Microsoft Research Podcast - 098 - Hacking the runway with MakeCode with Dr. Thomas Ball and Dr. Teddy Seyed

Episode Date: November 13, 2019

Computer programming has often been perceived as the exclusive domain of computer scientists and software engineers. But that’s changing, thanks to the work of people like Dr. Thomas Ball, a Partner Researcher in the RiSE group at Microsoft Research, and Dr. Teddy Seyed, a post-doctoral researcher in the same group. Their goal is to make programming accessible to non-programmers in places like the classroom, the workshop… and even the runway! On today’s podcast, Tom and Teddy talk about physical computing through platforms like MakeCode, a simplified programming environment that makes it easier for young people – and other computer science neophytes – to start coding with programmable microcontrollers. They also tell us all about Project Brookdale, where they did a collaborative fashion show that gave emerging designers the tools to embed technology in their garments and produce wearables you’d actually want to be seen in! https://www.microsoft.com/research  

Today's two-guest podcast is all about physical computing, programmable microcontrollers, computer science education, and fashion. Join me along with a veteran researcher and a newly hired postdoc researcher for some perspective on the magic of making with the power of code. You're listening to the Microsoft Research Podcast, a show that brings you closer to the cutting edge of technology research and the scientists behind it.

I'm your host, Gretchen Huizenga. Computer programming has often been perceived as the exclusive domain of computer scientists and software engineers. But that's changing, thanks to the work of people like Dr. Thomas Ball, a partner researcher in the RISE group at Microsoft Research, and Dr. Teddy Syed, a postdoctoral researcher in the same group. Their goal is to make programming accessible to non-programmers in places like the classroom, the workshop, and even the runway. On today's podcast, Tom and Teddy talk about physical computing through platforms like MakeCode, a simplified programming environment that makes it easier for young people

and other computer science neophytes to start coding with programmable microcontrollers. They also tell us about Project Brookdale, where they did a collaborative fashion show that gave emerging designers the tools to embed technology in their garments and produce wearables you'd actually want to be seen in. That and much more on this episode of the Microsoft Research Podcast. I want to just start by saying that I've been waiting a long time to get you two in the booth with me. Tom Ball, Teddy Syed, welcome to the podcast. Thank you so much. Yeah, thanks for having us.

Tom. Yes. You're a partner researcher in the Research in Software Engineering, or RISE, group here at Microsoft Research. And you've been here a while. Teddy, you've just been hired as a postdoc here at Microsoft Research. One week ago. Yeah, I'm kind of fresh out of the gate. Although you've been at the computer science thing for quite a while, and we'll get to that shortly.

But you're early on in the game. So we've got both ends of the MSR spectrum represented here. And because of that, I think this is going to be an exciting version of this question. What gets you up in the morning? Why do you come to work? What's your motivation, as they say in the movies? Well, for me, it's the Microsoft research environment. We get to probe fundamental questions about computer science and work with an amazing set of people. Over my 20 years here, I've been able to hire people ranging from empirical software engineering researchers to people doing automated theorem provers and compilers, programming languages. And Teddy is an HCI expert, human-computer interaction, working in the intersection of computing and wearables. And so I just find it really exciting to be working

in an area where there's an huge amount of diversity, a great amount of talent, and it's just fun. For me, every five years or so, I like to sort of switch things up and change things up. So I'm at one of those phase chains, and Teddy is like the phase change agent. Glad to be that. Something like that. The face of phase change. Teddy, what gets you up in the morning? I mean, I had a really good internship experience.

I mean, I was pretty driven and entrepreneurial kind of before I came here. And then coming to MSR was a really good research experience for me. I remember one of my first meetings was sitting with Tom and a few other researchers asked, I think, as well. And you're kind of like, oh, like, wow, these are all people that I have, like, read about. And they're so open to collaborating. And it was a very eye-opening experience for me. And then just the internship was fantastic. And so now when I wake up, it's like, who am I going to work with today? And like, what kind of things can I do that I can't do anywhere else? So it really drives me. Tom, go back just a little bit, this umbrella of RISE, Research in Software Engineering. I've had many of your colleagues on the podcast, and it seems to be a kind of phrase that encompasses a lot of different threads of

research. Is that by design? Yeah, that's really by design. When it was started, really by Wolfram Schulte back almost, I think, 2009, it brought together a lot of different groups. And, you know, there's so many aspects to software engineering and software quality. There's lots of different ways to look at the beast, and it's very complex. So we look at empirical methods. We study how people program. Danae Ford Robinson just joined our group. She studies people on GitHub. We have people doing formal automated theorem proving to help us find defects in software or prove that software doesn't have certain classes of defects. So in general,

there's a wide range of problems, software quality, software performance. And for the most part, we're looking at improving the life of the professional software developer. Although in the last five years we've expanded, we do work really for bringing the programming tools and technologies to end users who may not be programmers. There was work that made it into Microsoft Excel called Flash Fill, which is a form of programming by example or program synthesis. And then the latest stuff that we've been doing with education is a simplified programming environment called MakeCode. We're making it easier for middle schoolers to start programming with embeddable computers. Teddy, I want to ask you about your PhD because it's an interesting one. It's the first of its kind in Canada because you came out of the University of Calgary. Tell us about your PhD

program, what it involved, and how it landed you here. So my early PhD, it was really heavily into software and, you know, building these Iron Man type interfaces because I'm a huge Marvel Comics fan and I love Iron Man. So I was really interested in like building all those things and I got really good at it. And then I was interested in the business side. So I was working for a startup in Calgary and I didn't really understand sometimes sort of the business lingo of how, you know, like a PM would communicate with an external customer. I'd done a lot of demos and things like that,

and I was quite comfortable, but it was different being able to communicate with, you know, VPs and execs. So then I was like, okay, I'll take some MBA classes. So I did that as part of my actual degree. And then I really liked it. And then I also found it interesting because it was very related to the research I was doing in terms of methodology, right? Where, you know, I might run a couple studies to get a paper, whereas on the entrepreneurial side, I'm running studies to actually like make the product better. So it was like very much the same for me early on. And then I got a bunch of experience doing, you know, hardware startups. And I got really interested in hardware.

And then I brought that into my own research. So I started doing more hardware projects as part of my dissertation. So I did a smartwatch and a smartphone, and both of them did really well. They got on Gizmodo and Forbes magazine and things like that. The watch was interesting for me because I was actually trying to build it. So I was pushing to VCs and trying to raise capital

and sort of struggling with that. But what was interesting in that journey was I wanted to make something that was unique, that was from my perspective, that wasn't something from Google or Apple, right, because they were doing smartwatches at the time. So I jumped into the world of fashion, and I got stuck there because I'm trying to communicate with people that aren't technical at all, and try to teach them concepts, and it was always very frustrating for me to do

that. And I just stayed there because it was a really good place to sort of learn and experiment how to create wearables. I was having impact on my community with working with kids and teaching them tech through fashion, so it was just a really good place to be. And then coming to MSR, I was working on a startup at the time that was doing a fashion tech kit. And then I emailed the MakeCode team out of the blue. I was not expecting an answer back from Tom. But Tom replied and we had a, I think it was supposed to be half an hour call and ended up being a bit longer.

It was a little longer. Then you came out just for a quick visit. Yeah, a quick visit. And then, you know, they were like, do you want to do an internship? I was like, yeah, sure. Tom, I always find it interesting from the perspective of two different people,

like he's a grad student in Calgary and he thinks I'm going to email the mothership, right? Microsoft. So how do you receive these emails? You get this from this guy. What do you think? Well, I mean, you go check out what he's been doing, right? And it's like, whoa, this is really interesting. And

we had been working with the BBC on the micro bit. So we knew about the embedded technology space and how it was being used by schools, but also had a little familiarity with its use in the makerspaces and fashion. And so this just seemed like a really great match. We had helped the product group stand up the MakeCode product, which started in research, and we were looking for what's the next thing. And, you know, sometimes, well, luck favors the prepared, right? And so we were prepared and Teddy was prepared and we were both, it was good timing, although we had to wait another year before we really got things going because we had some stuff going and Teddy had some stuff going. So, you know, the email was the seed, but it got a lot of thought processes going about taking this idea of democratizing technology and really making it broader than just CS education in the schools. But many people need to control and use computers as part of their work.

And one of the ways you control the computer is through programming and trying to simplify that. So you may not be a computer scientist or even ever take a computing class or a programming class, but you're going to be faced with using the technology or making use of it. Well, let me go upstream with you there because downstream we've got some tangible products and tangible projects, but I'd like you to talk a little bit more upstream about making computer programming accessible. Here we're talking about people who might not think of themselves as programmers or even able to begin thinking that way. So how have you attacked this from a 10,000-foot view? Right. Well, at the very high level, I feel like people want to be

in control of their lives and in control of computers. They don't want to feel like their lives are controlled by computers, right? And I think that the way that you gain control is through education and through understanding. And to be educated or to have control, you need to be more educated about how computers work and then how to make them do your bidding. So I think, you know, fundamentally that nations around the world have recognized that CS education is just as important as an education in, say, reading, writing, arithmetic, physical sciences. And Microsoft, from its roots, starting with BASIC and DOS, giving people the foundational tools to be able to control the computer, that's part and parcel of what we do. And the great part

these days is it's just much easier because of the internet, because of the web, because it's easy to get access to all these technologies. If you have access to a computer hooked to the internet, because of the web, because it's easy to get access to all these technologies. If you have access to a computer hooked to an internet, which is not a given everywhere. It's not a universal. It's not universal. But if you think about today compared to when I got started in the late 70s, there's so much more available for people who want to learn about computing and then become educated to learn how to control the computers themselves. Let's talk more in depth about Microsoft's CS education platform, MakeCode. Peli Dehulu was on the podcast a while ago and talked about MakeCode back then. Why don't we review a little bit? Tell us what MakeCode is,

how and why it came about, and where is it getting traction right now? MakeCode actually started out of a previous project from Microsoft Research from the RISE group called TouchDevelop. And the idea there was very simple. Everybody has these smartphones, but in order to program the smartphone, you need another computer. Why can't you program your smartphone using the same smartphone? And Nikolai Tillman had this great idea, and Michal Moskal and Peli joined in, and I came in a little later. And so we created an app to make it possible to write little programs that controlled your smartphone, but using your smartphone as the programming environment. And from that, we got to essentially a touch-based

code editor that you could use a single finger to program with. The BBC, a few years later, had this idea of the BBC Microbit. And this is a little embeddable computer that has its roots in the Arduino system, sort of a visible printed circuit board with some flashing lights and buttons on it. And they wanted a programming environment that was web-based to help take the micro bit to all the fifth graders or year seven in the United Kingdom. And so we use TouchDevelop to essentially deliver that system in 2015, 2016. And when we were code complete with TouchDevelop for the BBC micro bit, we said, okay, we're going to create a compiler from the TouchDevelop system into the MakeCode system. We demoed it internally and externally,

this PXT with the editor, the new editor for the micro bit. And we got interest from the Microsoft execs as well. So they funded the make code product group. So we have now some developers on that side, a program manager, and on the research side, Peli Mihau, myself, Steve Hodges, now Teddy, who are thinking more about what can we do with this platform moving forward. So it started with really simplifying programming for complex devices. And our first complex device was a smartphone. Then we went to the BBC micro bit. And now with MakeCode, we've expanded from the micro bit

to a number of other devices. But these devices are generally battery powered, like your smartphone, but not quite as resource rich as a smartphone. They're much lower power and smaller. Well, that's an excellent segue into what I want to talk to Teddy about next, which is using programmable technology microcontrollers in fashion. So tell us about your personal hack-the-runway vision

and how you decided to take high tech into fashion and what you call haute tech couture? The vision before I started, I mean, when I had emailed Tom and the MakeCode team was, you know, I had done a lot of work with Arduino and a lot of workshops with kids and designers. And just a lot of time was not used optimally in the workshops and it was just becoming really frustrating. So that's kind of how I came across MakeCode and that was really great. And then during the internship, the vision that I had wanted and sort of my bigger vision is I don't really think a tech company should be making wearables. I think it should be coming from the fashion world.

We should be providing tools for them to be able to create these things. Right. But in order for us to get there, we need to create something that designers can use. And like, if you think of them as developers, we need to create developer tools. But these developers are fashion designers. And we can solve all this on the technical side. And in the meantime, when things are ready and things are washable and stuff, they'll have that experience to be able to make the things that we want. That's kind of the vision that I had to start.

I want to press in a little bit there because to me, you're making developer tools, but you're giving them to people that aren't developers. So how does that work? That's, I guess, a little bit more on the HCI side. So on the HCI side, you spend a lot of time working with designers, working with people that are not sort of technical, trying to figure out what their challenges are, how they're thinking about design, what they're thinking about fabrics and materials. So I spent a lot of time sort of before the internship just thinking about fashion, really, and how that world works and understanding the lingo. Because obviously one of the biggest things is communication. You have people that are super technical and people that are not technical at all.

And bridging that world is very difficult. But one of the ways to do it is through communications. Okay. So let's say I'm a complete neophyte. Do you have to teach me computational thinking? With the work that we had done, I typically don't start with technology at all. Actually, it really starts, but what's the story you're trying to tell? Every garment has a story.

It reflects the person wearing it. So what is the story you want to tell? Then from the story, it becomes, okay, what kind of technology can you use? So for example, if you wanted to do a story around how you felt that day, then you probably want to use some sort of sensor that measures your emotion. Then we can talk about how to measure emotions. Then after that, we can talk about where you would place the sensors, and then it becomes a question of where you're placing it on the body,

how you're sewing it in. Then it's like, how do you want to power this thing, right? So it's sort of a narrative, but it's all starting from the story. And on the platform side, I think we have to have these components that can meet those needs, and that's where the complication comes. The designer might have something that's really not technically possible. We can meet them halfway or say, these are the sort of things that we have in mind. We might have a set of animations that are pre-built that they can choose from and parameterize visually, not by programming, but by drag-and-drop menus, by demonstration.

But there would be a more visual interface to start with, especially if it's a visual you're designing. You know, I think you also see that in the classic visual basic, the forms editing. You know, you have people creating these forms applications who would drag and drop widgets onto the screen, and there was the code behind. And I think for a lot of domains, that sort of paradigm of having a visual editor with domain-specific representations, in this case it might be about the sensors or the garment or the lighting and the animations, backed by code, the designer wouldn't necessarily have to edit themselves. But it would be there for somebody else in the design production pipeline to sort of go and maybe tweak. So you have to bring new people into the design production pipeline, probably. Yeah.

That's part of what we're trying to do in the MakeCode system is to use JavaScript and TypeScript to program these embedded devices so that we have access to this world of web developers who know JavaScript for doing web pages, but may never have thought of using it for programming a fashion garment, but now they could use those same technologies for doing web pages, but may never have thought of using it for programming a fashion garment. But now they could use those same technologies for doing so. Let's talk about Project Brookdale. Now, this is why I wanted you two in the booth. It happened not that long ago, but it was exciting enough for me that I started to ask, when can I get these guys on the podcast?

So I could give a big run up for this, but I'd rather have the two of you tell us about this research project. What is Project Brookdale? How did Microsoft Research get together with the Brooklyn Fashion Academy? And what were the results of bringing microcontroller technology to the atelier? So the nice thing about this project was I don't think I could have done it anywhere else. One of the things I was trying to do in my grad school was like be able to properly build something and give it to fashion designers and deploy it on a runway and see what would happen, right? The nice thing about the runway is that it's a very practical way of teaching you about wearables, right? It's all the conditions of heat on the body, things breaking, people moving, things being packed, all the sort of crazy conditions that you

would typically have to test for individually, you can get all in like sort of this nice runway idea. And so when we came to do Project Brookdale, it was really about how do we make a system easy for people that are not technical. In this case, that audience was fashion designers. And you had had a contact, I guess, with somebody in this fashion academy, right? Yeah, exactly. And I had contacts with the Brooklyn Fashion Academy already. and so it just worked perfectly, the timing of being able to build the hardware and the software and going there and sort of deploying the system. I should say, on the MakeCode side,

we had been working on this new MakeCode Arcade, a game engine, and James Devine, who was another intern, started working with us in the summer of last year on JackDack, which is a way to network together microcontrollers. So we talk about programming the microcontrollers with web technology, but how do you program a web of microcontrollers? Well, you need to be able to hook them up.

And originally that was just to hook up two game machines so you could have multiplayer gaming. But then when Teddy came along and said, oh, I have this fashion thing, we were like, oh, well, maybe we can also wire them up on the body and have this easier way to connect together different microcontrollers that would then be part of a garment.

And so sort of a plug-and-play microcontroller-based toolkit to make it easy to prototype. And so in MakeCode, we had a bunch of new technology. Teddy brought the application, and then we all brought it together. And it did come together. The hardware, the new software, the designs from the designers, and this team in Brooklyn in a last week mad dash to the runway. It was an experience never to be replicated. We'll do it again, Tom.

We'll do it again, but not in the same way. Not to be, tell us... Not to be forgotten, but not to be... Right. Not to be done in exactly the same way. Well, Teddy, tell us a little bit more about how this instantiated on the runway. Yeah, so the theme of the show was the Herrera women of Namibia. So the fashion was sort of influenced

by the Victorian air style that they had brought forward into the future. So each designer brought their own sort of theme and background into how they were using technology. So over a two and a half month period, rather than building every possible sensor, we picked sensors that were based off of their story. One of them had a really personal story about fabric and mental health issues. So she had tears kind of jumping down her dress and that was

one that Tom had worked on. Some of them use motion. There was one really cool one where it was sort of a yellow fabric and he had two lights coming down on the side. The Herrera women always worked in the farms and with cows, so she had this theme of sort of the women being strong like the cows and sort of walking forward. So there was a lot of themes like that. So Teddy did a lot of interviewing and there was education both ways about learning about the designs they wanted to create and then about the technology and the possibilities. So there was a tremendous amount of interaction in the lead up while we were scoping out the software and the hardware.

Yeah. And I mean, even the hardware, for example, we didn't design it. The brain, for example, is round because it was going to be on the center of the body according to most of the designs that we had worked with. And then things like the beads we designed to be sort of to the body. So it's going to be placed on any part of the body. And then we color coded them because we wanted designers to be able to use them without us being around. So you would know that red would be light and yellow would be environment and things like that. So it was very heavily driven

by a lot of the work at the very beginning, just interviews and communication. All right. So this is both a research project and a runway project. What were your findings? I mean, I would imagine you came away learning a lot after you tried this with these fashion designers and fashion design students. What have you come away with in terms of where you'd go next with something like this? Well, I think from the platform angle, we proved out some new technology. So the Jack-Jack networking, the software, the ease of use of the plug-and-play, and sort of learning about failure conditions. Whenever you try to do a rocket launch, a lot of the rockets are going to end up not making it into space,

and you learn from that. Similarly, with the garments, there were failures, and we learned a lot about power delivery. We learned a lot about the networking. And really, this was just the first prototype to get to the next step, which is to try to make a design that's more robust. It's still a work in progress. But, you know, proving out the technology and understanding the failure modes was, I think, a major aspect. So I'm just talking from the platform. Yeah, I mean, we didn't do as much programming with designers as we'd like. We didn't have time, and it turned out actually maybe having a laptop to code the garment is not a good idea.

Maybe we should actually use a mobile phone was one finding. One of the other ones was, you know, the audio jack stuff was really great for prototyping. But what is it going to be like if we do wireless, right? I mean, anything Bluetooth or wireless is kind of the ideal condition for wearable. But the unique thing about the runway is it's so noisy. There's cell phones going, there's all these radio signals going around, so that might not be the best idea for something like that, right? Okay.

Doing some of the stuff more integrated into fabric is kind of the next phase too. All right. So all of this is available open source, am I right? So MakeCode is all open source, the software platform. The hardware that we prototyped for Brookdale is not available yet, but there are devices like the MicroBit and the Adafruit Circuit Playground Express that you can do similar things with that are available from our hardware partners. So there is a lot available already, but not the actual Brookdale bead and brain yet. It will be coming soon. Yeah, that's the plan. It will be coming soon. Yeah, that's the plan.

It will be coming soon to a runway near you or someplace else. Yes. Well, let's talk about open source for a second because it's a huge trend over the past, I'm going to guess a decade, but maybe more. What are the factors that have influenced this shift in your mind, Tom? And why is it important to the broader software ecosystem, especially when it comes to research? Yeah, it's really transformative for the industry. I mean, every company has to think about what its secret sauce is and its special value that it adds. And so much can be found on open source that can be used that you can then add your value on top of that.

So, for example, MakeCode uses two editors that we didn't have to create. One is the text editor codename Monaco, which is part of VS Code. We just extract that component and use it. The visual programming with the colorful blocks is from Google. It's Google Blockly. These are both open source. We've actually made contributions back to, for example, the Google Blockly code base. We have regular calls with them. That was unimaginable 20 years ago when I joined Microsoft. As well, in research, it's great when a PhD intern or any intern is coming to Microsoft.

They can see the project we're working on before they even arrive. They can have a much more productive three months. And then after they leave, they still have access. So contrast that to 20 years ago, where often, you know, people came in, they worked on something great and left their mark. But then once they went back to the university, it was much harder to have access. And so I think just overall, people realize it's a win, win, win. There's still opportunities, obviously, for every company to choose what they retain sort of behind their firewall, which is their secret sauce. And I think every company does that to some extent. But I think there's been a realization for collaboration, for productivity, and for so many domains, it's much easier if you can see the code and be open from day one.

It's time for What Could Possibly Go Wrong on the Microsoft Research Podcast. Oh, goody. Oh, dear. And at this point in the show, I like to have a little conversation about the possible risks associated with innovative technologies that we showcase here. So are there any concerns floating around in your worlds, anything that keeps you up at night? Well, yeah, I'll go back to sort of the control aspect. I think that there's a real danger that we are losing control of the computing technology. And when I look at what's happening around the world, social media and the use of AI and all these computers on the edge that are going to collect all this great data, you know, there are great benefits to come from that, but as we've seen, risks. And, you know, that's the sort of stuff that really, you know, 25 years ago when I was getting into industrial research,

I didn't really project that we would be where we are today with sort of wrestling with these issues around, really, are we in control of the computers or are the computers in control of us? So what are you thinking then? I guess the better question is— Are we in control of the computers or are the computers in control of us? So what are you thinking then? I guess the better question is what are your people doing about it? Well, this is why I think make code and things like education, CS education are important. Part of what we're doing is making it easy to make this invisible fabric that runs our life visible. Like to talk about what is a computer, what is networking, security and privacy. People need to become more educated and starting at an earlier age

to understand with age-appropriate content, you know, what is it that this smartphone is doing for me? What are the dangers in being out on social media? And how do I take charge of my data? How do I take charge of my programs. And I think we're still sort of in the infancy of thinking about that at a scale. And so I think the education is really critically important. And that's what we're helping to do through, I think, things like make code and giving people the tools and the vocabulary to talk about computers and computing so they can be just informed citizens, really, and they can help inform policymakers.

So I really do think, you know, we need to start earlier so that the people become educated. Right. Did you address any of this when you were completing your PhD? I started working on it sort of towards the end, I guess. I had run this really cool little workshop in Nova Scotia with Kathy snow from Creighton University I think she's moved now I what we did was we just got girls in their parents so the elders as well in that community to basically make traditional indigenous garments with

technology in them and you know I took a step back in that workshop and interesting because I'm usually the one leading it and so I let a couple of my friends angel Alex sort of run that workshop and, it was just amazing to see the difference it had made for those girls. These are huge questions. And I'm glad you guys are thinking about them. It's story time. As we mentioned at the outset, you're both working here at Microsoft Research, but at different points along the career path. So tell us how you landed here. Just give us a kind of short story of your journey to MSR.

What brought you here and why? I've been in industrial research for 26 years now, 20 years at Microsoft Research and six years before that at Bell Labs. Bell Labs was amazing, but I was thinking, well, programming languages and programming tools, where's the future for that? And Microsoft was clearly the answer for me. And so 20 years later, I have to say that was a pretty good bet.

And it's still, I think, the great place to work on programming tools, programming languages, software engineering. But really, I decided I was a programming language researcher. And around 99, I decided it was time to sort of go someplace where that would be really valued for the long term. How about you, Teddy? I mean, as I mentioned before, I was always kind of entrepreneurial throughout the PhD, and then I came to MSR as the intern, and I basically continued sort of in my entrepreneurial spirit, and that's basically what sort of rocked the boat for me kind of working here. And then obviously working with Tom Peli-Mihau asked us to, I mean, everyone was just fantastic to work

with. So, I mean, I really couldn't say no. Right. Well, let's go a little further on the personal aspects. Tom, I happen to know something interesting about you that I just learned like 40 minutes ago. But is there anything that people might not know about you and might not assume about you that has influenced your life or your career, Tom? Well, people know I have a really loud voice and I'm prone to sticking my foot in my mouth. But yeah, it goes back to junior high where you could get out of one gym class a week if you did something in fine arts or singing. And I was like, oh my gosh, there's nothing I hate more than the gym locker room. So let me give it a chance in

choir. So I've always had music in my background and my parents were amateur musicians. And so sort of a musician in computer scientist's clothing. I like to compose. I've sung in the Seattle Opera Chorus. I play some piano badly. I play electric bass in a rock band. I even composed a song about the micro bit, which is still under production. And then we went to a conference in the UK and I worked with a band there to premiere it at the Micro Bit Education Foundation's first conference. So yeah, the music has always been something for me that is a creative outlet, but also an inspiration for new thoughts. And I'm thinking of more ways to weave that into my research. Soon. Yeah, sooner than later.

Teddy, what don't we know about you? I think I've mentioned I'm a huge comic book fan, right? I mean, that's like my whole jam is comic books. That's what's driven me on a lot of things that I make, especially Iron Man at the beginning. Huge Wolverine fan because he's Canadian and he's short, kind of like I am, right? And he's from Alberta, which is also where- Wait, wait, wait, wait, wait, what? Yeah, Wolverine is my favorite comic book character. Yeah, but he's from Canada?

Yeah, he's from Canada. He's from Alberta, in fact. Wait, they're that specific? Oh, yeah. I had no idea. If there's a scene in one of the X-Men movies where he's actually in Alberta. So I actually got into computer science through this computer animated show called Reboot,

which was from this company called Mainframe in Vancouver. And it was like teaching you basic computer science concepts without ever knowing it. And then when I first got into university and I was learning all this computer science concepts, I was like, oh, this is actually all the characters I loved in, like, Reboot are actually concepts.

So it made it really easy for me. And that one, yeah. So I really love Reboot. That's awesome. I'm learning so much. I didn't know that. I did not either. That's the point of it, right?

At the end of every show, I ask my guests to share some parting thoughts with our listeners. And this could come in the form of advice or, I ask my guests to share some parting thoughts with our listeners. And this could come in the form of advice or inspiration. I think this whole podcast has kind of been that for me. Or you could predict the future. Oh, boy. Definitely don't want to touch that one.

No, I just wanted to say, because I didn't get a chance, you know, why make code? We came up with that name, and I think it really ends up talking to the idea that what is it you want to make? What is it you want to create? And the idea of putting creativity first and then bringing the computing and the coding into that context. A lot of times when we lead with coding, I think there's a certain set of people who get turned off. But if we lead with making, whether that's making an art project or music or a fashion garment, or you want to measure something about the environment

to realize a better irrigation system, you have a system, you have a problem in mind, and the computing and the coding are part of that. So I think it's really good to think about domains and problems first, and then bring the coding in and support as just one tool. I'll follow up with what Tom said, thinking about domain. I think for future researchers who kind of want to do what I did in terms of entrepreneurship, I think the biggest lesson learned for me was

at the beginning, I was probably more focused on the entrepreneurial side and business, I would say, as was for the last couple of years, it's more focused on impact, where how can I go into community, how can I make change? One way is going through MakeCode and using that in workshops in Nova Scotia or working with designers. It's really about impact for me, the entrepreneurial side. So emphasizing the impact of the things you're working and thinking sort of the bigger picture is really important.

MELANIE WARRICK- Tom Ball, Teddy Sayed, thanks. You're welcome. This has been so fun. Yeah, thanks for having us. Yeah, this was great. It was pretty awesome. To learn more about how software engineering researchers are working to simplify the programming experience, visit Microsoft.com slash research.