Episode 77 Transcribed === [00:00:00] Sean Tibor: Hello and welcome to teaching Python. This is episode 77, the power of bots in the computer science classroom. My name's Sean Tiber. I'm a coder who teach. [00:00:26] Kelly Schuster-Paredes: And my name's Kelly and I'm a teacher who coats. [00:00:30] Sean Tibor: And this week we're joined by Tom Lauer's from Birdbrain technologies and a fellow CMU alum like myself. So a bit of a, a Pittsburgh connection there. We're going to try not to geek out too much about, our favorite pizza places in squirrel hill and late night coffee shops and things like that. So, you know, Kelly can, can stay with us, but we're going to talk about. Robots today and how we can use bots in the classroom to empower and engage learners. Tom, welcome to the show. It's great to have. [00:00:59] Tom Lauwers: Yeah, thanks for having me guys. [00:01:01] Sean Tibor: We're excited to get going. And we're going to talk more about your role with Birdbrain technologies, the Finch robot, and some of the other things you've been working on and how everything got started, but we'll start in the same place we do every week with the win of the week. So something good that's happened inside or outside of the classroom. And because we like to make our guests squirm a little bit, Tom, we're going to make you go first with your one of the week. [00:01:21] Tom Lauwers: All right. So I think. My one of the week is the, it's more of a wind of the month, but it's the continued expansion of our Python with the Finch robot course. So, uh, this is something that we launched last month and we had hundreds teachers sign up for it immediately. It happened so fast that the emphasis foundation, which is funding the distribution of the. Uh, kicked in funding for 150 more. So we are continuing with that. So I think that's, that's more than just this week, but it's definitely a big win and it's been great. It's it's, going to be great to see everybody using those robots to learn Python. [00:02:01] Kelly Schuster-Paredes: That is awesome. Oh, I wonder if that has anything with this, you know, this whole push of, uh, what is it, the Tio B E index or something Python's the first time in 20 years has surpassed Java C. So, know, having bots that run Python, Sean and I are always like, Ooh, yay. [00:02:21] Sean Tibor: Exactly that definitely helps. Uh, people understand why we teach Python in the classroom because it's so versatile and you can use it for so many different things. I mean, sure. Maybe you could squeeze a little bit more performance out of something like that's closer to see or, or uses a framework that's optimized heavily, but Python has that optimization to go from. I have this idea about how I want to make my robot do something to it, actually doing it right. It doesn't happen. Done in nanoseconds. If you can out how to get it, get the code and get it working fast. [00:02:54] Tom Lauwers: I mean, it is the perfect language, to introduce students to, because not only is it, you know, compare it to something like Java or C easier to learn and has kind of cleaner since. But it is used all over industry and all over the world by professional software engineers. So it's not a toy language, right. It's something that everybody is using all the time. Um, so yeah, I, I, I am all on the Python bandwagon with. [00:03:22] Kelly Schuster-Paredes: That makes us very happy when we started this, three, four years, once it's been it's our fourth year now of teaching Python in the middle school. And I, I, was in shock that we're doing this. Everyone kept asking us. So, you know, it was hard not having. Um, manipulatives that use a lot of Python at that we knew of at the time. So it's, it's such a great move that there are more things out there. Like the French. [00:03:46] Sean Tibor: All right. So Kelly, would you like to go next? [00:03:48] Kelly Schuster-Paredes: Yes. I have a really fun win for myself. It was a personal goal. We embarked on. Um, I embarked on the AWS. Um, you Udacity. Scholarship. What was it? Um, machine learning scholarship course in June 25th and it ends October 11th and I finished my quiz. I'm not sure what I got on the score cause they don't tell you until October 20th. But I finished it. It was forever long course. It said it was only supposed to be 11 hours, but it took me two months to complete and I did it and I took the quiz. It was a 30 minute quiz and I had to take a machine learning questions and questions. And it was just a huge personal goal for me to actually. Go into a quiz and from AWS and go, oh, I pretty much know all this stuff. So, um, huge when I just to get it done, we'll see what happens if I'm one of the 425 out of what? 440,000 people that took the course, I'm not crossing my fingers, but just doing it was, was nice. So. [00:04:56] Sean Tibor: Well, and that's a huge accomplishment because, um, you know, when we first started working together three years ago, you were learning coding for the first time, not just Python, but how to code and to go to the point where you're now taking AWS courses and machine learning and deep learning is a huge, you know, a huge level of progress. So it's pretty awesome to see. [00:05:16] Kelly Schuster-Paredes: Yeah, I'm not sure I can write my own models yet, but I can use some, [00:05:21] Sean Tibor: And while you can wield them, right? like that's the most important thing. [00:05:24] Kelly Schuster-Paredes: something like that. Yeah. [00:05:26] Sean Tibor: Yeah, So for me, this is the big win. This week has been the return to soldering in the eye lab. So we've been doing a lot of, uh, work with our classes. Kelly's been working with her sixth grade classes. Our seventh grade classes have all been back in the eye lab. We had to take a hiatus last year from soldering. Primarily just because of proximity. We couldn't guarantee safety and, and social distancing and the iOS due to COVID. And this year we have some new protocols and some new ways that we can handle it. Some better understanding of how co COVID is transmitted. That allows us to be in closer proximity. And so our students are back in the, in our innovation lab, they're using soldering kits. They've been making a really cool kit that Kelly found that is a, an RGB led with three potential. Geometers hooked up to it and a battery. So it's all that. Right. Super simple circuit, great for building skills, but then when they're finished with it, they have this RGB led that they can. You know that they can light up and change colors on it. And we're using that as a lead in to teaching RGB color theory for other applications. So when they're using a circuit playground with a Neo pixel on it, we can talk to them about that red, green, blue mix, and doing that digitally versus doing an analog with the dials on their potential auditors. So it's such a, great natural fit. I think this is the kit we've found so far in. How it fits into our curriculum. most importantly, the kids love it. They love the soldering. They love the hands-on, the focus. It becomes like a meditative state for them. That they're just hyper-focused on, on soldering and getting that perfect solder joint. And it's so much, it's so much fun to do. And even when I'm sitting there trying to help them do solder stuff that they started in backwards, or, you know, with way too many solder, globs and everything, use it as a way to teach about making mistakes. And I keep telling them like, look, the only way I know how to fix this is because I've made this mistake too. [00:07:23] Kelly Schuster-Paredes: Yeah, it's a, it's such a, it was such a good find. Um, and they're not say locally, but the kids are made in California and they're a little bit more costlier than the mass produced ones we were using before. They are super quality. It was called the, um, learn to solider solder. We'll put it in the show notes, but the hue, octagonal board or something, it was just great. And just be able to show them the four prongs on the led and talk about that. And, uh, it was good. Fine. Good, fine. [00:07:57] Tom Lauwers: I mean, that sounds wonderful. That sounds like the kind of thing that's going to stick with those kids for a long time. They're probably not surprised. Unsurprisingly, I'm a big proponent of hands-on learning and tangibility, right. Even with coding. because I think that for [00:08:13] Sean Tibor: Okay. [00:08:13] Tom Lauwers: kids, like something work in the real world is, is what it's about. It, it makes the connections in their mind in a way that a virtual environment doesn't. And, and so I'm glad that we're getting back to that. I mean, I think that was the big, well, not the big loss. That was one of the losses of, of last year. [00:08:33] Kelly Schuster-Paredes: Yeah. And you know, on a, on a side note, on a teacher note, it with soldering because then they leave computer science for the quarter going, oh my God, it was so much fun. And everybody it was the best, the best class we don't want to leave. And I'm like, yeah, I forgot about all the crying that happened seven weeks ago. [00:08:52] Sean Tibor: And they take home this little gadget, they've got an actual artifact from it. It's not just a file that's on their computer with some source code. It's something that sits on their desk or on their dresser. And they look at it and they think, think good, positive things about what they learned in computer science. So that artifact is also really important. [00:09:10] Kelly Schuster-Paredes: So [00:09:11] Sean Tibor: So we'll, we'll go and re yeah, we'll go in reverse order for the fails so that Tom has a little bit of time to figure out what he wants to, to share. , for me, the, the big fail is. Israeli the fail of Google CoLab in K through 12. So I am really disappointed. Google changed a lot of their education policies on September 1st, and I think it was done in a less than transparent way for educators. , there's a lot of things I love about Google and a lot of things I love about their products, but I think this was poorly handled. Um, essentially what happened was. Google changed their policies around their workspaces for education and what apps could be made available to students under the age of 18. And one of those apps that was affected by this, because it wasn't in the set of Google apps. In this admin console that you can specifically enable for students under the age of 18 was Google CoLab, which we have been using extensively for teaching computer science. It was a big part of our curriculum this quarter already for students to explore, submit assignments. Unfortunately, it wasn't like it all went away at once for all students. Some students had it working, some students didn't when we look at an, our admin console for our workspace for education, looks enabled and turned on. So for some students, it simply didn't work. So I have been spending the day going through and excusing assignments for students who simply could not access. It could not finish it in time. Couldn't get the assignment done and it's just been a big fail. Um, and I'm really disappointed. I mean, if anybody from Google is listening, like I know you can do better. Like I've seen you do better. This was. So I hope that it comes back at some point, but in the meantime, going to have to re factor all of our assignments to make sure that they are running independently of CoLab and that we're not as dependent on an outside service as we have been on co-lab [00:11:12] Kelly Schuster-Paredes: Yeah, that was huge because we had just made the switch this year. We used a little bit of it and we had the kids make their own, but we went in and. Um, we converted all of our, we have what's choice board, so kids can choose what activities they want to work on out of the ones that we give them. And we had transferred them all from word documents into this co-lab with running code and I mean, hours and hours and hours. And it was just a trickle, triple trickle, effect of kids going. I can't get into my co-lab. I'm like, I don't know what you're doing. Go to the tech lab. Another person was just like, what's. I had all. And, and the funny thing is, is we have, like Sean said, we had all of our copies from the students. So it was something that as, as an adult, you didn't really notice what was going on or you couldn't see the fix. So it was a huge fail. But speaking of sales, we had another fail to we've. We failed a lot this quarter, I think. So how did big fail with the circuit playground Bluetooth? And it was, that was my fault completely. I didn't know that ADA fruit had, um, Sean kept talking about putting files on the ADA fruit and I was like, whatever. I didn't realize that it was the packages that the ADA fruit needed to run For the Neo pixels for the board, the digital, I didn't know that you had to actually import and he kept saying, should we just put the files? I'm like, no, the kids can do the UFT, the UTF file. No problem. We'll get it. I wrote all this, this choice board up, and I didn't realize that the blue fruit had changed dramatically. I think it changed a lot from the little black one that I practice on and we wrote up a choice board. Totally ditched. I was like, forget it, kids. I'm taking that away. It's my fault. Don't worry. But I owe a huge learning mistake on that one. the board before you assign it. [00:13:08] Sean Tibor: Yeah. on the, on the plus side, Scott Shaw Croft from, uh, ADA fruit has given me some suggestions for how we can package our own. You have to file to flash those devices that would come with. The library. So I have some testing to do that, to simplify and streamline the process a little bit. And then I think if we tweak the choice boards, the instructions that we want them to follow and, and read, I think we'll be back on track with it and it'll make it a lot easier for everyone. But there was definitely something that, well, Yeah, we probably could have done that a little bit better, you know, and made it a little bit smoother, but the kids have been persevering and making it work and they still get it to light up. It just means more work, uh, to, to get there. [00:13:49] Tom Lauwers: I think, I mean, when it comes to that kind of electronics in the classroom, robotics in the classroom, Um, you know, there's the challenge of designing the hardware and the software, but there's also the challenge of removing as much friction from the process for people who aren't electrical engineers and software engineers, like in, I would say that's actually the greater challenge. It's something that in the 11 years that I've been working on Birdbrain technologies and on our products, it has been a slow and steady improvement over time. I've just like, oh, this is a pain point. This is a problem. You know, this could be made better and we still have a couple, like you have to load firmware on the microbit. You know, we could do that in house maybe, but it would be hard to figure we haven't figured out how to do it yet. Um, but yeah, it's, it's, [00:14:42] Kelly Schuster-Paredes: Yeah. [00:14:43] Sean Tibor: a trivial problem. it's [00:14:44] Tom Lauwers: it's not. [00:14:45] Sean Tibor: you want to make sure that there's. You know, you're maybe changing one thing to make it easier, but you might create three other problems of complexity and other other areas. So there's definitely that tension between the ease of use well as the distribution, as well as the learning experience. Right? Sometimes the, the easiest path isn't the most, uh, pedagogically. [00:15:07] Tom Lauwers: yeah, the, the analogy or the phrase that I've heard to describe it. And this is not from me is to. To have like the good hearts remove the bad hard. So the bad hearts are like, you know, things that don't really add to the learning experience, but that maybe trivial information that an electrical engineer knows that no one else really needs to know. Don't make other people learn those things because you don't. [00:15:33] Kelly Schuster-Paredes: Hundred percent. I, um, Sean loves hardware. I am a phobia of hardware, you know, the worst. The last thing that a teacher who is learning to code. Or who is not a CMU alum. You know, I was a bio bio pre-med major. The last thing we want to do is have to 30 kids screaming I can't get in, you know, and it's just like, just want to throw the hardware out and just go back to putting pretty graphs on, you know, on the, on the editor, because in the end, those are easy. The code when you're starting to get into code is the easier thing to find. For, for some of us, me, um, but trying to figure out, like, change the wire, change the cable, check the date, you know, so many things, so many things. So yes, I, a hundred percent agree. And the, but the microbit is awesome. So thank you for using that. [00:16:29] Tom Lauwers: Yeah, well, one, once I saw it, I realized how awesome it was. And I'm like, let's build a lot of stuff around this because I'm not going to replicate this. I'd rather just incorporate it. [00:16:39] Kelly Schuster-Paredes: Yeah. [00:16:40] Tom Lauwers: Um, [00:16:41] Kelly Schuster-Paredes: Before [00:16:42] Tom Lauwers: yeah. [00:16:42] Kelly Schuster-Paredes: you want to do your fail? I don't know. We want to give you your failed [00:16:44] Sean Tibor: Yeah. [00:16:44] Kelly Schuster-Paredes: we [00:16:45] Tom Lauwers: Oh, oh, [00:16:45] Kelly Schuster-Paredes: want to talk about your butts, but. [00:16:47] Tom Lauwers: I have one. As a response to the pandemic, Getting vaccinated and feeling like I can go out in the community again, I've started doing, um, workshops and volunteering. and I did one in July that went well with finches. And then I signed up to do a sequence of eight, one hour workshops with a cycling club for disadvantaged youth. And the first session was on Wednesday and we're going to build like robot legs out of hummingbird kits that model the muscles in the leg. Since it's a cycling club, kind of teaching both anatomy and robotics. Uh, and I just was under prepared. I feel like I ha I I'm a little rusty, especially working with students. I mean, in my, in my background, I've mostly done professional development for teachers in the last five. Um, and you know, it's 7:00 PM and they've been cycling for two hours, so they're kind of tired. And we got to the point where they were all connecting to the robots and they moving servos. But what I should have done was bring a bunch of printables because you know, the area that we're in, I don't have a projector. It's not a regular classroom. It's like literally like a bunch of tables in an indoor cycling place. Um, so I should have brought a lot of principles and maybe thought a little bit harder about what happens once they connect. Cause they got to that part pretty fast. And then I had sort of 20 minutes of running around trying to teach each individual group more instead of doing it, you know, as a whole group. So I have learned from that and next week will be better. [00:18:23] Kelly Schuster-Paredes: That's a hard one. Definitely. Especially when, when you're in that situation, when you're trying to run around and solve everyone's problems and it turns into this, uh, this downhill slope of everyone lining up for your health. That's one of my classes this year, Sean always lasts this one class because they, as soon as they come in, it's like 10 people in a line at my desk. And I'm like, I miss zoom. Bring back, zoom, away, go sit down, go sit down, go sit down. So yeah, that's a hard, that's a hard lesson, but it's always good to see and you can reflect on that learning experience. So. [00:19:00] Tom Lauwers: Yeah, I mean, I think, I think next week there were a few kids who'd done some scratch programming who actually figured it out very quickly without much help. And so I'm going to recruit them to help me also in, in addition to bringing more. [00:19:16] Kelly Schuster-Paredes: Nice. [00:19:17] Sean Tibor: Very Nice. So why don't we jump into our main topic and we've kind of been hinting at it throughout this, the winds the week and the fellows the week, but, why don't we take a few minutes just to introduce you and who you are, that you, that you've created and the products that you offer, just so we get a sense of, what you're bringing to education and how that relates to teaching and. [00:19:38] Tom Lauwers: My official title is founder and CEO of Birdbrain technologies and, Birdbrain technologies mission is to inspire deep and joyful learning for all students through creative robotics. And so we do that through the robotics products that we provide the Finch robot and the hummingbird robotics. And those came out directly from my doctoral work at Carnegie Mellon. So I, I spent quite a while at CMU. I went there as an undergraduate and graduate students. So 11 years total, almost, um, and the research we were doing at the time was all about in the, in the case of the research that led to Finch, it was about developing a tool for, um, computer science education. And this was in the late two thousands. So it was focused on high school, AP classes and college introductory freshmen level classes. And that led to the Finch robot. And then the, um, the hummingbird came from a project called arts and bots, which was all about providing kind of initially it was about providing an alternative craft-based afterschool robotics activity, but as teachers started using it, it actually became more about integrating robotics into the curriculum. Um, in kind of upper elementary, middle and high school, uh, and that led to the hummingbird kit. So in 2010 I was graduating and, um, we had these two research projects that had shown success in classrooms, and I decided to start Birdbrain technologies to commercialize them. We launched the original Finch robot in 2007. And the original hummingbird kit in 2012, we're now at this point on generation three of the hummingbird and generation, two of the Finch with the new Finch having come out, actually during the pandemic. So in, in late 2020, and it's been really interesting, like as an evolution of the company, as I said, like, my background is very [00:21:33] Sean Tibor: Um, [00:21:33] Tom Lauwers: in engineering. My undergraduate work is electrical engineering, not computer. I was actually turned off by a lot of my computer science classes in the beginning, which is part of the reason I think that I like that I'm interested in, in providing tools for computer science. But as we've grown, it, it's been interesting to see the progression from what I would consider kind of early adopter teachers using the tools because, you know, I wrote technical support and. Documentation that was of written for engineers. And should I, you know, I didn't have that much background in like curriculum design, but as the company grew and as more people started using them, we were able to hire a lot of people in the education space to kind of design good curriculum and good learning materials. And I would argue that at this point, materials are. One of, or perhaps the biggest differentiator terms of what we provide. We provide free professional development courses for teachers. We provide lots of curriculum examples. Um, we also think about the classroom when designing our products, right? So from a software perspective and a hardware perspective, what works in the classroom is what we're aiming at, what works in somebody's home. You know, we don't really sell into a consumer. We sell into the education market. [00:23:02] Kelly Schuster-Paredes: Yeah. [00:23:02] Tom Lauwers: yeah, that's, uh, that's 15 years in a very, very brief nutshell. [00:23:09] Sean Tibor: Yeah. And our experience in our classrooms, we've worked pretty extensively with the Finch robot. One of the things that Kelly and I designed last year, because we were using kits to do the competitive first Lego league in our middle school. we made the decision at the start of last school year during the pandemic that we weren't going to be engaging in competitions. And so. Mostly Kelly, but I helped, um, put together this amazing exploratory program for robotics, where it was all in the classroom. We could control the experience in terms of was working with which robots at what time, and keeping social distancing and, and all of those things that we needed. And we everything from exploring, you know, electrical engineering hardware. Like here are components, sensors, things like that that are foundational, and doing some coding with them to doing some prototyping. But one of the things that was a kind of a big hit for us was the Finch robot, the French 2.0, when we got it, uh, was something that was, was well engineered. Right. So we could rely that it was, it was not some. Uh, some kit from Amazon that was put together for, to hit a cost And build materials, what it was designed for a classroom. it had enough activities that were engaging for the students that they could really see and envision what they wanted the robot to do. And I think, you know, my, if for example, my favorite feature is the fact that you have that little rubber stopper that you can pop out and put a, a marker in there and draw on the table and they can see. Where their robot is going and moving, and it helps them with their design and intentionality. [00:24:50] Kelly Schuster-Paredes: And a great, so it was the other thing. And a lot of, lot of listeners always ask us to share our Korean. And cause we did something similar to your hummingbird kit. We wait, we hugged. And we had so many parts from I don't even know our w whatever we had, all those parts, tons, like boxes and boxes of led servos and cables. So we started trying to put together these kits that kids can have in a plastic bag that no one else could touch. And we started throwing Servo's in there. Sean printed out, know, a 3d printer. Um, linear actuators. we kind of did with your, your hummingbird in a little bit of a Haji pot, it wasn't as, as good as some of your products that you guys produce, it was something that we could do. We could, could get it out. And if anybody's looking for a curriculum, I love, I love it. I've been watching your, your website, just spend through the hummingbird activities. It's, it's something that once you get your. Students teachers in, in low-income can still do all these projects. You have the guitar where you have these little lights and a stoplight. I mean, even the, the, those stars, you know, you can start learning about astronomy. So whoever your curriculum designers are I'm, I am a curriculum designer. From years past and I love building curriculum so well done. Um, it's just like one of these things that everyone's asked us, you know, it, share what you do. And I'm like, oh, you don't have to go to bird brains. We're not allowed to share what we do. I mean, we share a little bit, but we can't give our curriculum now at being where we are in our schools. So pretty cool. [00:26:33] Tom Lauwers: Yeah. [00:26:35] Sean Tibor: so tell us how this started. Like, so, you know, you're, I think you and I were probably at CMU at least for a few years the same overlap and you're an undergraduate electrical, electrical engineer. How do you go from being an undergrad, electrical, electrical engineer to thinking want to work on. You know, improving the education space with robotics. So was there a moment that that happened or were there kind of a series of events or a mentor? How did that take place? [00:27:03] Tom Lauwers: Yeah, I think, um, so as an undergraduate, I was interested in democratizing robotics. So like I was in the Carnegie Mellon robotics club and eventually became the co president of that club. And we, we developed like a standard platform. This is pre Arduino and started a course called fun with robots where we had something like 50 CMU students build a robot. With our kind of standard platform. So almost what you guys are doing with, you know, hodgepodging materials. and so I was interested in the educational applications of robotics, even, even back then. the first couple of years, as a PhD student, I worked on something completely different. I worked on a robot that, um, balanced on a single bowling ball, just it's called a ball bot. Um, Theoretically interesting, but I didn't see the applications in the timescale that, that I've kind of learned about myself, the timescale that I was interested in. Right. So I, I learned something about myself in that process, which is that I'm not that interested in basic research myself where the impact could be decades now. I would rather work on something where I can see impact even in the research. And so I naturally kind of gravitated toward a lab at the robotics Institute called the create lab, is all about creating, um, or finding applications of robotics technologies to empower communities of practice. And it's all about working with those communities in the design process to figure out okay, how, you know, as, as how can we design something that helps this community necessarily. Imposing a tool on them and saying, oh, this we've, we figured this out for you guys. Here you go. Um, you know, that's you, you need that bi-directional feedback. And so the projects in the lab at the time were educationally focused, which I also, because of my background was already interested in. And so I started working with teachers and students to develop the things that became the Finch and the hungry. [00:29:18] Kelly Schuster-Paredes: That's super cool. um, I do what I do every time we have these podcasts. So I, I spy on you on the internet I pulled up your, uh, dissertation. I, I love this. Piece that you put in there in emerging technological domains that contain educational potential partnerships with educators and content creators. At the very beginning of the design process may lead to educational relevance more quickly, percent, a hundred percent because you can, a lot of people try to design things and have never been in a classroom and, or they try to make ways that kids or teachers should be in the classroom. And haven't been in the classroom in decades. a hundred percent is that partnership. And that's what makes you know, your product. You have to know that if the Finch, you know, we don't want it falling all the time, but if it drops, it's not going to splatter into a zillion pieces like our Legos, you know? you have these durable things, you know, that's what I love about the BBC microwave. They get thrown around left and right. And I think I've only lost a button maybe three or four times. So being able to work hand in hand with educators to see what they need and how you can change it quickly is impressive. So that, you know, is only got halfway through. It's a very long [00:30:37] Tom Lauwers: yeah, that's halfway through is pretty impressive. Um, [00:30:42] Kelly Schuster-Paredes: read. [00:30:44] Tom Lauwers: yeah, I think that's like three or 400 pages. So. [00:30:47] Sean Tibor: So as you were, as you were going through this design process and working with educators and working through curriculum, what were some of the things that emerged that were kind of most important or critical to designing for education versus. Designing for other fields. I mean, we've alluded to durability, which is especially important in middle school ages and elementary ages, but what, what elements came out like I noticed for example, that the Finch is very much by an actual bird. Right? So it looks bird like it has that kind of those characteristics and features that help students relate to it. But was that something that was like, discovered, was that something that you designed for branding or because it was actually helping in the classroom? [00:31:31] Tom Lauwers: with the fence. The newest Finch is sort of a refresh on the original design from an aesthetic perspective, but it still has those bird-like elements. Um, but the original robot, we have an industrial designer sketch up several different designs. And we were already working at the time with a group of something like 20 high school teachers. It's as well as a couple of community college faculty. And we sent the designs over to them and asked their students and themselves to kind of. And they came up with these combination of like, um, a race car or, well, there were nine designs and the ones they gravitate to were kind of this race car and this Finch, um, this very bird-like robot. And so the final design was sort of a merger of that feedback. Um, so we definitely went out and asked people and I think the aesthetics, know, are important. I think. Uh, it's very much intentional to make it something that is attractive and can be anthropomorphize so that you can kind of make a connection with it without making it too much like a kid's toy. So that, that was the other thing, right? At least the initial research was all about high school students, but even with the Finch too, we were really imagining something that could be used from kindergarten to college and where, you know, Uh, 18 year old would wouldn't be like turned off by it because it looked too kiddish and a five-year old would still be engaged with it. And so making that balance I think was really important. other, the other of the original of aesthetic thinking was gender. Right. We don't want it to be gendered in one way or another. A lot of robotics stuff is kind of male, gender. If you look at, you know, so we weren't going to call it like the Hawk or the Raptor or something aggressive. We weren't going to put in too much of an aggressive like styling. Um, but you know, we don't want it to be like pink either, right? Like the kind of stereotypical female. Um, so there's, there was a lot of thought given to the aesthetics and the design. [00:33:42] Kelly Schuster-Paredes: Yeah, I like, um, well, I always tell people, it reminds me of the war of the worlds, those big silver, you know, the silver things that come in. And I'm always like every time I think about that, I remember that movie. the first time I really thought about bots when I was little watching it. So not that it's silver, not that it's like coming out to get you, but. It does have that kind of nose in the front end. So it's pretty cool. The other thing I really have to say, and going on back when Sean was talking about putting the marker and one of the first things we noticed, and we appreciated after um, when, to the CMU robotics for Lego Mindstorms back in, when did we go? three years ago. For Nancy and what was very frustrating for Sean? Not for me. Um, cause he wanted this precision 90 degree turn and it was killing him and you know, and he kept and everyone else is moving on and he's my partner and I'm just like, it's fine. No, it's 89 points. Degrees and those getting very frustrated. So the first time we plugged in the fence, the first thing Sean does is make sure it turns exactly the angle that it was supposed to. It did really well. So he was quite impressed. So, and it came from CMU. So he was even more in love with it at first, but it does have a great precision. It does have. [00:35:03] Sean Tibor: I would like a chance to defend myself on this one. Um, like in, in movement navigation, especially with, the Lego competition, like errors are cumulative, right? So over time, you have errors in variability, it will grow over the course of it. What's important to me is that students shouldn't feel frustrated by the fact that the robot won't drive to the same place. You know, if they can get it to within a certain amount of precision, then they can focus on actually solving the problem. Instead of trying to do what I did, which was to just get it, to do the same thing twice. Right. And so I think, I think that that's something that the, the Finch does really well. And I have to say that the new Lego spike prime does really well. So that's, that's one of those things where it's like, you want to get rid of the bad, hard, right. Getting it to dry. In a repeatable fashion is a bad, hard, but the good heart is being able to navigate and put together all the pieces. So I apologize to Kelly many times over the last four years for putting her through that. But, uh, but I, I now learned that about myself as I that's. One of the things that I look for is, is removing that variability and the cumulative error so that the students can focus on what is actually really valuable learning the process. [00:36:18] Tom Lauwers: we, we sweated that, uh, quite a bit, the movement of the Finch and trying to get it as, as accurate as possible while keeping the, you know, the cost to a reasonable amount. Um, and that is actually something that came from feedback from the original Finch robot, which was tethered. And so, because it was on a tether and the reason it was tethered came out of the reality of like technology in 2010 and actually was a direct consequence. Of designing for the classroom versus the home in 2010, but for the new Finch in 2020, we were like, yeah, we can definitely make this untethered now. So now that we can make it untethered and let's make it really accurate because clearly everybody wants that. Everybody's asking for that, um, from our feedback from the original Finch. So we, we spent probably a year iterating on the motion algorithm to try to get it as, as close as we could. [00:37:11] Sean Tibor: It's it's really impressive. I mean, I think if you look at the price point of it, it has that reproducibility, you know, where students aren't, you can get to the point where they're not really noticing, too big of a very, very incident, then they're able to move on to the solving the next. [00:37:26] Kelly Schuster-Paredes: Yeah. And it's just for the cost is $139, which is really impressive. Um, versus our Lego $350 kits, but know, one, I know we're going to short on time, but one of the other things, what I really want. [00:37:44] Sean Tibor: Uh, [00:37:44] Kelly Schuster-Paredes: For us in our, in our curriculum. lot of times I go into, turtle module and that's my second unit. And moving over to the fence, you can still talk about, you know, here's your object, we're going to name the French birdie or, or whatever. it's, it's a little bit of a, an easier move I've done turtle because they get the idea and I hate to say it, but if they can draw. Dar over and over again and make that darn spiral on a desk. Cause that's all they make is with the turtle over and over again, the spiral. So when they can do that, got to win. You've got to win. So. [00:38:22] Sean Tibor: Yeah. Tom what's what's next for Birdbrain like, I mean, I, I don't think. We should expect to Finch 3.0 right away. Right? Like there's definitely some room to run there, but, what's next more curriculum design, more, more lessons, professional development, new bots, you know, where are you spending your time and attention when you look ahead now. [00:38:43] Tom Lauwers: there's a, there's a lot planned for the next year in the shorter term. We are releasing a, a small AI and machine learning module to go with the Finch that it will work in Python and in snap, which is one of the blocks based on. And it's been pretty fun to design that so far. You basically like you use Google teachable machine to train image model, like models to classify images, models, to classify sound, and then also poses. And then you can use that in Python to control your Finch. So you can speak to your Finch through the computer and make it drive. If you've, you know, if you've taught the model three words, you can usually say forward, backward turn or something. Or if you can do the image processing, you can like basically have different people's faces, make the Finch do different things, so that that's fun. And that's coming up pretty soon. I've already mentioned the Python with the Finch robot free professional development, but I mentioned that again. So any us public, uh, and charter school teacher in middle or high school take this free course. And also receive a free Finch with it. And we've got something like 110 robots left to give away. And that's, um, through funding from emphasis foundation USA. And then related to that, we're on note. Um, our loan program should be starting back up in January. So the loan program is something we've been running for a very long time since 2013. And in the loan program, essentially, um, teachers and librarians in the U S can apply for a flock of robots for, you know, for a few months to catalyze some computer science activities. It's completely free. We pay for shipping both ways. Uh, and we have been using mostly Finch ones in the loan program so far because we've had supply chain issues, which have kept us from transitioning that over to Finch, to. But we expect those declare up soon, and then we will have a thousand Finch, two robots in the loan program and ready to go out to schools to use. that's exciting. I can't wait for that to get back up. It's it is sort of been running in kind of an altered state all through the pandemic, you know, and then the transition to Finch too. I'm excited about getting back to kind of a more normal. Loan program. And then the, I mean, the thing that we're trying to launch next year is not, it's not computer science related, but we're [00:41:23] Sean Tibor: Yeah. [00:41:23] Tom Lauwers: math manipulatives for elementary education that are, um, uh, digital manipulatives. So they're tied to apps that you use on a tablet or a computer, but then you are also manipulating cubes or dials, you're kind of getting exercises and feedback from the app about what your. Um, so we'll see how that goes on. I'm excited about that. That's called outlet and, um, we don't have anything like on the website about it yet, but, uh, look for it probably sometime next year. [00:41:58] Sean Tibor: I sweat. [00:41:58] Kelly Schuster-Paredes: something. I saw something out when I was Googling and saw. So I saw it was like a tray of cubes [00:42:05] Tom Lauwers: Yeah. [00:42:06] Kelly Schuster-Paredes: Yep. So it looked pretty cool. I know, I know we love manipulatives and the lower years, because just clicking on an ad. Doesn't do a lot for that processing from the short-term to the front, to the long-term, but using your hands and crossing over your body, we know that that helps students really retain that knowledge. That should be fun. [00:42:26] Tom Lauwers: Yeah. And we've had, uh, we had, uh, an NSF grant that we sh split with Carnegie Mellon to develop. And it was the same process of working with teachers iterating. We actually had pilot, uh, in, in classroom pilots last year with it, uh, with six teachers and over a hundred students and they managed to use it despite last year, school, years restrictions. [00:42:50] Sean Tibor: Yeah. [00:42:51] Tom Lauwers: So we didn't get quite the research study we expected, but we got an interesting research study that shows some, at least that it can be used in a lot of flexible environments. [00:43:02] Sean Tibor: Very nice. Well, we just about out of time and need to wrap up. But, um, did, uh, for many of our international guests, I did want to point out just a couple of things. Um, so Carnegie Mellon is located in Pittsburgh, Pennsylvania, which is also the site of the pie con in, I believe it's 20, 24. they're going to be bringing it back. So, we'll all pile on Tom to show us around and, and get. Uh, an introduction to the city that he loves and the robots that he adores. Tom, we, we really do want to say thank you for taking the time to talk with us about all the work that you're doing. for both Kelly and I we've been using the Finch robot particularly, and I think we probably need to get some hummingbird kits just to, to be able to play. [00:43:45] Kelly Schuster-Paredes: Yeah. [00:43:45] Sean Tibor: Um, but we can, we can definitely say that. The, uh, the experiences that we've had have been very good for our students to learn and to grow. And we appreciate all the hard work that you've been doing over the last 15 years to make that happen. So thank you very much. [00:43:59] Tom Lauwers: Yeah, thank you so much for inviting me on. And it's just great to hear that, you know, it is really all about the experiences of the students in the end and having them have, you know, deep, joyful, positive. [00:44:13] Kelly Schuster-Paredes: Awesome. Just out of curiosity, do you ship to do is Bahamas is not the USA, but do you do any, you do shipping to, out to the homeless and anything like that? [00:44:22] Tom Lauwers: yeah, we do international shipping. Uh, we also have resellers. I don't think we have one specifically in The Bahamas, but we have resellers internationally as well. [00:44:29] Kelly Schuster-Paredes: We know we, we're trying to like [00:44:31] Sean Tibor: uh, [00:44:32] Kelly Schuster-Paredes: for some of our friends that have underprivileged camps [00:44:35] Sean Tibor: um, [00:44:35] Kelly Schuster-Paredes: um, try to look out for the community. Those are building. Access for everyone. So just handing it out there to our listeners and stuff. So [00:44:44] Sean Tibor: Yeah. So if you'd like to continue the conversation, if you have some experiences that you'd like to share, you can always follow us on Twitter at teaching Python. at SM Tibor on Twitter and on Peloton. So if you want to cycle with me, that's the place to find me. I'm not on there very often. So maybe a little nudge is all I need. is at Kelly Perez on Twitter. Our website is teaching python.fm. And you can always send us a note through there. We love to get listener emails. Uh, we are also on Patrion. there's a link to the Patrion page in our show notes. We have a lot of wonderful Patrion supporters who keep the podcast going and keep it funded. thank you to all of you. I don't think we have any extra announcements and I have students about ready to get learning here in just a minute. So I'm going to us off by saying for teaching Python. [00:45:33] Kelly Schuster-Paredes: And this is Kelly signing off.