Embedded - 1: Start Tinkering

Episode Date: May 16, 2013

Featuring Elecia "El" White (@logicalelegance), Jen Costillo (@rebelbot @r0b0ts0nf1r3), and Star Simpson (@starsandrobots). This show was recorded at DesignWest, the embedded systems conference. Board and parts vendors: Sparkfun and Adafruit (both have great tutorials) Getting started boards Arduino  (and AVRFreaks) and Raspberry Pi Light things up with ThingM Find components (and datasheets) at Digikey. And Mouser, Future (Octopart). Avoid Alibaba.com. Amazon has a wide range of electronics tools at generally ok prices. For sharing: Make Magazine (ideas in writing), Github (software) Open Design Engine (hardware)

This is Elysia White, and you're listening to Making Embedded Systems, the show for people who love gadgets. We did our inaugural show as a session at Design West, the Embedded Systems Conference. Star Simpson and Jen Castillo joined me to talk about how we each got into tinkering and why we loved it. Next time, Jen and I will talk about voltmeters, oscilloscopes, and logic analyzers. My name is Jen Castillo, and I'm the founder of RubbleBots, which is a software consulting firm. I am primarily a software engineer with 16 years of background in firmware engineering.

I've done my own boards a couple times, but I primarily do firmware. The reasons why I'm here today is because I started off in tinkering because I was working as an assistant to a CTO, doing prototypes and patents and not a lot of hands-on work. And it turns out that that's what really gets me going. So he recommended that I get some hobbies. So I started doing some crafts and more electrical work that I really needed to do to satisfy what I needed. But that quickly gave way to me realizing that what I really wanted to do was create a startup. So I went back for my MBA and what I found was all this electrical tinkering hands-on work that I was doing helped me illustrate my points very effectively

as I was doing my entrepreneurship classes and some of my ethics classes. So it all started rolling together. When I finally finished my MBA, I focused back on my projects, but mostly from a technology and art perspective. So I've done a couple of projects, mostly in wearables, wherein one is called the RFID Dance Project, where I'm translating dancers' motions into real-time MIDI music, and the other is making dancers' costumes that react to light and motion. My name is Star. I am the creator of a project called Taco Copter. Some people may have seen it. One-click taco delivery to your location via flying robot.

As a result of starting that project, I'm a hardware engineer in general. I've gotten more into flying multirotors, so that's my current hardware fascination. But more broadly, I've been tinkering with hardware since I was 14, and I tend toward the hardware side of the scale here. And as I said, I'm Elle White. I wrote Making Embedded Systems for... I'm sorry. You need to bring the microphones closer in.

All right, closer in. I wrote Making Embedded Systems for O'Reilly, and in my professional life, I am an embedded software consultant and founder of Logical Elegance. I've worked on DNA scanners and children's toys and a gunshot location system. Just really, really cool stuff. My free time, I've always wanted to be a mad scientist.

Not so much to rule the world, but more just to amuse myself and my friends. I've noticed there are a lot of things that I want to have that don't quite exist. And so I try to make of them. And that's how I get into tinkering. The first time I got into tinkering was a TTL UART to thumb drive, a data logger that would log information from TTL so that my embedded systems could spew data.

It was when USB was new, USB 1. And from there, I went to light up anything, light up shoes, some nice high heels. Last year, I presented on Christmas lights that stay lit all year long, changing for the different holidays. And that was at Design West. We had a good time, but it was a project I'd been working on for years. Lately, in my professional life, I've been doing more health-related things. And that has certainly bled into my tinkering, as I've been working on a t-shirt that helps with posture. I do work on my computer most of the time, and I certainly hunch over like everybody does. But before I talk too much about that,

Star, why did you get into tinkering? I started tinkering because I grew up in Hawaii in kind of a rural area, and I really wanted to learn about hardware. I loved Star Wars, so I wanted to know how the robot worked. And teaching myself through tinkering was pretty much the only way to get access

to understanding how to deal with hardware. What about you, Jen? Like I mentioned, I've always been very hands-on from a young age. This is probably going to date me, but I'm used to having to build my computers from a kit and then getting a chance to program them. And that was just the household that we grew up in. So it was pretty natural that when I got out into the real world, being very cerebral wasn't going to cut it. So you mentioned your dance project.

I want gory details, you know, protocols, gory details. So in 2007, I came up with this idea, I think it was New Year's Day, that I was tired of seeing dancers always being forced to respond to the music that they were hearing. Instead, I wanted to flip that paradigm and make it such that the dancers made the music respond. So that's where the RFID Dance Project came from. Each dancer was strapped with an accelerometer, a Zigbee wireless, some batteries, and a microprocessor, or microcontroller rather. And the data was then sent over to different modules on the dance floor. And that would allow you to contextualize the movement that was happening.

So there was a protocol that happened between those boards on the floor and the tags on the dancers. In addition, then all that data was sent back to a central processing unit that changed all that data from movement into recognizable MIDI notes that were then played out in real time. As you can imagine, one, there's a lot of latency potentially happening. Two, there's a lot of power issues that you need to think about, particularly if you deploy this system onto a stage anywhere in the world, you need to make sure that you have batteries or power readily available when your batteries suddenly run out five minutes before you go on.

And then, well, it turns out those technical issues were the least interesting things. The thing that took me the longest time was the gesture recognition that I was working on. That took a lot longer. Even though I was managing a team of engineers across the country, I had hardware engineers in Detroit and Utah. I had advisors elsewhere in California. I had a total of three engineers I was working with on the gesture recognition portion, and that took a really long time,

not only because I was working with small microcontrollers that didn't have that much memory, but they also didn't have that much processing power. In addition, I didn't know much about how to work with physical computing in that regard. So one of the problems is how do you do gesture recognition on a small microcontroller where you don't have that much insight into the system? I didn't purchase JTAGS to help me do that. But what I did find out really quickly is I needed to figure out a way to capture motion samples so we could test out new algorithms and ideas that we had.

So I would spend most of my time with my team in California, buying them dinners, buying them equipment, giving them LinkedIn recommendations to keep them going. And so that's at least during that time, we're talking about 2009, a lot of people were out of work. So it was very easy to find people who are eager to do something if they were unemployed or they didn't feel like they had enough experience to get a new job. So LinkedIn recommendations worked really, really well at that time. But that's pretty much what happened. Well, and taco copters.

I just discovered that you actually have delivered some tacos with copters. But tell me more about the hardware. True. The hardware is interesting. Mostly the hardware that we were using was off the shelf, actually, for TacoCopter because you have a lot of robustness and reliability issues. And we wanted to start with a good, stable platform. So that meant going commercial. But for that, we did prototype the taco dropping mechanisms, because you can't really buy that off the shelf, per se.

It doesn't really exist yet. So for that, that was fun. I got to meet mechanical engineers and aerospace engineers who could give advice about the history of parachute design, something I'd never looked at. You think about it, right? You're going to deliver a taco to someone. How do you get it to them, right? From a miniature helicopter. Well, you can't get the vehicle too close to them, so you're either going to have to lower it or drop it to them, and there are issues with

parachutes. They blow around in the wind. They don't quite land straight down under where you drop them. The winch has other issues because now you have, like, a double pendulum that's not anchored to anything. And if you're delivering a reasonable number of tacos, that pendulum has a lot of mass to it, and the vehicle doesn't because it's as light as possible, so they can actually drive each other around. Every time you say tacos, I want to laugh. That's a laugh. That's totally fair. So for my project for the posture t-shirt, I have five accelerometers and five motors.

And they're placed across a T in the back. And a microcontroller monitors the accelerometers to see if you're not in a good posture. Initially, there was calibration. I thought, well, for now, I'm just going to have to require people when they turn it on to sit up straight. And later I'll figure that out. Because, you know, with tinkering, you get to do the fun stuff first. But when I talked to a physical therapist about that, she was amazed at the idea that everyone could then have their best posture.

Because not everybody's posture is the same. And when she uses braces to make people sit up straight, it goes bad because the braces don't work for everybody. And the idea that your posture changes over the day and you could readjust what was good was really good for her. And so I got even more excited. I got the accelerometers working despite the fact that they all had the same I squared C address and I got one motor board working with the help of a very friendly and out-of-work tech but I'm not so good with the soldering skills and the motors and the FETs they're a little complicated for my circuit building ability and I did some things on the business side too. I did the slide deck and showed some of the clinical studies.

And I even talked to one of the consumer fitness companies. They were pretty interested in the idea, interested enough to make me start wondering, what is my tinkering time worth? And how do you go about selling an idea? It was very cool. They eventually passed, but I'm still kind of excited about it. The next step, though, is to fab some boards so that I don't have to build the motorboards with solder balls and hot glue, because that's

my medium, except fabbing real boards means sinking actual money in and I don't know if I want to do that. I kind of want to make 10 shirts but I don't want to do manufacturing. I don't want to do large-scale production. That's not fun to me and tinkering is supposed to be fun. I'm not even sure I want to do a Kickstarter. That's too many. So right now the idea is shelved as I figure out what my next steps are. I hope that SparkFun's going to make a FET board. You turn on a GPIO and it makes a little coin cell motor go.

But we'll see. Maybe I can talk to someone here in Embedded System to doing that for me. So that's our projects and that's our introduction. but I think now we're going to go into the chat portion. Jen, tell us about your gear. What do you think is the most important gear for people to get when they start tinkering? I know for many of us it takes us years to amass the soldering station that we love,

the little tips and tricks that we have for doing reworks very quickly. For people who haven't soldered before and are trying to get into this type of work, I usually ask them, how comfortable are you with soldering? How much money are you willing to invest? And if you really do want to solder, are you going to be doing a lot of through-hole or are you going to be doing a lot of through hole or are you going to do surface mount? Because there's minor variations in all that equipment.

If you're just starting out and you're really uncomfortable with burning yourself or making a lot of mess, I highly recommend breadboards. They come in many, many sizes. You'll have to get a new breadboard for each project. You'll probably invest in a hot glue gun, at least I know Elle does, to keep the wires in. You'll buy a jumper wire kit and the parts that you need. But for me, I'm much more of a solder person. I get lots of, I have alcohol, flux, different, you know, different soldering wire, solder paste, some horsehair brushes that I usually cut off to make them really short so I can actually get rid of all the flux at the end I've invested I

moved up from my from my radio shack soldering iron to a Weller and now I have an oaky I can't remember which number it was that I have but I have some nice fine tip work that I can actually do in 0402 at home. Star, what do you have on your table? I absolutely have the full soldering setup because I think it's totally invaluable for tinkering and prototyping. But one thing I think I would add to that list is a hot plate because at a certain point, so many components today are available cheaper or smaller in SMT packages

that if you have a little hot plate, you can stick your circuit board onto it and melt the solder and deal with all the parts at once, or make your own surface mount boards. And I find it really useful. I probably use mine on a weekly basis. It's a little challenging, I think, at a beginner stage, but it's something to think about and look into as you look for gear you want to tinker with. I was always hesitant with the soldering iron, although I do have one,

and I am capable of putting on my own headers and even occasionally modifying more than that. But I was spoiled at work with some awesome technicians, and I know that I'm pretty incompetent with the soldering and the kitting and the cable making. Oh, I spent a lot of time cable making last summer. It was horrible. Even choosing the right parts, I'm not that good at. So I like having a hardware engineer at EE. So it was really great for me when boards started coming as components, when SparkFun happened, or Makershed, or Adafruit,

any of the ones where you get a board, and it has the capacitors and the resistors and all of that other stuff that I just don't care about on the board. So I love using the Lego blocks. And I'm intimidated by the soldering irons. I think there's some more stuff that are in my toolbox. You didn't mention having a voltmeter at all. Being able to own things out? Tell me you don't live without that.

So the digital multimeter is probably the easiest and cheapest thing that you can add to your toolbox that won't burn you and will save you time. It can be as cheap as $5. It can be as expensive as $1,000. It doesn't have to be a fluke. It doesn't have to be that fancy. You're really just looking for voltage, current, and resistance. And if it beeps, even better. Another thing to consider is whether you should or should not invest in an oscilloscope.

That's a really big expense. I saw someone's head turn. I borrow mine. I borrow mine from a friend who doesn't know what's in his garage. I'm pretty sure he's going to ask for it back today after that comment. Oh, he probably is. But there's a product out there by Seed Studios. That's S-E-E-E- e d and it's called the dso i think there's a version two out it's basically a pocket oscilloscope it's one channel but and

it's about a hundred dollars but it's ready to go another option out there that's relatively low cost for logic analyzer is a product called Logic by Salig. It's basically 8 or 16 channels of logic analyzer. You plug it into your Mac, PC, Linux box, and you run the software and you plug in this USB

pod and it will give you out of the box logic analyzer. That is a protocol analyzer too, doesn't it? It does do protocol. So you don't have to do the bits and the hex. Yes. That's the fun part though.

For some people. Some other options, some other things that always come into play when you're tinkering is how do you power your device, especially when you don't want to go through a lot of batteries and you don't want to deal with custom batteries out of the box. You can get a bench power supply on Amazon. They're about 70 bucks. But if you don't want to spend that, you just go get a wall board, cut off the end, plug it into your breadboard, and then you have the power that you need. If you need to temperate... Wait a minute. You have to check the voltage in the amps, because I've done that, and it's been really

bad. Would you care to elaborate? On the Arduino, I had a wall wart, and the Arduino takes 5 volts, and I don't know how many amps, but I had 2, and I got the 3 volt and the

5 volt mixed, and it didn't go well. So be sure to read your wall warts before you tear them apart. And check with your DVM. Yes. That's where the DVM really is great. I mean, I do also have a bench supply. $70 used, and I'm pretty sure it was used in the space program in 1960.

So you mentioned the Arduino. What other eval kits have you used? Which ones do you like? Well, I like the Arduino. I used to think it was silly, actually, because it's an AVR. It's an Atmel chip. There's plenty of those out there. Why have this super spiffy board running this not quite C language thing.

But then I found a person who had built a system, kind of like I do, had taken parts from SparkFun and put them all together and built a pretty neat little gadget and didn't even realize, asked me then later what embedded systems were. And I pointed to the gadget and said, that, that's exactly what it is. People get into Arduino and don't even know

what they're doing, but it turns out just fine, thanks. So I like the Arduino, but I also like any dev kit I get for free. That's kind of my bar. Embed has a nice one that they're probably giving out for free, and they have a good compiler. Oh, I also like it when the compiler's free. So MSP430s. Expresso boards are probably the one dev kit that go from goofing off tinkering to working professionally, I keep those around in great quantity

because I tend to give them to clients because I use them up. What about you, Sarah? Well, just before Arduino, there was a really great board called the SDK500, which is about as futuristic as it sounds. It had eight buttons and eight LEDs and could take any size atmel chip and you could really test your programs on it. It had, you know, eight buttons and eight LEDs and could take any size Atmel chip and you could really test your

programs on it and it was great. But since that, I think the Arduino has filled in a lot of spots, honestly. And I'm curious, given you guys both work really hard, how do you find the time for all of this? How do you find the time to think? Well, like Jen was saying,

if you... It depends on what you're doing at work. If I'm programming at work, I am not programming and tinkering. If I'm writing FDA documentation at work, at home, I want blinking lights and moving motors. So tinkering provides a balance for me. And it isn't about finding the time. The time's there. It's about choosing to tinker instead of choosing to, well, for me, it's read, sci-fi, mystery, anything. What about you, Jen? You're really busy. You have two jobs, really.

Yeah. So, like I said, most of my... So my second profession was a hobby that turned into a profession, was dancing. And so what I do is I just try to merge as much of that artistry into my technical work. So that's where that comes from. And what about you? I mean, you have a job too. I do, it's true. I guess you make a really good point about, you know, the time is there, and it's just how you choose to use it. I think when you feel that something needs to exist, it's not so much a matter of, you know, how will how will you make time for it but when will the time be free jen was talking about having people help her i don't think they

like to be called minions but that's what i call them mentally um how do you find people to help you uh and how do you bribe them i might be simple, but the pizza bribe is pretty much the best one I've found. A lot of people are happy to see things work. They're friends of yours, they have hands, you can tell them how to use them. And an evening of pizza and building will typically work pretty well for making a project go. You told me a little bit about the quid pro quo. Exchanging services. Yeah, so that's another fun one. So coming from hardware, I tended for a long time to think of coding

as something that was sort of like a necessary issue to deal with to make the hardware work. Just writing the software just makes the, you know, makes the voltages do the right thing. So I decided finally that I wanted to pay more attention to software. I wanted to learn more about, you know, what the best practices were and how to do it right. And a friend of mine who's a software engineer happened to come to me and say, Star, I want to learn about this electronics thing. What do I do? Do I take apart a fan? How does this work? And I said, why don't you just come over? But I have this software project that I don't know how to finish. So you come over. I'll teach you what

I know about electronics as much as I can, as much as you care to know about. And in exchange, I'd love, you know, if you just, like, can read over this code and tell me what you think of it, that'd be really great. Tell me what you think. And I see a lot of projects get done that way, actually. That's how most of mine have gotten done, although some of the favors are outstanding.

You know, people who have helped me, who I know someday will call in the favor when they get their brilliant idea. So, shall we take a question from the audience? Anybody? No one? Okay. So Star, I know that you recently came back from China.

Can you tell us a little bit about your experience of tinkering in China? Sure thing. So for some background, I went to China to start a hardware startup Can you tell us a little bit about your experience of tinkering in China? Sure thing. So for some background, I went to China to start a hardware startup making plush electronics for children to learn to tinker. Highly topical. Turned out it's a strong interest of mine. And while I was there, I did the thing where you're in Shenzhen, China, and you go and shop for electronics.

And everything was different. The whole experience was totally different than what I could possibly have expected coming from the US. It's, you know, the Shenzhen electronics market, which many people have heard of, is a building at least as big as this one, but many stories taller. And it's a complete, like, bazaar in that there are, you know, stalls and vendors. And you can go around to different people and, you know, check prices. But everything's live. Everything's happening in front of you.

And I don't want to talk about it too, too much. Because, Jen, I know you went to the same market. So what's the name of the market that you went to? The Shenzhen Electronics Market. So was it SEG or SEC? Yeah, I've heard it called SEG, but I never knew what that stood for. I still don't know what it stands for. But much like she had mentioned, it's a bunch of stalls for people.

But the one thing that... So if you've ever go to Shenzhen, it's like electronics heaven meets Vegas. There's lots of lights. There's lights everywhere. But they all turn off at 10. There's lights on cars. There's lights on people's homes. Just everywhere. But they all turn off at 10. There's lights on cars. There's lights on people's homes. Just everywhere.

And if you want that look, you go to this market. And you go up to the top floor. And there are people that will sell you reels upon reels of weatherized, addressable lights. You should bring me some back. You fit them all together. And you can even buy a remote control, a wireless remote control that will let you control them to whatever color, whatever pulse rate, and so forth and so on.

The problem is that when you go to Shenzhen, you have to verify everything. People will sell you a USB flash drive that you think is 8 gigabytes. It will be 8 gigabits. You have to verify everything. A friend actually was sold, I think, like a terabyte SD card, and he knew it couldn't possibly be for real, but he bought it anyway because it was cheap enough and found out that the SD card had a circular file system on it.

So you could just write and write and write, and it would, you know, loop around and start overwriting from the top. And sure enough, you could drop a whole terabyte worth of system on it. So you could just write and write and write, and it would loop around and start overwriting from the top. And sure enough, you could drop a whole terabyte worth of data on it, and you'd have, you know, whatever, 500 megabytes later. I think I've written that file system. Yes? When you go to Shenzhen,

do you have problems about the communication? So the question was when you go to Shenzhen do you have communication problems? That's a really good question because yes, I don't speak enough Chinese to nearly be able to claim to buy electronics effectively. So I developed strategies. At first I actually drew pictures of components that I wanted and values to go with them, because I don't know how to say resistor in Chinese at all. But what I found worked in the long run was DigiKey part numbers. Actually, like Lingua Franca, they're common, they can be looked up. There's DigiKey Hong Kong that you can look up,

you know, the translated name. But that said, probably my proudest moment in Shenzhen was the day I bought a handful of 7805s without having to resort to gestures or part numbers. Yeah, I had a friend that spoke on my behalf and was much more familiar and had their preferred vendor that he'd gone to several times, but still always was checking to make sure that everything was working.

I found it similar to when I go to Akihabara in Tokyo to buy electronics parts. You really need to be just very clear and test, although I generally don't test as much when I'm in Tokyo, to be honest with you. Yeah, and because I'm a aerospace engineer in Seattle, and I, even my boss traveled to China, and the person there, if the English is good, do not understand engineering. But engineering do not understand good English. So it's hard to communicate when we do the Chinese airplane project. It's hard, yeah.

So the comment was that if the locals' English is very good, they're generally not very strong on English. And if they're really strong in engineering, but if they're very strong in English, they generally are not very adept at speaking English. And so that ongoing communication problem is there. My only relation to going to China, my book came out in Chinese last week. Congratulations. Okay, so I haven't been to China,

but I have been to the Expo floor at Design West here. What are you looking forward to, Jen? Where are you going to go first? I'm an Atmel person, so I usually go there, see what I can get for free. Some other things that we didn't talk about

were awesome vendors. We mentioned digikey and sparkfun and ida fruit but there's some some places to avoid i found uh many in many cases alibaba was not always the best place to go to get electronics in bulk but but when you search for coin cell motors there you can get them for a penny. You mean I can't actually get them for a penny? You may never get them. Never get them. They say order 1,000, and you think, oh, they're just a penny, no big deal.

No, they never really arrive. So I agree with Jen. I haven't had a good experience with Alibaba, but what about you? Good, bad vendors? Any luck at all? Has anyone received anything from Alibaba? I think it's because of communication issue.

Because Alibaba has some good supplier and a bad supplier, but sometimes it's hard to identify them. Yeah, I agree. Sometimes Alibaba does have good suppliers and bad suppliers, but sometimes it's hard to identify them. Yeah, I agree. Sometimes Alibaba does have good suppliers, and sometimes they have bad suppliers, and the communication is very difficult. Yeah, last month I helped start up in Seattle

to get some components from China. We looked at Alibaba, called for five suppliers, but choose one supplier, the price is good, the communication with the vendors is critical. I mean, you don't just plunk down your money to buy 1,000 motors. You do need to vet them. But you were telling me about Octopart as a website. I've never heard of it. There is. Octapart is one of my favorites. I'm sure many people have had the experience of attempting to search for a part on DigiKey, and it's certainly sort of the pre-Google

search experience. My coworker once said, never search. Never search DigiKey. Use the product index. Just click through. Never search. Octopart set out to solve that problem, and they are basically an aggregator. They search Mouser, Newark, DigiKey.

They search many different distributors, and through them, you can also do price comparisons, which is super useful. So you can find the best price, you can find the best supply, and then I end up, people say, do you use Mouser, do you use Newark, what do you use, do you use Jamco? And I say, well, it doesn't matter, I just pick per part and go from there. So it's kind of like kayak, but for electric travel.

Exact parts, precisely. What's kayak? Kayak is a travel... Why am I hawking travel? I would add one note about buying parts from China and this is something that you would absolutely need someone who spoke Chinese fluently to

get through but while I was in China there's a Chinese version of eBay called Taobao and I've had the best time with Taobao. It's like a local seller thing, kind of like maybe like an internal to China eBay, or sorry, internal Alibaba-ish. And while it was in China, a comment about the Chinese web, you tend way more often to be able to get a live person through the website, live chat.

It's just you expect it from a website in a way that you expect to be able to search for the thing you're looking for on an English or U.S. website. And so by being able to talk to people who are your sellers, you can get a much faster sense of what their supply is like, what the quality you're going to get is like, and buy things via Taobao. Do you think we could convince ST Microelectronics to do that for us in the U.S.? Because that would be cool. Yeah, that would be pretty great. Taobao and Alibaba are the same owner. Same brand here, please.

Ah, Taobao and Alibaba are the same owner. Well, that's good to know. And I know it's possible to get shipments from Taobao to the U.S. via resellers. Because China labor is cheap, so that's why we have to live for our customer service. So going back to your experiences back at Shenzhen at the market, were there other things that you were buying other than parts? I know that I spent a lot of time buying tweezers and Kapton tape and things that usually cost $10, $20, $30 here cost $5, $6.

So did you come back with a suitcase full of Kapton tape? She totally did. I did. She showed it to me triumphantly last weekend. Look at my Kapton tape. Yes, that's great. So how do you make sure your Kapton isn't just yellow scotch tape? Well, one of the problems that you tend to have, I mean, obviously that's the same problem.

So you end up peeling off some. You end up seeing whether there's electrical conductivity. You usually kind of go to these markets with some way of testing in some way. The main problem that you have with the Chinese Kapton tape potentially is usually the adhesive. That's usually the bigger problem is the cheaper adhesive. At least that's what I've found thus far. But I got a two inch bolt of it for about six6, which is unheard of. Does it work? Well, it worked when I was there, but I haven't had a project

to do yet. So if anyone has any project ideas that might involve Captain Tave and an Altoids tin, see me after the session. Let's list the gear we have and let you choose what we're going to make next. Where do you get your ideas, Star? Tacos. Where do I get my ideas?

I would really love some firm and confident answer to where I get my ideas. I don't know. I think they show up on Tuesdays at 2 in the morning for me as frequently as anyone else. I think they show up on Tuesdays at 2 in the morning for me as frequently as anyone else. I think one thing to recommend is just look widely at the world of what people are building. Read Hackaday, read the Spark Fund blog. Look at what's out there.

As often as not, some project will pop into my head that's like, oh, what if you crossed Project A with Project B and you had you know some wearable you know what you would call and then you'll try it and sure enough you're tinkering what about you I I find I have a similar kind of inspiration usually it's because I have a particular problem that's irritating me that I want to solve or I want to solve it in a creative way usually for the benefit of the arts but the dancers with the dancers yeah I do everything for the dancers of the arts. But the dancers. With the dancers, yeah. I do everything for the dancers. But I find it curious that you're mentioning, like,

oh, you see someone else's idea, or you see how if you can merge two ideas together. I'm wondering, you know, do you do a lot of collaboration? Do you try to, you know, when you see something, do you decide to make or buy it? You know, how do you go about, when you see parts in the world that already exist, do you go ahead and replicate them

or do you decide to either procure them or buy them? I'd actually be really curious to hear Elza's opinion on this. When I'm working with a client, with a startup in my professional life, I always say buy. Startups are too time constrained to be able to make everything they want to. So you should buy everything you can. But I tinker because

I want to learn more. And so make versus buy on that is different. And I buy most of my hardware. I don't make it because I'm not very competent at that. But even with software, there are some, do you buy it off the shelf or even just get it free off the shelf and have to invest your time in understanding it? And for that, you know, I write a lot of my own, even library sorts of things. And I find that it helps me in my professional life because then when I need to go have a circular buffer library, I have one. I know exactly what the licensees is and I can just pop it in and use it right away. I don't have to worry, is it GPL? Do I have to look up what it is for

a client being able to use it, I know. And it helps me, it not only keeps me sharp technically when I'm not programming, I get to program at home, it helps me even with interview candidates, or interviews where I can show them, here's my project, here's what I did. I did this part of it. It's not a question of, you know, you're in a big team and what, you know, did you write the driver or did you write the whole operating system?

Because that's a good question when you're hiring someone. And if I have a project, I can just say, here, I did this. And sure, they'll see the solder balls and the hot glue, but they can also see that I have a good coding style and that I can put it all together from ground zero and you're making an excellent point these days more and more startups and even later stage companies are interviewing by having you present so having something that you can talk about in depth without violating your employment agreement is really important not only does it show a level of

maturity in your work, but it's something you can talk about very honestly, what your mistakes were and what you would do differently, and particularly because it was your money that was probably on the line at the time. Employers are going to take that much more seriously because of the level of investment that you put in there. So for several of my projects,

I have presentations that are ready to go at a moment so I can go present because there are a lot of issues that come up during those interviews that will make you rise above what other applicants have. So particularly when we're in a recession, these are the types of things that will help you greatly. Well, and you took one of yours to a job fair and got to show it off to a bunch of people, didn't you? Yes. So one of the things that you can do with these tinkering ideas that you do, maybe you don't turn it into a startup,

but instead you just share it with the world. One thing you can do is there are hacker fairs. I know that Hacker Dojo has pretty much an annual job fair that basically the employees, the prospective employees basically have a table, much like a science fair, to show off what they can do. And employers walk around and try to find their prospective candidates. Kind of like a science fair for adults. Yeah. Right? Have you found that tinkering helps you

get jobs, keep jobs, be amused at jobs? Yeah, I suppose so. I put a lot of my products on my website, which I think is a really important, like, end part of having a project. I think I find a lot of value in documenting projects. I used to write a lot of instructables. And the part of me that really loves tinkering is amplified that much more when someone reads something I've written about my project and goes off and makes their own version of that. I found that exceptionally rewarding. So there's that.

Tinkering children. Tinkering children. Tinkered children. That doesn't come out right, does it? The idea that your projects live on beyond you. It's pretty cool. And then you have to find the one that they tinkered on your idea and somebody else tinkered on their idea, and then you'll have grandchildren.

So I guess in as much as those projects have, you know, people have talked to me about that or whatnot, it's become part of who I am and what I do. I suppose it's been helpful. So some of the places that I show off my work is on GitHub. There's also opendesign.net, which is more geared towards hardware,

so mechanical design, PCBs and so forth. Those are places that you can essentially build your portfolio that when you're talking to employers or just talking with your friends or looking at people to work with in general, they can go up and look at that, grab a copy, and create derivatives from there for the value of your work. So you have these projects, and you start them,

you're taking them through, you're collaborating with people, you're documenting them. How do you say when you're done? When I get bored. When I get bored. That's the difference between tinkering and professional life. I like to ship things. I really like to ship things.

But at home, not so much. I go from project to project because this is my free time. This is my fun. And when a project stops being fun, I put it on a shelf. It doesn't always stay there. But I like the design

and I don't like the production. So once it gets to the spot where it stops being fun, I get bored and wander off. What about you, Star? How do you decide when it's done? Yeah. I mean, similar. When the fun part's over,

when you've learned what you came here to learn, when it's documented at best. I used to want to be a CEO and have my own startup, and I thought this tinkering would lead to that. I thought I would be in charge. But I did the management track and the director and got to tell people what to do. I liked the mentoring and I liked the program management, but I didn't like the politics and I didn't really like the in-charge-ness.

I liked to get my hands dirty and to do stuff. So I think I've given up on the idea of founding a startup, maybe a CTO as a co-founder, but I don't want to be a CEO. Jen's got an MBA. Maybe she has a different answer on this. No, I don't. Actually, one of the things that we gloss over really quickly

and I want to bring back is the idea of you create your own modules, you've done all your own work, so you know that you can give it away freely. One of the things that we didn't talk about is if you accidentally borrow things from work. Oh. Oh, scopes, code, whatever. You borrow the oscilloscope,

you maybe accidentally might have violated your employment agreement. A couple of resistors, they'll never notice. Hey, I work for myself. It's okay. The point is that I just want to make sure that the audience is aware of the fact that if you borrow the oscilloscope, if you borrow something that later you end up using as a basis for a patent or something like that, you should check your employment agreement to see how that will be affected, because you may inadvertently be giving your patent away to your employer.

California's got some fairly strict guidelines on that, but it's better not to have to fight the fight. If you're going to borrow stuff from your employer, get permission, preferably in writing, and let them know what your idea is and how it is separate from your job. And if it's borderline, be careful. I mean, they may say, no, we don't want you to work on X because it's too close to our product Y and we don't, you have a non-compete, so you can't. But if you're making light-up shoes or writing instructables or a taco copter or dance RFID thing,

they're not going to care. They're going to be happy. You're learning stuff, and that's good, and it's keeping you happy. So most companies are good about this, but ask, don't just assume. Any further questions? Well, I think we're about out of time. I would like to invite you to come tomorrow at 1 o'clock, and Jen and I will be presenting with a few other people

on sensors in health-related applications, gory details. And I believe, Jen, you've got another presentation after that. Yeah, and at 3 o'clock tomorrow, I'll be presenting essentially top-to-bottom approach on Android sensor subsystems, how to design them, how to use them. It's going to be pretty cool. And Star, you'll just be wandering around the expo floor? Yeah, this is it for me.

Thank you very much. Thank you all for coming. Thank you very much. Thank you all for coming.