a proper engineer now and re-engineer it in some other material. Um, and, uh, anyway, I kind of like, no, it's like, eh, but the geometry is working in Lego. So, um, I ported basically Lego. I may, I designed my own Lego pieces in like an open S CAD.

Chris Gammell: In a totally non trademark infringing way. If, uh, if the Danes are listening. Uh, no, well, all of the designs for the, um, I'm trying to save your ass here, Jason.

Jason Huggins: So I checked on this. So, so all the patents for the, the parts, uh, they've like, they've expired. And then there's utility patents and design patents. The design patents, um, have expired for the parts that I'm, uh, checking out. However, that hasn't stopped Lego from suing people, even when their patents have been, um, right. They're really good at that part. They're really good at the, at the suing thing. But, um, anyway, um, it, it is interesting. So, yeah, so I basically made my own Lego and I can 3D print them and it's all awesome. Uh, and so whenever I'm starting a new project, um, I, I kind of, it's, it's almost like the, uh, like there's nothing more terrifying than a blank white sheet of paper, right? Like it's completely blank. Uh, it's the same thing with like constraints, right? With anything with mechanical, it's hard to know, uh, kind of where to start. And so if, when I'm first trying to kind of sketch out an idea, um, I have all these beams and they're the same kind of dimensions as, as Lego techniques that are eight millimeters wide and the holes are eight millimeters center to center. And so I can kind of quickly kind of create some kind of boxy form. Um, and then over time I can, you know, uh, be more intentional about it, get rid of the holes I don't need, whatever. But, uh, there's a certain level of my design processes. Like I need a bucket, I need it like a box or a bucket of parts on my desk and I just start building. And then after the fact, I'll design it from scratch.

Chris Gammell: Uh, right. And this is actually, this is a brilliant piece of prototyping advice. I think in general is, um, use what you know, right? So that's obviously what you're doing. You're using standard parts, uh, readily accessible, you know, you could pay for them, but you're doing it for cost savings to just print them. Um, right. And yeah, I think, I think that from a prototyping perspective, that's fantastic. And I, I, oh, sorry, go ahead. And that, well, I'm saying that that's, that's served you all the way up through your current iterations, right? I mean, like, so your first one sounds like it was more of a rep, rappy style, but then the Delta robot that people can see online, it's, it's very, you know, it's got all these pieces in it and it's got some other custom pieces too, but all the hold down to the phone and stuff like that are adjustable.

Jason Huggins: I think one of the first thoughts around that was like, I had to make, I had to design parts that Lego would never make, like to hold like a classic, uh, high tech three 11 servo. Like it doesn't make a part for that or a, an Arduino mounting plate, right? So all those things I made, like a Lego compatible thing of that. But, um, um, anyway, slight segue here. Uh, this is why I'm such a huge fan of the fritzing project. Oh no, we're going there. If you start where, where you are at the very first robot that I sold, besides being, you know, again, this is the angry words version. Um, besides being made out of wood, uh, the circuit board was literally an Arduino and then a shield and then a, uh, one of those little mini breadboards and like six wires to connect up. Like, so I, anyway, so yes, you can start a robotics company, uh, with robots made out of wood. You can also sell circuit boards where all you're doing is, um, giving them a breadboard and wires. Now, granted I've moved on. I'm way more professional now, but where I started was like breadboarding. Yeah.

Chris Gammell: It goes back to why Jason and I enjoy, uh, uh, you know, talking to each other and, or arguing over, uh, fritzing versus KiCad. Uh, so I, I, I actually, the, the, the, what did I call it? Uh, shine on you crazy KiCad was my even simpler version of a, uh, you know, getting started tutorial, basically just a blinking light on top of raspberry pie. That was fully inspired by Jason. And then a couple months later, it's out there. I, I'm like, Hey, have you watched the, he's like, Oh yeah, I'm not going to watch that. It's still too long. I'm like, it's a 20 minute video to get a, to a completely, uh, done circuit board. And yet for some reason, it's, uh, you versus Brian Benchoff until the end of time, you, you still stand by your, your fritzing.

Jason Huggins: I, I will tell you why it's because the, uh, the fritzing folks put together a video. This is like how to make an Arduino shield in like one minute. Um, or one of one, the one minute Arduino shield, which I think, uh, it's false advertising because I think actually the video is like two minutes long. Uh, it's pretty obvious to kind of fact check them on that. However, I think that was basically where my attention span was at. It's like, I, like they got rid of everything else and, uh, right. And that was it. Um,

Chris Gammell: right. Well, it also points to the fact that, so you are not really doing all that much electronics, right? I mean, you're using a lot of the off the shelf stuff using Arduino. So the, um, right.

Jason Huggins: The, um, I did make my first, uh, or I, right. Yes. So, right. The robots are, um, off the shelf components. Although now I have, um, I think the design, what do you want to call it? Um, I've got the things that I want in the next version. I might have to make a custom circuit board up that I'm using like, uh, all right,

Chris Gammell: we'll finally get them. We'll get, I'm going to get them folks. Don't worry.

Jason Huggins: So the gory, gory details right now is I moved from servos to steppers. And by moving to going into that move, the, the big risk of the project was, it was all, all the mechanical and electronic changes. Um, I, I stuck with off the shelf, like CNC, um,

Chris Gammell: yeah,

Jason Huggins: uh, circuit board. So it's like a, the CNC shield and then a Pololu, uh, uh, step drivers on top of it. One thing I noticed is that, that, uh, I need to disable each motor individually so I can do a separate calibration routine. And there's only one global, uh, on or off enable disable. And so now I'll either, I'll have to go in and, uh, figure out how to do that or potentially make a modification to that board. And that'll probably be my first custom board. Uh, also I need six steppers for the, like it's kind of an involved robot. Um, and so I don't, I haven't seen any boards yet. That's kind of rep rap family of sorts that controls six. Uh, anyway, uh, but I have done circuit board design.

Chris Gammell: Yeah, I know. And I think, but I think what that really points to, and probably what I should have brought up at the top of the show is that robotics is, and I think you mentioned this too. It's like, it's this holistic kind of design thing. It's like, like you said, mechanical software is a huge piece. Obviously you're doing, from the browser visualization all the way down through the actual control of steppers, you know, utilizing some library, stuff like that. But that's really the piece that keeps coming back to it. I mean, I don't know if you saw that. I mean, I think about the, uh, the Boston dynamic stuff like that, that backflip robot they just released, but like, I mean like layers upon layers upon layers of abstraction or like robot operating system. I think you've talked about in the past, same thing. It's just like all of these abstraction layers because you know, we operate as, you know, domain experts, right? You were a software person who's gone down through the layers of abstraction and figured out all these different pieces. Right. But most people don't do that, especially in big companies, right? Most people aren't building, you know, you know, bespoke robotic testing solutions. Right.

Jason Huggins: Right. I, I actually, ironically, I'm, I'm kind of a skeptic of the, of the Ross, the robotic operating system. Um, and, uh, that's not a popular opinion, I guess is the unpopular opinion. Uh, but one specific reason is as you, as you pointed out to it, it's like, as I'm trying to understand the whole stack, I look at Ross as kind of like this huge, uh, foggy, uh, window that I can't see through. Like they're purposely trying to create this abstraction layer that you don't have to go to the, to the whole stack or through all from top to bottom and know everything. Uh, but as I'm trying to, at least for, for my needs, um, I, I want to be able to go from, I want to have a clear line of sight from the browser that's controlling things all the way to the motor that I'm, uh, really.

Chris Gammell: So you want to be able to like, so maybe that's a good question. So can you single step a stepper from the browser, like just a jog command?

Jason Huggins: Right. I mean, it's complicated, but like, you know, sending a little, you know, web socket commands and going all the way down to the Johnny five thing. And then it saying one step, right? Uh, granted, that's probably not the most efficient thing. Like I'll find at some point I'll say, Oh yeah, these abstractions are great. Uh,

Chris Gammell: I'm sure you, I mean, you already said you have, right? You're doing Johnny five, right? You use the Johnny five library. So that's already an, I think,

Jason Huggins: I think what I've realized is, and this is actually a lesson that I learned from the software world that I guess I'm kind of accidentally now it's intentional because now I just thought this thought, but, um, I, I used to, um, when I was first getting started in web development, I, um, you know, took an off the shelf, massive framework. And I realized for the next year of my life, all I was doing was hacking away parts of the framework that I wasn't using, um, because it just was unnecessary and getting in the way and whatever. And so now I've gotten to this place where I find it's a lot easier for build to build things up, uh, using as few, like no framework, but like as few libraries as possible and bring things in when I need it. Um, and so I'm a huge fan of like very small modular things you can add to, like in, in programming land, this is kind of like an argument between like using Django, um, or rails as like a all in one framework or something small. Uh, and Python land, it would be like a flask. It would be like this micro framework or something like that. Yeah. Anyway, jumping around. It's interesting.

Chris Gammell: No, no, it's interesting too, because so I, like you, like you said at the top of the show, like I, I, I work with a lot of software people. The trend that I've seen that's really interesting is like, so I work with the software people that are building a board. They say, all right, I'm going to make a board and I'm going to have a processor on there and maybe a couple of LEDs. Like I'm gonna have two LEDs and that's all it does. Right. Yeah. And, and I start saying, well, you know, you want to throw like, you know, zero ohm resistors in there so you can wire to stuff later. You want to have extra headers broken out, all this stuff. And software people push back on that very harshly. They say, no, no, no, no. It's, I'm making this one thing for this one task. Right. Right. And, and I want, and, and, you know, some of them say, I want to understand all the way down the bottom as well, but like it's, but from my hardware experience, I know that I have to get as much mileage as I can out of a circuit board. Right. I'm only going to be bodging wires in there and I'm going to be trying to tie in other things later. That's where my experience comes in. Basically. I'm not sure how, if it ties to this specifically, but that's a big software versus hardware thing that I've seen.

Jason Huggins: Yeah. I might, I might be cursed to repeat, to repeat this, you know, history of, you know, lessons learned. But like right that this board that I designed it fritzing and it was awesome and it worked. But I, I put an 80 tiny 85 chip on there or 80, 84, 85, 84, 84 has the 14 pins because I counted up how many, how many pins I needed and I only needed the pins that it had. And that was it. Like great.

Chris Gammell: Yeah. Done. Problem solved.

Jason Huggins: And yes, I could get the 80 mega three, something, whatever that has 24 pins and it has way more features and it would be the same cost. But it was like, I wanted features on that at 18 mega parts. Yeah. Right. But I wanted like literally like the minimal chip that I needed and nothing extra. Of course, since then I've realized like, Oh, it would be really cool if I had like a, I could add a rotary encoder and like, Oh, I need an extra.

Chris Gammell: I think, I think the, the it's, it's all tied to the iteration cycles. Right. So for me, I know that an iteration cycle might look like a month, but it's actually six months. Right. For hardware. Right. Because it's not just, it's not just getting that first board working, but it's also building up the other ones and blah, blah, blah, blah, blah. Right. You need to get as much out of it as you can. I remember a former coworker of mine, he designed in an entire extra channel that was unpopulated of a, of a device. And people would look at it and be like, why is this board half empty? And he'd be like, you just wait. I've been here long enough. And sure enough, you know, six months later, they, they designed a two, two channel version and he looked like a goddamn hero.

Jason Huggins: Right. So something, something related to that, I guess, like, and again, this is like showing massive like software development biases, but like I, I have an, I have an allergy, a severe allergy to compiled languages. So C, C plus plus Java, C sharp, anything like that requires, you know, waiting a second for it to compile. Oh yeah. It's so, so inconvenient. So this is like, I don't, my, my personal rule basically is like, I'll only program in programming languages with a read about print loop. Like that's JavaScript,

Chris Gammell: Python. And I guess Ruby would be out there. Right. Um, that's a ripple. If people remember Tony being in the show, he talked about repls a lot. Right. And, and so,

Jason Huggins: um, one of, actually one of the things that kept me back, like held me back for years, getting into electronics was the fear of not having that instant iterative feedback loop with circuit boards. I was afraid that I was going to do something and not find out that my mistake until a week or two later. And that's, no,

Jason Huggins: no.

Chris Gammell: It starts on fire. Like two seconds later, you just got to wait two seconds.

Jason Huggins: So, so I, I have, so I finally just got over that fear by watching a lot of, you know, five minute or less YouTube videos, by the way. At least you're consistent. Yeah. But the other part of it was like, it is nice to see now that like other mill is out there. I've got a, uh, an inventables, uh, X carve and you can do some milling on there. I think there's another thing that's, I'm really excited about, although it's not open source. So that's a dig against them, whatever. But, uh, Volterra is like a, um,

Chris Gammell: you can personally, I mean, I like the people that run it, but yeah, I, okay. So I, I, I haven't used it, but that's kind of, this is who they're selling to though. Like, so this is specifically, yeah, this is exactly what we're talking about. So this, this fits into something for people because software people are like, yes, I want to close that loop. Right. Right. So even if it's not something with it, software person says, I'll just iterate faster. I need to have a board, you know, thing. I need to do this. I need to do that. Yeah. And then me and Dave usually say, well, if you really need it, pay for overnight shipping. Right. Most people don't actually need it. It's better to just check work and yada.

Jason Huggins: So actually I did learn that like design rules, checking like that actually say, I think that was like clicking that little button that just says like enforcing design rules. Like that kind of got a little bit of that dopamine hit. Yeah. Yes. It was in fritzing. Or it was saying like, Hey, you know, this, you know, all of your traces are not far enough apart from each other and whatever. It was like, Oh, okay.

Chris Gammell: This is kind of naturally in Kaiken. I'm just saying, you know, if you would have just watched a tutorial for two hours, you would have seen this. Yep.

Jason Huggins: Yep. Um, I have watched some, uh, Kaikad, sorry, Kikad, Kikad, whatever. Kikad. It's Kikad. Uh, we're like that, that, uh, I don't know what the fritzing sound is. Anyway, um, where like there's that bump and move. What do they call that feature where you can kind of push and shove? Yeah. Push and shove. Right. Fritzing doesn't have that. And, um, uh, you're more likely than not.

Chris Gammell: Eagle doesn't have that. Come on, man. Right. So,

Jason Huggins: so that's, that's a, that's a cool feature. Uh, it's a cool feature. Yeah. Uh, Kaikad's got that on everybody, you know, like, take that away from you. Right. So anyway,

Chris Gammell: I, I, I've already, you know, like, like we say often on the show, uh, PCB packages are like religion. So I'm, I've, I've given up proselytizing to, uh, to Jason, uh, but it's still fun to, uh, you know, poke fun. And of course I always have, I always have bench off on my side. Uh, Oh my goodness. Yeah. Yeah.

Jason Huggins: Uh, so, um, yeah, I, I don't know where to go with that. So if people, if people are right, right.

Chris Gammell: Yeah. So, but this is also, this is another interesting point. So I've asked you many times in the past, Jason, why aren't you doing injection molding yet? And it's because you are still prototyping. In fact, it sounds like because you continually update now. Right. I'm not sure has that impacted your, uh, business versus, uh, you know, making a one product and then just sticking with it. Right. Is this necessary for the business you're in?

Jason Huggins: I think it's, so I'm luckily in a spot where it's, uh, you know, for all the kids in the garage working on, uh, you know, their latest iPhone app, they should get into robots. You're like, it's a sweet, um, no, it's a horrible field. Don't go into it. No, where I'm at, where I'm in now is like, it's like,

Chris Gammell: stay away. Right.

Jason Huggins: It's, it's right now it's currently, but obviously not the future when we do, um, take over the world. Uh, it's low volume, high margin. And so therefore, um, uh, it's important to kind of get it right. Um, and I guess I get in that context, I could pay for overnight shipping, I suppose. Um, I don't have to drive. It's not like I'm making a million things and I have to figure in every penny.

Chris Gammell: I mean, it's really industrial. I mean, I think this is the industrial version of software testing,

Jason Huggins: right? Right. Right. Um, so, um, there is, there is a kind of a dilemma of sorts there where like, I do have a limited capacity to make certain number of robots per month. Right. I've, although I do have eight, I'm looking in my office here. I've got eight 3d printers and, uh,

Chris Gammell: yeah. So Jason has a, uh, basically a print farm. I mean, that's the easy way to think about it. Yeah.

Jason Huggins: Ironically, my, my job, uh, when I was at Google and the previous company, uh, it was the Selenium farm, um, where you, the whole thing was dividing stack that you have a hundred virtual machines. You divide your tests across a hundred machines. They get done a minute, blah, blah, blah. Um, so I'm applying the same, uh, divide and conquer kind of thing to hardware. Like, well, if 3d printing is slow, the solution to that is obviously get a hundred 3d printers. Duh. Why would you, with injection molding, like you have to anyway. So, so there is a, there is kind of a tension of sorts of like, okay, yes, we're a two person company. I've got 3d printers, but there's, there is still a max on how many I could put together in a month. And so there's that dilemma when I wake up in the morning, like I could go sell more robots or I could go, make more robots. Like there's a certain cap on how many I can sell just based on, you know, my capacity. Right. So there's a, there's a tipping point I think where, and I, and that's still in the future, um, where it makes more sense to go for that super efficient injection molding, you know, outsourced every little part thing. Uh, but you know, it's kind of an adventure of sorts to kind of find there. Um, well,

Chris Gammell: it sounds like you're iterating your way there though too. Right. So you, you basically have this close, it seems like pretty close relationship with what consumers need, right. Or consume all your clients need. Right. And you keep tweaking the product to get towards that. Right. I assume at some point you're going to, you know, basically you're going to run out of robots in a month and be like, Oh, I have to do something here. Right. Not that it'll be a fun month, but I'm just saying that like, you're going to hit that product fit. That's going to be good enough that you're gonna be like, Oh, well now I just make a thousand of these. Right.

Jason Huggins: Yeah. And so I'm still kind of discovering that. Right. I mean, so like there's a classic kind of Silicon Valley ethos. So like fail fast. And I'm, I'm a fan of the other side of like fail slowly. Right. Like, you know, like have lots of, you know, have a lot of money. Control burn. Right. Right. Like, like, you know, don't live in Silicon Valley, live in Chicago. Like have a, you know, I was in my basement for a while. Then I finally got a, an office, but you know, like fail slowly. It means like, yeah, increase your burn rate, whatever, blah, blah, blah. So just, sorry, just rambling around, but there's like one, for example, the most recent tweak to the product. I thought it was amazing and it can go to any XYZ location and click any, any, you know, physical or virtual button on the phone. And the feedback from the customer who had that version was like, you know, it's great and it can do the thing, but I want it to tap faster. Right. Speed is always a feature. And anything and including robots that tap buttons. Right. And so I had a, it was like a lead screw. It can kind of go up and down. I could still, you know, double tap, whatever. Okay. So like, okay, I want to solve this problem forever. So I changed the fancy word for the, where the hand, where the finger is the end effector on a robot. I swapped that out with a solenoid.

Chris Gammell: And so now I'm just the finger, the end, the tip of the finger extends instead.

Jason Huggins: So actually there's now two Z axis movements, basically like there's the slow, there's the, still the stepper that's driving a lead screw that can kind of go up and down and get to the side buttons or whatever. Kind of like slow. Oh, I see. The slow twitch muscle.

Chris Gammell: And this is for people that are thinking about like a, like the Z axis is the slowest axis on a 3d printer as well. You just have to do it.

Jason Huggins: So, so when you have to get to somewhere, uh, I do, I still have that access, but when I just, when I'm hovering over the screen and now I just need to like tap as fast as possible, I have a solenoid right there with a little, you know, conductive tip there. So can I've clocked it now. I, I changed the little, made some software tweaks. I, I'm now up to, uh, 800 taps per minute. So, so that was, yes. But to your point, like that was customer feedback. And because I didn't have, uh, you know, uh, a factory in a foreign country and a massive volume, uh, that I, you know, of inventory that had to sell down or whatever, like I could just instantly change the product, you know, take a, you know, make a tweak and then send them a YouTube video. Um, right.

Chris Gammell: And this is all super niche still, right? I mean, people know they need this. The ones who know they need it. Like you said, they had secret robot labs, but then they had to start from scratch. Now you're basically saying, I'm in your domain, at least let's talk about what you actually need. Because like you said, I think you said the, this, some of them are touching the buttons on the side versus the actual screen. That's a totally different action than, you know, tapping a screen or different size products or, you know, like you said, cars, cars don't have, you can't put a, you can't put a robot flat on its, you know, like on the ground, you have to mount it to the actual touch screen or, or inside the car or whatever. Right. So like all of these things are different constraints.

Jason Huggins: Right. I, I mean, so there is a certain, uh, what do you want to call it? Envy of sorts. Like it would be, it would be cooler if I was working on something that had the scale of like the Arduino project. Uh, cause like that, who wouldn't want to be the folks in the Arduino project or whatever, like where, um, whatever. I, I, I wish I was working on some bigger, more higher impact thing, or I don't know. It sounds all just annoying. Um, but I mean, there's a certain level of though, of like, I have to kind of focus on like, you know, what do I want out of this?

Chris Gammell: Business at a certain point, right?

Jason Huggins: I mean, at some, at some point, like, okay, fine. I've never been on the cover of make magazine. That might never happen. Although my robot was in popular science. So I do have that, but Hey, so there's, so that basically that, that envy and jealousy will kill you. Right. But like, I have to keep focusing on like what I want. And the thing is, is like, I've got all these art projects and I kind of have to get comfortable with the fact that like I might create them and no one will know they exist and it'll, you know, it'll get no likes on Instagram or tweet or Twitter or Facebook or whatever, but I will know, but I will know at the end of the day, I may, I wanted to make something and I took me 20 years, but I finally learned how to make it with electronics and software and mechanical stuff. But, but I, I set out to do a thing and then I did it. Right. Um, and so that's like the thing that like, that I really want to get to and whether, you know, um, you know, yeah, however it works. Like I, I'm, I'm lucky that I found like I could leverage my software testing career into this hardware testing thing. Um, and I also kind of look at it as people pay for it. I mean, that's the main, that's the main thing that enables all this. That's the, that's the key thing. So I'm also kind of like looking at it from a, uh, you know, all these lessons I learned from, uh, you know, Silicon Valley, like I'm basically doing the opposite. Right. So instead of like go big or go home, um, I went home. I worked out of my, I, I, I, I worked out of my house for a while. Um, or, or there's this thing of like, you know, it's in grand slams or, or die. It's like, I would rather, you know, you can win games on doubles and triples. Um, that's right. Right. So there's a certain level of like,

Chris Gammell: Oh boy. Right. That's a tough thing in Chicago, man. Sorry. Um,

Jason Huggins: so there's, there's this feeling of like, I'm instead of taking a venture capital, um, right.

Chris Gammell: I wanted to come to this. So tell me about your venture capital world because you, you were in Silicon Valley. You've seen how it goes. Your first company did take VC or no?

Jason Huggins: Uh, it took a, took a lot. Um, okay. I think the last, the last round I think was like 75 million bucks or something like that. Holy balls.

Chris Gammell: Okay. So, so what about, what about tiny robot company then? So, right. So, so,

Jason Huggins: um, uh, so, so I, I did, um, disclosure. I did take, uh, a seed. What do you want to call it? Precede, whatever the words are called. A little bit of money. Uh, Bryce Roberts at OATV started this thing called Indie.VC. And it's kind of weird paradox thing where a VC is basically saying like, screw VCs and whatever. He was like the one VC.

Chris Gammell: I like, no, I like Bryce. And I, I like, I like this model too. It's basically cashflow is more important than big sell at the end. Right. That's kind of the short pitch.

Jason Huggins: The Indie.VC program is like the anti-accelerator, right? Instead of focusing on a demo day and working on your pitch, it's all about profitability and talking to customers.

Chris Gammell: 20 year, 20 year timeline kind of stuff. Right.

Jason Huggins: So, so this was, uh, the, uh, January of 2015 is when he launched this and I was already going to, um, incorporate Tapster. I think right around that same month, I got a, an order for 10 robots from a very large car company. And I'm like, wow. And at that point, this was just a hobby. And I really needed it. Now at this point, treat it as seriously as this car company was. Cause like, if nothing else, you know, limited liability is a thing. Um, I'm sure.

Chris Gammell: Yeah. Anyway, so I mentioned that you're also now playing in the hardware world where another difference I've learned between software and hardware world is like, Hey guys, don't forget about the cashflow. You got to buy a bunch of shit, you know, like you need to have stuff to build stuff.

Jason Huggins: That's another benefit of having low volume, high margin kind of stuff. Cause I don't have to, I don't have like half, half of my cash just sitting on the warehouse space. Cause it's your office, right? Right. Basement. Right. So I was, uh, I had, uh, I got so many, uh, it's not the amp two hours. Sorry, but, um, the, um, we're cool,

Chris Gammell: man. We got time.

Jason Huggins: Yeah. So I've, so I, Bryce was basically the only VC that I kept in touch with, uh, when I was doing all the pitching, all the, all those years for, uh, for, for sauce labs. Um, and I remember one conversation, this was before he started India. I was like, you know, why is it that investors are always going for grand slams and, you know, the, the, the go bigger, you know, or, or die? Like, why can't you, you know, like dividends? Like, why is that such a dirty word or something like that? Right. Like, or like cash just, you know, just like, you know, why can't you just focus on what your earnings are? Yeah. Why can't you just focus on profits, right? Like, and if nothing else, like if you're profitable next time you go for a fundraising round, because you're profitable, you don't have to take any horrible deal. Like you could walk away. Um, if you're, if you're doing that huge scale thing, you're always at the risk of going out of business every six months. Right. So, um, anyway, but the weird thing,

Chris Gammell: you're in good, good, uh, good company here as well. I mean, we talk about that stuff, self-funding, all that. Yes.

Jason Huggins: So in that conversation with him, he was agreeing with me and I was like, I was there for like an argument, like the old school, like Bonnie Python, like, no, I'm here to argue with you. Why are you agreeing with me? Like, I wanted you to articulate the VC point of view. He was like, no, no, no. Like they, he was frustrated as well, uh, from the VC side. I was like, that was amazing. So when he actually launched indie.vc, uh, when I first saw it, like, oh, this is exactly the thing that I remember talking about with Bryce. So I was at midway through writing an email saying, Bryce, did you check this out? I scrolled the bottom. It said like Bryce Roberts, like, oh, you punk. Anyways. So, um, so he, so I applied almost as kind of like a, just like a plus one, you know, thumbs up to like, this is kind of the future of juice,

Chris Gammell: juice, juice stats and be like, oh, you got 201 applications that is 200.

Jason Huggins: Right. So, right. So, so I ended up, uh, you know, they, but out of a couple hundred, uh, applications, they did pick, uh, Tapser. Um, and so we're one of the eight, uh, founding, whatever you want to call it, the first batch of sorts of the, of the companies. Um, and, um, anyway, uh, so,

Chris Gammell: so how does that, how does that impact you though? Like, so you said you didn't take much VC, but so that was, it was, so the only funding,

Jason Huggins: so that was like a hundred thousand dollar convertible note. Um, and it's also interesting cause you can actually kind of quote unquote pay it back. I'm not using financial technical terms here, but, uh, it's, it's,

Chris Gammell: you can buy your way out of having meetings with VC in the future. Right. Right. Right. So in the sense of, and,

Jason Huggins: and more importantly, the, the ethical, like the, the unwritten promise when you take any funding is like, you're going to go raise a bigger round next year. Right. Um, like you're going to go, you know, you're going to try to get this sold or IPO in five years or less. Right. Um, with this, it's like, no focus on making a profitable company. Like, so it's, it's much of like the implicit promise of, of, you know, the strings attached with that money as much as the actual money. Um,

Chris Gammell: more like grandpa giving you a loan rather than, uh,

Jason Huggins: I mean, a well-connected grandpa, right? I guess like,

Chris Gammell: of course, grandpa's loaded, you know, right. I mean, so the thing, the thing I love about it and he knows about business, you know, you're still going to ask him business questions.

Jason Huggins: I guess the thing I love about OATV and Bryce and, you know, out of all the VC firms, but one of the things like, you can't take this away from them. They funded Chumby.

Speaker ?: Yeah.

Chris Gammell: And Chumby is company. If people don't know, yeah, go,

Jason Huggins: go Google it. Granted, it was obviously dead in the water. The second, the iPhone and the Android came out, it was like, obviously like, Oh, they're dead.

Chris Gammell: I don't know that Amazon, Amazon basically made the Chumby. That's ironically. Yeah. It's, it's coming back, but it was,

Jason Huggins: it was such this, this wholesome, good electronics kind of like, actually, I think even one of the first, I can't remember if their circuit boards were open, if there were like open hardware before it was cool.

Chris Gammell: It was eventually open. I don't think it was open when they launched, but I think, I think bunny was allowed to open source them at the end.

Jason Huggins: Sorry. I'm getting my facts wrong here, but, but dear or fuzzy, but I just, I loved them. And then therefore like,

Chris Gammell: I saw one on a desk at the coworking space. I'm at the other day. I was so excited.

Jason Huggins: So there's a certain level of like that blends between like, like any, any kind of piece of hardware that is not just like a boring piece of industrial equipment. Like it's something that puts you a smile on your face. Like I, I respect people who kind of go for that, that extra step. Um, I don't think the Amazon chubby, uh, modern day remake is, you know, is that smile inducing, I guess. Um, got it. I guess you get your smile. Goofy. Right. Right. Right. Anyway. So, so, so they, they funds, uh, uh, hardware companies, if nothing else, and the kind of hardware companies that are quirky that I kind of like. Right. Um, so whatever. Um, but the nice thing is, is that, uh, you know, that was 2015. There's been, uh, zero conversations or pressure or whatever to ever, uh, do any, or more fundraising beyond that. I think the way I look at it, my tagline for that is like, it's just enough VC BS. Um, every quarter they email me and ask for the financials, um, which means I have to have financials. Like I have an accountant, I have a bookkeeper. Like we, we have profit, you know, cash flow statements, all, all that stuff. It's nice having kind of a, what do you call it? Accountability buddy kind of a thing of like.

Chris Gammell: Accountability buddy. Yeah.

Jason Huggins: Right. Like that, you know, it, it,

Jason Huggins: it makes things, like seem professional because they are. So when I try to go higher, like, it's like, Hey, I do have an investor. They, they are like, um, you know, they are interested in making this thing go through. But the nice thing is, is that we don't have all the stuff that you don't need, like, um, fancy monthly board meetings, uh, in fancy expensive places with catered lunches and whatever. Uh, and all of the drama that goes with like just perfecting your pitch deck to the expense of actually making the product or all, all of this, you know, stuff like the other stuff. Yep. All the other stuff. It's like, no, I like my, like I obsessed with the product and talking to customers and, and so does, uh, Bryce. And so that's nice. Um, anyway, I could ramble on about a million topics. No, no, no, this is good. So,

Chris Gammell: so ultimately though, what did, what did this impact of? So basically you had to focus on cashflow of actually making a profit, like basically getting paid, right? I mean, like you were making a, I mean, so what, what is, I mean, do you have a retail price on your, you don't have a, a site that sells just robots anymore, right? You quote jobs now, but.

Jason Huggins: Right. Well, so I actually, I did have pricing up. Uh, I just recently took it down. Tapster.io. If you want to go check it out. Um, so hip.io. Dot.io. Um, also ironically, Phillips electronics owns tapster.com because they had a, uh, pair of headphones that was out like 10 years ago. Obviously, uh, it didn't do well. No one. Anyway, but so they own the, I think they own tapster.com. I'm going to someday I'll have to figure out how to get that from them. Um, whatever. Uh, Oh yeah, that's right. That's the downside. Um, anyway, but there is, uh, when I, my very first angry birds robot was like a hundred dollars and it actually costs more than that to make. Oh yeah, totally. So I was losing money every time I went to the post office. Um, since then I've learned to raise my prices. Um, now the robot is hovering around like 5,000 bucks, which is like, what? Uh, a bit ridiculous, but at the same time,

Chris Gammell: I bet they gladly pay that at like you, you are solving an issue, right? You were, you were basically replacing, well, you're replacing a person, right? I mean, you're replacing a finger, right? So there is a finger. I can get you a finger. That should be, that should be your company slogan, right? Like, like Walter from a big Lebowski.

Jason Huggins: There's a bunch of, there's a bunch of puns out there. I've got kind of a text file of, of various, uh, versus this, you know, we're giving the finger to, right. There are, there are other companies that are specifically doing quality control, robotic finger testing things. And, uh, one is based in Japan. Uh, and I think their retail price is like 50 to $60,000. No, 50,000 just for like old, like nine button Nokia phones. I think it's like the touchscreen version is like 75,000. There's another company. Uh, and actually they were, uh, just acquired by a company in China, um, a couple of months ago, but there is a company in Finland, I think. Um, and their, their robot sells for like a hundred thousand dollars. Um, and that's just before the software. Once you add the software, then it's even more, right. Um, and so there's also a certain level of like, I'm coming in classic kind of disruptive, whatever, maybe this is the part of the VC or the Silicon Valley. It's showing. Yeah.

Chris Gammell: Your slip is showing,

Jason Huggins: man.

Chris Gammell: Yeah. Right, right, right. So, uh, come in, come in 10 X,

Jason Huggins: it can be 10 X better or 10 X cheaper. So I guess I went with the cheaper part. Hopefully it'll be better because it is, you know, a very high quality testing robot. Um, other, uh, kind of footnote to history. Um, this, uh, thing I like to point out is that a lot of people look at HP, uh, and specifically, you know, the whole idea of a garage startup, it started with HP and Palo Alto, uh, the specific, their first product, um, 200 day, 200 was a piece of testing equipment. Um, so there's a certain level of like, uh, you know, uh, I, I'm more, I'm more hipster Silicon Valley than anybody.

Speaker ?: I mean,

Chris Gammell: you're not going to find listeners to this show. We don't think HP and think test equipment. It may be calculators as well, but I mean, amp hour listeners know that HP is truly a test equipment company. And then it got bastardized and started making computers and whatever the hell they are these days.

Jason Huggins: If you want to, if you want to make, um, if you want to make,

Speaker ?: uh,

Jason Huggins: if you want to start a company, you know, uh, testing is not a bad way to do it. Sure. Yeah. That's a good point.

Chris Gammell: So the, the pricing that you do though, so you, do you actually do the, uh, uh, the, uh, the kind of application engineering as well? Do you, how much do you work with a customer? To actually get a robot implemented in their system.

Jason Huggins: Right. So, so this is kind of a repeat of sorts with, um, with the, my previous project, like just running the infrastructure is such a big enough of a job that, um, uh, I don't write the tests for them. I will get to them to the point where they know how to do it. Maybe I'll give them like kind of a hello world test. Um, but, uh, I won't do that for them, but, but it depends, right? If, if we put together a package, um, then maybe I'll write a couple of tests for them.

Chris Gammell: Right. Right. Cause I mean, so this is interesting too, because, you know, we've talked on the show before about like productizing services and making services out of products and all that stuff. Right. So like the classic thing is for products. If you want to make it a service and you add like a SAS component, or if you want to, if you're a consultant, you want to make it whatever you make it into a product. Right. And so that's like what Dave does with testing gear, stuff like that. So this is kind of in that weird middle ground though. Right. So especially if you keep changing the robot, I assume there's probably not, there's documentation, but it's not like, you know, full product documentation. Right.

Jason Huggins: Right. I mean, so,

Chris Gammell: so yeah, who are you interfacing with as well? I guess it's the, that's the question that I'm really getting at.

Jason Huggins: Um, this gets kind of really wonky, I guess, but, uh, I'm taking a lot of lessons learned from the previous experience doing test automation. I'm kind of seeing some patterns either whether I'm seeking them out or I'm kind of falling into it cause I can kind of pick up on them. But what I have noticed is that there's like a typical kind of archetype for a person who's on a project. And usually it's not the end line engineer writing the test. It's their boss. And, uh, you know, a good manager, uh, if you're heard, I don't know if you've heard the, um, the umbrella theory of management. Um, a good, a good manager can kind of basically hold an umbrella over their workers and kind of keep all of the, uh, the gunk of the world from falling on top of the engineer. I've had, I've had that boss, but I've never heard that. You can, you can hear it. You can use a different, you know, four letter word than gunk, but at a certain level of life, it's the gunk umbrella, uh, right? Um, you either let it flow onto your engineers, flow right down onto them, or, you know, kind of protect them. The best engineers, and these are the kind of the best managers, but they're manager engineers of sorts that they can hear their engineers coming up with like mentioning their, they have these quote unquote untestable or impossible to test scenarios. Like you're either testing, test engineers, or they're working on the product and they're pointing out, they're like, Hey, um, there's a couple of scenarios. And like one specifically like, Oh, when, when our little device disconnects, uh, you know, Bluetooth and then it loses connection to what iPhone, um, we don't have an automated test for that. And we don't have any automated test rig for that. Right. Um, and so I think we need a robot, uh, because there's no other way to do it. And so that robot or a really smart dog. And so either, either that engineer will go and do a day of Googling and find me and then task their manager, or like, Hey, go talk to that guy and see if he's like, uh, legit or, or the manager. It's, uh, we'll go on the quest, uh, themselves and go start finding out if it's possible. Uh, and then when they start searching, like testing robot button, clicking, whatever, um, they find me. Um,

Speaker ?: they find me.

Jason Huggins: And so I, I, I kind of, it usually gets down to like, let's say out of like a hundred test cases for whatever random thing you're building. Uh, I think the most interesting ones, at least lately are like, they have a Bluetooth device and an iPhone app and they're trying to, you know, how do those two things talk to each other and automate, you know, when wifi drops or Bluetooth, whatever. Um, uh, those test cases that were, they need a robot, uh, are interesting. Right. And so we'll kind of like explore that space, if you will, from a conversation and then say like, well, actually, well, you know, you could, you could totally just use the iPhone simulator for that. No, but this other one, yeah, you totally need a robot. Let's, let's talk about that. Right. And then sometimes they'll say like, Oh, but like we're testing on an iPad or something and like, okay, well I'll make, I'll make a bigger version of the robot for that or whatever. Um, or multi-touch or whatever. Anyway, so it becomes kind of like, you kind of explore kind of like what they actually need. Um, and, um, kind of go down there, but, but then I deliver the thing and then, uh, there's probably some, you know, I actually, I guess my, my customer, they're, they're more likely they are, uh, software engineers. Um,

Chris Gammell: so that's like an API access kind of thing or what does that look like?

Jason Huggins: Uh, so the dream, uh, coming from the Selenium project, or if people are familiar with that, uh, it has this protocol, the web driver protocol, like the dream actually is to, um, if the dream for the Selenium project was, you could write one test and it could automate any web browser. Uh, the idea would, would be for my robot to kind of be sitting there and it would, uh, it would speak the web driver protocol. So, so really you're talking to it over like a Jason plus HTTP interface. Uh, and then, and then at that point it then tells a robot to go to a location. So, so I guess maybe web driver is my robot operating system. Um, as far as like my level of abstraction,

Chris Gammell: but it's still pretty low level of like go to position, you know, 42 comma,

Jason Huggins: right. Well, six kind of thing. Well, it depends on what level of abstraction, right? So at a very basic level, it's like playing a game in battleship, right? Like you go to XYZ, whatever. Um, the one that's what they want though.

Chris Gammell: That's what I, that's what I'm really asking. Yeah. Yeah. That's, that's,

Jason Huggins: that's the minimum they need, uh, to get going. Uh, one level above that is, um, computer vision integration, which gets kind of awesome pretty quickly, or at least kind of fun to hack on. Um, where I use the open CV library, I'm using either a video feed off the device or a camera. And then I'm using, uh, open computer vision magic to find elements on the screen, converting that to an element and then clicking that. Um, Oh wow. Okay. Uh, super fun stuff. Yeah. That's super cool. And doing it all in JavaScript.

Chris Gammell: She's kind of working on that too for, uh, so from embedded FM, she's doing that for, she's doing computer vision for typing on keyboards, I think. Oh yeah. Cool. Yeah. So I might put you guys together at some point. Yeah. Uh, okay. That's, that's cool. So that, but that is the interface is basically, you're basically, so I've talked to Mike Osman about this too. I said to him at Defcon, I said, he's basically making software interfaces for the RF spectrum, right? That's what he's basically making. You're basically making software interfaces for a finger.

Jason Huggins: So the way, the way I describe it, it's like, it's like the reverse Tron. Um, I'm taking like, the way I would describe actually Selenium all those years ago was like, it's like a virtual robot that sits inside your computer screen. And types feel a text, types, text in a fields or clicks buttons on a web browser. And, uh, so I think that's how I got into robotics, robotics, uh, partly was because like, I wanted to test this metaphor and like make a quote unquote, real, real world Selenium robot, uh, that can type text on real keyboards and click real buttons and things like that. So there's a, there's a certain feeling of, um, uh, there's automation in the software space is kind of like a known thing. Like anything you see on your computer screen, you, um, you can write some automation script for that. Um, whatever. Um, ironically, actually kind of little things that I noticed that no one ever notices. If you have an OS 10 machine, uh, there's a program called automator, and it's kind of like a macro script writing tool that the logo for making an automator script is a little robotic, you know, hand finger thing. So that's kind of funny. There, there's this other, other someone somewhere at Apple also saw this relationship between like virtual robots on your OS on your computer screen and real robots, clicking buttons, whatever. Um, yeah, but it's, it's, uh, the way I look at it is like, I'm, I'm taking all these lessons learned from software automation and applying that to the real world. Um, and, and if you boil it down, like web driver is a very complicated, but ultimately that like there's two classes of, uh, API methods. There's like finding things and then doing things to the thing you found. And if you think about robotics,

Chris Gammell: that's how I live my life, man. That's pretty great. And I go find a sandwich and then I do a thing to the sandwich, which is eat it.

Jason Huggins: I guess that's three things, right? That gets very philosophical. Okay. I want a sandwich. Okay. Now I have to go find the sandwich and then you,

Chris Gammell: someone needs a want robot. That's what, that's the,

Jason Huggins: and then you put that sandwich somewhere. Right. So I haven't figured out, you know, I can't make more people want more robots, but once you, once you have a robot or the classic thing of like, you know, the beer in the fridge, right? So you have to, um, you have to navigate to the fridge and navigate back and then also find, uh, the, uh, the beer in the fridge, whatever. The thing I've noticed that like, we'll probably never actually have like, it'll be very impractical to have beer fetching robots that look like humans, because like the most efficient, cheaper version that you're going to go buy at Best Buy or Home Depot someday. It's good. You're going to have something that looks more like the beginning scenes of like back to the future where it's just, or it's like a vending machine. Like it just, Oh yeah. It's, there's no robot. It's just a lot of mechanical industrial launcher almost.

Chris Gammell: Right. Yeah.

Jason Huggins: Yeah. Or, or a drone will just, you know, fly over your desk and just drop it or something like that. It's not going to be an articulated arm.

Chris Gammell: Right.

Jason Huggins: Or it's just a fridge with wheels. Or, well, actually it might be the Boston dynamics.

Chris Gammell: Uh,

Jason Huggins: they're creating their,

Chris Gammell: their latest version. I was going to say, that's not a, that's not a practical robot for just about anything. The best, the best application I've seen is the, uh, is the, the DARPA stuff where they, they go and try and like rescue like danger situations. Right. Right. Right. That they're doing as a test, but the robot is truly doing a human replacement thing. So,

Jason Huggins: so side notes to side notes. Uh, the one thing I found fascinating about, uh, like basically there's like four countries. If you're, if you're going to plop down on the world anywhere, um, you want to find your way to one of, you know, Japan, Germany, us, or China, if you want to do robotics. Uh, and the interesting thing culturally in Japan for years, which has now changed because of Fukushima, uh, the Fukushima nuclear accident, the Japan, uh, Japanese robotics, mostly focused on, uh, like protocol and like humanoid robots. Like, you know, the more often YouTube video you'd find of a Japanese robot would be something that could like, SMO, like it could like dance or walk or, or wave. Hello. So the kind of culturally embarrassing thing after Fukushima was that none of those robots could go into a nuclear reactor and, uh, you know, find the missing uranium rods, right? Like they were all like,

Chris Gammell: they could, but could any robot do that in any country? Uh,

Jason Huggins: no,

Chris Gammell: I mean like the, the Atlas robot, which is the one we're talking about with the DARPA stuff, right? Like, right. That's the idea, but right. Right. Right.

Jason Huggins: I think culturally though, I think, uh, uh, American companies were more focused on those kind of industrial, uh, purposes. I think I have to go. I see. I have to find my, so there's a kind of a cultural difference of like, um, uh, practical robots versus kind of entertainment robots. And I, and I think now that those, those cultural differences are, are changing. I think like specifically because of Fukushima, there's now more Japanese companies, uh, robotics companies specifically focusing on those industrial, you know, get in and like open a door.

Chris Gammell: Right. Well, maybe there's money for it, right? Like the same thing of testing. Right. Now that that thing has happened, that has driven a new, a new class of need. Right.

Jason Huggins: But on the other end, you have also, you know, American companies doing more entertainment robots, right? You've got Sphero and a bunch of other,

Chris Gammell: what's, what's an entertainment robot. I guess I didn't understand that.

Jason Huggins: Well, I guess it's something that just doesn't solve a problem other than maybe putting a smile on your face. Um, anyway.

Chris Gammell: Yeah. Um, yeah. There are other countries we should say before you write in, we know there's other countries that are doing robotics. Please don't say we don't care.

Jason Huggins: Yeah. I'm, I'm, um, how do I say this? I, I can get in trouble here. I, I, sometimes I take an argument, I, I, I push it to an extreme and then, then obviously once you get into the, the gunk of any particular point or argument, it's, it's all, it depends. It's all, there's never a black and white line. Um, what I did notice though, at least within like there's usual suspects for robotics companies. Um, for a long time it was, it was mostly, um, Germany, us, Japan, China is coming up there, but I've, I've shipped my robots in lots of different places. And the other thing is I'm a huge fan of open source. Like that's a really key part of what I'm doing now, open source software and open hardware. And, uh, one of the key reasons. So yeah, for the, for a person who's not living in Japan, Germany, China, uh, or the U S. Um, part of the reason why I wanted to keep my designs as 3d printable as possible is so people could download the plans and make it locally. Uh, I totally drank all the Kool-Aid when it came, when people were talking about the Nirvana of, you know, 3d printing and the next wave of industrial revolution, all this stuff, like I thought it was being amazing. And I have seen the frustration of, uh, making something in one country and shipping it to another place. Like I shipped a couple of robots to Brazil and getting, Oh man, tariffs like customs in Brazil is, uh, uh, incredibly difficult is to be charitable. Uh, and so it would be easier if someone was there with a 3d printer, they could just download the plans from GitHub and just make it locally. So there's a certain feeling of like, um, I, I, I'm, I am, uh, I do not want to think that only robots can come from four countries. Like I would love it if, uh, they could be made everywhere and open source is kind of like a, a cool kind of infection vector for that.

Chris Gammell: I think I was watching Lamar and Phil talk on their, on their, their, uh, weekly show. I think they were talking about Adafruit or not Adafruit, sorry. Uh, Arduino opened a factory down there just for that purpose. There's like Brazil only, uh, Arduinos now because of the tariffs and the, the hassle to get down there. Right. Yeah.

Jason Huggins: I'd heard about it and then I actually experienced it firsthand. Right. Right. You're like, Oh yeah,

Chris Gammell: I'll just open a factory. That's way easier. Right. Uh, well, so what is the, the next stage of, I mean, like you said, I mean, the, the growth isn't like the main focus, but obviously you want to keep, you know, putting more automation and more testing into the world. So what is the next stage? When do you ever pull that trigger on injection molding or anything like that?

Jason Huggins: I don't know. I think my, my biggest, my biggest focus is, um, I haven't, uh, I've got those art projects that I keep saying the company is funding those things, but like some of those things are, they haven't gotten farther along. Um, so I'm trying to figure out which, anyway, so I'm mostly obsessed with the art side of the business. Uh, the, the, um, the business side, I think the, like the bigger problem I think is to, um, I need to get to position where I can hire and my, and I, uh, I specifically want to hire a software engineer. I, uh, ironically, um, you know, one, like, I think I, you can go down this whole rabbit hole here, but, uh, like, you know, do I try to find someone who like as a mechatronics person who can do everything, do the mechanical electronics and the software, um, that would be the perfect person. Um, but absent that it would be someone who can work on more of the software engineering sides of, of things.

Chris Gammell: And so really mechatronics, you also mean someone that software is also in that stack you're saying,

Jason Huggins: right? Right,

Chris Gammell: right.

Jason Huggins: I'm realizing that, um, I think I have, I mean, it's always kind of like a three legged stool, uh, of software electronics and mechanical stuff. Um, but I'm realizing I need someone who's really strong on the software, uh, for some of the stuff that I want to do. Like I, I haven't done, I want to do, I want to get more progress made on the, uh, on the computer vision, open CB stuff, right? Whatever. Um, so, uh, I haven't had to, how do I say this? Like, this is maybe a little too much in the weeds here, but, uh, the, um, I've improved the business model, if you will. Uh, in some of these categories, uh, with talking with customers there, it's not just the, it's not, as you were kind of touching on it, it's not just the hardware, it's the helping with the writing of the tests. Yep. So there's a certain level of like,

Chris Gammell: consulting on top of it or whatever, right? You're selling a package.

Jason Huggins: Yeah. It's like a support plan, kind of like the red hot metal, the red hat model where, uh, it's like a quarterly payment and like, you're, it's almost like a, like you're, you're paying for everything, not just the physical good, but like,

Chris Gammell: um, eventually you would have a training manager, you would have a, you know, a software consultant that could fly in if they needed that. Right, right.

Jason Huggins: So I'm putting together these packages where it's like, yeah, I'll, I'll do an onsite quarterly visit. I'll do a training workshop. All these things are just you right now, right, right. But there's a certain level of like that, that pay, like when you start talking in terms of fractions of a person's salary, you quickly get into tens of thousands or hundreds of thousands of dollars. And so at that point you can kind of kick the can again on not having to scale up to do, you know, thousands of robots because right now it's ironically, you know, robotics company, but like labor intensive, right? Um, cause it's like you have this robot that can click buttons on a phone, but you still need a skilled engineer to figure out how to consume that API and like make an automated test for it and whatever. Yeah. And plug it into their, uh, you know, when code gets checked in on GitHub and that whole build pipeline and that this kind of fits in at the end of that, like it's still a programming heavy, uh, jobs. So I'm kind of like, uh, as I think like this, the current snapshot of the company is, is focused on like kind of that, like that engineering services, if you will. Uh,

Chris Gammell: it almost sounds like machine building too. So like, so my, my, when I used to be on the engineering commons, Jeff, my co-host, uh, who's still doing it, uh, he was a machine builder, right? So he would go into factories and they'd be like, well, we need to, you know, put this sausage into this wrapping, right? Right. You know, like that kind of thing. And it basically would make that thing happen. Right. Or if people has watched, uh, Jeff Kaiser, the, the stuff that like that, uh, the valve hardware thing, like they have that at, that was the, uh, the facility out here where they're actually doing all these different test things on the steam controller. Right.

Jason Huggins: This is actually touches on something that I'm still puzzled with. Um, I, it's like, I haven't quite found my place in the world, uh, without not to philosophical, but like more is like product focus. Like if I go to a software conference and I show off my robot, I'm a little bit of a, like a circus, um, you know, a side show. Um, and granted almost intentionally. So like the first one, it was like specifically like, Hey, can I play a video game? Right. And so now, but there's this thing where like I've, I've realized that most iPhone or Android developers, they, they don't need the robot. Like you can do, they can do is a lot in a simulator. It's the people making these devices, uh, when they sit between the hardware and software world, they have an iPhone thing and a Bluetooth device. Um, and the weird thing is about those people, um, or even let's say the car companies, um, or device manufacturers, like, I guess I could go to a device manufacturer conference or a car company conference. But the other weird thing at the other end of the scale is like, um, the robotics conferences. Um, the people who are finding me and needing the robotic finger are not necessarily going to an industrial robotics conference. Specifically for that point, most robots are deployed on like an assembly line, uh,

Chris Gammell: in kind of a very super specific physical goods,

Jason Huggins: moving kind of industry. And like the, I would love to go and be a full fledged, you know, card carrying member of the robotics world and go to the robotics conferences. Um, but most of that, the, you know, um, is physical good stuff. So if I'm now actually making a robot that's plugged into kind of a software, um, development pipeline, um, the software people who are getting my stuff are not going to robotics conferences. And anyway, it's just like this weird thing where, uh,

Chris Gammell: I think points to a great need though, too. Right. I mean, that, that's the idea is that there, there aren't other people in this space, right? So people were going to those robotics conferences and they got sold big equipment and they went in their secret robot labs.

Jason Huggins: And then that's actually one of my customers. They were like, they bought the a hundred thousand dollar pick and place machine and was like duct taping a little iPad stylus to the end. And they were so glad they didn't have to pay a hundred thousand dollars. Granted, I realized like one, there's a market for my stuff and two, I should raise my prices. Uh, right.

Chris Gammell: But then also the scaling thing is the same thing, right? So this is the same thing you talked about with 3d printers. If they want to go from one tester to 10 testers, that's literally impossible for them with you, with a smaller thing, then that's totally possible.

Jason Huggins: That's a direct repeat of the lessons I learned with, uh, software. Granted, like there's this idea where it's like, you can make things more efficient, um, or you can just throw more hard. I was going to say,

Chris Gammell: you're like a three trick pony here. It's great. It's not a one trick pony, like a three trick pony, but it works. You know what I mean? Like this is great. This is, this is exactly right. Right. I'm a,

Jason Huggins: I'm a fan of the worst is better philosophy. So it's like, yes, you can make an efficient algorithm or you can just like buy more computers and just, you know, like computers are cheap. Right. So now the robots are expensive. No, that robots are cheap. You'll find robots doing more things. Yeah. Right. No,

Chris Gammell: I think I, uh, super tangent here, but like I, I look at like the ESP 8266 as like an enabling thing. Literally. I mean, yes, it's capable. Right. But like, it's all cost. Right. I mean, like just when you start having costs drop like that, you just never know what's going to happen. So if you, if you have to pick among those five, 10 times better, 10 times cheaper, I'd probably say cheaper if you can, you know, like, right.

Jason Huggins: Right. I mean, cause the thing is like everyone is embarrassed when the thing sucks. Right. So like you will, you will always be pressured to make the thing better. Um, and you will go there cause you want it to be better. You want to sell more and whatever, but it helps. Um,

Chris Gammell: well, I think from a, from a community perspective though too, right. So that if you have something that a hundred thousand people are going to can afford versus a thousand people, then you start playing those numbers of like, well, if a hundred people have it, then 10 are going to try and do anything with it. One person is actually going to like, you know, contribute to an open source project. Right. And then that just builds the software base from that. Same thing happened with raspberry PI. Same thing happened with ESPs, all that stuff. Right.

Jason Huggins: I think there's one other aspect of this, like the scale between a, you know, cheaper, better. I have to be careful. Like, you know, like, wow, he makes really crappy robots, right? There's a certain level of like, you can kind of, uh, you can put very quantifiable kind of bookends on your risk on what is, you know, considered cheap and not good. Like, uh, I used a certain kind of motor and one particular customer said, you know, what's the mean time between failure on that, on that thing. Right. And I'm like, I don't know. And like, well, you know, have you run an environmental test? Like put it in like a humidity chamber. And like, I'm like, I probably should do that, but I don't know. Like how, how about this?

Chris Gammell: I will give you this. You tell me what an environmental chamber is. And I, no, no, no, no. I mean, actually I, I, I,

Jason Huggins: on my to-do list is to go buy a, like one of those like wines, uh, seller, like a refrigerator.

Chris Gammell: They have them at M hub. You can just come over there. Okay.

Jason Huggins: Wait, my point is, uh, what I told them is like, okay, look, I don't know. Um, but let's see, I'm going to add like, you know, an infinite warranty on that part. If it dies, I'll replace it, um, immediately and for free. And I'll even drive to you and replace it. Right. And you, and you don't know how good or bad these parts are. You can kind of just say like, Hey, look, it's this price. And if anything, literally, if any part on that breaks, I will replace it for free. Um, and then of course you get that feedback and hopefully that next version, you don't have to do it. Like in the sense of like, I'm willing to like, uh, learn that lesson, uh, of like, Oh, I should have picked a better part. I'll willing to learn that, you know, lesson. Uh, yeah. With dollars. Yeah. Yeah. With dollars. Yeah.

Jason Huggins: Like one, well, one lesson is like, yeah, if I, if, if someone in Brazil wants to buy one of my robots, I will find another way to get it to them other than the previous way. Um, don't ever use the postal service to,

Chris Gammell: I was going to say, you, you go down, you, you suddenly start taking trips down for, uh, uh, for carnival. Well,

Jason Huggins: you get a DHL account and you, you know, become best friends with your account manager. You do not. Oh yeah. Anyway, anyway. So, so there's, there's ways that you can mitigate the risks of, uh, being that quote unquote 10 X cheaper. Um, also I think the other thing I've lucked out with is like going into robots. I'm starting out with something the size of a shoe box. Like if I had to have something that was the size of an SUV robot. Yeah. I mean that would, I would have to bring in all this other stuff of like, you know, what's what, you know, um, how would I get it there? I would have to come, you know, have to think about, right.

Chris Gammell: They're making crates and yeah.

Jason Huggins: And then the safety cages and, you know, like my robots right now, at least, uh, are incapable of killing people. Um, as I mean, I haven't tested this, but it's all like, you know, five volts Arduino stuff. And I, I don't, I, maybe I don't want to know this, but like, I don't know if an Arduino has killed anybody yet. I hope not. I hope I'm not the first, um, but five volt electronics seems to be kind of safe. So starting in kind of a safe spot. So even if it's a, even if it breaks, um, you know,

Chris Gammell: we all know that your robots can be dangerous if it starts wielding a knife as an end effector, right? That's, that's the real,

Jason Huggins: I, I mean, I do have this theory of like, you know, the end goal for, for testing. You have two extreme scenarios. Uh, one, it's not just a robotic finger at the size of a shoebox. It's something that really to get, to make a robot that can test anything the way a human would, you really actually need a terminator. Um, I mean, like it, like it, it walks and it can make a cup of coffee or open a door just like any human, right? To the point where we came.

Chris Gammell: You started, you started in 2000, you have a tapster. So when you eventually name your robot, the T one, T T, Oh, T 1000, T, sorry. Right.

Jason Huggins: It's even funny. T could have been for tapster, right? Who knows? But I like to say that, you know, um, I'm probably not the inventor of Skynet cause no one's come back from the future and to try to take me out. Um, right. Yeah. Yeah. Yeah. So anyway,

Chris Gammell: you only get one shot, you know, right.

Jason Huggins: So where are we here? Uh, well, you got, you got terminator robots really are actually, they were invented for testing. If you go into Google, probably the backstory on Teradyne or whatever it was, it was probably, uh, or what was it, Scott? What was the company that might've been Teradyne?

Chris Gammell: I don't remember. Yeah.

Jason Huggins: Anyway, I think that's actually a real company. Cyberdyne. Teradyne is the, is actually ironically a testing company that acquired universal robots. Um, they are an actual company. I think it was Cyberdyne systems is. Yeah.

Chris Gammell: Okay.

Jason Huggins: Right. So if we, if we change the name of Tapser to Cyberdyne, then, then you should raise it.

Chris Gammell: Okay.

Jason Huggins: Anyway.

Chris Gammell: Um, well, I guess we, and a terminator, right? We use the terminator.

Jason Huggins: The other end of the spectrum is the, is the matrix. It's a pure simulator. You can't even tell that you're in the simulator. So it's either that it's either the matrix or the terminator are like the two end goals for, um, for robotic testing.

Chris Gammell: I guess the two questions I always ask at the end are, uh, who would you like to contact you? Usually we say that in terms of jobs. So if you are hiring someone, you want to hire a software engineer, but who do you want to hear from? If you, uh, you know, someone's listening to this.

Jason Huggins: If I could hear from anybody, I would love to talk to the head of testing at Tesla or SpaceX. Uh, my, my observation of the universe is like everything, literally everything that gets manufactured has to be tested. Um, so instead of testing, you know, washers and widgets, like why, why am I not testing, uh, sports cars and spaceships, right?

Chris Gammell: So you don't want to sell them a robot so much as hear about their testing methods or what?

Jason Huggins: I think I do. Like I am fascinated by like these impossible to automate scenarios. And often when you're inventing some new thing, um, uh, it's kind of hard. Like if you just like, or, or, uh, talking with like, uh, the Oculus Rift team, how they tested that. If you say some of their, um, little Kickstarter videos, they've got an Oculus mounted to the end of a universal robots robot. Like, why are they doing that? It's probably for testing. And I bet those engineers are like trying to, they're just figuring it out, right? Like there's no.

Chris Gammell: And that's the thing. Like I, I think so many of these companies, they have test groups, but first off test, people are just usually so flat out, you know, busy that they don't have a chance. And then no one's allowed to, especially in the hardware world, no one's allowed to talk about it because this is like, this is the magic stuff, right? This is the stuff where, you know, this is getting the product out the door. You also see all the warts when the things don't go out the door. And like,

Jason Huggins: yeah. So, um, I have had this exact conversation with the company that bought the 10 robots. Um, um, and I wagged my finger in their conference room, um, as I was going and delivering these 10 robots. And I said like, Hey, look, uh, Hey, it would be awesome if you could talk about this someday. Cause like, it's kind of weird being this guy doing all this stuff and I can't talk about it or any of who my customers are because it's all secretive. And they're like, well, we don't want that other car company to know about it. Yep. And I'm like, well, you know, it's probably not, uh, a trade secret. Oh,

Chris Gammell: you told me about this too. And then eventually in the software world, you said that started to come together, right? Right. Right.

Jason Huggins: So my, my lesson learned from the software world is like, you know, for automating browsers, uh, we got Mozilla, Microsoft, Apple, Google to all kind of standardize on the W3C standard for, for browser automation is, is suspiciously identical to the, you know, Selenium web driver protocol. So we've done that. Right. And it's not giving away a trade secret. Um,

Chris Gammell: right. And I think that's the main thing is that like, there's so much like secretive nature to it, but it's like, look, you're all doing the same thing. You're all going to benefit from this, but everybody thinks that it's like, Oh no, we're so special. It's just that no one wants to take that first step. So you need that third party to come and be like, Hey,

Jason Huggins: what if we did this? So, so, I mean, I think there is a lesson learned. I think as, as the quote, if the, you know, the eye rolling phrase, the software's eating the world, I think specifically the software culture, uh, hopefully the good parts and software development process processes, uh, there, there will be more openness. Um,

Chris Gammell: but that also is driven by money, right? I mean, like, that's also like, Hey, I'm going to save you money. If you all do the same thing, you save money.

Jason Huggins: Yeah. Yeah. I, there's, um, I, I collect weird things, but one of the things I collect is like weird robotic stories. And one of them was like, uh, T-Mobile and it was a robotics company. And they like sued one of their vendors because someone like broke into their secret robot lab and was like stealing, like how they set up the robot lab. Right. Um, it was just like a New York times story coming a couple of years ago. Uh, it's insane story, but there's a certain level. You could make the argument of like, how much money were you spending on the secret robot lab? We're really expensive robots. And all of this like secrecy when you know, uh, Samsung or, you know, and all these other companies are also doing the identical thing in secret and super paranoid. And really all the secrets that you're protecting are like, make a robot go to X, Y, Z location and click this button. Right. Like you could, you could just focus on the argument for open source is you could, you know, we all collaborate on the same, uh, button clicking routines. And you don't have, anyway, you can kind of shift those costs. Anyway. So, um, so that, that, that's whoever is, uh, you know, doing testing and secret robot labs. I'd love to talk to them. The other one though, is I would love to talk to, uh, you know, the people who worked on the inform, like that, that motorized pin art project at MIT, uh, or anyone like kind of working on that similar thing. Like,

Chris Gammell: so there's this kind of linear actuator people, uh,

Jason Huggins: cheap 3d printed, uh, you know, linear actually, like basically people who, um, I need to find my tribe, uh, or my, my people, um, um, uh, in the kind of the art world of sorts in the sense of like my, my tagline for the art that I want to do more of is art. You have to plug in, um, you know, the stuff that every, like, and doesn't just blink, come on, blinking is great. Like Mike Harrison does great. No, but this thought exercise of like, what's the modern art museum going to look like in a hundred years when they look back at now, um, and whatever. And I think there, because of all of the cheap components, thanks to the, you know, the, the global, um, uh, effect of all the miniaturization and, uh, cheapness of all these parts that go into phones, you can now buy 10,000, $50,000 or 50,000 LEDs for not that much money or little motors and all kinds of things and make these huge projects. Um, and,

Chris Gammell: it's just a new medium that's available kind of stuff.

Jason Huggins: Yeah. I mean, I think maybe a more for a hundred years though.

Chris Gammell: That's the main thing.

Jason Huggins: A more dusty word is like kinetic sculpture. Like I'm interested in like art that moves and interacts and all kinds of stuff. And, and I think there's, these people are kind of all like scattered around the world. A lot of people are in San Francisco. A lot of people are in New York, uh, in that scene of sorts, but I think, you know, there's this internet thing. Um, Chicago is not a shabby city, but like, there's a certain people, a certain thing of like, I would like to find, uh, more people who want to work on those kinds of things and, you know, do that stuff. And hopefully, you know, tapster can be an ongoing profitable entity that can let me, uh, go, um, work on that stuff.

Chris Gammell: Yeah.

Jason Huggins: Okay. How do people get ahold of you? All right. So I'm, I'm hugs on Twitter, H U G S. Uh, they can email me hugs at tapster.io. Uh, you can go to tapster.io and, um, view our amazing webpage. Uh,

Chris Gammell: I was going to say, you should add more pictures of robots. You know, people like pictures of robots.

Jason Huggins: I think that the best criticism is that, uh, there's no video. You should never trust, uh, just a picture of a robot. Like it should, there should be video, but you can do a YouTube search for tapster robot. Um, I'm trying to beef up the demos there. Um, specifically going down the path of trying to make more, uh, uh, music demos. Like, uh, my favorite one was I had two, uh, tapsters, uh, playing heart and soul on the iPhone garage band.

Chris Gammell: So I've seen that. Yep.

Jason Huggins: I need it. And I need to get those on the website. Um, whatever. But that's like, that's it. Who does that? Yeah. Oh, a lot of stuff. What about the, uh,

Chris Gammell: GitHub is also hugs?

Jason Huggins: Yes. Also. Yeah. So, um, uh, H-U-G-S on GitHub, but also a tapster bot is that, is the GitHub handle there. Uh, the latest version of the robot, the sidekick, I call it. Um, that is not up there yet. Um, any day now, uh, hopefully before the end of the year, I'll get that up on the, on the website. Cool.

Chris Gammell: All right. Well, uh, thanks for being on here, man. We covered a lot of, a lot of stuff.

Jason Huggins: It was almost the amp two hour. I don't know how you're going to edit it down here. Uh, I won't,

Chris Gammell: you know, I won't, uh, but we should also mention you were also talking more about the software side of things at another podcast. So if people haven't gotten enough of Jason, you can go over to developer on fire and listen to two 87. We'll link that too.

Jason Huggins: Thanks. And thank you to everyone who's, uh, listened this whole way. Uh, you're, you're, uh, you're amazing. Yeah.

Chris Gammell: Cool. Well, I will see you at the meetup on Tuesday. Hopefully. I hope, I hope so. Yeah. And if other people in Chicago, we'll hope we'll see you there too. Thanks to all our patrons and, uh, we'll talk to you soon. Bye. Bye.

Chris Gammell: Bye.