Embedded - 338: Working With People Is Terrible

Episode Date: July 24, 2020

In a surprising turn of tables, Christopher White (@stoneymonster) joins the show as a guest to talk about his career, burnout, and musical instruments.  Christopher attended Harvey Mudd College for his undergrad mathematics degree then got a Master’s degree in physics at San Jose State University. Some things he has worked on include: Multicast OSPF LISP OCT His current band is 12ax7 (12ax7.fm). The outro music is a track called “Solstice”.

Welcome to Embedded. I am Alessia White. I'm on my own as a host this week because my guest is Christopher White. What? Hello, and thank you for being on the show. How did I get here? Could you tell us about yourself as though we just met over, I don't know, conference lunch table? You want me to do it as if I was actually at a conference lunch table?

Because that's just going to be five minutes of silence. Okay, I'll act as a person who would do that. Hello, my name is christopher white i'm an embedded software engineer that's a lie i'm a software engineer i've been working for about uh 25 years something like that something like that uh on a variety of things at a variety of companies um some of them very small little devices and some of them bigger more expensive devices and some of them scary devices and some of them boring devices okay so we're gonna talk about your career going through small and scary and big and boring yeah so this is your opportunity to turn it off now and but before we do that i want to prove that it is you.

Okay. You're looking at me, so I'm a little worried. Yes or no? Do you? No. Oh, sorry. Okay. I'll cut that.

Do you know the month and date of when we started dating? August 26, 1994. How do you know that date? I took the Farside calendar day off my Farside calendar and kept it. How did you tell your brother that we got a dog? I butt-dialed him on the way home from picking up the dog, and the dog was just howling in the car. After we attended a book completion party, as we left the pizza place, do you remember what you said? It was some variant of, wow, we're really cool.

I don't think it was we. I think there was a pronoun there. I'm pretty cool? Yeah. Well, anyway, there's a whole story that goes with that that makes that sound less ridiculous than it sounds. What skateboarding injury caused the most amount of blood loss? Blood loss? Well, I've only had really one skateboarding injury.

That was one. So, there are these trees. I don't know if you know about these trees. But they drop little things that look like the bombs in all the Mario games with little spikes coming out of them, and they're hard. Like sea urchins. Yeah, but they're less long, but they're pokey seeds. And our college, there are a bunch of these trees that drop them,

and they were great for getting stuck under skateboard wheels as you were going. And so what happened in this particular incident was I was going and the skateboard wheel hit one of those and the skateboard stopped. But as you might know from physics, that doesn't mean that you stop. And so I kept going until I landed on my elbow. It was great. But as I was flying through the air, I was reassuring my roommate who was skateboarding alongside me that I was fine. Were you? At the time when I was flying through the air, I was, but I admit that I was premature in the assessment doing so before I hit the ground. What's your favorite instrument?

Favorite instrument? It's okay. You can look around. There are hints around here. The one I'm playing at the time. What is your least favorite Star Wars character? Least favorite Star Wars character. See, the obvious go-to of course is

the one we're all thinking of. Jar Jar. I really don't like one of the Emperor's Purple Pie Men guys. Those guys, they were weird and they're kind of creepy. I didn't care for them. Return of the Jedi. Favorite

number between 273 and 477 inclusive? 301. Do you have a tip everyone should know? Always check your references. I stole that from a movie. Don't always panic. How did you get the moniker Stony Monster? You really want me to go into this story?

Well, I'm going to ask you about going to school next, so go ahead. Okay. So, it's really boring. Okay. Am I going to reveal this to, to all 10 of our listeners? Okay. So in college, uh, we had this like pre period before classes start, I think like a week or so everyone was settling in, whatever. And,

uh, they had these activities and things for people to relax, get to know each other. They were, there were some nice things and then they had some things where they mixed it up and, you know, made me question ever going to college, ever. But anyway, one of these trips, little side trips, was to a lake near school.

It was probably a reservoir, but Lake Puddingstone is what it was called. And so a bunch of people went, either here or there. Nothing happened there. But at breakfast, I think the next morning, i was making a joke and i said hi i'm stony the pudding stone monster and and from then on i was stony the pudding stone monster and i just personally shortened it to stony monster and turned out that nobody on the internet had that at the time well there weren't that many people on the internet at that time amongst the 50 of us on the internet had that at the time. Well, there weren't that many people on the internet at that time.

Amongst the 50 of us on the internet at that time. There was already a Chris White. There was already several Chris Whites. But there wasn't a Tony Monster, so I used that for emails and stuff. And then I started using it for user IDs and things, and I kept using it. Even though it sounds vaguely druggy, that wasn't the connection. So the school you're talking about is Harvey Mudd College. Yes.

Why did you go there? I was a very lazy student in high school, even though I did well, which set me up for really great things in the future in school. That was sarcasm. When being lazy wasn't great. And applying to colleges was hard. So my guidance counselor in high school was really hot on Harvey Mudd College. And he encouraged a couple of us to look at it seriously.

And around the same time, I got their ridiculous ad in the mail. Because back then, I don't know if it's still true, the schools inundate people with ads. So you don't apply. And it was their junk mail ad. I don't remember what it exactly was. It was a postcard that said junk mail on it, which was quite a bit different than it was very tongue-in-cheek and self-deprecating and it appealed to me as a sarcastic 17 or 18 year old at the time so yeah i did some more research into them and i i think i i applied to them early decision but i went and visited and it seemed like the sort of place that kind of connected with me.

It was small. My high school was very small. This was bigger, but still small compared to most colleges. I think our whole entering class was 150 or 120 people, something like that. So it didn't seem scary to me to go from high school to that versus going to, say, Berkeley or something or another public university with 20 or 30,000 students. So that was nice. And it felt like going there and seeing and talking to professors and stuff and seeing the little fakie classes they let you attend and things. They were small.

So it felt like we're going gonna be lost in the sea of students so that was probably the biggest factor uh and the other thing was i was interested in science and they that was all they do pretty much and seemed like okay i should go someplace that focuses on this interested in science but you majored in math well yeah but that was the later decision and now you write software. That's a rumor. Was a math degree from Harvey Mudd good preparation for you, for your career? Yeah.

But, I mean, for a variety of reasons. One, I didn't just take math. Let's go back. I started out thinking I was going to be quote an engineer. Harvey Mudd is an engineering school. That's how they used to advertise themselves. I don't think they do anymore, but the focus was definitely engineering. Engineering was the largest department, the biggest funding and everything. And so, you know, my brother was a mechanical engineer in college and engineering seemed cool. You had to build things and learn about various cool applied areas of science.

So that was what I was going to do. Until I took my first engineering course, which I think was first semester. It was called E4. And it wasn't even really an engineering course. It was a projects course to kind of get you to learn a little bit how companies might work and how working on a project on a team might work and how to give presentations to companies. And so, you know, you get on a small team of four or five people and you have a project for the whole semester where you have to design something. It's not even something real. It doesn't have to be. Some of them are more real than others. I think mine was some sort of modular connector system for plugging. I don't even know

what it was for, but it was just a connector idea. It wasn't, we weren't building anything that did anything. And it was a nightmare. God, working with people is terrible. You know, I guess I came in there assuming everybody was going to be super excited and motivated. And it turns out even, you know, they weren't. And a couple people never did much, and there was a lot of arguing. And I just, at that stage of my development, finding out even little bits about how managing things works was a huge turnoff. And I think I just went, engineering? I don't want to work at a company. This is stupid.

And so I started looking at other things to do. And math was cool. And it seemed like I could do a lot with it. And I didn't have to necessarily decide on my career instantaneously. But you also worked for the computer science department. Yeah, well, I took a lot of CS courses. So my math degree, they used to have, before Mudd had a CS department, which was relatively new when we showed up there, they had a math slash CS degree in the math department, which meant you took a bunch of

math classes and then a prearranged set of CS courses. And I sort of did that. Not quite. I did a pure math degree and then took basically all my electives in the CS department. Anytime I had free time or free space in the semester, I would take a CS course. So I took a lot of CS courses. So that got me on to working for the CS department. What were your least favorite classes? Ooh. working for the CS department. What were your least favorite classes? Ooh, physics of all kinds.

Because I did very poorly in them. And the ones I did poorly in, it's kind of a cop-out, but I don't feel like they were taught that well. Although it was definitely on me that I failed them because I kind of gave up and didn't engage and didn't do the homework. And going back to the whole lazy high school student thing, I hadn't quite figured out how studying worked. And it turned out that even though I thought I was doing cool stuff in high school and learning things, took a bunch of AP classes, did really well, college courses were slightly more challenging. And that bit me.

So physics, I did very poorly in physics. And a physics-like course in the engineering department called STEMS. Systems Engineering. Systems Engineering. And it was basically the same kind of stuff from physics, you know, from freshman physics.

It was basic electronics and stuff. And what was the other, like mechanical systems, so spring systems and all that stuff that you get in, you know, your first and second semester of freshman physics. So they taught that in the engineering department again for some reason.

Maybe they just didn't trust the physicists. I mean, it was about differential equations. goal was yeah it's physics x and x dot and x dot dot that was it's all the same stuff yeah but it was totally and since i failed it the first time going in and being presented that with the engineering department's attitude which i might remind you i was not hot on very quick, very early on. I didn't do so great in that one, but I got a D in that class, so I didn't have to take that one again. Music. You played music then? Yes, sort of. Yeah. Yeah. I know you had an electric bass.

Yeah, that one. Yes, he pointed out. For the audience, there's a black electric bass hanging on a column. That makes sense. That totally makes sense. What did you play? I played a lot of, I played bass. My brother and I had been kind of playing music together a little bit back then, toward the end of high school and when I started college. So that's when I started learning bass. So we had a few songs we were working on even back then. And I played along to a lot of Rush records and stuff. That was pretty much what I did in college. I didn't have any formal band or anything I was doing back then.

And when you say you played together with your brother, Matthew, how did that happen? Did you physically in the same room? Did you record? Yeah, yeah, back then. Yeah, there was no way to do it. I suppose we could have shipped tapes back and forth. I don't know if we ever tried that, but no, back then. And we didn't do much during college.

He was busy in grad school i was busy in undergrad um so yeah okay and so skipping ahead a few years wait we're done with college no no no okay how did you get your summer internship junior year junior year right summer internship um starts with a C. No, I know where I went. You just asked me how, not where. I'm paying attention to the questions. Yes, it was at Cisco Systems. And it was kind of a last minute thing because everyone was trying to get an internship junior year.

This was the thing to do. And basically everybody who got them either went up to the Bay Area or went to San Diego. And there was very few other places where companies were hiring. And so I had looked around and looked in the Bay Area, applied to a bunch of places and didn't get anything. I think we had some interviews, right? So there were some interviews. There were on-campus interviews. On-campus interviews.

So I interviewed with Microsoft. That really didn't go well. It's because you didn't play Solitaire enough. I just didn't understand cards, which apparently was the whole point of the interview. The answer is always seven, even if the math works out to 6.5. I don't remember that. Yeah. So I interviewed there. I interviewed some other places. Didn't get anything. I got rejection email, rejection physical mail in my mailbox from places I hadn't applied to, which was really nice. Um, but then I ended

up getting something. It was with an alumnus who was working on, I think like a wedding planning website or something, um, in the Bay area. And he had some funds to do, to have a, somebody come in and help, uh, during the summer. And I was like, all right, that's work. I'll do it. It wasn't very much money. It was barely enough to probably survive and may not have been in retrospect. And so I kind of was signed up for that.

I was like, I don't really want to make webpages, but whatever. You know, it was back then webpages were, if you don't remember, or if this was before you were alive, this was 1994. So it was, you know, type in HTML, and that was it. And if you wanted anything to happen on a server, it was something called, oh crap, what was it called?

Anyway, you called C code. Like, it wasn't, there was no JavaScript, there was none of this stuff. Um, uh, I can't remember what they were called, some sort of scripts that your webpage could call out to, but it was highly insecure and a lot of typey typey, not a lot of visual design. Um, so I wasn't super excited about that. And then I just went to the career center and started looking around again toward the very end of the time when I could have found something. And there was an email in one of the binders in the career center saying, Hey, my team at Cisco is looking for summer intern, blah, blah, blah, blah, blah. Have them send me their resume. And I said, well, I'll try it

because I'd been doing some networking stuff in class. I enjoyed networking. The concepts were cool. I think we had our clinic project, which we can talk about later, was networking-based to some degree. So I had done some things. So I threw my resume out in an email, and they responded and said, why don't you come up for an interview? And I was like, what? That never happens. So I flew up there for a day and did an interview and did well enough that they had me do the internship that summer. How big was Cisco then?

Thousands. A couple thousand? Yeah, probably something like that. I think my employee ID was like 4,500 or something like that. So assuming some people had left, they were probably in the 3,000 or 4,000. And you went there, I was going to say you went there after college, but the truth is you worked senior year for them. Yeah. I don't know if I'd finished what I was doing or I kind of had finished it, but there was more bug fixes and I was working on some other things. And they just let me stay on as an intern remotely.

But still at a much higher rate than the college paid us to do. Well, the college paid us minimum wage, which in 1994 was like five or six bucks an hour. You'd buy a lot of ramen for that. So yeah, that was great. I got to work remotely, which, you know, it didn't seem like a big deal back then because it just seemed like, oh, okay, I work at a networking company. Of course I can work remotely. So I would SSH into computers up in the Bay Area from Los Angeles and had a little, you know, we had a two-factor authentication thing. Even back then, I had a little calculator doodad that gave me the one-time numbers to log in. I thought that was super cool, like I was in the CIA or something.

But yeah, it was really nice. And I didn't have to work a ton during class. It was just, you know, fix a bug here and there, keep up on the bug reports coming in on the stuff I'd worked on. And then you went back to Cisco for full-time after school. Yeah. Did you like working there? I did, for quite a while. It was a good place to learn because things were established and it wasn't like the startup environment being thrown into the fire immediately.

They had written their own operating system, so learning some operating system concepts from something you could read the source to was interesting. Had some good mentors there, some very good mentors, people I've continued to keep in touch with and work with for the 25 years since and gotten other jobs or worked together on contracting things. And it wasn't a sweatshop. The tech, normal tech companies then weren't driving people

to work 80-hour weeks and stuff. So it just felt like a normal job, and I learned stuff. And even when you lived in the Bay Area, you still worked from home sometimes. A ton, yeah. They put an ISDN line in our apartment, and I had an actual X terminal as my second computer that just did X windows to inside the Cisco network. So I could get to my computer on my desk at work,

which back then was a Sun workstation, and to do whatever. And it was great. And of course, back then, windowing systems were so simple that 128 kilobits per second was just fine for, you know, basically not quite VNC, but VNC-like behavior.

And you worked on multicast? Multicast routing, yeah. What is that? So normally on the internet, you have an IP address for your device, and there's an IP address for a thing you're talking to, whether that's Yahoo's data center computer or a computer in your house that you're in your server or something and so when you send packets back and forth you uh tag them with the ip address of the thing you're talking to and yourself this is me and i want to talk to b or whatever and so that's the the

things in between the switches and the routers, look at those addresses, and they have tables that tell them how to get to where that address lives. And there might be multiple hops in between, multiple different kinds of interfaces and wires and satellites, optical fibers, whatever. And those devices know how to figure out how to take that bit of data, that packet of data, and get it to the thing it needs to go to. But that's one thing to one thing. And back then, video was kind of just being thought about over the internet. There were some things you could do. Real Player was just coming out.

I don't know if anybody remembers that, but it was one of the first kind of internet video things. It wasn't like YouTube. It was more like RTSP, real-time streaming, which some of us are still using to our great horror. The problem is with that kind of one-to-one thing is if you want to, let's say you wanted to run a television company or a cable company over the internet. Let's say you have a million people that you want to send a TV show to. If your server has to send a separate copy to each of your million customers, it really doesn't work very well. And that's what would have to happen with Unicast delivery. Unicast being the one person to one person.

Multicast was built into the concept of IP, the IP standard, but it hadn't really been explored that much. Multicast lets you have an address that corresponds to a group of people. And then the routers in between only spread that out at the last moment they need to. So it might take one packet to go 90% of the way to the group that needs to get that packet. But then the last router says, oh, I'm at the

neighborhood now, and I need to split this packet and send it to this house, this house, this house, and this house. So it compresses all of that replication down to very, very little, depending on how things are structured. And it was a big efficiency advantage. And so what I worked on was those tables, getting those devices that know how to route things from one place to another, how they communicate to each other, and how they develop a map of the network so that they can figure out how to take that packet from one person and send it to the groups who are interested in it. Is multicasting used now? Yeah. For what? Not for Netflix, because that's got to be point-to-point. If I press pause, It's point to point. It turns out that, you know, people built a lot of network infrastructure,

but there's, um, it's used for like cable TV for actual cable TV. Uh, all that stuff uses it, um, for live casting of like SpaceX launches. I don't,

I don't think so. I mean, it depends. Maybe. Yeah. I don't think so. YouTube doesn't do that. Um,

uh, but it's used inside places more than over the internet. The original vision was over the internet, but I think it's used more inside places now than outside. What instruments did you get while you were there? at cisco um at cisco i got that bass in the corner which the upright bass which you got me full big size bass yeah and the acoustic big of course yeah acoustic acoustic that's symphony like thing yeah uh and i got the green bass over there the six freles. I believe that's all I got. What kind of music did you play then? At Cisco. I don't remember.

Was that when we were hanging out and doing bluegrass stuff? We definitely were hanging out with bluegrass people. Yeah, so I might have been pretending to play some bluegrass. Oh, I was in a little blues band. Yeah, yeah, yeah, that's right. I forgot all about that we played a couple of couple of things in mountain view there's a there's a there's a betamax or vhs or whatever was alive back then videotape of that performance why did you leave cisco uh well

i've been there four years three years four years four years. And as all things do, things change. And I had gotten to the point where I was not working on kind of big self-contained features anymore. I was chipping away at smaller features and tons of bug fixes. Some of the people I worked with had left and it had gotten quite boring. And at that point, they were starting some new big projects and I got shunted over to a new team of people I knew, but a different team doing basically a rewrite of the operating system, which I think they eventually released,

but it took a very long time. The typical very large company, oh, we want to rewrite the operating system that three people wrote in 1988. Let's put 600 people on it. So I wasn't having fun with that. I wasn't having fun fixing bugs before that. A bunch of people I knew had left,

and I got a call from one of them and he was at a new startup and basically asked if I was interested in coming over and talking to them. And I was like, well, uh, sure. Why not? Everyone I know is already there. I might as well go. Well, and your mentors were there. what was Procket like yeah the company was called Procket um they were making what's called a core router which is what the biggest most powerful uh routers that drive the internet there were things at the middle of the network so where all the traffic ends up going before it has to split out and go somewhere else um to kind of aggregate all the traffic ends up going before it has to split out and go somewhere else, kind of aggregate all the big wires and optical cables and things. So Procket was working on a core router with custom silicon, full custom chips, which had never been done before for networking.

Full custom meaning like a microprocessor process from scratch. So that was interesting. No software had been done.cessor process, like from scratch. So that was interesting. No software had been done. It was just all from scratch. Very small when I joined, a couple dozen people, something like that. Maybe I think I was employee 19, which for a young person was cool and strange and maybe not appropriate. Yeah, and when I got there, there was nothing.

I mean, very little. The chip people were off in the corner doing their chip stuff, but we needed a software stack to run the router, which meant all the routing protocols had to be implemented, IP routing stack. We needed an operating system. We needed an architecture. None of that existed.

And it was a night and day shift We needed an operating system. We needed an architecture. None of that existed. And it was a night and day shift from sitting and fixing little three-line bugs at Cisco, you know, once a week. If I could manage to get through the process to, hey, we need all this stuff. And, oh, yeah, you were doing routing stuff, but now you're doing operating system design and this and that and get on it. You became responsible for many things while you were there. Yeah. And it was sort of sudden because you went from being a junior engineer to not having very many people tell you what to do. Yeah, and having to decide what to do. It wasn't that I was a manager or anything or

even a tech lead. It was just there weren't very many people. Procket acted like a startup. Oh, it acted like? Well, I mean, Procket was a startup, but they acted like it too. There was free food, movie nights, exhortation to work weekends and long hours. Video games, ping pong, expensive offices. I had my own office, which was about the size of your office in the house, with a door and glass windows outside and inside. Two desks. Yeah. Do you, well, the offices aside, because the other parts, because... And the other parts.

Do you think it's effective to getting people to work more? It's not effective for me. It's probably effective for some people. I didn't find it particularly useful. I found it kind of annoying that some people decided to start coming in at lunchtime and then stay through dinner so they could just get free dinner. And then it looked like those people were working late. But I usually got there early in the morning and tried to leave at a reasonable time. So it was always a source of annoyance and friction.

No, I didn't really like those things that much. But you liked working at Procket? Some. For a while. What was the most important thing you worked on there, or the most important thing you learned there? Oh, God, that's a hard question. It was the first place I had to think about software

from a clean design point of view. At Cisco, the code I had as examples to work from in retrospect wasn't very well written or it had been written many, many, many, many times. Kind of rewritten and overwritten. So it may have started out well designed, but had accreted lots of fixes and things.

And so the code I was writing there wasn't very good because I was taking that as an example, and I was new, too. I mean, it's not the example's fault necessarily. You know, I tended to do things like very long if-then-else chains, taking many pages to do logic instead of the state machine or something like that.

And part of the fault of that is that I didn't take probably the CS courses that would have taught me better software design. I took the ones that seemed fun to me. I know you're shaking your head as to if they didn't exist. Maybe they didn't exist. So yeah, I had to learn that real quick because I didn't want to make a mistake. So, you know, I had Knuth books for stuff. I read a lot of books, read a lot of algorithms books and tried to figure stuff out. So that was the biggest, most important thing I learned was, okay, this is a clean sheet project from scratch. What would it look like if you could do what you thought was appropriate? And being able to think about a project that way. Being able to do a design from the beginning without writing any code.

Writing a very long document about how I'm going to design this thing. Have it reviewed by other people before writing any code. That was really important and huge. this thing, have it reviewed by other people before writing any code. That was really important and huge. Yeah. And you did something to Juniper routers from Procket. How did that work? Well, the main thing I was working on after getting the OS stuff together,

and I wasn't the only person doing the OS stuff, but I did the memory management and designed how we were doing message passing and things like that between processes. Because we were trying to do kind of a new thing. Most routers back then, the operating system was cooperative, multitasking, single address space. If your routing protocol crashed, your whole router crashed. Which, you know, was a problem.

And Juniper had actually done that too. They had a real operating system. And they were a relatively new company too. Yeah, they were the first post-Cisco routing startup to take a bunch of Cisco people and start, and Procket was a follow-on to that. So after doing the OS stuff, I was tasked with writing what was called OSPF,

which is a routing protocol. Like IP is a routing protocol? No, IP is a packet protocol. IP is just the definition of how the packets get put together, the headers and things and addresses and checksums. But it doesn't do anything if you just

throw that out there on a wire. Would I have heard of any other routing protocols? Well, have you heard of BGP? Have you heard of ISIS? Have you heard of DVMRP? No, that's a multicast protocol. That's a lot of acronyms.

Yeah. There's a few of them. For different purposes. This is how the bytes go places. How it decides where to send the packets. Is that right? Yeah. So remember how I said

each of these routers have tables and things? These protocols are how the tables get generated. So if you have a network of devices, routers, and switches, but let's just talk routers because there was a distinction. There was. The routers talk to each other to find out how to get to the things they know about. So you might have a router that connects directly to your house. And so it knows, oh, I have Alicia's house and I have her neighbor's house. And so those are directly connected to me.

So I know about those networks. And so it might talk to the next router upstream and say, hey, I know about this group of houses. And so if you want to the next router upstream and say, hey, I know about this group of houses. And so if you want to talk to this group of houses, you need to go to me. And by the way, the wire between us costs X amount of time or something. There's a metric associated with the wires to say which links may be better. Maybe they're faster or something like that. So you say, okay, I'm router A, I connect to this neighborhood. And hey, router B, I got a neighborhood and I can get to them with

cost, blah. Here, stick that in your table. And so you aggregate all these things and it's very complicated and there's actually a lot of algorithms associated with it. But at the end, all these devices have tables which map out the entirety of the internet or pieces of the internet. Most of them don't have a picture of the whole internet. They have a picture of where they need to get to and then kind of aggregated things that say, well, if I need to get anywhere else, I just throw it over there. He'll know how to figure it out. So OSPF was one of these routing protocols. there's kind of two classes of routing protocols interior and exterior um the exterior ones are kind of how big things big organizations get to

each other like how one isp gets to another isp and the interior is how do i get to things within my own isp or my own company? So OSPF is an interior one. And they tend to have kind of more complicated algorithms because things might change faster inside an organization. And if something changes, you don't want routing to be pointing in the wrong direction for very long. So the algorithm is for how to take a change way over in one side of the company and propagate that through all the routers within the company quickly, figure out which paths things should take due to that change. Those are the kinds of fun parts of routing that I got to do with OSPF. Okay. Answer your question, whatever it was.

I don't even remember what the question was at this point. Okay, no, I asked you about crashing Juniper routers. Yeah,

so, right. So I wrote OSPF and as part of doing this, an organization might have lots of routers and they might have lots of routers

from different companies because it's good to have diversity within your hardware and everywhere else. But, so one of the things that we would do was interoperability testing, and I think at some point they had some of our devices.

Oh no, maybe some of ours were on the network, but I don't remember the circumstance exactly. But yeah, it turned out that I had made a mistake somewhere and was sending a slightly malformed OSPF packet to the neighbor routers. And one of the neighbor routers was Juniper's and it crashed them every time. I thought that was great. I was like, well, fix your bug. Shouldn't be crashing. You should sanitize your inputs. It's not my fault. Is OSPF still used today? Yeah.

All over? Is it? Yeah. Oh, okay. Okay. Yeah, all those routing protocols are still the same ones. BGP is still the thing that runs the backbone of the internet, and it was invented in 90-something. Music while you were at Procket. Yeah. Do you remember anything about that very little let's see rocket i was there like 99 to 2002 i might have gotten kind of a frankenstein drum kit then i think i think that was around the time i got a frankenstein drum kit, somewhere around there. But I wasn't doing too much, I think.

Well, I think I tried to play in the HP Blues Band. Wasn't that? Yeah, the HP Blues Band. We had a couple of gigs and I played bass in that. And then I think I did an open mic night at JJ's Blues Club in San Jose, which was my first time playing live in a small club. I'd played live before, but it turns out you can really play badly when you can't hear yourself at all. So that was a demoralizing experience.

Yeah, I was at Procket, because I remember getting the call from the guy in the HP Blues band who was forming another band that I think my open mic night was kind of an audition for him. I remember him calling me in my office at Procket saying, nah, yeah, no. Went well, huh?

How did you leave Procket and why? Okay. So I burned out completely. It turns out three years at a high, Procket was not one of the 40-hour-a-week companies. It was extremely high intensity. It was one of the darlings of the dot-com era. I went through a tremendous amount of funding, was valued at billions of dollars at one point.

Very high pressure. Things were not going well for a long time. Turns out it's very hard to make six full custom chips. Really? Yeah. Who could have guessed? But it did come together in the end, and they had a product, and it worked, but unfortunately the.com boom busted

right around that time. And I was just tired. Real tired. But you left before the bust. Sick of software. You left Quittle? No dot-com bust was around 2001 2000 2001 i think and i left in 2002 so stocks had already gone to nothing uh yeah so i completely burned out decided i hated software and engineering uh just like i did freshman year and was gonna go

back to school because i'd been doing some astronomy astronomy was cool astrophysics seemed cool i'd been reading about stuff and seemed like i wanted to learn something again so uh yeah i so your manager pulled you into your into his office to give you more stock options and more stock? Yeah, bonus and stuff. And I decided I couldn't in good conscience take that with a smile and then quit a week later. So I just said, yeah, you know what? Thank you for all of this, but I'm out of here. I didn't say I'm out of here, but. But you meant it. I tried to be polite. Okay. So you went back to school.

I went back to school. You ended up with a master's degree. Master's degree in physics. Physics. Physics. A few minutes ago, we talked about physics. It was a long time ago, like 30 minutes ago, but yeah. And you didn't like physics.

Did not like physics. And then you decided to get a degree in physics. So here's the thing. I'm a very strange person. And even six years later, I still was mad about failing physics undergrad. And like I said, I was interested in astrophysics. And you can't get an astrophysics degree without the physics part. Sadly. It's just astrology then. You know, I decided... Astronomy. Well, that's pretty funny, because when I left, the CEO of Procter & Gamble said,

so why are you leaving to study astrology? And I was like, yeah, I'm just going to go. But yeah, so I was mad about failing physics. Mad, you know, internally mad, not at physics. I was mad at myself. And I kind of wanted to, to, it sounds strange, but I kind of wanted to clean that, clean that part of me up and prove to myself I could, could learn it. And so I, yeah, I talked to San Jose state. I thought, okay, I can go there, do a master's degree and then do a PhD

somewhere else. Um, if I feel like it and And talked to the department head there, and he said, yeah, come in, take some summer classes, because this was around summer 2002, and take some undergrad courses. And if you like it, then we can talk about, you know, officially getting you in the master's degree program. So I took courses in the open, whatever you call it, open course system they have where anybody can take courses took a few there and then and then signed up for

the master's degree yeah um yeah and so i did uh em i think over the summer undergraduate em which is the class i had failed in my undergrad experience and then gotten a C- the second time. And I did really well because it turns out when you're not mad at yourself and you actually study, you can do well. So I kept at it and enrolled the following fall and took a lot of – they had me do the entire undergraduate physics program before I could take graduate courses. So it took me a while. I didn't have to, it wasn't as if I was a college student, undergraduate college student,

but so I didn't have to take like electives outside that. So it was a little faster, but it was a lot of work. But you only were in class for like a year, a year and a half before you went back to work? Two. Two. Two years. And you went back to work and you took classes still.

Yeah. So I hadn't finished after two years because it took at least a year, year and a half to get through the undergrad garbage stuff. Thanks. Undergrad courses, the cool undergrad courses. And I've been doing research i'd started my master's research sometime around there toward the end of that period and it wasn't going well and i was starting to sour a little bit on the idea of the phd uh and i think there was a summer where i was supposed to be doing research and i was just struggling really hard.

My advisor had left the country, so I was working remotely with him. It was just real hard to sit at the house all day alone and try to do physics. So I kind of didn't do great with that and decided, hey, at the same time, somebody I had worked with at Procket said, hey, I'm at this new company. We need some help with something. Would you come in and contract? So that was that summer, somebody I had worked with at Procket said, hey, I'm at this new company. We need some help with something. Would you come in and contract? And so that was that summer.

And I said, sure, I'll come in for a little bit, do some software. Maybe it'll help with my mental state. And it did. And I enjoyed it. I had fun. And I was writing software again. I was like, well, I can do both. I can finish this master's degree and work.

And maybe that will keep me busy. What company was it? A company called Reliant Technologies. And you knew people there? Just one person. The person who kind of brought me in, yeah. Okay. And how did you know? I worked with him at Procket. He'd also been at Cisco before, but I don't think I crossed paths with him at that time.

And what did Reliant make? it was a dermatological laser. It was a medical device. And you scanned it on people's skin and it did a variety of things. It shot, well, it shot lasers into your skin. One laser. One laser split up by a bunch of optics. And the laser made little dots in your skin. Not burns exactly, but just heated spots.

And it turns out that did something to erase wrinkles, scars, birthmarks, all kinds of cosmetic issues, some of which would help people who were really bothered by them, some of which were kind of rich person age-related stuff that probably wasn't that important to be spending a lot of a lot of effort on that's where the money comes in from something was generally helpful so it wasn't like i felt like we were doing something completely just targeted at the super rich um but the coolest part about it was the technology was cool. There was a lot of, it was my first real embedded thing, you know, where I had software that controlled motors and lasers and power supplies

and had to do safety stuff. And I had to deal with the FDA. And, oh, and I was a manager. They hired me as a manager, so I was running the software there entirely. So that was a huge challenge coming out of, well, basically two years of grad school and being in a startup before that. So I had to learn on my feet there as well. Yeah, that was my first encounter with things like RTOSes, SPY, I2C, all that good stuff I learned on my feet there. And there was a lot of really weird stuff at that company too.

So I learned a lot about weird companies, weird choices in technology, and, you know, how to put your foot down. Because it was a medical company. Yeah. And it was possible to hurt people. How did you learn that? How did I learn which? How did you learn how to be a manager and how to say no? Saying no was a little easier when the FDA existed outside, knocking on your door.

Because you could always point to an outside agency and say, we're not going to do this because it violates this regulation, or it doesn't live up to what we've said in the specification, or stuff like that. Or it's, you know, this is likely to be a safety issue. When you have that to fall back on, it's a little easier than when you're making a toy or something. And it's like, well, we're not going to do this because it's not a good idea. Okay. When you don't have a legal argument to back it up, sometimes it's not as easy to convey

why you might put your foot down about some technical decision. Learning management, I don't know that I did very well. It was a small team. I just kind of did what I'd seen other people do. Tried to protect my team from things they needed protection from and tried to inform the rest of the company about what my team did and make sure that we got, you know, a place at the table. But I probably didn't do a lot of the things one should do. The people you were managing were relatively senior? Much more senior than I was, yeah.

What kind of music did you play at Reliant? At Reliant, I started in a real band with my brother and some other folks it was called the ballistic cats uh started that around 2006 sort of half-heartedly and then it sort of snowballed from there for another oh five six years what kind of music is ballistic hats uh sort of hard to describe rock and roll but sort of uh with an older vibe to it so a lot of surf uh the blues rock um americana kind of stuff so we played rockabilly rock some rockabilly stuff yeah that kind of flavor but all mixed up a lot of original stuff we had i think we did three or four records or three records in an ep i wasn't on one of them but um yeah it was it was it was

intense for a while and what instrument i played drums in that band so you'd officially moved from bass to drums. Yeah. With Reliant, was there anything that was really important or cool that you learned? A lot of things. I mean, Reliant was a very screwed up company. So I learned what bad companies look like and shady companies look like to some degree. Are they still in business are we gonna get sued no they got bought by someone else um it wasn't super shady but you know there's nothing illegal it was just the marketing sleaziness i guess of some things just being being at that level and seeing being in those meetings and seeing how

discussions about that kind of stuff happen and stuff was eye-opening it wasn't all you know technology and cool cool cool here's this new device we're making you know there was real stuff happening and real customers and patients who weren't the customers the doctors were the customers but the patients you had to look out for first because they're the ones that your device is being used on. And I'm using too strong words with stuff like sleazy and stuff, but you see the market you're in. And dermatology, even though we were treating things

that were helpful for a lot of people, it was still what you know, what you think of the kinds of doctors who are buying stuff. There's a lot of LA doctors and things and they have certain, they had certain ways of wanting things and they weren't necessarily compatible with the way we wanted to make the devices too. But in the networking, you weren't involved much with marketing. No. And even if you were,

the people you were marketing to were very, very technical. Extremely technical. And so this was a very different application, customer base. Yeah. And the company culture was not great. It was very... customer base. Yeah. And there were, you know, the company culture was not great.

It was very sexist. Things that bothered me a lot there. And that was a surprise coming from Cisco and Procket, which were definitely not that way, or at least not where I saw

or even was able to see at that time. So I had visibility into stuff that I didn't particularly care for. And while I learned a lot and while I had a lot of control over technology, I didn't have a lot of control over those other things. So the learning experience was, yeah, I learned a lot about embedded very quickly and how to make large-scale device software from the ground up and FDA stuff. But I also learned not all places are necessarily well-run with good cultures. It showed you what a company culture really can do for a company.

Yeah. So you left there. Yeah. And you went back to Cisco. I did. Why? And how was it different? I have no idea what I was thinking. I think I really wanted to get the hell out of there. So that was part of my thinking.

You've been there for about three years, relying. 2004 to 2007, something like that. Yeah, yeah, 2004 to 2007. Yeah, so Cisco, I think I interviewed at some other places. No. No, I didn't. When did I interview at Apple? I don't remember. After Cisco. After Cisco.

After the Cisco. So I went to Cisco because some folks at Procket ended up there. Procket had bought Cisco. Nope. Nope. Nope. Cisco had brought Procket for a very small fee. Basically for the employees and a little bit of the IP.

And some of those employees stayed there. And after several years had kind of revived the Procket chip architecture within Cisco, they were going to redo it and have it be the basis of some new Cisco switches and things. And Cisco had already taken Procket's software and made it the basis of a new OS. So that was kind of cool. So they were doing the chip thing,

and I'd been talking to some friends on that team and wanted to get out of Reliant, and they were looking for some ex-Procket people who knew things. It sounded fun, so I interviewed there and ended up heading over there. Did you enjoy it? Not at all.

What happened? So the bellwether was I got there and they didn't have a computer for me for three or four weeks. I had a desk and a stack of papers. What do you need? More as a software engineer. So that was OMEN 1. OMEN 2 that I should have paid more attention to was they hired me to do a different team. And I was seconded to the X-Procket team. So I was supposed to only work on the X-Procket stuff, but I was on a different team

that was doing just normal mundane switch chip. There's a network processor in the switches and they were doing normal driver stuff for that. So that's what that team did. The X-Procket, started out okay, but then they just went really slowly and people stopped showing up and, nothing was happening. And I'd end up going for breaks with the people,

the other couple of people who came in every day and, you know, I have a lot of coffee and talk about the impending financial crisis and long walks, British politics and long walks. And, uh, and the team I was on,

the official team I was on started wanting more from me. So I started working on very boring driver code, um, and then discovering that, uh, Cisco software development processes were a bit molasses like. So anytime I wanted to make a one line change, I had to go through

24 to 48 hours of static analysis. Builds took four or five hours. So, by the time I got back static analysis

saying it was okay to commit, the tree had already moved on, so I would probably have to start over. So, I don't think

I actually successfully committed anything in the eight months I was there. Well, I don't need to ask you what made you leave, but you were still going to school at this time. Still working on your master's degree. Mm-hmm. Yeah.

And then... Close to finishing, but not quite. And then you did some consulting after that. No. I interviewed with Nuvation, which was a consulting house. And I got talking to somebody I'd worked with at Reliant about another medical device company. And ended up going to the medical device company. Sawtooth. Sawtooth Labs, yeah.

What did they make? They made what's called an atherectomy device um and ather from from arteries uh atherosclerosis treatment device um what it was was a catheter based device that could go in somebody's leg arteries. And it had an imager on it, so the doctor could see the inside of the artery to see what kind of disease was present. And then combined with the mechanical part of the catheter,

the idea was you could see where disease and plaque was, and you could remove it. So that's an itty-bitty little camera? Not a camera. Optical fiber. So the imaging sensor was optical fiber-based. I used a technology called optical coherence tomography, which is sort of an interferometry technique. And you shine a laser into the tissue, and some of it reflects back.

That interferes with the light at the end of the fiber. And some of it comes back all the way down the fiber and you get it on a photo detector. And then by manipulating the laser in certain ways, that ends up being the Fourier transform of a scan, a single one-dimensional scan into the tissue you're looking at. So you inverse Fourier transform that and you get this intensity profile. And so you scan the fiber around, and then you build up, kind of like ultrasound, an image, a complete image out of these one-dimensional sort of slices. So yeah, we did that. That was super fun.

I got to use my physics background because I had to read a lot of papers about the optics and how to do the image processing, how to clean up noise and stuff. And there's a lot of high-speed data acquisition, really high speed, because we wanted, eventually we wanted like 30, 40 frames per second of a whole rotated slice. And if you think about how fast you have to be doing things to get high resolution and spin that fast,

it's quite a lot. So that was fun. how fast you have to be doing things to get high resolution and spin that fast. It's quite a lot. So that was fun. Until it wasn't. Well, how big was the company when you started? Just a few people, maybe five or six. And the CEO, the funder? The founder.

Founder. Yeah. And funder. Wasn't involved for the first bit. Yeah, I think for like the first six months, he wasn't. I never met him. Didn't involved for the first bit. Yeah, I think for like the first six months he wasn't. I never met him. Didn't really talk to him much. I think he called into a couple meetings, but that was it.

And then things started to change. Yeah. So he took on an active role. Company's focus got a little defocused. They weren't sure what they wanted to make. Um, he didn't like the answer that whatever we made was going to take at least two years, especially with regulatory involved.

Uh, so we kind of floundered and did a lot of lab stuff and experiments for years. Um, which was fine to, you know, wrote some good software. Um, it was a good learning experience again.

And, you know, I got to hire somebody who I really liked, a couple people. No, just one. No, did I hire? I can't remember who hired the second person. It might have been Dennis. But I hired Dennis, who is great to work with. And so that kept me going for a long time.

But, yeah, it was kind of a slog, and then, you know, it was definitely a startup, intense startup environment again, and I got really burned out again. And my mental state was very poor. How did that, I mean, what did that look like? Well, what do you mean what did it look like? Like, how did I feel? Yeah, like, how did that come out in your life?

A lot of physical kind of distress, like my stomach didn't do well. I had a lot of problems with my stomach. I had a lot of problems sleeping. You know, your general anxiety and stress, physical symptoms. Yeah, it didn't feel great. But before all that, you did get to see the device work. How is the OCT catheter different than just a normal catheter? I mean, it's got this imaging, but how do they usually do cleaning out plaques from arteries? It was rarer.

These devices existed before imaging, but it's super hard to do without good imaging because you might take something you don't want out of the artery. Seems bad. But, you know, they do similar things with x-ray or ultrasound. There was an ultrasound that could go in the arteries as well, but the resolution was quite a bit lower than the optical-based system. But yeah, usually x-ray guidance and stuff. But yeah, for the disease, it was targeted at peripheral artery disease,

which is something that happens to a lot of older people, especially with diabetes. Your leg arteries get plaques in them and your circulation gets really bad. And a lot of times they have to do kind of pretty invasive bad stuff to resolve that up to, you know, amputation and stuff. Um, so this was like, okay, if we can go clean that out, we don't have to, we can save people's legs. Um, so that was the main goal to start with. And I, we did that. I mean, the product worked and did save a lot of people's legs and they,

they wanted to go on to cardiac stuff after that, which was a much bigger challenge. But if it worked for that, that would be a huge game changer and, you know, replace open heart surgery in some cases. I don't know if they still exist to some degree. I don't know if they've gotten there. And around this time, you also finished your degree. I don't know if they've gotten there. And around this time you also finished your degree. Yeah. So there was some additional stress maybe floating around. I seem to recall one question that was important in your degree.

Yeah, yeah, yeah. How many electrons are in a chicken? I don't remember. It's a lot. Especially if you start out with a 10-kilogram chicken. So I decided not to do a thesis because I wasn't enjoying research, and I went back to work. And a thesis is a different kind of work. So to finish my master's degree, I did an oral exam instead. And so you had to get a bunch of physics professors in a room, and they just pepper you with questions while you're at the white blackboard.

And for a couple hours, it was pretty intense. But the first question was, how many electrons are in a chicken? And being a good physicist, I approximated things to nice round numbers so I didn't have to do a lot of arithmetic. And I started with a 10 kilogram chicken. And they thought that was a little large. I mean, you've been a vegetarian for a long time and not a cook before then. But 10 kilograms is a lot. It turns out I use the metric system exclusively in my physics courses.

But I never really got a working sense for what a kilogram was compared to pounds. These are just numbers. Who cares, right? So 10 kilograms, what's that? It can't be more than 10 pounds. So as things with Sawtooth, now Avenger, became more startup and you decided to leave. How did that work?

I basically said, I'm either leaving or you're converting me to a contractor. Uh, cause I thought that if I switched to a contractor, I'd be less emotionally invested, less mad all the time. I wouldn't have to be there at the hours they demanded. Uh, I could work fewer hours. I would feel less guilty about taking off if I wasn't feeling well. And, uh, they said, yeah, let's do the contract thing. And I did that for a while. And did it work? Yeah.

Yeah, it did. I'm still mad, but I wasn't mad mad. You weren't as invested. Yeah, yeah. And you didn't feel bad about leaving. Right. And I passed it off, you know, I passed the team off to Dennis, who was much better than me at being a director of software.

So it all kind of worked out. And if you want to work for Dennis now, you should contact Cruise. Yeah. Let's see. You started consulting in general then, not... Yeah, I started taking on other clients. And you went back to Cisco. It's not going back to Cisco if they're a contract. For the third time. going back to cisco if they're a contract if they're for the

third time that's not going back if they're a contract uh yeah so another mentor who was at cisco and had gone back there from from procket and i think he bounced around a few other places first but uh was doing something cool and needed some help doing a mobile implementation of a new protocol. Not exactly a routing protocol, but sort of a means of putting a network

on top of the network, an extra layer of indirection. This is called LISP? LISP, Locator Identity Separation Protocol. The idea was to kind of, you know how everything has an IP address? It's kind of bad to have that be an identity of any kind. It doesn't really go, it goes with a device, not with, it doesn't even go with a device.

It goes with a device, comma, the network you're on. Like, if you have your cell phone, you get an IP address from Verizon, but when you're in your house, you get an IP address from your Wi-Fi network. But your phone is still your phone. And if somebody wanted to get to your phone, they can't just go to an IP address. There has to be some out-of-band mechanism to figure out who you are.

And there's lots of applications where that might be nice. Here's this identifier for your phone, and it sticks with the phone, and that can be used like an IP address. So that was this layer of indirection protocol that he developed and Cisco was working with too, and needed a mobile application for demos. Like, okay, here's this working on a cell phone.

Here we switch from Wi-Fi to cell, and I still have the same TCP session even though my IP address changed. Is that in use now? Yeah, I don't know where exactly. It's definitely in use. It had a lot of data center applications, it turned out. For your VM, you could give it one of these locators

instead of an IP address. And then if you moved it from blade to blade throughout the network there wasn't a whole renumbering dance you had to do how did you find consulting jobs a lot of them found me a lot of them were just prior clients so like stayed consulting for reliant for not reliant uh avenger sawtooth which became the oct artery place uh stayed consulting for them until 2016 or thereabouts so for a long time um and there were some other ones that were follow-ons to that other companies doing oct that needed help and there were very few people who'd done the software for that so that was kind of a niche thing I could do. But yeah, just network, word of mouth.

I didn't go seeking stuff. Have you ever gotten a job where you didn't already know somebody who worked there? Aside from the first Cisco internship? No, I have not. I have not gotten plenty of jobs where I didn't know. So I've applied to lots of places, but I've never worked somewhere where I didn't know at least someone, I think. And that's, I mean, yeah, interviewing is hard, especially if you don't know someone. You worked as a consultant at Fitbit. Yeah. And then you a consultant at Fitbit. Yeah. And then you switched to full-time.

After all that, pulled on Fitbit as a client around 2012, something like that. Yeah, 2013, 2011. Seems so long ago. Damn long time ago. And then they became my primary client after a few years.

And then many years after that, they gave me an ultimatum and said we'd like you to become full-time now not because we're going public we don't want to know no they were already they weren't no they weren't public yet i think that's why they were converting everybody public really yeah yeah yeah they were public already cut were public. Really? Yeah. Yeah. Yeah. Yeah. They were public already. Cut this part. No, it's fine. They were public already. Um, but,

um, as something happened with contractors and they decided they didn't want any anymore. And so it was unstated that if I didn't go full time, they would terminate my contract. But it was a, you know, it was a nice,

nice offer. And I said, as long as I can work remotely and nothing really changes, I don't care. Then that lasted a little while. As my track record of full-time might have indicated to you, after consulting there for five or six years, or whatever it was, it did not last as long as a full-time employee. And you worked on Embedded-time employee. And you worked on Embedded, and then you worked on iOS there. Yes. And you had worked on Android

for previous clients. Just for the Lisp, Cisco stuff, yeah. Okay. And then you left I'm sorry, I was going to say Cisco, because it seemed like it should go back to Cisco at any point now. Nope, not doing that.

But you left Fitbit. I left Fitbit last year. Pick a reason. I had been working on iOS. The iOS team was not pleasant, was not learning things, was not contributing. Everything was an argument. The company was not doing that great. And yeah, I'd run out of my full-time,

my acceptance of being a full-time employee. And one of the reasons you went was the people there were amazing. Yeah. And the people there were leaving. Yeah, a lot of people left. The culture had changed a lot. Not for the better.

And so that was kind of disappointing. When I started there, it was a very diverse team. Great, great team, best people overall that I've worked with. Just everyone was really, really good. A lot of diversity of experience, a lot of diversity of kinds of people. It was just really nice. And then that just kind of slowly, slowly and then not so slowly kind of whittled away. And it became more normal, normal, boring, annoying company. What was the most important or interesting thing you learned at fitbit boy i don't have a good answer for that ir doesn't count the ir sucks yes uh i don't know i did a lot of stuff there but it was it was weird stuff i did graphics drivers and graphics and got pretty deep in and application layer and embedded graphics.

I mean, I don't know if that's, I mean, that's kind of like saying got really deep into the Commodore 64. It's not really applicable and except in a few places. It's not what most people think of when they think graphics. They think Halo 3, not trying desperately to draw a firework animation on a 250x250 resolution LCD screen. Just think, anytime you get that firework animation, Christopher did that by hand. Yep. And are you still enjoying consulting?

Yeah. No, it's nice to have gone back to consulting. I'm getting to do another diverse set of things, which is great. I've kind of gotten bored I'm doing iOS which is really cool being able to do a whole app by myself from the start

which at Fitbit I was excited to go on the iOS team but then it turned out it was like our team was like this little we took a bunch of first firmware engineers and put them on an iOS team because we were supposed to be able to go back and forth better and uh we didn't get to do much it was really hard to learn and the app existed and was years old so it's an existing code base it's always

hard to make a contribution um so that wasn't really fun i didn't really really felt feel like i learned ios very well even after a year of doing it there. So yeah, I've got a contract that I'm doing that now and felt like, oh, okay, I did learn some things at Fitbit. At least I know where to look for things and what some of these words mean. And so that's been fun and doing some ML stuff, different and fun. Yeah. And I like not working. I like just not being beholden to a company. Instruments and music. Oh my God, so many instruments and music.

So during the last two years of Fitbit, I bought 637 synthesizers. It's not quite that bad, but pretty close. One, two, three, four, five, six, seven. it's not quite that bad but pretty close uh one two three four five six uh at least one bass guitar drums uh yeah yeah yeah and since leaving fitbit i've bought i get i have a lot of instruments. You bought me a guitar.

No, you bought that before I left, didn't you? That was last year, May. I thought it was after you left. Oh, okay. On my birthday last year. So you bought me an acoustic guitar, an electric guitar a few weeks ago. Yeah. Eventually, the whole house will just collapse in on itself, but the gravitational pull of all the musical instruments.

I had to take a video today of some testing I was doing. I'm sorry. Were there instruments in your way? They were totally. I was like, oh, don't mind the guitar. Sorry. No. And let's see, you do a podcast.

I've never heard of it. Why? I don't know know you asked me to that's not true um the dog asked me to nope try again um it was all your idea i didn't even know what a podcast was we got a sponsorship from nike for millions of dollars and they pulled out at the last minute we'd already recorded so we decided to just do it anyway is that the story? no I think I said hey why don't we try podcasting because you

wanted to do something I'd had a bad class I was doing community classes and the teacher was bad and I was like I don't want to do this again but I need to learn something let's do six or so episodes 350 something the teacher was bad and I was like, I don't want to do this again, but I need to learn something. Let's do six or so episodes.

350 something episodes ago. You didn't ask me what music I'm playing. What kind of music are you playing? I play in a band called 12... I'm going to do selfless... this is a very selfless self-promotion. No. Shameless self-promotion. That's the word. Not selfless. This is a very selfless self-promotion. No, shameless self-promotion. That's the word, not selfless. That's different. That's totally different. Shameless self-promotion, I'm in a band with my brother, again.

It's just the two of us. And we make kind of weird art rock, hard rock instrumental music. And the band is called 12AX7, named after the tube. And we have a record we're finishing up, and we have several songs out there on iTunes and all the other places from that record, but the whole record

isn't there yet. Yeah. And we do that remotely. So we ship stuff back and forth over the internet and mix stuff together over Zoom of course now, because that's all you can do. What do you wish I'd asked you about? I don't know. This is take two.

So was there anything you asked me last time that you didn't ask me this time? Not much. Although I guess that's why this is take two? Yeah, well, it turns out, and what happened was, you see computers. We hate them. When you use a computer, it expects you to do certain things. Every time. And if you do not do certain things every time. And if you do not do those things every time, different things happen.

Sometimes the different things are good. Usually the different things are bad. And in this case, the different thing was that I recorded the entire podcast through the tiny internal microphone of my MacBook Pro, which almost kind of sort of could have worked, except Alicia is not facing the MacBook Pro and she's about eight feet away. And so it was mostly me talking to a ghost voice with a lot of echoes. And yeah, we, we, we bend it.

I suggested releasing it as a Patreon goodie that was just weird, but he bent it. Do you have any thoughts you'd like to leave us with? You know, I think I had a glib thought last time we did this podcast, but this time I actually thought about something. Everyone's relationship with work is different, and it took me a long time to figure out what I was comfortable giving to a company and to the concept of work. And I'm very lucky to be able to kind of find a relationship with work that is acceptable to the way my brain works and, uh, and my feelings about things. So I don't know if I have advice exactly,

but don't feel bad if you don't feel like the company environment necessarily fits you. There's other things you can do and may be hard to do them. It may not be successful. But there's other ways to engage with tech work that isn't the standard path. Our guest has been Christopher White, my partner here at Logical Elegance and co-host of this very podcast, Embedded. He's also my sweetie. Thank you to Christopher for producing and being a guest, and thank you for listening. You can always contact us at show at embedded.fm or at the contact link on embedded.fm. I do want to point out that we never said OSPF actually stands

for Open Shortest Path First. There'll be a few links in the show notes if you want to know more about those acronyms. And now I don't have a quote to leave you with, I'm going to leave this spot open so Christopher can put some music in. You think I'm going to put music in? I have an hour to edit this. Oh, okay, wait a minute. Fine, I'll put a 12x7 song.

Do you want to? Or I can just use the OSPF thing as a no I'm leaving this hold this meta discussion and I'll put a 12x7 song this is hot off the press yeah we're releasing this in an hour wait till last minute

it only takes a minute goodbye everyone bye Bye. At this time, our sponsors are Logical Elegance and listeners like you. guitar solo We'll be right back. Thank you. guitar solo guitar solo guitar solo Bye.