Embedded - 228: Pedantic or Andrantic

Episode Date: January 3, 2018

The Amp Hour and Embedded join up to send a holiday letter to listeners. Chris G is ever improving Contextual Electronics. Chris W has a new band: 12AX7. Elecia still has a book: Making Embedded Systems. Amp Hour episodes mentioned in this one: 372: Where Chris and Dave talk about 2017 304: Alexa jokes 281: The first Amp Hour / Embedded show, with call ins 256: The first time Chris W was on the Amp Hour 187: Elecia joined the Amp Hour for the first time Embedded episodes mentioned: 223: Where Chris talks about his new synth habit 227: Talking about Udacity and learning 203: EE Charlie talks about good design We talked about teaching which led to: Short mention of Dreyfus model of skill acquisition of which Chris G’s friend Mel did a great explanatory comic Daniel Spalding’s How to Teach Adults (pdf) Dan Luu’s Learning To Program post Udacity’s Self Driving Car courses Computer vision with Python OpenCV Article on how the difficulty is the point of teaching literature The new art and engineering Function Podcast Hilarious World of Depression podcast Books we are reading! Build Your Own Transistor Radio by Ron Quan The Hobbyist’s Guide to RTL-SDR by Carl Laufer Spineless by Juli Berwald about Jellyfish Lost Art of Reading Nature’s Signs by Tristan Gooley Into the Drowning Deep by Mira Grant (terrifying mermaids) Catseye by Andre Norton Teach Beyond Your Reach by Robin Neidorf Mastery by Robert Greene Understanding By Design by Grant Wiggins and Jay McTighe Making Learning Whole by David Perkins Elecia got a JTrace Pro Cortex-M for herself for Christmas. Chris W got a Moog Werkstatt and an assortment of Teenage Engineering small synths. Chris G mostly got sweaters because Chicago is very cold. BMW now sends YouTube ads via snail mail

🎵 Welcome to Embedded. I am Alicia White alongside Christopher White here with Chris Gammill. Hi, I'm Chris Gammill from the Amp Hour, and we're doing our annual crossover episode, Amp-etted. Amp-bedded. Amp-bedded. It sounds more like embedded. I mean, you guys really get the better. Well, we could call it the Amp-bedded Hour.

It could be the Amp-hour. Amp-hour, well, yeah. It doesn't sound right. Anyways, yeah, Happy New Year. i think we're we're at the end of the year we made it 2017 that dumpster fire as we talked about in the empowers last episode the dumpster fire every year so we made it through that's the odd number of years yeah exactly exactly so what's new out there what's uh anything on fire or uh what's what's going on out in california oh we didn't get the fires the southern california got the

fires we got a lot of smoke for a while and okay which is impressive since oh yeah yeah southern california is 300 miles from us so oh it was like it was specifically that smoke yeah oh well later okay i got the wine country smoke too but and i hear it is very cold in chicago oh yeah yeah it's uh it is uh four there's not enough digits in that temperature yeah right so yeah i was looking at i was i was flying in today and i was like oh oh, 4 and minus 18F. So, I don't know what that translates to C, but it's starting to convert. You know, when it starts to converge, when F and C start to converge, then that's a bunch of F and C, you know?

That's when you start looking at California and going, hmm. Yeah, yeah. Maybe I do want to pay that much for real estate out there, huh? Hey, it got down to 32 last week for a few minutes oh wow okay yeah probably had a water main break or something all right um so in the last year of that enough of that somebody listens to this show uh chris you left supply frame and went to Hologram. Oh, yeah. We have to update our respective listeners if they don't cross over.

Yeah. Like a true crossover. Right. Yeah. So I left SupplyFrame, who also owns Hackaday and FineChips and a bunch of smaller Tindy, stuff like that. And I was doing product stuff there. And now I'm at Hologram.

And I am a developer advocate are you both developers i'd like to advocate for you it's like evangelist yes that's another terrible title okay yeah you know i've been thinking about it's like an app engineer but like without needing to necessarily it's just the enthusiasm not the actual applications so you just exude positive feelings that's right right i'm all positive all the time what does hologram do uh we do connectivity for cellular so um yeah devices that need to connect to cellular towers without uh without a cell phone present so i i o i o I O T yes. Telemetry as well. That's a better term.

Cell modems at one point. Yeah. That's what, I mean, that's what I'm putting on boards these days. So like little U blocks, modems and sequins and, uh,

what else is out there? Uh, there's, uh, uh, I guess, I guess that's an integrated board is the Pycom. Yeah.

So there's like, I'm learning a whole bunch of new stuff and wireless stuff and everything so yeah it's been going good and hologram does the whole thing i mean it's i mean when you have a cell modem if if i were to go out and just buy a cell modem i would have to then get a contract and then it would have to go to a server and i'd have to deal with the cell modem to server translation and then the server infrastructure and hologram does all of that part right yeah yeah there's three pieces so there's hardware there's connectivity and then there's software services right and those it's basically one two or three you could pick any

you don't have to use all you know you can use whatever you'd want because like people don't want to necessarily um use our servers right so they want to bounce it over to aws iot or they don't want to use our hardware and that's fine you can just stick in one of our sim cards into someone else's hardware and then you use our connectivity so it's kind of like you can or you could just use our hardware and use someone else's connectivity but that's kind of weird um so yeah there's lots of choices there and it's uh kind of like you can, or you can just use our hardware and use someone else's connectivity, but that's kind of weird. So, yeah, there's lots of choices there. And it's kind of just offering all three pieces because, have either of you done cellular projects before? Yes. Yeah, it's pretty painful, right?

Oh, yeah, probably ShotSpotter, right? Yes. Yeah, yeah. So, you know that pain. Well, when we did it, we got the hardware. We worked with the cell modem companies. We dealt with the servers. We did all of it.

It was hard. You needed a team to understand both the operations side and the software side. And now that team is being outsourced. And I'm not sorry because as solo providers, it was hard to get the cell modems to work all the time. But as a group having different boards, I can just see how hologram and particle IO sorts of things make so much more sense than having to go it all by yourself. Yeah, there's lots of really...

I mean, particle's lots of really, I mean, Particle's great for hardware. Like I said, Pycom's another one that's great for hardware. You know, and like, yeah, there's tons more providers. I just got a board sent to me from a Tindy seller who's doing stuff that I had never seen before. Like, so yeah, just like lots of people trying stuff. And, you know, and the really crazy thing is it's not, so I posted a bunch of links in the subreddit. i'm looking at laura's stuff as well laura wan and

like i'm excited about tying that stuff into cellular tying that into wi-fi and like there's a bunch of youtubers that are doing that stuff now it's just kind of rf everywhere aren't we gonna run out of uh spectrum space for all these little devices that are hogging all the limited bandwidth for for cellular i mean well 2.4 ain't exactly uh you know the wide open wild west anymore it's pretty pretty crowded in there so yeah it's uh what would i see there was an article about that a while back it was like there's they're starting to use new white space so i think it was 800 megahertz region um because that opened up when the tv stuff went away maybe it was lower than i don't remember what that's somewhere around there it's below it yeah it's yeah it was and um and so

that's kind of open now and then so like laura can do 433 890 through the 886 or something like that and then 900 so that's all in there too and so there's still like the different bands but you know it's just about how much you're actually using. You're not going to be doing like streaming video. I mean, what's the point? Right. Right. Right.

How do I, how do I let, how do I let my delivery guy into my house? Yeah. Right. Well, so, okay. So I've, I've obviously explained myself too much. Please explain yourselves to my listeners. Who are you, Alicia and Chris? I am Logical Elegance, and I used to have a partner with my small consulting contracting. Still do, you just don't pay me.

Well, I mean, you went and got a full-time job. No, it got me. Legally, too. I mean, yeah, you guys are still partners, I think, right? Yeah. Yeah. Yeah. So I'm an embedded software consultant or contractor. And yes, those are different.

With contracting, I usually help small companies implement their napkin sketches so that they have prototypes. And then they can go off and either try them or get VC funding. And with consulting, usually I help small to medium companies who don't necessarily have another embedded software resource. And they can call me and either chat about their mystery bugs or I can mentor their engineers. That's really fun. So yeah, that's what I spend most of my days doing. Contracting, writing software, writing stuff.

Stuff. Well, sometimes I write documents and sometimes I write encouraging little notes that say, globals are a good thing sometimes. You can do it. But let's try not to have quite these many globals. Yeah. And Chris, you took a full-time job? I did. Oh, I guess I should have showed up for that.

This is like the college test. I was supposed to start this year? Yeah. Not only did you start this year, you shipped something. Yes, I went from consultant doing similar things that Elise was doing to full-time firmware engineer at Fitbit. So if you don't know what Fitbit is, get your head out of the sand.

Fitness trackers and smart watches and fitness watches and smart trackers. Yeah. It's great. Yeah, lots of tiny embedded things i'm sure so yeah consumer embedded is fun consumer is a different world if you haven't done it before we had uh well some of your former guests as well of our own gen uh who are the reverse engineering podcast came on and they talked about a lot of that stuff so that was that was insightful

that was good i mean and i had, I thought about it after the fact. I was like, oh, I haven't really talked to many people who have done consumer stuff. And of course, I talked to both of you and just totally gapped on it. Yeah, I mean, I did toys for LeapFrog and that was even more consumery than Fitbit was because their profit margin was tiny.

And a penny really made a difference. I think about that. consumery than Fitbit was because their profit margin was tiny. And a penny really made a difference. I think about that. So, so I have young nephews now and, you know, they just have like so many blinking loud things. And I can't imagine like having to test that stuff all day long. I can't imagine like having to like repetitively hear, you know, what is it called?

The dogs. What are the dogs these days? Pup Patrol? I don't know. Yeah, I'm like learning all these new brands as well. They're terrible. It's all terrible. So, at least you were doing educational stuff.

So, they weren't terrible. They didn't make any noise. They were still chiptuning, right? It's like you can only take so much 8-bit where you're like, yeah. I had, well, speaking of gifts, I bought my parents an Echo and an Echo Dot. And I think I'm more in love with it than they are. Yeah. The story repeated over and over throughout the country.

Yeah, we bought one for my parents last year. I'm not sure. I've never heard about it again, so I'm not sure it actually got put in here. Yeah. But even though I didn't want something listening all the time, I have to admit I do like our Echo. It plays music when I want it to play music, and it tells me jokes and it

occasionally will tell me facts yep and sometimes it'll uh order alexa order a dollhouse uh no i actually used it anyone listening yeah that was the one where that was the story where a bunch of people actually had it order a dollhouse when they were on a podcast or a new show or something so yeah uh you got a neat gift this year christopher i got many neat gifts well i was thinking the gift i got you yes the best gift come on chris uh so what'd you get well if you've been listening to our show which why wouldn't you of course embedded.fm you can go there right now and subscribe i've been messing around with analog synthesizers and she got me oh yeah she got me this uh semi kit although it's mostly assembled from moog or moog or however you pronounce it m-double-o-g if it's got two o's

it's gonna be ooh yeah except it's not no it's actually yeah yeah it's moog um it's a it's not. No, it's actually, yeah. It's Moog. It's a very small analog synth, one oscillator, but it's got their usual filters and stuff, but it's for tinkering. So it comes with a PCB and a case, and you have to put it all together. But the PCB has a bunch of test points on it, and the silkscreen on the PCB is like a block diagram so each section of the synth

whether it's the oscillator or the filter or the envelope generator all that stuff is silkscreen and marked so all the components for each block are ganged together so you can kind of tell just looking at it how it works and then it's got all these signals brought out

so you can experiment with it it looking at it how it works. And then it's got all these signals brought out. So you can experiment with it. It's even got, it had like a piece of perf board as part of the PCB. Oh, nice. Yeah, like prototyping area. And I guess they produced this for like a conference for engineers a number of years ago. And everybody freaked out and said, you got to sell this. So they started selling it. So I've been playing with that and plugging it in my other synths and making some very strange noises.

It's been pretty fun. It's great. Yeah, yeah, yeah. And is that like one where you actually have a keyboard that controls that or what actually does that? It's got some buttons that do a full octave on the front. And then since it has all the control voltage inputs, you can hook it into control voltage equipment.

So there's a lot of keyboard controllers that output both MIDI and control voltage inputs, you can hook it into control voltage equipment. So there's a lot of keyboard controllers that output both MIDI and control voltage. So you can control it with a keyboard. It takes some work and you have to calibrate it, but it does work. Nice. How many different synths did you get for these holidays? Was it four or was there one I missed? It was four. It was four it was four okay yeah because i got three my parents got me two of the little teenage engineering pocket oh those are so fun and yeah and you got me one so so i have way too many john park which showed a picture of this uh box

he made he's at adafruit doing project stuff he had a box he made to take on an airplane to compose stuff to compose chip tunes on the airplane i thought that was awesome also it would be super freaky what are you doing yeah they they look a little weird so they kind of look like old calculators yeah but yeah that's great where do we hear your music uh that music nowhere yet yeah but there is a place you you did start a new music thing. Yes, my brother and I have been working on it slowly. We released one single.

Nice. If you care, you can go to 12ax7.fm. 12ax7. Like the tube. Yep. That's the band name. Nice. You'll get people just for that, you know? It's instrumental guitar rock, and that tune is sort of mainstreamy, Nice. You'll get people just for that, you know? It's instrumental guitar rock, and that tune is sort of mainstreamy,

and the other two we're currently working on are getting weirder. Nice. Yes. Maybe another month or two before we release those. So I was listening, and you were talking about this kit on your show at one point, too. Which episode was that, the latest one? Probably not. Which kit? You were talking. Oh, wait, maybe you were talking about it somewhere i don't know i had a different i heard you oh well i heard you

talking about a synth at some point so yeah i want to refer people to that that one is uh is the he's a little noisemaker synth uh and that was a full electronic kit you had to solder and oh okay do the full deal um that was the Bastl Castle, I think. Was that? Yeah, yeah. Oh, okay. Yeah, it was back, show 223 of yours. Yes.

Okay. So, Alicia, what'd you get? Oh, I got a fantastic picture of a narwhal roasting a marshmallow on its horn. Okay. That was why I got it, too, is because we looked at it we both started cracking up like right away and every time i look at it it's hilarious is this like a like a full like living room print or it's it's a hand-done collage of different papers so it's it's an original but it's not like a super expensive super large original it's

it's you know eight feet by ten feet no it's it's like 24 inches by about 10 inches okay um and then uh my company which you might say is just me because people have abandoned me. I was going to say the IRS would say what? Yeah. I got a JTrace Pro, which is a super fancy. So, you know, there are programmers, like for Arduino, where it just programs it, and then you have to use printf or something to debug the chip. And then there are more proper debuggers like jlinks or stlinks or even the black magic probes

where you you can download code you can flash the code and you can also use gdb or some other debugger in order to look inside and see what's going on the j trace, and there are a few other things like this, but the J trace not only does the flashing and the debugging, it stores a lot of extra information so that you can back trace to where you were, where you can back trace everything. So instead of saying, show me the current variable X, you can say, show me what X was 10 cycles ago or a thousand cycles ago or 20 000 cycles ago and how much how much memory is there is it like a ton it is just an absolute ton i mean it's it's insane and uh and the idea is that you can you know everything about what's happening and you don't have to fiddle usually when you're debugging you

have to find you have to break in just the right spot and you have to to maybe change a few things that you go down this path um the idea with this is you just let it run and then you sort it out and it's all done well i mean it does a lot of other things too right it you use for profiling so it'll tell you where you spend all your time. And it's instruction accurate, so you can really tell where things are. And it's got some onboard current monitoring. Oh, I didn't know about that. I think so.

I don't know how to use it. Connected to what? Well, I mean, this is one of those things that if you had a 20-pin connector, it would be really nice you'd get all this information but even with a five pin j tag style connector or even a two or three pin swd style you get some you still get some of this yeah so for most trace like the double e has to bring it out for you yeah and it's only on the arm cortex processors yeah the one i got was only for cortex which i i spent a long time trying to decide whether i was okay doing cortex m um or if i thought we were going to be doing a lot more a's and r's this year

and i finally decided like the the more linux style? Is that right? R's have redundancy in them. So it's multiple processors in one chip. And they're good for automotive and medical and space, the areas that you really don't want it to screw up. But the R's are still A-series. Right. And then the A-series run Linux usually. They're big, big full mobile phone chips.

Not the M's, which are kind of small and very, very cool and very cheap, but kind of

small. Right. Yeah, resource constrained, right? Yeah. Well, they don't have

memory protection. Well, they sort of, they have some memory protection, but it's not enough to run a Linux-style

operating system. I mean, you couldn't run much else besides a tiny stripped down linux uh what about you chris what did you get for christmas um or gift yeah yeah boring stuff from uh you know just like warm clothes for me um but uh i had bought myself a hack RF. Uh, so I finally jumped into that and that has been, that's, uh, Mike Osman and great Scott gadgets. Uh, it is their SDR.

So it is basically a troubleshooting tool using, uh, GNU radio. And so as often listeners of the, uh, of the amp hour would know mike osman's been on our show a whole ton and so he's always talking about that stuff but basically it allows you to to do sdr in a graphical or so you can do rf stuff in a graphical manner so you can basically you know make a oscillator and then you can push it in you know then you can uh down sample stuff and you know basically push around in the spectrum and then you can actually you know then you can uh down sample stuff and you know basically push around in the spectrum and then you can actually you know change sample rates you can what do they call that uh all my signal stuff is failing me decimate decimate thank you uh yes decimate and interpolate

all that stuff and basically do dsp type stuff without much hassle so is that that sounds like software though yeah yes so there's a hardware part i assume yeah oh yeah so that this so the thing i bought is like a 300 basically it's a it is a um nxp part that then basically goes to an rf front end and then an antenna and basically it's just like so i like like've explained it for Mike, when I've talked about for Mike before, I said it's basically like a software interface to RF. And it's great for that stuff. So it's pretty cool. I mean, it's, and so the,

I didn't realize GNU Radio is actually basically just doing a graphical block and then writing Python on the backend. So it's just generating Python and you can hook things together and um yeah it's really great for troubleshooting and hacking and um you know kind of figuring out what's going on in the spectrum so that's pretty neat uh my asterix radio too i have uh one of the rtl str dong. Yes. Which is a little $20 thing that does something.

I haven't played with it that much. Right, yeah. So that one's basically, that's a TV tuner. That's an old TV tuner. And then you basically can look at, yeah, you can basically kind of slide around. You can slide the frequency around so you can tell what frequency you're shifting from right so you know if you're trying to look at signals up at 2.4 gigahertz you basically tell the the device basically says 2.4 gigahertz and then you you know shift it down into baseband and then you can look at it in something like str sharp right

that's right i assume what you're doing with it yeah um that's just all the hand bands and stuff yeah it's just safe so that's just received the HackRF, you can actually transmit as well, like low power. So that's pretty neat. You are responsible for using your HackRF1 legally. That's right, yes. I think the low power really helps a lot of that stuff. But yeah, it's not hard to change low power to high power. Yeah, right.

Amplifiers, huh? um but yeah it's not hard to change low power to high power yeah right amplifiers huh so yeah i um i have i had a lot of trouble getting it set up to be honest it was it was much more difficult than i thought that would be um they basically say you should use pen 2 which is a distribution of linux which is based on gen, which itself is not super friendly. And my mistake is I can't... Why did they choose that? I think it was just what they were on. It's like a security-based Linux thing. So it makes sense that, you know, the security people would be doing that. My mistake was I kept trying to force the issue.

I was like, oh, I'm going to have my MacBook Pro with me. I'm going to try and put it the macbook and just trying to get uh you know stuff installed as a secondary os and it's just it's a hot mess and plus it's got the usbc ports so like i'm just i'm i'm like all right well i just need to buy a computer just for that i know that's not actually like if someone heard me say that who was with an r hack rf they'd be like oh get a windows computer or Linux computer. But yeah, my experience with Macs has been very, very poor so far. But otherwise, yeah, you know,

I'm sure that you can both commiserate on using Macs. I'm not sure it's the Mac's fault that a completely obscure distribution of Linux has been chosen. I don't know. You ever tried like installing like just even Ubuntu on it? Like, Oh yeah. Maybe I'll have to ask you more about that after the,

after the show. I'm having a lot of trouble with that. So, and I'm talking like no, no VMs though. Like I've, so I like, Oh,

you do it. Yeah. You can't use bootcamp. You have to use, you have to basically use like a, a different boot start thingy. I'm sure that people are going to write in and be all angry about this stuff but i'm just i've really struggled with it so if you can't use a vm that gets trickier

yeah right i did a vm but like basically it's so the couple the hardware is so coupled that like you get all these over overruns really really quickly so it's um yeah i'm just gonna try it elsewhere so with linux i mean so with ubuntu even it there's a there's a build for it now as well and that's and that's what dominic uh one of the other software people at great scott told me to use so um it's better uh yeah so i got that and i got a bunch oh um well so yeah the uh so i'm playing with laura modules like i mentioned at the beginning and uh so basically just having troubleshooting for that and just kind of seeing what's going on there obviously i'm doing cellular stuff

so just kind of peeking at the spectrum and having a better grasp of it all bought yourself a work tool i know i bought myself a work tool i mean yeah my, my business bought me a work tool. So yeah, I think it's a similar kind of thing. Yeah, and it's, I don't know. It's an interesting, I wish someone would have showed me this. So like you can just go and watch, I think I mentioned this on the last amp hour, but you can just go and watch Mike Osman's like intro video for it. And I just wish that something like that would have been around uh when i was learning signal stuff like just

just show show me that right show someone hooking stuff together you're basically you know you're demodulating a fm signal you're using some blocks so yes of course you're hand waving a little bit there but man at the end you have like an audio signature like it's just it's fantastic you know it's it's it's context and and uh and that's what i i really missed about the my learning process in school and i think we've talked about this you know endlessly now but like but like these kind of tools are just make it so clear even even at the beginning. So I was listening to your show. Your last show was with the Udacity person.

Yeah. Yeah. It was great. And I sent you that. So you told him about the Dreyfus model, I think? Yeah. We talked about it to him, which was a little odd, because I was hoping for a pedagogy from Udacity.

But that wasn't Anthony's bottle of wax, so that's fine. Oh, yeah. And I meant to tell you this on the secret Slack channel that I'm on with you guys. You can't say that out loud. Oh, no. You have to type it to us instead. Sorry. I learned from Mel.

So, I had sent you both a about about the dreyfus model uh it's a by by a friend mel and mel told me uh or something very interesting pedagogy is the wrong term is it did you know that it's andragogy andragogy oh because we're adults exactly i was like oh my god yeah of course right but it's like that's just the the accepted term is pedagogy oh because we're adults exactly i was like oh my god yeah of course right but it's like that's just the the accepted term is pedagogy right but it's actually yes pedagogy is for children are you being pedantic or andrantic now pedantic or and i think we just found a title for the show uh yeah uh i don't know um i'm sticking with pedagogy because because people know what you're talking about yeah or at least they they have heard that word right but i i will accept that it is wrong just like how when i say lead sentence christopher corrects my spelling on that

and i'm like yeah still don't care right well this is an exciting tour through linguistics uh yeah so the dreyfus model though i love the dreyfus model right and and you were both talking about that uh but the dreyfus model is great because it's like these different stages of like i the thing that i loved about mel's comic is basically like you're you're going to cook a thing right you're going to cook a thing, right? You're going to cook a thing, and the first thing that an absolute beginner asks is like, you need to say like, crack an egg into the bowl, and here's what an egg looks like. Like that's the stuff that you don't even think about.

Like, oh yeah, of course, someone who's an absolute beginner needs to know what an egg actually looks like. And I love that. I just, I don't know. It was great. I will post that comic, so. It's a great comic, and it doesn't, it shows you the Dreyfus models. And it kind of, I mean, there's been a few people who have been talking about how you don't teach experts the same way you teach novices.

Yeah. Yep. So Daniel Spalding is a book about teaching adults as well and he talks about that i've talked about that i think before so what is it daniel spalding he has a it's i think it's just an e-book um daniel spalding is a book called teaching adults and it specifically covers that so it's funny i'm looking at a book how to teach programming and other things which i don't think is the same person. It's How to Teach is his site.

Okay. But he talks about how a novice with computer programming may not understand that there is no real useful difference between cutting and pasting a program into something and typing it in character by character. That those are identical to a compiler or to an interpreter. They're not even wrong. There's just no mental model and how you have to build things up. And I think if more people know about the Dreyfus model, we can start talking about the things inside there. And how you can... Oh, you sent over a link, the advanced expert.

So the Dreyfus model starts out with the novice, where you really don't know anything. And then the advanced beginner, where you start to know things. And then the third level is competency. So that kind of most people know what those three things are. And you sent over an article that talked about the advanced expert, where you, instead of becoming competent, you wander off into your own little branch, and you can never really become competent from there, because you have defined your own science, And so you can't go beyond that.

You have to know that you're a beginner in order to learn more. It's never like Dunning-Kruger lives. Yes. It's the lane that you wander down, right? Yeah. Well, and I think this is actually pretty poignant because I'm kind of in this realm of like, I'm trying to relearn software. And I'm really, I'm really struggling. Like, I'm really having a tough time.

And I don't really know where to go next. And like, you know, people point me to like, oh, coding bootcamps. Yeah, yeah, yeah. That's all one size fits all stuff. So, I would ask you, since you didn't ask for advice, but... Sure. Hasn't stopped you in the past, man. What do you want to do with it?

Okay. Would be the first question. And how are you attempting to learn now? The answer is, I want to do my job. Okay. And how am I attempting to learn is going like just like screaming and and googling stuff um pretty much so so this is this is a case of definitely um do what i say and not what i do but getting frustrated doesn't help

oh okay well i'll just stop doing that click yeah just turn off like that uh when uh data in in star trek he just like turns off his emotion chip you have just oh i'll turn off your frustrator chip yeah that was easy but for your job i mean what do you need to what what so like my job these days is no no no it's not secrets it's like basically like making tutorials like i just have to go out and like so like the hardest thing is when there's a new platform right you have to kind of just go and make examples for people and you have to learn stuff you have to struggle you're basically just struggling for people so that they have a

starting point and um and some of it is just i mean some literally is tenacity right and like and i the thing i always emphasize is that like, when I think of someone like Ben Krasnow, who I, you know, hold up as, you know, the great prototyper, right? It's just tenacity, right? Or like Sammy Kamkar, right? Sammy is like super tenacious. And people like that, you know, they just, or Micah, Micah just, you know, just keeps trying different things and is methodical about it, right? That whole like, so I've been meaning to laser cut a thing above my bench, but just say experiment first, right? So like the best prototypers, the best people doing this stuff, they're not necessarily like, I'm just kind of floundering, but they're they they basically say i want to try i have a i

have a i have a thesis or i have a what's the word i have a goal uh no scientific on how this works i mean hypothesis hypothesis that's the word thank you i have a hypothesis and here's what it is i'm gonna try xyz and i try it and then do I meet the hypothesis? If not, you know, cycle back through. Like basically applying the scientific method to prototyping is like a goal of mine. And I really struggle with it. I mean, like it's because there's so much information out there as well. Like, so there's also the whole, you know, tuning out the world and not getting super frustrating and just Googling everything.

Yeah. That's a radical. I mean, you're talking about learning coding and that's not my approach that's not my approach until I'm debugging something or there's a problem um so I'm trying to think if what well and that's what I wonder about like should I be going back to basics well why why are we the worst people to ask I don't remember how to teach somebody how to program i mean i mean i mean you i know how to teach somebody how to learn stuff but not right i mean somebody here's where i'm headed with this there's the tactical part of programming right there's the

i'm programming in c and i know about these data structures and i know how pointers work and then there's the mindset of programming which is I know how to structure a program. I now know fundamentally how a computer thinks. And before I even start typing, I have some idea for how this program or library or set of things I'm working on are going to be laid out and communicate with each other and what the memory is going to be doing. So my worry is if you start with the prototyping stuff you're kind of in the tactical zone and you know that that seems like an easy place to get into trouble um if you don't like some of it is is like a blank sheet of paper problem right like that's that's kind of what i'm talking about here is like sometimes it's just blank sheet of paper it's like okay well and uh there's a link that i i put on our list here from dan lu as well he he talks about uh at one point

where he basically started having he had a mentor he went to and he kind of just kept uh he basically built confidence of like the blank sheet of paper problem right of being able to like go and well he basically was kept being presented with harder and harder problems and then at a certain point, he's like, well, I can just go and take on any problem. It's like, yeah, of course. I mean, with enough time and the internet, you can pretty much do anything. And I think that's the blank sheet of paper problem. But the rest of it is the, okay, now you have a blank sheet of paper. What are you putting on that paper?

And what are the generally accepted methods? Or even where are the generally accepted methods or even like, where are the resources to go towards so that you can pre-fill some of that sheet of paper? You know what I mean? So I listened to a new podcast, um, that one of our listeners,

uh, is the host of the function podcast. Oh, Tom Anderson. With Tom Anderson. Yes. And, um, Oh, Tom Anderson. With Tom Anderson, yes. And one of the ones they had was talking about notebooks

and how if they get nice notebooks, they'll never use them. And they now have decided that when they get nice notebooks, they put something on the first page, and thereby it's ruined. It doesn't even matter what they put on it. They're going to ruin their notebooks so they can use them i feel that way about the blank sheet problem put anything down put how are you going to debug it put put how are you going to use printf put hello world and now it's not a blank sheet anymore write the problem down write write down the problem statement use the scientific method you know what is your hypothesis what is your problem um and or or write down something utterly generic

like many of my projects have a debug console at one point in their lives for manufacturing or for debugging or a serial console so i start with a serial console and that means I need a UART. Now, what are my other hardware inputs and outputs? And so I need drivers. Now I can test them all from the serial console and build up my whole system. And it's a lot about not having a blank sheet of paper. It's about having these Lego blocks that you toss on the paper and then you can figure out how to put them together and what glue logic you need. Is that helpful or are you like, where did the semicolons go? Oh, no. I mean, like a lot of that stuff. I mean, well, I think these are two different problems too. Some of it is the, where did the semicolons go? Like, what was it? Oh,

it was at Supercon. I opened up Mike Harrison, you know, and so that's the opposite problem, right? So one's the blank page problem, right? Of just like, all right, what's first, what's next? The other problem is, okay, now you're starting from someone else's project and you open up and you're like, oh my God. Yeah. That's a different problem. Yeah, right.

And that's another problem. I mean, that's probably a little bit, well, I don't know, actually, what would be more likely these days. I guess if I'm Googling around and I'm trying to, like, you know, piece together different things, it's kind of how fast can you take in what someone else has written or, you know, is trying to do and then make it your own um and that i think some of that is just where does that is where does semicolon go it's what what the hell is that what are they doing there that becomes a lot easier if you've ever implemented anything similar because then you know what the structures are you know what the overall like if you're doing signal processing, you probably need an amplifier. You probably need some sort of filter. And so when you're faced with a giant schematic, you start breaking these things down into blocks and the blocks you understand. And you have to do that with software, too. If you're looking at a giant wall of software, hopefully they've broken it into some blocks for you just by files.

But it's still going to have stuff that's idiosyncratic. I mean, if you've never seen a linked list implemented in C, you're not going to know what you're looking at. Right. And they're going to call it ll instead of linked list, so you're not going to be able to look it up. Why would they do that? Pointer stuff, array stuff.

Yeah. I think diving into the deep end is not probably a healthy activity. Well, there are some good sites. I mean, SparkFun and Adafruit both have a huge amount of code, and they write their code with the intention that it is read. And that is a lot different than some of the open source projects that are written with the intention that it's my project, you can just get off my lawn. I don't think you can learn by reading code.

Not initially. I think initially that's really hard. I think you need to do exercises and implement stuff. And that's the hump to get over because you can't really do anything exciting yet. And if you try to do something too exciting, you're just going to get frustrated. I mean, that's why stuff is, that's why courses are structured the way they are, right?

I mean, you do a bunch of boring exercises. Well, sort of. I mean, so I've been working on that for, so for my course as well. If people don't know, I have an online course and I teach people hardware and thinking about, and this is how I met Mel as well and learned about Dreyfus Method. I was kind of starting to restructure some of my course and thinking about, and this is how I met Mel as well and learned about Dreyfus method. I was kind of starting to restructure some of my course and thinking about this because, I mean, well, so both of you, you know,

are learning and or have learned hardware. And so, you know, as well, it's a similar kind of problem of like, like why is, why resistor? Why is capacitor? And right, I mean, it's just, it doesn't make any sense at the beginning and you know and obviously people start with math and they go oh well let's let's do some calculus

and so yeah um so yeah i mean i agree that like some rigor is helpful there but it's that balance of like where do you introduce it i think that's the think that's always the problem. Just in case nobody has heard, what is this thing that you're talking about, about teaching electronics? Is there a name for it? Yes, there is. It's called Contextual Electronics. Thank you. Yes, it's a course that I teach online. And there's a lot of free stuff, too.

So, like, Getting to Blinky is the – it's meant for getting started in Ki kicab but it's also a good introduction to doing layout just in general and so it's basically a 555 timer with some uh you know like what do you need to to get a small circuit blinking because blinking is the hardware equivalent of hello world um and then there's a small one as well for doing a uh a blinker for raspberry pi which is even shorter and then there's a small one as well for doing a blinker for Raspberry Pi, which is even shorter. And then there's a bunch of courses around building devices that aren't blinking, just blinking. I thought I would go through, I went through one of the short ones with getting to Blinky and installing KiCad and all of that. And I thought I would just do it and learn it enough that it was something I could see if I wanted to do more of.

And I got to the end and there was no way that I could fail to send it off to Osh Park to get the board. I didn't actually stuff the board, but I did send it off. But you held it in your hand. Yeah, there was something about getting to hold it in my hand that was exciting. That's super exciting. That's a great feeling.

That's like a dopamine rush. It's all about the dopamine, right? So, yeah. That's great. Yeah. So, I liked the parts of contextual electronics I have taken. I think that it is very cool. Unfortunately, I have gotten very sucked into more software instead of more hardware lately.

So say we all. I like the comparison, though, about looking at schematic and blocks. Because that was like a... So I think there's different moments in hardware. Like for my own journey, where certain pieces of the matrix kind of fell into place. So actually being able to visualize current in a circuit is a huge difference. journey where like certain pieces of the matrix kind of fell into place right so like actually being able to visualize current in a circuit is a huge difference um and then being able to actually pull out like being able to visualize blocks in a schematic and saying oh well that piece is just

an amplifier but with other stuff around it that's a filter blah blah blah right and and like hooking things together like that in my mind is super important so i assume that that's an analog to what you're talking about with the the code piece as well whereas i just i haven't yeah i haven't i haven't figured out the blocks yet so i i can't just look at it yeah exactly well but that's that's one way to start analyzing it too because code is naturally structured as blocks if it's written well. You know, you have individual functions that are supposed to do one thing. And then you, so you can find the top level and start diagramming it.

But you really do have to diagram it. You can't try to hold it in your head. But yeah, some of the, I mean, at interviews, some of the most deep questions I've ever gotten have nothing to do with code. It's been like, well, we're building the system to do this uh give me the block diagram and yeah and you can you know you can start pretty high level and then start get to a lot of details that way right yeah and that's like pattern matching that's human stuff right it's like okay well i've seen this a hundred times

right 90 of them have been this way it It's probably going to be this way, but maybe not. And you can discuss it there. That's experience. And are you, what languages are you looking to write in? So I'm doing C. I'm doing some circuit Python, MicroPython. And then my goal this year is to, well, I mean, Python as well. Just like higher level Python um but maybe do some javascript yeah i mean i don't love javascript but it's very useful right and it's it's between

i was inspired by micah too yeah i mean like as someone who making works on a product and stuff like that where you know. Right, your apps are written in JavaScript. The SDK is JavaScript. I can say nothing negative. You can say lots of things negative. No, it's actually, it's really cool. I've listened to your show.

It's the fastest damn thing to get going to write an app for something. They did a really good job. So I can't, I really can't say anything bad about it. It's great. Yeah. Yeah. Yeah, sometimes we make fun of JavaScript, but it's only because we're C programmers and we fought hard for that knowledge,

so we have to protect it. I think some of it is like, I get a little frustrated when I see like shoehorning it right so like like running hardware with javascript i get the draw it's like oh pull more people in and i'm more for that than i used to be where i'm like oh okay well yeah more people doing hardware is just a generally good thing but it's like all right yeah you could say the same thing about python i suppose or circuit python but like you know eventually you gotta dive dive down it doesn't go away people are trying to shoehorn things and all the time yeah right says the person who is trying to block his shoe says

the cobbler the cobbler's assistant a lot of rust going around oh yeah interesting a lot of people really interested in rust and trying to trying to make it go places it probably shouldn't yet. My eyes just glaze over when I hear about that stuff. Raspberry Pi and BeagleBone both made JavaScript so easy. And Python on those are so easy that I understand you start using those, you start wanting to do more hardware, and you don't want to have to deal with C. So yeah, I totally understand that. But I agree with Chris.

A block diagram, totally the way to start. Like start teaching or start projects? Start projects. You don't need to know what's in all the blocks, but... And analyzing other people people like you said when you are faced with code you've never seen yeah um that's true yeah i uh so another thing that i've been so um i think we have goals on here when we talked about goals do we have goals for 28 yeah we do have goals for someone wrote it i didn't i don't know if it was me it might have been me definitely wasn't me i just saw this diagram an hour ago um yeah uh so my goal for

2017 which did not go super great was to get better at iteration and uh i think that's a really big piece too because it's kind of what you're talking about it's like i i think a lot of this is still residual school stuff personally like you need to get it right it needs to be right the first time and it you know you know it's like that that's a really bad i'm very against that because the best people just iterate and understand that there will be iteration including for hardware right like so yes i want to get my circuit board right the first time but it's not the end of the world if it's not right the first time um you should send it out right um and a lot of people get hung up on that piece especially because it's so cheap these days um same thing for code it's like yeah

it might not work but then you just recompile it and try it again um so there's some of that i mean i i've i agree with that iteration is important and getting good at it. I do think sometimes easy iteration becomes an excuse to not think. I'm a monkey typing at a keyboard and I will just type all the options. And that's not an accusation. I do this all the time. When debugging, it's like, okay, I'll just change this line. Well, there's this parameter.

I'll just change it. You know, there's only 600 possible combinations. So how long can I at one minute compile and download? That'll take 10 hours. So I'll just do that. But, you know, or I could go sit and think for two hours, and, oh, that's the actual problem. So, yeah, there's a balance.

Yes, it is a balance. I have a theory that every time I see people using an other mill, I think the other mill, which I love as a tool, I think that that company has been bolstered by software people. They see hardware and they go, need to iterate faster and i will do anything to to do that and i think i think feather mill is a path forward i don't know if it's the only path forward but software people are like one of the main clients there um because i look at it i'm like i can't use that for much. I would use it for jewelry and fun things, but not for boards. I never really understood the board thing.

I love solder masks, so got to have it. Not everybody does, though. Okay, so that was the not goal for 2017. Do you have actual goals for 2018, or are you just carrying that one over? I have themes, which is like the biggest cop out ever. But basically, when I came up with, I should probably publish it this year, if I'm going to talk about it here. So yeah, I had other ones for 2017. So iteration was the first one, I did okay on that. Was that the other one from last year flexibility last year not great and then

temperance i did good on that uh and so this year i have experiment slash risk taking that's uh the big one uh produce efficiently and then tell stories those are my my three themes for 2018 and um produce efficiency efficiently is something that i'm very about i have been about because like like just basically reducing overhead on projects you know being able to iterate faster means you need to lower your overhead so like video editing was a big one in 2017 um so but you know that's like workflow stuff so like just improving workflow so that you can iterate faster and you can you know it's a there's a lot of stuff in there or improving workflow so that you can iterate faster and you can you know it's a there's a lot of stuff in there or layers in that that thing but uh tell stories is pretty

self-evident i think um and then experiment risk taking i think is also uh you know but that's also like trying rf rf scares the shit out of me to be honest excuse my french uh it's i don't know why it's just always scared me i've never you know i have a ham license and i've never used it so uh 2018 is the year of rf so i have a ham license and i know how to use it no all right you're not afraid to use it huh no uh do yin's have any goals i don't do that good policy i have a couple of goals i don't know if they're 2018 specific i remember one year i had a resolution to have a resolution every week for the year that's right that was pretty amusing that every week I would try something different. And there were things like go out after dark to do things at least five days out of seven or not eat any sugar for a month was probably what killed the whole thing because a month was way too long to go without sugar. And that was pretty cool.

That was very adventurous and led to lots of different interesting things. Between eating grains for at least three times a day or baking bread. It was fun. But I don't think I have anything like that this year. I've been on a kick of learn hard things. Learning easy things is fun, but everyone can learn easy things. If I want to be relevant career-wise, I need to learn hard things. Because not as many people will be able to learn hard things.

And if I can mix them together the right way, then it will be very useful. Do you have examples? Well, I think I've been doing computer vision and robotics and machine learning. And I have been trying to learn all three of these things. And it's really hard to do all of them. And I took the Audacity self-driving car class for the term one, which had some computer vision and some machine learning.

But that was very much at the Dreyfus beginner stage. What does an egg look like? No, it got past the novice of what is the egg. I'm just doing some pretty, I mean, I don't know how easy these things are, but some of the exercises I was looking at what you were doing going, whoa, what, really? Like, oh, the car is, you know, it's marking all the lanes and you have to take their video of driving and mark all the cars.

And it's like, wow, this is really sophisticated stuff and you're only a couple weeks in. So it seemed cool from the outside. It was cool and I learned the cards. It's like, wow, this is really sophisticated stuff and you're only a couple weeks in. So it seemed cool from the outside. It was cool and I learned a ton, but I was following their recipes. Right, right. So it's like cocktail party cool, right? It's like, oh, wow, that has like some flash to it,

but you're saying that you couldn't, if you didn't have the recipe, you wouldn't necessarily go far? Well, I am saying that i started that way um i have been working on my own robotics ai computer vision project and without the information from the class i would have a whole bunch of things i don't know i would have 95 unknowns but because of the class, I learned Python much better. I learned OpenCV much better. And so now as I'm trying to solve my own problems, I'm down to 40% unknowns, which that I can start looking up, that I can Google, I can read the

books, and I'm not completely bewildered. And that's been leading to some neat things that I didn't expect. That's the thing I was thinking I didn't say well when we were talking about learning, is you kind of have to have a hook. You kind of have to have a hook to learn. Knowing what to Google for? Well, that, but you need to learn C or Python. You got better at Python by taking this course about machine learning and machine vision, not about Python. And so you had a, you had a good hook to get you interested and,

you know, kind of backdoor teach you Python. So I think that's really important. And it's hard to find the right hook sometimes. So you mean like a project though? Like that's a project or, you know, a course or something, but it's not, you don't go out and say, oh, I want to learn Python, so I'll take the Python course necessarily. No, I totally agree. Right. No, exactly.

And I read a lot of their Python. And so some of the things like some really bad ways of doing lists and slicing them, I got better because I was looking at how they were doing it and it was much clearer and simpler and probably a lot faster. I mean, that was, that was part of the thing. I wrote a book. I don't know if your listeners know that.

Heck, I don't even know if my listeners know that, but I did write a book about embedded systems. We've all plugged something this episode. As you say, yeah, it's good. It's good.

You're doing that. Also people may remember this from when you were on the show the first time. I think we talked about it then. Episode 187, Werewolf Worshipping Wookieist. Back when we did that whole mess. But a lot of the pitch for my book was that you would only learn half the book. The other half of the book you'd already know. It was for people who already knew some computer science or people who already knew some electrical engineering.

And the idea was you can't lever off the knowledge you had into the other half. And I think that's important. I think it's important for me as I learn because I need to attach it to something or I just don't. I see it and it goes by and I don't catch it. So yeah, with the Python, it had to be part of machine learning because I really wanted to learn the machine learning part. And even the machine learning had to be attached to something, which is why I have my little typing robot. And that's really just attached to my sense of humor because it's so ridiculous

right so uh is that do you think that's enough of a draw like if you had okay so say your dogs needed like computer vision help to like do something right would that be a different draw like is the typing robot itself enough of a draw it is for me it wouldn't be for everyone it is for me because career-wise the ability to identify and react to environmental stimulus through a robot in which i can explain the kinematics and the drive mechanisms that's all career wise, very good for me. But if I didn't, if I was a home hobbyist and I really didn't care about being able to talk

about these and in job interviews, then yeah. Being able to open the door for my dog when it wants to come in totally different set of things. Maybe I wouldn't be spending nearly as much time doing matrix math. Maybe I wouldn't be spending nearly as much time drawing out what I'm learning and trying to make sure that I can explain what I'm learning. Different paths are great for different people. But for me, for 2018 and for last half of 2017, it's been learn hard

things because that's what I need to do i think i spent too much time learning easy things learning fun things and i i forgot some of that tenacity and some of that resilience that i needed to to stick with it yep i uh i always tell people about that too with like you never learn hardware like when you're actually manufacturing it like in in volume if you can but if not i mean even if it's low volume you just you don't realize how much stuff will go wrong and and that's why you always read about people like oh well our stuff's 18 months late because we forgot about a b and c it's like we didn't forget about it you probably just didn't know about it you didn't think to ask people because you don't know what to ask so that's that's tough and so you try it i mean that's the iteration part right but well yes but i think that the

the thing is like finding a mission also helps right for some people it's like oh i have a kickstarter i want this thing in the world and they then they finally go for it yeah oh yeah applications always matter to me i I mean, the typing robot, I'm hoping at the end it is nearly performance art, even though there are plenty of other robots that can type. The combination of the cheapest possible robot arm and the ridiculously expensive intelligence and the two cameras that are required,

it's just going to be very silly yeah it's really ridiculous and that who specified this it it makes me laugh and that's what i wanted to do laughing right right uh there was there was an article i posted too about uh it was i mean it was definitely an auxiliary um industry it was about reading though and about like people teaching like literature and it was kind of in the same realm right of like the struggle is the important thing right the what they say the the difficulty is the point uh you know so like finding these hard problems that you're talking about is like its own reward i'm obviously you get to solve the problem but you also get the stuff by

proxy you get the information get the stuff by proxy. You get the information and the knowledge by proxy. So that's a good, I think that's a really good goal to have. I mean, or goal or theme or whatever it is. I think the other theme is mental health this year. Yeah? Yeah. A little too much work last year?

No. No, actually it wasn't too much work. I don't know what it was okay was this oh you posted another podcast that i started listening the uh thwad someone posted this oh oh yeah no that was that was uh yeah so i started listening to this apm nPR podcast, The Hilarious World of Depression. Yeah, it was really good. The one with Jeff Tweedy was really, really good. I haven't heard that one yet, but it's really high-functioning, really funny, smart people talking about their struggles with depression. And the thing I have most taken away is that I need to stop pretending that I don't have a problem or that I have a little problem and I need to go get some help because I clearly have a big problem.

These people are talking about their problems as having a big problem and I'm thinking, oh, that's not that big a deal. I deal with that all the time. Maybe I need to see somebody about this. Yeah. No, I think it's very – I very i think in engineering especially like you know like smart people have this stuff and they're like well i can think my way out of it it's like well sure that's not always that is not always sometimes it's chemical so i've been to therapy if it's you know that's i'd say wonderful thing i i love it i it was a great time for not a great

time it was a terrible time but like know, it was a very helpful thing. And I think that, you know, if people, you know, if it's out there and people realize that it's more common than they think, yeah, I think it's good to go talk to someone. See, you're so lucky. I've been to two therapists in my life and they both sucked. Like, really. That's too bad. Like, one, after I left, I was way more messed up. The other one kind of looked at me as a high functioning engineer and was like, oh, we don't need to see each other anymore. And I'm like, yeah, I haven't even started to tell you my problems because honestly, I'm not the sort of person who can just tell you on the first date.

So you're right. We don't need to see each other anymore. Right. No, I've had bad ones too. no i've had i've had bad ones too so um i've had good ones and bad ones so you know how many yeah and well in that hilarious world of depression talks about it's you do need to go through four to six in order to find one and even that is that's average not you have to find one that can deal with you. Yep.

Yeah. So, yeah. That's good advice. Mental health, 2018. I like that. See, I have questions from listeners for hardware things that we couldn't answer. Oh. I will join you in not being able to answer them.

Go ahead. There was one from Tyler. His question is, how can he make prototyping cheaper and faster? See, this is such a softball question for you. He says, my college taught wire wrapping. What was it, like 1920? And it was basically torture. What is it with the DVD commentary?

I'm currently considered using KiCad because it's free and ordering PCBs from something like Oshpark. A lot of the homemade PCB solutions don't look appealing. Maybe we can recommend something. Can we recommend something? Sure, yeah. That's pretty easy. Yeah, you're right.

That is a softball. But here's the thing I'll say, right? Whatever you pick as your first CAD program you're probably gonna use for a long time so keep that wisely choose wisely i mean like it and it's not i i i have gotten past the point where if people aren't using kai cad key cad whatever i don't care anymore uh i still want to help people with their electronics problems but i teach with it so that's why and i think it's actually pretty great so yeah i think tyler would definitely benefit from that i would also caution though as well i jump into using pc uh layout programs way too fast um you know don't necessarily wire

wrap but man uh skywire and like just duct tape stuff together and like the best prototypers right so like i i like talking about the best prototypers i talked about like ben and micah and everyone like that like they they bodge stuff at first right they take existing solutions so alan uh yates who's also been on our show before and i think your show yeah your show too um so he talks about as well uh jerry all these people that were jeff all these people that have been at Valve and elsewhere, the best prototypers, I think they take what they can, they do the most they can with the hardware they have. And so if that's wire wrap, that's fine, right?

If it's, you know, a development board where you're then wiring in a sensor, right? So you have a sensor board breakout and you're just literally wiring two boards together. That's great. If you're using a breadboardboard i'm not a huge fan of breadboards but sometimes that is the solution so uh plugging stuff together with those wires and up until like about 10 megahertz or so you don't really have to worry too much from there for sure yeah the layout and uh learning that stuff and start small like we talked about in the show already. So get something to blink.

And, you know, maybe KiCad's the way for you. So, Osh Park, of course, I love as well. Yeah, that makes sense. I agree. If you can find a dev board that's even the least bit similar to what your goal is, start there. I mean, you can go pretty far. far the world is pretty extensive these days right so you can get yeah i've been pointing people that tindy tindy very often i mean obviously i used to work at the company that owns tindy but like more and more people are putting stuff on their tindy ebay you know uh you know maybe

even looking on like a hack day io or a Hackster and like finding people that are already doing it and be like, Hey, do you have any extras? Like just like reaching out to people too. You never know. And, and they might point you to stuff that they've used too. So a little bit of extra legwork can save you lots of time because nothing's worse than, you know, like getting all the way through a design cycle, right? You're designing a breakout for a sensor and then you make a mistake on the board you got to do another cycle there it's like if you can pay a little more even i know that's they asked um cheaper and faster right so cheaper sometimes it's worth it it is cheaper to pay a

little more that sounds stupid but well it'll save you some good axes right right yeah well there's even just the cheap and fast i mean yeah, if you get a dev board from SparkFun, maybe it's $30. Maybe all of the parts on it only cost about $5. Well, if you make your own and you don't even count your time, by the time you have made a board and stuffed it And quantity one, that's the difference between prototyping. With prototyping, buy off the shelf if you can, because it will be cheaper. Now, when you start making 10 and then 100 and then maybe a million, economies of scale are not the same as they are when you're making one or three. Right.

So, yeah. And at that point, someone else might be doing your layout anyway. So, yeah yeah it's somewhere between about five and a hundred you're probably gonna have to make your own board and you probably have to learn some methods that you might not already know um but that won't be faster yeah yeah yeah so um i don't know if we had did that answer the question or no i think so i think my problem is that I've gotten into the, I've gotten so far into like the pay, pay to solve my problems stuff these days.

Like I pay for software now, like happily, like I have a smile on my face when I pay for software. Like 10 years ago, me is like, what are you doing? Just struggle with GIMP for a while longer. No.

Inkscape isn't so bad. Yes it is. Yeah, right. a while longer um no inkscape isn't so bad uh yeah right um so yeah and like these days though like i just love paying a little bit more and for like solved problems right so like both of you are fans of the salia i think salia is fantastic that's like a logic analyzer with like a really intuitive interface that's a solved problem now same thing with like i like the analog discovery too you've talked about that too like from test equipment side of things pay for that no problem software pay for that no problem breakout hardware pay for that no problem granted the caveat here is hardware is expensive it's so expensive so just keep that in mind that's cheaper than it was i

mean a logic analyzer cheaper than it was you know the first time i used a logic analyzer was this giant thing from hp that probably came with a personal support engineer so yep yeah and it's getting better and better like i mean like well like like just even looking at the number of tindy sellers there's so many people doing that or if you watch uh so i was just watching that's the funny sounding swiss engineer he calls himself uh matthias i think uh one of the youtubers he does he does mailbags and then mick makes does mailbags and dave my co-host does mailbags and like i'm just thinking about like how many people are sending boards around everyone like what a what a wonderful world we're living in where there's just hardware floating through the postal system to to random

people well obviously youtubers but like we we got we got random hardware and it wasn't because of the show whoa really they're just like sent us oh yeah package oh god and in this is a different reason though you see open up the container. It's a little booklet. And then there's this booklet. And it's like a hardbound picture book. Not super hardbound, but it's cardboard. It's solid. And then you open up the right flap, and then you open up the left flap,

and a video starts playing, and music starts playing, and there's a video player in this little book thing. They snail mailed and a YouTube ad to us. They snail mailed a YouTube ad. And when I pushed skip, there was no good video afterwards. It's like a five inch 800 by 500 display with a speaker. And it had a pretty big resolution in there. Yeah.

Wow. Yeah. So she tore it to shreds and, uh, you know, it's got a, there's pictures that will go online at some point, right? It's got some ARM chip from AllWinner, I think.

And, you know, flash and RAM and display. And a USB port, which I will so not plug into anything. And you thought that you could change out the video on the usb port apparently this is some sort of video greeting card that has been going around this whole thing yeah somebody's got i've seen these before yeah but it was but they're not cheap they're like 15 10 15 bucks each but the thing is if you buy a bmw i mean like so okay let's let's do some you know we are not analytics bmw customers yeah i've never heard well you are fancy car customers let's be honest about that we are specific fancy car customers well i'm the sort of fancy car customer who's not going to buy a car because you sent me a video

player i it was a negative to me i'm like come on can't you have like put this into your car i mean the i8 is a beautiful car. I don't know if you've seen it. It's a beautiful car. Kind of looks like a 911 in the back. I mean, I think it's beautiful, but yeah, I can't afford any of these cars. Also, I don't have a car, so there's also that. Did you give up a car going to Chicago?

Oh, you didn't know that? Yeah, I haven't had a car in a year and a half that's fantastic yeah i drive a car every like six months now i just drove one in the snow that's a great way to get reintroduced to that to that experience um uh yeah no it's uh it's i'm i'm on a train line here so it's it's great yeah wait a minute you you were texting with me and you said you were in i thought somebody else was driving yeah my dad was oh all right uh yes we were texting and yes i did say i was driving i was not actually driving at that point i was in a car a lot this weekend or this week don't text and drive that's right i agree with that um yeah yeah i'm all about the self-drive i'm i'm so excited

for the self-driving car thing me too like i don't need that i like i mean i have self-driving cars it's just the the compute i've said before it's a meat controller right it's not to disparage you know uber and lyft drivers but like they're getting replaced and like i'm not gonna miss them that much you know like they're yeah the conversation and I'm not going to miss them that much. The conversation isn't that great. You have fantastic conversations. When you come to the Bay Area, I hear every single one of them is an entrepreneur who tells you all about their ideas. That doesn't happen here.

I guess in LA they tell you all about your screenplays. Yeah. In Chicago, they talk about sports. So, yeah, hard pass. Yeah. So, I'm excited to see the pictures of this video player, but, you know, it's just economics, right? If they send out $1,000 worth of video players to random people, and, you know, you're being targeted somehow,

and one person buys it, that pays for itself. So, boom. Advertising. Merchandising. person buys it that pays for itself so oh advertising what a merchandising it's a colossal and i hate throwing it away i mean we'll take it to some electronics place and throw it away properly and maybe i'll clip the lipo off because you can never have too many yeah it was a couple hundred milliamp hours no it is that wasteful. 4,000. Really? Yeah. It's sad that that's like throwaway because it's like, okay, well, that's some harmful chemicals.

And yeah, most people are not going to recycle it. So that's a big drag. And I'm not even going to attempt to recycle the display and anything I want to use because I don't want to deal with writing your driver for that display or sourcing it. If I want to make more than one, I'll just buy one off. Right. Yep.

Spark fun or whatever. Uh, another question, uh, from embedded listeners for hardware. And, uh, this one is from Stanley on our savior board episode with Charlie lad.

He liked the discussion of what good hardware documentation entails, but he worries that his documentation is lacking. And he wanted to know if we would be willing to send an email sample so he could get a better handle on what he should be shooting for. But Stanley, we didn't do that. And I think you probably sent this like six months ago. My email is so, I'm not good at it. 2018, you declare email bankruptcy? Let's just say anybody who wanted a sticker had already emailed me before December 15th, has not gotten their sticker by the end of January, should email again.

Because I am, right now, I have sent out all of the stickers that I am supposed to send out. I have not replied to the emails. I am declaring bankruptcy for emails sent before January. Or before June, sorry. Because I'm going to admit, I'm not going to reply

to them. Sorry, Stanley and probably Tyler. We're replying to his email right this instant. I mean, yeah, they're getting reply to them. Sorry, Stanley and probably Tyler. We're replying to his email right this instant. I mean, yeah, they're getting front and center. It's just they probably stopped listening because they felt slighted. Yes, exactly. Yeah. Do you have ideas for good documentation?

Good documentation. For hardware, oh. I'm trying to think of the best yeah go ahead when we talked to Charlie we talked some about how the hardware engineers have to read a lot of the data sheets and as they're preparing

the schematics they have to do a lot of the address lookup and how are you going to program it in order to verify this part will work. And that the firmware engineer should check all of that. But oftentimes the firmware engineer has to redo all that work because the hardware engineer doesn't document it. And I still find that happened recently with the project. And I do find that happened recently with the project. And I do find it frustrating. I mean, you want me to change this dual pot to control this voltage and current system.

I'm happy to, but can you give me like five values of what you expect so that when I do it, I know whether or not I'm doing it right? Maybe I have them swapped. Can you just give me some examples? So like testing criteria almost? I want a few testing criteria for the things that software can change. I want to know GPIOs, should they be inputs or outputs?

Turned out I had an output because it said out and so I naively thought that if the net name had the word out in it, it should be an output from my processor, but that wasn't true. So I think this is this is a place where the software hitting the hardware world is really good because this is kind of like an api idea right so like ultimately what what knobs can i turn that kind of thing like on a board um detailing like all this stuff like what could

possibly go wrong is kind of tough because i don't know everything that can go wrong so well certainly things you've seen right yeah you can call those out to people who might not have have that experience so i'm probably not a good example of this because i'm not very good at documentation but um yeah what is some good guy? I guess I've seen some in the past, but it depends, too. So, like, development board documentation is pretty bad, usually. But, like, my favorite development boards are, like, the ones, like, so the different layers of abstraction, almost.

It's like, okay, the first thing that I get with the development board is how do I make it do interesting things? And usually there's apps or some kind of thing that interfaces to that. And then it's like, okay, now I need register maps. Like, what am I going to do if I'm going to go program it myself? And then layout stuff so that I know

that if I go copy it, then I can just replicate the thing as easily as possible. So like those kind of three layers of of abstraction kind of um did i answer any of that stuff i don't think i did i'm sure hearing that no i do we know what stanley's working on i guess that's the other question no i know uh also who's charlie you're talking about? Charlie was the guest. Charlie and I worked together like 20 years ago, and then again like 18 years ago. He's one of my favorite double E's,

partially because he would send me Excel sheets that would tell me what the GPIOs should be inputs or outputs and whether or not they were special, which I guess like cube mx outputs that now um which is nice but even then if if there were things i needed to know that i might miss like it's really important that you start these chips in this order that sort of thing was always on the schematic. It was like the schematic was the interface between hardware and software.

And there were support documents. But I didn't have to go learn it all myself. Yeah, that's good. I mean, I guess, yeah, the power of stuff is very important. And I've seen some really bad schematics. I've probably mentioned them on a show before. But like some where they're just like they were in some competition to squeeze as much stuff into a sheet as possible. I was really proud of a student of mine who started like documenting and adding pictures into KiCad.

I didn't realize that was possible. So like, here's what this waveform should look like. And it was a screenshot of the waveform. That was killer. That's nice. And it was like in like a bounding box. And like this part is doing like a sine wave generation. And here's the sine wave.

And this is what it looks like. And then it was all hierarchical as well. Like that was great. So I've seen different levels of that, you know, uh, I'm not sure, maybe it'd be better to like build a database of good versus bad, but I, I don't really have that, you know, because not everything is open. Right. I think that's what I want. Uh, and maybe that's, maybe there's somebody out there who already knows that,

I want an open hardware project that has all the documentation done well. And sure, there are schematics, and sure, it's really nice to get a Gerber file when you're talking about open hardware, but what about all of the documents that go towards making the software more possible? And that doesn't work for small dev boards that are like accelerometer test boards. This only really counts if you have a processor and a couple of some things.

Right. And then it would be really, yeah. So listeners, do you have an open source hardware that you can point us to that you would be like, this is good hardware documentation. This is what software? Because that's a totally different thing of like, oh, well, you know, here's the markings and make sure this part is really tricky, blah, blah, blah. Like the footprint on this is really bad. This one always flips up, you know, tombstones in assembly, right? That's what the manufacturing engineer wants to know versus software engineer doesn't care about that. And there really are different perspectives to this. Right.

It's like an empathy thing, right? That's what Chris Speck talks about all the time. It's like empathy for different engineers. What do they care about? Yeah. Because someday you may be the software engineer, you may be the hardware engineer. You never know what's happening here.

We do seem to change around. Yeah. I guess mechanical engineer would care about like part height and like things that are. I remember making a board and I had a EE friend help me and it turned out that neither one of us had looked at the part height with the case. And then I had to machine out part of the case because the screw holder. Yeah. Here's the thing. Mechanical engineers are all.

Let me tell you about mechanical engineers, since they're not here to represent themselves. They always try and make the tightest, sleekest-looking damn thing, and that's always their problem, right? And, of course, it looks nice. But, yeah, it'll get you. Don't get me started on chemical engineers. Yeah. Do you actually?

I used to work with a bunch of them, too. No. Okay. Yeah, I used to work with a bunch at Samsung. They're very process-oriented. They're pretty great, actually. Yeah, civil and chemical engineers are probably, and then mechanical engineers are probably the rigor,

the rigorous ones, and then double A's and then software. Yeah, we'll talk about long cycle times, right? They go change a bridge right or uh or a chemical process they literally could just blow up so yeah all right uh did we answer that question or no i don't know if we did i i think the real way to answer it is to ask for our listeners to answer it for us. Oh, okay. Fancy. Fancy, yeah. I think that covers everything on our list.

Really? Well, I was skipping a few things. All right. I don't want to talk about predictions for the future. I don't really have any. I don't either. Things will get worse and better. Things will get mostly better, but people will complain more. Predictions are very hard, especially about the future.

I don't know who said that, but... What are you guys reading, maybe? Maybe that's a good question to end on, because that's always like futurist stuff. Also... All right, all right. I'll be the appropriate for the show. Did I say something? I'm reading Build Your Own Transistor Radio,

which sounds like a book for four-year-olds, but I'm already totally confused after about eight pages. Oh, that's Ron. That's Ron Kwan, former guest of the Amp Hour. It's very cool, but it's not simple. It's harder than it's... No, that's a hard book, yeah. Yeah, so I'll probably go through some of the exercises in there

and build a few of those things. And Ron doesn't make simple circuits either. Like, I mean, they are simple circuits, but there's another prototyping method. He does like dead bugging and stuff like that and clapper clad. Yeah, he mentioned clapper clad. I'm like, what? No, I'm not doing that.

Mm-hmm. Yep. Okay. Yeah. what else am i reading uh you have an str book too yeah yeah yeah a book about uh things to do with the rtl str which also has a lot of detailed detailed kind of background electronic stuff so but more software okay so yeah cool stuff like that what was the laughing about is that she's usually the one reading the kind of books that i just mentioned so i so what are you reading so my kindle so so i have i have a kindle, which I usually read at night. And it usually has nonfiction on it. And for that, I recently finished a book about jellyfish. And then I am currently reading a book about surviving in the wilderness.

Because, you know, these are things that are important. Well, you guys live in the wilderness, so you never know. But the laughter was because on my iPad, which is where I do most of my reading, my Kindle app is broken and it won't sync to Amazon. And then, so I need to delete the app and then reload it entirely. But I have many downloaded books this is essentially can think of no method of keeping track of which books had been downloaded and re-downloading them later other than to read them all before deleting the app this is my to be read pile and so yes i i have i actually have taken a few days off with the holidays and have just blown through a large number of books.

You only have 57 left to go. As of today, I only have 52. Are these non-Kindle books? I don't get it. No, these are all Kindle books. So, would they re-sync when you download again? No, because... She doesn't know which ones they are because there's 800 books in our library. Actually, there's something closer to 1,500 in our library.

Can't you mark them in the Kindle app? I suggested perhaps writing them down on a piece of paper. That works too, right? That wasn't acceptable. The old ways are the best ways. These are my to-be-read books, and it's time for me to read them. And I have just been clearing them out, okay? That's great.

Okay, so what else is on your list? I finished the nonfiction book about elephants, which I do sometimes read nonfiction on my iPad. I read a terrifying book about mermaids. Yep, you heard that right, mermaids. That one's nonfiction? No, that one was fiction. Can we really be sure? I don't know, although it went really strangely well with the jellyfish book that I was reading at the same time.

Yeah, right. You opted not to download the book about cannibals because I didn't want to hear about it. You didn't let me read the book about cannibals for reasons I don't understand. Did you think that I was going to become a cannibal? No, I thought that you would provide me with cannibal facts, which I care not to hear. Right. Yeah, Chris does often have to hear about whatever I'm reading about. I finished Cat's Eye, which is an Andre Norton book.

She is a very famous science fiction author, but older science fiction, not current. And that was, sometimes the older science fiction, you just read it and you're like, wow, the world sure has changed. And this was not like that. This was very good. Yeah, I don't really know if I'm going to go on because it's sort of embarrassing. Yeah, if you ask Alicia what she's reading, we could do a whole podcast. Yeah, it really could.

Let's just say yesterday was a three novel day. Wow. That is. But I should be reading Tim O'Reilly's WTF, What's the Future book. Because he will be on our show soon. Yeah, I know. And, you know, I really was excited about this book because it's what's the future and i'm like okay let's talk about robotics and machine learning and cars and flying things and

and rockets and there's so much future that's awesome and for the first like 100 pages it's all been about history and i'm like i don't care don't care just don't care that's not even the future that's the past right Right. And WTP. Then we, then we made a website and I'm like, yeah, don't care. I was,

I was in college then. Yeah. I don't care. So that, that's been a little tough, but I'm still looking forward to talking to him in a week or two. That'll be good. What are you reading?

I am chugging my way through the Isaacson, uh, Ben Franklin book. And, um, uh, Ben Franklin book. And, um, yeah, it's great. Um,

it's going slower than I thought. I, the thing that really bugs me is like, it's getting towards the end of his life and I'm only 55% through the book. So I'm just hoping there's a lot of footnotes at the end. I haven't looked ahead and I probably should, but we'll see what happens. Um,

and then I was real stuck on, so I, I, I went through five different books about learning uh so teach beyond your reach uh mastery by george leonard that's a great book that's actually by a kickboxing coach talks about like becoming a mat he talks about loving the plateau that's that's a great that's a great lesson that's pretty much the summary of the book is to love the plateau. Learn to love the plateau.

So, like, as you're learning, like, computer vision, you just need to, like, work on the practice. And then you make these, like, leaps and bounds forward by getting better at the practice. So, that's how you learn hard things. Thinking fast and slow is... That's really hard. Of course. Well, I mean, sometimes that sometimes that's like the opposite of the message of peak which is you can't get better if you just keep doing the same thing you have to

find ways to work hard to get better i guess his plateau is probably still working hard at it yes it is and it's having a directed plan so i used a lot for like my piano practice um thinking about that stuff because like and so like so i used a lot for like my piano practice um thinking about that stuff because like and so like what i did is i went to my teachers who were taking it way too easy on me and i was like look no i need i need a practice regimen and and that has been killer you know the hard thing getting music instructors to to to give you stuff like that is so hard they just listen they're like yeah they'll be like oh whatever you want to do you know just play a song a song. It's like, no, no, no, I need a practice. And so that was great. I've been thinking, I have a list going of like, what does it look like for an electronics practice, right? But then that kind of balances against, with piano, it's like, oh, scales, I need to know scales. Okay, great. so maybe bold port is like a good a good uh prescription there for like soldering every

day or not every day but you know on a regular basis but like i have a list going of like what does electronics practice look like you know and then so that has been a very useful book for just kind of thinking through those things uh another one is understanding by design that's not very foreign because it's a i got sold a pdf book on kindle what what how where where i could have done that anywhere why did they do that i can't read it on my my kindle um or my yeah my that's what it's called right the kindle yeah i couldn't read it on a kindle because it's a pdf um making learning whole that's one that mel uh my friend mel uh suggested and then uh how to be miserable the 40 uh 40 uh strategies you already use that's a self-help book

like crazy uh but that's actually from the cgp gray um video if you haven't seen it oh he's got some really neat videos oh he's great um uh he has one about seven ways it's based on that book but seven ways to be miserable it's like a summary of it, and it's so good. It's about depression, too. So, it's, you know, basically like stay inside, don't exercise, you know, all the usual things. So, it's very tongue-in-cheek. Seven Ways to Maximize Misery. Yep. All right.

That's the one. Yeah. And then you started to say Thinking Fast and Slow. I don't, I am not a fan of this. Maybe I'm only 10% in. Oh, here we go. Here we go. No, I can see not being a fan of it. Maybe I'm only 10% in. Here we go. No, I can see not being a fan of it.

It is a trudge. Yeah. Does it get better after 10% in? What do you like about it? 10% is only 1,000 pages, right? Yeah. So, having heard you, I think you like to know how to use something before you want to be bothered whether to dig into why it's important. Correct, yes. I'm very, I'm engineer, not scientist, right? So, I don't want to know the basis of all the universe. I want to know how do I use the universe to my advantage so go to the first page of each chapter and then

take a page back at the end of each chapter he tells you how to implement all of the things he's or he gives you high level implementation on the things in that chapter and so like in the chapter where he talks about uh i'm a little hazy on exactly which chapter is which, but he talks about arguments and diffusing them. He gives the words that he's explained why these are the important words.

And so if you don't care why these are the important words, skip to that last couple pages of each chapter and see if this is interesting. Because it is, I because it it is i mean it's like a thousand pages it's got really good stuff if you read the whole thing of course these other things are going to make more sense and they're going to stick a little better but if you're the sort of person who wants the action before you want the understanding

then do that it's like context or something huh yeah yeah no that's great yeah that's that's really good advice i like that a lot actually so um okay i have to uh okay i will be a little bit of a fanboy of my own company they have like the best which we copied totally from buffer but it's the best perk ever any kindle book i want i can just buy wow company will pay for it personal or professional it is it's a so it's copied can i can never work there can i work there just just just for the benefits she will bankrupt your company i know right and you don't you don't even have to just i mean because i i buy almost all of my books from kindle daily deals i almost never

still bankrupt the company yeah it's awesome it's awesome yeah it's a great that is the best perk i've ever heard of too like because it's just so out like oh i don't have like i told my family for christmas i was like sorry uh you know you can't like i always ask for books right it's just a thing i've done since i was zero uh i didn't ask but you know like it's just always been a thing when i do holidays and stuff like that whenever someone's giving a gift but now it's not really a thing so yeah uh do is this off of your Amazon account? Or do they keep the book? Yeah, that is how they...

No, no, that's how they do it. So, it's my account. It's just a reimbursed thing. But it's just a mental thing. It's so weird, right? It is just a mental thing. Because it's like, you see a $30 Kindle book, you're like, oh, $30.

But now it's like... I have no problem with that. I want to learn that thing. If I want a book I'm just going to buy it I don't care how much it costs I was so excited when Chris finished

the book he was reading and saw that Artemis was out by the year and because there was no way I was going to buy it at full price

and he just pushes the button now I have it and I can read it as soon as I finish the other 50 books I'll totally be on it. Do you two share? Yeah. Yeah.

A Kindle account too? Okay. Yeah. So that means Chrissy is what you buy on daily deals each? Yeah. I don't like the deals. And I get the ads that are definitely targeted to her because she probably reads 10 times as much as I do.

Maybe you'll like sea anemones. Yeah, that's not what I get. Yeah, I do read romances sometimes. Oh, okay. And if you are the sort of person who even reads like one romance out of 10 books, they market heavily because most romance readers read a lot. Yeah, so my Kindle usually has some bare-chested dude staring from the front of it. Got it. It's fine. Now, the fact that I also read mysteries and sci-fi and nonfiction, they don't care. It's

all about the romance novels. Right, right, right. It's where the money comes in. Wait until that video ad comes in the mail. You open up the little video portfolio thing and it's it's fabio okay well i think we should end on fabio what do you think let me just write that down as a title yeah right i would like to actually have a spoiler uh a star wars spoiler discussion in the outro is that okay with you okay yeah that's fine so okay so now i just say thank you all for listening thank you to christopher producing and co-hosting thank you chris gamble for being our guest or for being a

host on his show we don't know if you heard any ads in this thing and you didn't hear them here you know like we didn't put them there we didn didn't. No ads. No ads. We still have Patreon. Everybody has Patreon. Amber has a Patreon. You don't get anything from being supporters, but we do love you slightly more.

We love you. Yeah. They're the best. Have a good year. Happy New Year. Yeah. Happy New Year. 2018.

2018. I think it'll be a good one. I think it'll be a good one. I think so, too. Okay. Spoilers, sir? Spoilers? Spoilers? Spoilers? Wait a minute. You were going to hook up one of the synths so that you... Nah, that was too far. Oh, man. You could have played us into, like, Spoilerland with

Star Trek sounds. Oh, we could... You know, if you want to come up with a sound, you can do it again. You can make a custom amp. You can make it a custom amp-edded theme if you wanted. You can intro your music like that. How much time do I have?

Amp power, embedded, and the Emperor's March. No, don't give me ideas like that because that's not actually practical. No, I mean,

you don't need to remix any of our stuff. I mean, yeah, no, use your Blippi stuff and just make it custom. That would be great. All right. It could be super simple, too.

I will endeavor to do that. Write us original music. In the next 48 hours. Right, right. That's right. Okay, so here is... Did you like the Star Wars movie? I loved it.

Oh, my God. Of course I loved it. Oh, good, because we know people who didn't. Yeah, whatever. And they just, you know, whatever. Yeah. What a bunch of, you know, whatever. Yeah. What a bunch of, yeah. Here's my thing.

Did you see BB-8 with the carbon resistors? What was up with that? I loved that. It's always with the resistors. You know, I have the same thing in, you know, I love Andy Weir, but, like, the whole, like, power drill thing, where he's like, oh, you just do, like, a resisted divider. Like, that's literally my only argument, my only point against the entire martian book is just the the the resisted

divider in the power drill thing you know why they do that though it's like the one component that at least a fair piece of number of people know oh that's some sort of i know oh resistor right right right did you notice how chewy stocked uped up the Millennium Falcon before he went to go save the refugees? It was good that he put all those porgs in because they were going to be delicious. Is this like the did Chewie actually eat porgs discussion? I don't care. I've read about that. Chewie's still the best.

I don't either. She thinks he stocked it with snacks. Yeah. That's why there were so many in there. I think so, too. I mean, like, they were talking about, it was like, they were nesting in there, but yeah. Yeah, sure. Self-regenerating snacks.

Yeah, right. Snacks on snacks. Yeah, so that's really all. I was just, I was very dismayed that, like, why resistors? I don't get it. Why not just make it look like future-y panel? They were sort of not, I mean, they didn't look exactly like resistors. They were resistors. They don't get it. Why not just make it look like future-y panel? They were sort of not. I mean, they didn't look exactly like resistors.

They were resistors. They were a little different. They were carbon film resistors. They were big. Or not even. They were like wire-wound resistors. They were like, yeah. They were space resistors.

Space resistors. Porgs are secretly puffins. Yeah, right. With cuter faces. Well, maybe not. Yeah, no, I loved it, though. So, I guess that's all I have to say. That's not how you fix electronics.

You don't just stick metal bits and... Stick your head against it. That's all that works. That's how I fix it. It must be why it never works. Yeah, right. Right, right. Like, they were doing it like...

It was like a slap doing it like it was like a like a like a slapstick like putting your finger in the in the leaky leaky wall kind of thing yeah wait a minute get it but like isn't aren't electronics just like water i mean that's that's a good point yeah i didn't think about that uh yeah yeah well okay i guess i'll stop recording now looks like my computer's slowing down it's a good sign