Joe Reis_mixdown.mp3 Joe: [00:00:00] As far as your question of job hopping and kind of opportunity costs? Yeah, it's a big thing. Like the one thing you don't get back in this role is your time period and the story I'm reading, David Deutsch. So maybe that actually might be proven false, but quantum physicist. But but you know, as far as you're concerned, as far as anyone's concerned, you know, you don't have a lot of time to. Harpreet: [00:00:30] What's up, everybody? Welcome to the Artists of Data Science Podcast, the only self-development podcast for data scientists you're going to learn from and be inspired by the people, ideas and conversations that will encourage creativity and innovation in yourself so that you can do the same for others. I also host Open Office Hours. You can register to attend by going to Bitly.com/adsoh forward slash a. Ds0h. I look forward to seeing you all there. Let's ride this beat out into another awesome episode and don't forget to subscribe to the show and leave a five star review. Harpreet: [00:01:11] Our guest today is a business minded data nerd who's worked in the data industry for two decades. 20 years. And in his two decades as a practitioner, he's worked on the full gamut of data tasks from statistical modeling, forecasting, machine learning, data engineering, data architecture and almost everything else in between. He's taken all that experience and started his own venture and is currently the CEO of Ternary Data. He's also the host of the Popular Data Show and Podcast, The Monday Morning Data Chat, and he also hosts the Data Nerd and a new podcast [00:02:00] that just started today, the Joe and Friends podcast. And if that wasn't impressive enough already he runs several popular meetups, including the Utah Data Engineering Meetup and the Salt Lake City Python meetup in addition to teaching at the University of Utah. He's also coauthor of the upcoming O'Reilly book, The Fundamentals of Data Engineering. So please help me in welcoming our guest today, the boss, mayor of the data Nerd Herd, and one fifth of the data heretics, Joe Reese. Joe, thank you so much for taking time out of your schedule to be on the show today. And man, I appreciate having any. Joe: [00:02:37] Time, buddy. Anytime. Harpreet: [00:02:39] Yeah, man, it's a good, good honor to have you on. I know we've been hanging out every Friday for almost a year. Joe: [00:02:47] Almost a year now. Yeah. Harpreet: [00:02:48] It's been like. Joe: [00:02:49] When you found out about your office hours, what was December kind of around Christmas time? And I was like, let's check this out. It's really cool. Harpreet: [00:02:57] So I've even come in and helping a lot of people with a lot of stuff and literally changing people's lives and changing the trajectory of their career. So I am forever grateful for your support in that as well. Joe: [00:03:09] Thanks for what you do as well. I think it speaks a lot to you that you've been taking time out of your schedule, I think as well, and doing this every week and now multiple times a week. So. So thank you. It's really cool what you're doing. Harpreet: [00:03:20] Thank you, man. Thank you. Enough about me. Enough about me. Let's let's let's talk about you, Joe. Let's get to know you a little bit better, man. I don't think like I don't think I've ever actually heard kind of your back story. And, you know, let's start at the very beginning, man. Where did you grow up and what was it like there? Joe: [00:03:37] Kind of grew up in a few areas. So, I mean, I grew up in I was born in Omaha, Nebraska, of all places. So in the middle of nowhere. And if that wasn't cool enough, then I also moved to another middle of nowhere place in Lander, Wyoming. I think it was nine or ten when I moved there. So that was cool. That was in the mountains in a what was that, the late eighties, you think? So [00:04:00] it just kind of this rundown town that, you know, I think the steel mill had just closed. My dad got a job there and so it was cool. But then, you know, that town ended up becoming rock climbing, got popular in the early nineties and then ended up becoming like one of the best outdoor destinations in the United States. So everyone knows where anyone who does anything out to outside knows your landers now. So it's kind of cool. Growing up there before and after it got popular. Harpreet: [00:04:27] I drove from Sacramento to Chicago, so I drove through that kind of entire stretch from Utah through Pass Wyoming and all that stuff. But I did a pit stop in Omaha for for I think just one or two nights, man. I really liked Omaha. I thought Omaha was a really, really cool place. I mean, three elevens from there, so that was. Joe: [00:04:45] Always three elevens from there. It's like, you know, Warren Buffet's from there. Harpreet: [00:04:49] So yeah, that's true. That's true. Yeah. Warren Buffett is out there, too. And I mean, Wyoming's, like, desolate, but beautiful. Joe: [00:04:57] So I-80, when you when you drive through Wyoming, that's like the best thing. It's great for the state because everyone thinks that that's what it looks like is just like a tattoo or something from Star Wars. But now if you go north, you start getting mountains like Jackson and the Wind River Mountains, and it's like probably the most beautiful areas in the whole country. But I'm glad I eighties where it is people just drive by and like this place sucks. Get me out of here. Harpreet: [00:05:21] So what were you like as a high school kid? Oh, jeez. Yeah. What did you think? What did you think your, like, future would look like? Joe: [00:05:30] Oh, man. High school. It was kind of funny. I think it's a mix between, like, Hunter Thompson and Ferris Bueller. So, like, definitely spend a lot of time partying. But I got good grades. I was actually a class president as well. For a few weeks. I ran as a joke and then I think the joke was on me exactly one. So and then what else? I did like debate student Congress as a speech nerd [00:06:00] and skateboarded the last I watched it a lot back in snowboarding so like a new sport so just did all kinds of stuff got a lot of trouble. So yeah, it was, it was a kind of town where I think, you know, in Lander where you the cool thing to do I think at the. I was crew's main. So you just take it up and down this main street. That was really boring. So I read a lot of books is what I ended up doing, right? So I think in addition to that, that's where I picked up a good reading habit. I think I was reading probably a few books a week. Yeah. Harpreet: [00:06:33] Yeah. It's one thing that that you're well known for is not only the data engineering know how, but the the eclectic reading habits. I have a. Joe: [00:06:42] Very eclectic reading habit and I read a ton. So yeah. Harpreet: [00:06:46] So when'd you make the move over to to Salt Lake City? Was that when you started working? Did you go to school there? Joe: [00:06:52] No, man, I wasn't working at all. It was climbing. So after high school, I moved to my senior year, actually to Bellingham, Washington, to finish up high school and start college early. Just because I was just tired of living in, you know, how it is in your teenager. You just get annoyed with everything. And I was annoyed with the town I was in, so moved out and moved to Bellingham. And at that point I was like, Well, jeez, I kind of want to go climbing, so I guess I'll move back to Wyoming. So I did, and I just spent my time climbing. That's all I did for a number of years, just rock climb. I had odd jobs, I was a floor installer, worked at a hardware store, did various odd jobs just so I could continue climbing, got some sponsors along the way. So that helped a lot in terms of I think getting free gear and also just getting a lot of publicity for whatever that's worth in climbing world back in the nineties but ended in Salt Lake as at the Outdoor Retailer Show, which was a big basically the world's largest outdoor industry expo. So if you're in the outdoor industry for selling skis or climbing equipment or hiking or whatever you're going to be at, the outdoor retailer shows as visiting some sponsors as back in Salt Lake had the show and [00:08:00] you know I kind of Salt Lake at the time I think had some weird stigma behind it. You know, it's kind of like it's a square town, kind of boring. And when I got here, I was like, okay, this place is amazing. It had like the best climbing I'd ever seen. And next to a city, the best outdoors that I've ever seen. And it was it wasn't a small town, so that was nice. So I decided to come here for a few weeks. This is in 1998 and yeah, almost, almost to the to the day actually. And then I ended up staying here. I've been here since 98, so yeah, that's cool man. Harpreet: [00:08:34] Throughout those years they kept a little bit of that, that teenage angst because he did eventually get tired of data science and and into data engineering. But we're going to get into that in a little bit here. You were in data science before. It was cool. So talk to us about how you kind of get into the field. Well, I guess, you know, if anybody wants to find out how you got into the field, you can listen to Kenji podcast because I remember hearing the story there. Ken's just on LinkedIn stream gave us a flex muscle but but you know, talk to us about I guess what the industry in general was like before data science was cool. What was it like kind of when you first started out and what drew you to this kind of field? Joe: [00:09:14] I mean, it's interesting back in the day, I mean, so I got into data in the early 2000 and it's really going to be an actuary actually, I think kind of share some kindred spirits with that. And that was a field I was going to get into. Excite, I think that was so. I studied mathematics at the University of Utah. Yeah, a couple of options back in 2000, 2001, you could go in to be an actuary, you can go work in the government or you can become a professor, or I guess you can go wait tables. It wasn't really or it could be an analyst, right. And so my senior year I got a job offer to go work doing a sales analytics at a company that soon turned into doing kind of more predictive modeling and optimization and that kind of stuff. And to me that was just doing like it [00:10:00] always had been looking for analytics jobs and I guess I landed that one pretty squared was doing that kind of work, you know, able to impact various departments in this company. And yeah, and I look back on it that that type of work was exactly what data scientists were doing, sounds without machine learning because you weren't really weren't doing machine learning to any degree because the computers were very low powered back then. To say like Excel I think had a 65,000 role limit or something back then. Joe: [00:10:25] So I mean anything fast I would tell would stop working. So it gives you an idea of like how much horsepower you had in your computers. It was like none whatsoever. So but, you know, you kind of made do with the the tools that you had and let's say in the late 2000. So I I'd always been interested in machine learning, you know, just trying to teach computers how to do pattern recognition stuff, I guess. But I just realized the hardware wasn't there. But around like late 2000, you noticed there's like cloud computing. And I had an itch to kind of get back into, to writing code. I used to actually write code when I was younger, like in the nineties and stuff growing up. I didn't even tell you that in high school. Yeah. So I mean, I had like an early problem on the first people on the Internet. I would say like 92, I had an Internet connection because my dad felt really bad for me. So he got me an Internet connection because there's like an outside outlet in the small town like that. So. You know, so it's kind of cool, you know, just getting into early networking and bulletin board services and just finding all kinds of getting into like whatever you call hacking back then it's very rudimentary compared to what you'd find now. And I wanted to get back into that. Joe: [00:11:31] I think in the late 2000, I just finished a stint as a chief operating officer. As a company at a company, I was kind of bored. It's like it's it's fun what I'm doing, but I really want to get probably back into something more hardcore. And so machine learning caught my fancy. And so I started diving into that and I'd say around there weren't a lot of machine learning opportunities back then. It was like, you talk to people about it like, I have no idea what you're talking about. It's the dumbest thing I've ever heard of. But, you know, I was lucky enough to to meet up with an old coworker [00:12:00] at a company I'd worked at and as well as a math professor who had thankfully been hacking on this interesting ensemble learning mechanism in his spare time and his basement. So we teamed up, I think, around 2012, around then maybe earlier, I can remember, but we were talking about starting a company that would just basically, you know, you could do automated machine learning. So company would give you a file, a user would give you a structured file, you build, make predictions on it. And so that that was cool. So I was hired as a software engineer to work on that, right? And I think that was when I started realizing very early on that the algorithms piece was actually the easiest piece, getting raw data and trying to, I would say, unforeseen raw data and trying to make predictions on that on any data set. Joe: [00:12:45] That's an engineering challenge. And very quickly, I realized machine learning and production isn't just about throwing an algorithm against a data set. It's like, okay, so how do you process the features in such a way that they're coherent through the algorithm in automated fashion? How do you serve models? Right? What about retraining, all this other stuff? So that got my mind thinking about it. You know, this is super early days, I guess now some of the stuff is ML ops, but that got me interested in data engineering. And then over time what I kept noticing with this is about the same time when machine learning was starting to take off in popularity. I think Andrew NG had open Source's first Coursera class, I think it was the machine learning class he was teaching an octave. I think that got I think the first class got 100,000 students, including myself. This is kind of cool to learn from this guy and he's teaching this class at Stanford and whatnot. So I think after that I was I sold on machine learning. But again, I realized that there's one aspect that you can do machine learning, but without good engineering, it's not going to really have any hope of doing anything valuable. Harpreet: [00:13:45] Yeah. Yeah. To a proper architects mindset back then that's says kind of what it sounds like. They're kind of before data science was their science. It had these other names like you mentioned, analytics or. Joe: [00:13:58] Data mining. Harpreet: [00:13:58] Data mining, things like that. [00:14:00] So I'm wondering in your opinion, where is the science in data science? Is there any science in data science? Is it scientism? Joe: [00:14:08] What a good question. I don't know. Sometimes I have this saying that data science is neither. So it's a kind of tongue in cheek. It actually comes back from the climbing days. There's an old kind of tongue in cheek quote that sport climbing is neither from a friend of mine, Jon Sherman, but he's poking fun at climbing. And but I sometimes feel that way about data science. I'm not sure where the science is because it's not really a science in the strict sense, like chemistry is a science or like physics is science like the amount of. So if you were to take a scientific approach, for example, that hypothesis, test it out, right? I see some data scientists do that for the most part. I see everything except that. So it tends to be just kind of throw whatever happens to stick to the wall and hope for the best. So what I have observed and why I kind of push back on the term data science and again, I don't really care what people call it, call it data science, call it whatever you want. But it just didn't have the rigor that I'd seen in the sciences, for example. Right. So and is that really like one methodology for it? I think. And I also notice that data science are incorporating lots of different aspects of what used to be very distinct disciplines in data by for all I know is, is data science. I think it is actually if I look at data science, job postings, analytics certainly is. So it's an open ended question. Harpreet: [00:15:29] Yeah. So let me speak in open ended questions like do you think I know you might be familiar with these ideas within philosophy logic, but this problem of induction, do you think that data scientists need to concern themselves with these type of philosophical problems, problem of induction, ethics, even things like that? Joe: [00:15:50] Yeah, I think so. Induction is a good skill set to have. Ethics is something I can't I don't think you can avoid. You should avoid it at your own cost for sure. So yeah, [00:16:00] I think it's just what I, what I noticed, you know, when data science is really taking off like in the early I would say like 2013 is when you started seeing the groundswell of data science starting to take off. I think Kaggle is starting to have competitions around around that time are the first Kaggle comps at least. And then I quickly went into. 2015. Window overdrive. Everyone's like, I need to do machine learning. I need to learn how to train models and all this other stuff, right? But what I think was really missing, it was like all the important questions, like training a model is one thing, but that's like one aspect of what a data scientist should be concerned with, in my opinion. There's also the question of should you even be doing this? And so I would see a lot of machine learning, especially in the in the old days, credit scoring models, for example, just really that kind of stuff. The data sets that were being used ethically nowadays, I would say you wouldn't be using those data sets. Joe: [00:16:49] I'll just leave it at that. So and just some of the approaches, I would say it just I'm glad that the field's maturing to to even ask these types of questions. I think it was definitely the Wild West for a long time. You can see this in the big tech companies, too. You know, I mean, some brazen approaches by certain companies that had their entire service knocked out the other day. It was a Wild West because I think it was people were enamored with the potential of data science saying, well, you know, we have all this power, so let's just make the world a better place by throwing machine learning at everything. And I think you now see the end result of some of that when it's not reined in with ethics and just thoughtfulness from the beginning. If you're only. But it's the old saying, right? Show me the incentive and I'll show you the outcome. If the incentive is basically just to make a lot of money, no expense, and what do you think you're going to do? Right. There's no consequences. Sure. Go for it. Harpreet: [00:17:45] Skin of the game. If you're a data scientist, you got to have some skin in the game. Joe: [00:17:50] I mean, if you try to do something, if you take the medical analogy to this, right. I mean, if you try to do this as a physician, just weird experiments on people. I mean, at minimum, [00:18:00] you're going to lose your license, right? And I feel like that's exactly what happened with society in general, with the application of a lot of these algorithms at scale in mass production. So and I don't think it's a uncontroversial thing to say anymore. I think when I would say this back in the day, people like crazy Facebook and all these other companies are doing such a great thing. Right. Harpreet: [00:18:20] And like that last point you're making that think every data scientist should read Skin in the Game by Nassim Taleb. Joe: [00:18:26] That's such a good book. Harpreet: [00:18:27] It'll solidify this this concept of of what that means, right? Like if if if you're the one thing you talk about in the book is like the the silver rule don't do to others what you don't want done to you. So think about what you're developing with those algorithms and think about if you would want someone to weaponize your own behavior against you in that particular manner or things like that. Just, just think about that kind of talk about a little bit, you know, what a recovering data scientist is? Well, not really. We talked about what kind of made you become disillusioned, I guess, with data science. But but what would you say is a recovering data scientist? Tell me what you mean by that. And then talk to me about what the road to recovery has been like for you. Joe: [00:19:13] Well, I snorted and smoked a lot of data science back in the day. I'm not proud of what I did. I was young. I needed the money. So I'm joking now. I think it was I can't remember who it was. I think it was Gonzo. Ben Taylor's business partner and I were joking about it because he he felt the same way. And I think we started calling each other reformed or recovering data scientists and just sort of stuck. Right. So because we had the same we had, we had. Exactly. You should get him on your show some time. This guy is fun to talk to you. But yeah, we had a lot of the same observations where I think there was just this mad rush to get into data science and we'd both work at this original auto company, and I think we had a lot of perspectives and he had been doing a lot of before. And I think we just we're like two peas in a pod. Like, you've seen the same thing. I am. I don't think I'm crazy. Right. And so that's when I [00:20:00] think sobriety as the analogy became an attractive option. But the road to recovery since then, you know, it's it's been fine. I think it was one of those things where I felt alone in the wilderness for a bit, so to speak, you know, especially from like 2015 to probably 2019, because everyone was like, why aren't you into machine learning? I mean, you're you're good at it and all this stuff, not to sound like Michael Scott, but, you know, it's I don't suck at it. Joe: [00:20:31] So doing it a while. But it was like most of the applications I started being used for is like that's not really like a machine learning application at all. You could make a report and pretty much accomplish the same thing at about a fifth of the time and effort. I just think it was like the fact that people are sewing, machine learning and everything. And so I just kind of sort of focusing on the broader picture of just the systems that would enable machine learning and analytic to function at scale. Because it wasn't like data wasn't it was growing a lot, right? It's like just because everyone's doing machine learning like data engineering was never a consideration. I just think there wasn't really a name for it. Early people like Jessie Anderson were giving a shout out to him. I think he's a pioneer in the space in terms of giving it some visibility, data engineering, that is. But yeah, I think I just turned my attention to the engineering aspects of it. And in the meantime I did work at another ML startup as well and help them out and but over and over I just kept seeing the same themes. It's like I don't think 90% of these things are trying to do ML with you don't need it, you really don't. Harpreet: [00:21:32] So talking about that, you mentioned we could do BI and report or whatever and get the result in the fifth of a time. Like when these companies are out here hiring, let's just say they they're out here, they're inspired to go to a conference and they're like, Yeah, we start doing data science. Let's, let's start a data science team. Do you think that these leaders and these in these industries, if they're not coming from a data science background already, if you think they know what to do with a data scientist, if they hire the first one. Joe: [00:22:03] It's [00:22:00] a tough question. It depends on the background of the person, I would think. It depends if there's very, I think, strict needs for data scientists. But if you just go into it saying you're going to hire a data scientist because you need to. I think that's not a good idea. So I've been in those positions. My business partner has been in this position. Right. Like I've seen this happen to countless people. You have the credentials on paper to be a data scientist. You get hired and then you end up just making dashboards and data pipelines to build a foundation. And that's precisely where I've been and what I've seen. So I still see this a lot all the time. I'm talking to data scientists on the phone like Zoom calls now, I guess. But because a lot of people approach me and just ask kind of in confidence, like, what's going on? Like, Am I crazy here? Am I being gaslighted or is there even a work for me to do here? Harpreet: [00:22:50] How do you think that negatively impacts a data scientist? Right. Because, I mean, nobody wants to get into a job and then like keep quitting every 6 to 8 months or whatever, nine months when they become disillusioned with the company and and realize that they're not doing any data science. Like what type of negative impact does that have on a data scientist in the long term? Like does that have the potential of setting you on the wrong path for career trajectory? Does it have the potential to make you just say, I'm done with that science? What are your thoughts on that? Joe: [00:23:19] I'm seeing it happen not just to me, I mean, but with others. I think there's just a sense of like, yeah, it's a weird because there's expectations in this reality, right? These are I just think the expectations of data science is way ahead of where it was and probably where it is for a lot of companies. And I would say if the company is data native and it has a big need for this kind of stuff, you'll know because you have to do a machine learning at scale in order to like just work as a company. And that's how Google and Facebook and all these other companies have to do things at scale. They couldn't do it with humans. So as far as your question of job hopping and kind of opportunity cost, yeah, it's a big thing. Like the one thing you don't get back in this world is your time period. And the story I'm reading, David Deutsch. So maybe that actually [00:24:00] might be proven false, but quantum physicist. But but you know, as far as you're concerned, as far as anyone's concerned, you know, you don't have a lot of time. Right? So I'm over the opinion. I don't waste time with people or companies if they're not going to if there's not a clear outcome that we can work towards. And I would say if there's not a clear buy in from people I know, I'm out as an employee because it's just there's so many other companies where you could be more beneficial and I think save your sanity because you're going to get frustrated. That's the thing. You go home stewing all night about how much your job sucks, how much your bosses get it, how much, how stupid the company is. I mean, we've all been there, right? It's like, I want to make a difference. Joe: [00:24:36] I don't think it's anyone who goes into a company and says, Well, I really want to do like a really crappy job and like add no value and just make this like the worst experience possible for everybody, including myself. Maybe there's people that are like that, you know, but I think most people have good intentions. But again, you're up against the inertia of a company typically, right? And that's not just your boss. Your boss also has things they have to navigate through politics, looking good for their boss, all this other stuff. So you need I think it's as much understanding the organizational dynamics as anything. I would say the more I've been observing it, the more I think that that is single handedly the best thing you need to do as a data scientist or anyone in any field, really. If you can suss out the organizational dynamics at an employer before you get there, hopefully the more the merrier, because again, you don't want to waste your time with people and nobody wants to waste their time on a bad hire. I mean, that's expensive, too, so but unfortunately, on a good day, hiring is a crapshoot. That's how it is. You can do all the assessments you want, but at the end of the day, like hiring is tough and the person you hired, even today, they may change. They may go through a life event that just throws them off. I've seen this happen to people like somebody seems like they're a great person to work with and all of a sudden something clicks and suddenly there's absolute it's unbearable to work with. You just got to get rid of them or just go somewhere else, you know, depending on the situation. So. Yeah. Harpreet: [00:25:59] So. [00:26:00] Let's get into your entrepreneurial journey now. You mentioned you and your business partner like to talk to us about how you started off on that path. How did you how how'd you guys kind of link up and decide to start ternary data and maybe we can even get the story behind the companies name as well? Joe: [00:26:19] Sure. Yeah. I think it's around, what, 2017? I was on my own just doing data engineering consulting and got linked up with Matt. This is partner shout out to Matt. Wonderful guy. He's my is my brother at this point. So. But he he he's a math professor. That's his background. So and he joined the dark side. I got into industry as a data scientist. I think he also he saw the same things I did, hired as a data scientist and doing, guess what, data engineering work, setting up pipelines, orchestration, workflows, everything except fancy algorithms and all the stuff that his he was supposedly going to be doing. So he wanted to go out and do his own consulting. And my friend introduced us and said, Well, why don't you guys just work together? Like, Sure, let's figure each other out, see if it's a good fit. And six months later, we started a business and I guess we're still around almost three years. So after so. Harpreet: [00:27:11] In that consulting kind of gig, like what are some big mistakes that you see companies make when they're trying to build a data science team kind of from the ground up or just let's say they already have established data science teams. What are some mistakes that you see them make when you get in there? What are some problems that you just see as a consultant pop up over and over? Joe: [00:27:32] So it getting in your team's way. It's a big one. That's a big, big one. So not giving them support by giving a lip service support like, yeah, we really like what you guys are doing. Yeah. Can we get more people in or can we get the support we need? You know, and it's like, yeah, we'll get to that next quarter. So that's typically what I see is just there's not the support. It's also not an appreciation for what they do. No understanding. It's of like, well, we have data scientists are they're really smart. They're working on [00:28:00] all of our problems and stuff. But I'd say it's just a lack of appreciation or recognition or anything. Harpreet: [00:28:07] So what's what's that kind of day in the life like for you as a as an entrepreneur or data science, data engineering entrepreneur? Joe: [00:28:16] My kids just got home, so I need to shut the door here. Harpreet: [00:28:17] Yeah. Yeah. Joe: [00:28:20] Day in the life. So day in life of being an entrepreneur or. I mean, it's changed. I mean, when Matt and I were first when we first started the company, I would say until recently it was just Joe and Matt show, right? So I would handle more of the kind of marketing and sales efforts. He would handle more of the day to day client engagements. I would certainly jump in if there's a need, but that's a powerhouse. You can do a lot of the work. And so that's how we divvied out the sort of the boundaries. And but yeah, I mean, you know, I would say for anyone who wants to start a consulting company or a business like and you think you're going to be doing your technical work, it's everything except that. So get good at business in general. It's tough. So. Harpreet: [00:28:59] So I can't get a business man. Like if we don't go to business school, can we still get good at business and. Joe: [00:29:04] Not I have to get an MBA. It's the only way to get it right. Yeah, I'm just joking. Yeah, no, no. I would say actually an MBA probably hurts you in a lot of ways if you're trying to start a new business. I think MBAs help you when you're operating an existing business, but it turns you into a a good operator. I don't think an MBA helps you in terms of understanding how to take advantage of entrepreneurial opportunities. That's what I've seen. And this is knowing a lot of MBAs and teaching at a business school as well. Right. Like it's not a knock against an MBA program. I just think it's the right degree for certain types of situations. But for entrepreneurship, the thing you need more than anything is just patience and persistence. And that can't be taught. Yeah. Yeah. Harpreet: [00:29:43] Can be learned though, right? You can't teach it. But you can learn it by going. You can. Joe: [00:29:47] Learn it. Yeah. And the only way you learn it is by doing it. It's kind of like I always say, it's like it's like learning a dating. Like you can read all the books you want about dating. That's not going to help you at all. When you actually go on dates like it might actually work against you [00:30:00] just come across like really awkward but no, no. I mean it's entrepreneurship is tough, right? But I think a lot of it is just you got to go into the right mindset and the right approach. So I made every mistake in the book. I've had other businesses before and most of them have failed. And I think a lot of it was just me not giving myself a chance to succeed. Right. You have these unrealistic expectations of when stuff needs to when you need to be a success. Think I need to be a millionaire by. I need to make $1,000,000 in this company by this time next year. That may happen, but the market really determines that. Not you. So, I mean, as a really wealthy person once told me when I asked him, So how do you make a lot of money? He's like, Well, this sells people what they want to buy. Harpreet: [00:30:44] Yeah. Joe: [00:30:45] It's true. It's also that hard. Harpreet: [00:30:49] So tell people what they want to buy. And I like that. I like that. So. So talking about again, getting getting good at business. What's like a time? Time is money, right? So if time is money, what's one cheap skill, business skill that a data scientist could pick up, that they can start practicing immediately, that will help them kind of be better in in their business. Joe: [00:31:14] Interact sales dealings. Harpreet: [00:31:15] Sales. Joe: [00:31:16] Sales, end of story. Because think about it every day, whether you're at a company working, you talk to, you know, your spouse or friends, right? What do you do every day? You communicate and you're also selling you're selling your idea, your version of it. If I'm selling it right now on selling, it's just how it is. Sales is by far the biggest skill I would say anybody should have. My kids, right thing I try and teach them is how to sell, how to communicate. Right. Because the old Jordan Belfort thing, like sell me this pen right from Wolf of Wall Street. And I think that I'll let you guys watch it. Less words in it, but it's a good movie. I think if that sort of thing excels is one of the things I would say is just that is the skill. If you're going to be in business, [00:32:00] it's not your technical chops. It's all about this the other day with some people. And I think there's a misconception that because you're technical and you have great skills, you're automatically going to have a successful business when you go out on your own. That is far from true. It actually, even in your job, your technical skills, I think account for like 10% of your success, probably less your ability to get ahead. It's about how you communicate, how you network, how you use all the soft skills that you think are lame and unnecessary. Those are the things that get you further and whether it's in business or in your career. So still. Harpreet: [00:32:33] It doesn't necessarily mean just like hand-to-hand, one on one sales. It could be it could be presenting, it could be communicating, it could be writing. Even if you're really good writing, you can have good written communication. So it's just your ability to clearly communicate your ideas and kind of just put yourself in the mind of somebody else, right? Kind of speak them right again. Joe: [00:32:57] Sell people what they want to buy, even in a business setting. Right. So I want to convince you that this model works and that it provides accuracy. Great. You can show me the numbers. That may convince me. It may not. Tell me why it's important. How is this going to make my life better at the end of the day? People want what's going to make them look better, whether it's at work, whether you're selling to a customer. That customer, for example, wants to make themselves look like they made the best decision when they bought you. That's it. They don't want to look stupid. Same with the higher ed. Nobody wants to look stupid because they hired you, right? And you want to make people look good so they get the promotion. That's how it works. How you get how you get ahead in corporations or in a company is you find somebody who your sponsor, they can pull you up. You can do all the tutorials you want and do all the Kaggle competitions you want. But finding good sponsor and networking and selling yourself, selling your capabilities is going to get you a lot farther than any of that stuff, I guarantee you. Harpreet: [00:33:54] So let's talk about. You mentioned technical skills there and now we're talking about creating value through your [00:34:00] skills. You had this awesome blog post about new data engineers add value. So talk to us about that. Did engineers add value and how should we think about a return on investment for the work that they do? And I'm sure that just I'm putting it out there. The ROI for data engineers probably ten Xed out of the A scientist 10.1. Joe: [00:34:22] So yeah, I mean, it's an interesting question. I would say that there's there's a couple of threads to answering this. So data engineers, any data team really is going to support two functions that can be external facing, which means you're going to be working on a product that customers use. So think about something like a recommendation engine for an E commerce app, right? Or something like that. So there's a tangible ROI in the sense where that engagement go up through your actions, was there a feedback loop that generates more money or something like that, whether it be consider ROI? So that's easy to calculate. Roi value is an interesting thing in general to zoom out where people always talk about, Oh, I'm going to add value. Like, What do you mean by that? Like Make me feel good or you make me money. Like, what do you. What's valuable? I have a lot of things that are valuable to me, but that might not be what's valuable to the company or this initiative. And so define value up front. Now, of course, when you're an internal facing data team or data engineer, that's a different story because now let's say you're supporting internal functions like making reports, making models that optimize internal processes. This becomes trickier. What's the outcome? How do you evaluate the ROI on that? If you can do attribution costs like ABC activity based costing, maybe that could help, but that's hard. Show me a company as doing activity based costing really well and I'd love to talk especially on data teams. Right. So it's kind of like the old saying like what's the ROI of putting on your socks in the morning? I don't know, ten at least, but. Joe: [00:35:51] So. So how do you evaluate it internally? Is that okay? So how am I affecting my downstream users? Like if I'm a data engineer, for example, and I need to supply data in a usable format [00:36:00] in high quality format to the data scientists and analysts, or are they getting the expected data on time when they need it, or are they able to do their jobs? Or is there a lag? Is crap always breaking upstream or our engineers with the applications have dependent upon? Are they providing data in a reliable fashion? To me? So it's all about SLAs, service level agreements, reducing lead time, reducing defects. So I take a very operational mindset using kind of lean thinking in these cases. I take that in general because this is kind of how I'm wired, but, you know, reduce time to value, reduce lead time and increase quality. So those are metrics that use internal facing, external facing as well. But again, just assessing ROI is always tricky and that's where I kind of take issue when people say, well, what's the ROI on my data team? It's like, it depends what your data team is doing at the end of the day. Like if you can attribute the salary to some sort of tangible output where you can put a dollar amount against that rate, go for you, good for you and go for it. But if you can't, it becomes tricky. So and I guess it depends then you have to look at what's the output of the end result, right? So if it's a, a data science team, like what are they working on, right? So everything should have a value, but sometimes it's really murky. Harpreet: [00:37:13] So I absolutely love that, man. Thank you so much for that. That's definitely one that to go back and rewind and listen to that. Take some notes on that. I really like that. Talk to us now about this this data engineering lifecycle, I guess. So this is coming from I believe it was another blog post or LinkedIn post I saw you write up is the data engineering lifecycle and I guess the, the different steps that go into it. Joe: [00:37:36] Yeah. Yeah. It's actually something that's gonna be coming out in our book coming I think earlier. The chapters might be out as soon as next week, so stay tuned for that. I think so, yeah. The data engineering lifecycle is an interesting one. We had actually first seen this in the with Google Cloud actually. They had a really good lifecycle for data within Google Cloud. And so the most part, I think they are right, except they got a few things wrong in my opinion. Just steps are really jumbled and they put technologies [00:38:00] under different steps and this is kind of a hodgepodge. So zoomed out and thought, okay, so for a data engineer, like what are the things you really need to know? What are the steps and data engineering and data engineering? What I, what I see it as is getting data from source systems, doing something with it and making it useful and then giving it to downstream users for reports and machine learning and stuff like that. So but the lifecycle, right, so it's like Source Systems is your first one. So getting data from source systems, what's, what's that? Well, it could be an API, it could be a database. It's using your application, ERP systems, so forth. So that's the first step. Get data, ingest it, it's the second step, transform it and then serve it. Notice though, I didn't mention storage because that undercuts basically ingestion all the way to serving, right? So all along the way, you're storing data in some way, shape or form. So that's that's a lifecycle in a nutshell. It provides a way of thinking about. So as a data engineer, like what areas do I affect right up to serving data scientists and analysts. So. Harpreet: [00:39:00] So mentioned, not mentioned I saw you mentioned this in a recent post. Was that is it LTE versus reverse LTE or that you had like some I don't know what that phrase. Joe: [00:39:13] Well, it's an interesting phrase. I mean, I'm pretty sure I'm friends with the guy who I know, the guy who developed a similar at high touch. I think he was a guy who came up with it. But at any rate, like it's something I take issue with it. I think it's it's a weird term in that it sort of describes what's happening. So reverse ETL right now. Basically, you take data from your data warehouse and plop the results back into your source systems. This could be lead scoring. It could be back into Salesforce, for example, so salespeople can get the best leads to contact and so forth. I was on a case dedicated yesterday and in normal fashion I just blab and I was talking about LC versus ETL and somebody wrote BLT and I was like, Oh yeah, bidirectional load and transform. And so that's actually like a rain band moment. You start not even thinking, you just blurted it out and just kind of went on my merry way. And it's interesting. [00:40:00] So yeah, I mean, it's sort of a tongue in cheek thing that I posted on LinkedIn and got some traction. I got a lot of comments on that and people took issue more that I was basically co-opting a sandwich to describe a data term, which I'm not apologetic about. All the data things should be named after sandwiches is would be my edict. So even had some guy named Ruben comment on it and I was like, Oh yeah, it's a good sandwich too. So but yeah, it's hard. I mean, I said it was more tongue in cheek, more than being serious, but yeah, but maybe it'll turn into a term. I don't know. I think it's, it's more, it's more funny just because I see a lot of these buzzwords right now to describe every facet of the data ecosystem, and I just get a kick out of it. Harpreet: [00:40:37] So that saw some of the comments on on that thread that you posted. I was like, Oh, that's pretty interesting. There's Robert Sievert shout out. Robert Siebert. That's how Robert he's watching LinkedIn right now. He said the BLT thing was him. So, Robert, thank you. Joe: [00:40:53] Give you the inspiration for that guy's awesome red dude. Yeah, he just joined a kind of a group of men, and I really like the guy, so shout out. So. Harpreet: [00:41:01] Yeah. So I was going through your blog, man. I really enjoyed reading through some of the stuff. Joe: [00:41:05] Oh, which one? Harpreet: [00:41:07] It was the. Joe: [00:41:09] Ternary data or my personal one. I have a personal blog that. Harpreet: [00:41:11] Yeah, the personal one and a personal one. I thought that was cool. Like how every Christmas you write your kids a letter and I like the ones you wrote for 2020. Joe: [00:41:21] So that was a messed up year. Harpreet: [00:41:22] But you also wrote this blog post about this concept of reputational capital that really resonated with me. Break that idea down for a. Break that capital. Break that concept down. What is reputation? Joe: [00:41:33] I mean, I got the notion really from from Buffett, not from him, like telling me like, hey, Joe, secret of the universe for you. But yeah, I mean, he's a cool dude, but no, it was what he did at the beginning of every Berkshire Hathaway shareholder meeting. He shows a video from the Solomon Brothers hearing. So he was CEO of Solomon Brothers for a bit and he went through some I went through some trouble, to say the least. This horrible stuff happened. But he shows this video of his testimony to Congress and he says basically, in a nutshell, [00:42:00] you know, use a newspaper litmus test is something that you're doing. Would you want that on the front page of your local paper? And then he says, if you don't if you don't need that for that advice, you need a fear. My other piece of advice, which is, do you lose money for the firm? I'll be understanding. If you lose a shred of reputation, I will be ruthless. Let's talk with me. Right. Reputation is everything. As he also says, it takes a lifetime to build a reputation. It takes 15 minutes to destroy it. So when we started our business, I thought it was interesting. We didn't really care about the money. We cared about reputation and cared about doing great work, meeting great people and just, I think developing good relationships. I always optimizing for reputation. I think we thought if we could build that pile of reputational capital, the money would follow. The reverse is rarely true, though. In the short term, you can build as much money as you can, but you can destroy your reputation. And then who's going to want to do business with you? So other people have reputations, I'm sure, but that's not scalable, right? And that's not kind of life I want to live. Harpreet: [00:42:59] So some good tips in there for sure. Good lessons, rather, in there. But do you have any tips for for, let's say, people who are kind of starting out, maybe they're not on an entrepreneurial path yet, maybe they're just early in their data science career. They're just starting their there. Their first job as a data scientist, how can they accrue some of this reputational capital? What are some things. Joe: [00:43:23] This be about badass do great work. The old saying is, you know, show up early, leave late. I think that's true. Not literally, though. I think office hours are stupid, but I think that mindset are just going in there and killing it every day. That is what sets you apart. You're going to get known for that in your career. Then when people leave jobs, right, you got to consider. It's like it's like pollen in the spring. It just goes everywhere, just blossoms and turns into new things. And that's people leaving their jobs, going to new companies. I think they're going to remember, though, is this this person right? They killed it at this job. That's the person I want that follows you. I can't tell you how many times I've gotten opportunities from stuff and [00:44:00] people I worked with like ten, 20 years ago now. Right. Opportunities. You've got to make yourself valuable value but make people's lives better. Right. And it's not just being about the it's not about being the smartest person. It's about making people's lives better and making them feel great about you and their experience with you. So that to me is reputation and, you know, and not cutting corners, not stealing, not doing shady stuff that's going to get you in prison. Probably you can get a good reputation in prison, I guess, but that doesn't really count for much. So, yeah. Harpreet: [00:44:35] So it depends on. Joe: [00:44:36] What what you're trying to do I guess. Right. So even like prison Mike I guess in the office. Yeah. Harpreet: [00:44:44] I like that. That's, that's it. Just show up, do the damn thing with a smile on your face. No matter if you don't like it or you do like it. Just do what needs to be done. Do what you have to do to. To make things. Joe: [00:44:58] All the time. All the time. And it's just that it's it's not just I think there's a misconception that you have to go in there guns blazing. It's just like it's a compounding rule. Just be 1% better. And it's the old cliche, like 1% better. It's going to make 330 some 7% better each year or something. But it's true, right? So, you know, just go in there and make good impacts and be a nice person. I think that's the other thing. It's okay to burn bridges, though, that also I think I should say if somebody is unbearable and they cheat you. Right, or you just feel like they're shady, who cares? Burn that bridge quick. There's other bridges. You don't need that bridge. So. That's because you don't want that. You don't want that reputational stain on you either being associated with a person who is probably less than reputationally desirable. Harpreet: [00:45:44] So yeah, I love that great, great life advice for sure, man. So another interesting blog that you got is about science fiction and technology, specifically how reading science fiction has made you a better technologist. So how is [00:46:00] that what science fiction done for you? Has it made you a better technologist? Joe: [00:46:03] I mean, you can see around corners in a lot of ways. Science fiction authors do a really good job at imagining the future. Right. Most of them don't come true, especially when they're both space aliens and going to faraway galaxies and with elves and stuff. But I was you know, I was a fan of like cyberpunk, for example. I grew up on cyberpunk, you know, as a kid in the late eighties, early nineties, like Neuromancer, all the William Gibson books that devoured those loved them still do. And you'll see even some still crash. Now you start to hear about the metaverse. It's a really good example. Right? So snow crash, it's about VR dystopia. Not too much like we live in right now, actually, except that we don't have VR, but almost everything else that I like. Yeah, I read read it a few years ago and I thought, yeah, this is shaping up to look like that 1984. That's the science fiction written in 1948. And I would say, yeah, that's that's pretty close to, to a lot of what we have, except not jackbooted thugs coming in the door. But geez, that was that was pretty prescient. I think it drives not only just seeing around walls in terms of technology itself, but also ethics and how you apply technology. So that's why I like it. I think it's effective. Harpreet: [00:47:05] Definitely. Like I watch science fiction. I haven't read much science fiction, but like I watch science fiction movies and stuff. So it definitely would need to I want to pick up something that I that I saw on TV is called the Foundation by Isaac. Joe: [00:47:17] Foundation is awesome if you read that, especially if you're into predictions. Harpreet: [00:47:20] Yeah, yeah, yeah. Right. The guy's like pretty much a data scientist, like doesn't want to check. Joe: [00:47:26] Yeah, that's a great book. I read that when I was a teenager and that stuck with me a lot. But Foundation Trilogy's awesome. Anything by Isaac Asimov actually is also awesome. Like, I was just a machine in terms of writing and thinking and I don't think there's any anyone like him since. So definitely I. Harpreet: [00:47:42] Have to check. Joe: [00:47:42] That out. Yeah. Harpreet: [00:47:43] So what would you say is the one sci fi work that's had the biggest influence on you as a technologist? Joe: [00:47:49] Try Neuromancer. It foresaw the Internet. Right. I still think. Harpreet: [00:47:55] It's. What's that quick rundown, a quick synopsis of the plot [00:48:00] for Neuromancer. Joe: [00:48:01] I mean, it has to do with like a guy who bit down on his luck. So he has to kind of jack in, so to speak, and a hack to pay off some debts and stuff. So that's about it. But it was a concept really. He came up with the idea of cyberspace, right? That was the first time the term was used. And so all the other books that he wrote are Burning Chrome, for example. This is a bunch of short stories that was cool, that just informed. Like, that's why I got into computers, really, because of those books, hacking, all this kind of futuristic stuff. People had skin suits and like, you know, look like the Blue Man Group or something. That kind of stuff had a really big impact on me because remember, I live in a small town. At the time I had nothing to do. I had no no outlets. So I would just sit there in my own imagination, coming up with like crazy stuff, listening to rave music that my friend gave me, like these bootleg cassettes from Manchester radio. I think those key 103 like a rave station play, jungle music and stuff. And I just like reading these books and listen to that. And with nothing else to do, you just kind of start coming up with your own stuff in your head. So it looks like that we're just fundamental and still are. And just being on the Internet at a really early age and just reading stuff like The Book of the Genius was like another book that I read that had a big impact on me. So if anyone knows who that is, I say in the comments, I'm curious who is that obscure in their tastes? But yeah, I mean it was also the forensic grew up with too. We all we all traded books, we're all just bunch of nerds. But I'd say everyone was pretty smart. So it's kind of. Harpreet: [00:49:27] Like it's like it's crazy how, you know, like at that age, your friends definitely have an impact on you, rub off on you. I was hanging out with people who were just too busy trying to be like Tupac, getting my ass in trouble, doing stupid shit with them, trying to be cool, man. I'm like, Fuck, man, I wish I had better friends. Joe: [00:49:41] No, I know it's crazy. And I was talking to my friend, one of my good friends last night. You know, he's he's one of my best friends. They were talking about like the the nineties skateboard scene because we're all skaters as well. And just how basically like that was like the last time I think you could go have fun because nothing was on the internet then. So you're just you know, it was a pre-Internet age, so you could just do whatever the hell you want [00:50:00] and nothing's there for posterity. It was awesome. Harpreet: [00:50:04] But nowadays, everybody's lives are documented. Speaking of like skater, like I love I love that skate culture type of skate. Growing up like the X Games and all that stuff, the OG skateboarders were like my idols. But it was weird, man, because I would dress like a skater, listen to, like, rap, and then to the opposite extreme of rap, like just punk rock and stuff like that. It was really, really weird. Just a really weird. Joe: [00:50:27] Kid, man. So a really fun thing happened last month here in Salt Lake. So they had Tony Hawk's vert alert event, which was so Tony Hawk set up a vert ramp free event down in Salt Lake. It was cool. So we had Christian Tsoi, Steve Caballero, Bob Burnquist, Andy McDonald, like all the skaters, like the Who's Who, and it was just cool, just a skate session, vert session with all these guys. And it was the coolest thing I've seen in a long, long time. So I'm just playing like eighties punk music. It was cool. Harpreet: [00:50:54] So speaking of music, man, you got quite the setup for those of you who are listening in on the podcast when this is released, I mean, and those of you who are part of the happy hour, you always see Joe's that background. He's got such a dope setup here. What's all this about? The keyboards? You got turntables, you got multiple keyboards, you're making your music. Do you got like you got any undercover Spotify. Joe: [00:51:18] Back in the day music club? Dj I still I still have DJ recently, but not with the book in the business. I just don't have time to post on SoundCloud or go DJ at clubs. But I'd like to again, I think it's really fun. So it's something I've always done. Harpreet: [00:51:33] So do you ever just like Friday night, have a couple of beers and just start, like, mixing all the time? You like that? Joe: [00:51:39] Yeah, all the time. It's good because I think especially nowadays being on Zoom calls like the last thing I want to do is start screaming. So I just buy like analog equipment. Yeah. So I just walk around, just. I just buy synths for fun and just play nerd out on equipment. So when I have like boxes of synths everywhere, so, and drum machines and stuff, so I don't know. [00:52:00] So my office is kind of a museum in a way. There's where posters, as you can see, and all kinds of fun stuff, tank shells and setlist from Tool. I like that. Harpreet: [00:52:11] But I love it. I love the setup. So are you into like I know you always call deep learning. Deep learning hurts my feelings. I love deep learning, but are you are you to generative models or anything like that or have you researched any of that to. Joe: [00:52:24] I can't say I found any utility, but that's just because I don't really have a use case for it. But I mean, they're cool the cool. So yeah. Harpreet: [00:52:31] Yeah. That's something I want to get into. Hopefully get an opportunity to to work because part of the job is me just creating a bunch of content. So some content I'm going to create is going to be all around deep learning. I'm hoping to do some generative music projects. I think that'll be a lot of fun. Joe: [00:52:47] And you know, heavy dope. Well, let me know as you come up with it. So I'd love to hear it. Yeah. Harpreet: [00:52:54] Let's do it. Let's do a question before we jump into the random round. It's a standard final question that I ask everyone. It's 100 years in the future. What do you want to be remembered for? Joe: [00:53:04] I think just being a good person who showed that you don't have to follow the standard script in order to have fun and be successful, right? I mean, that's what I tell my kids and try and inspire them is just to do what you're good at and do what makes you happy. I think the rest will fall into place. So what I want to be around for as well is just raising great kids and grandkids and great, great grandkids that follow the same path. Harpreet: [00:53:29] So I love the. And are you well on your way? And not not only just for the kids, but you just helping a lot of people everyday. Joe: [00:53:35] Yeah, that's a lot of fun, right? I mean, that's what it's about at this stage in the career is just giving back and helping out the next gen. So yeah. Harpreet: [00:53:42] I've never seen a post you wrote a while ago about how, you know, it's awesome at this point in your career that you're able to just kind of selflessly give back to the community and help people out. So I know it's appreciated if you guys aren't already following Joe, definitely smash that, follow, you know, look him up. He's awesome, dude, man, let's jump into the [00:54:00] random round. All right. So first question, when do you think the first video, first video to hit to 1 trillion views on YouTube will happen? When will that happen and what will that video be about? Joe: [00:54:14] How many what's the what's the video with the most views right now? Harpreet: [00:54:16] I'm pretty sure it is a baby shark with maybe 9 billion. Joe: [00:54:22] That is such a good point. It's probably going to be something like baby shark or like a Jake Paul fight that's like on for free or something. So yeah, I don't know. So but yeah, baby shark, that's surprising but not too surprising as they get played that about 1 billion times alone for my kids. Harpreet: [00:54:37] So I haven't I haven't gotten my kid into the baby shark stuff yet. He for some reason loves cars and is obsessed with garbage trucks. He's only a year and a half and he's just cool with garbage trucks. So I literally just put on like an hour and a half thing of just garbage trucks and he's just learning that exists. Joe: [00:54:54] Yeah. Harpreet: [00:54:54] Yeah. It's crazy to hour long videos of just garbage trucks driving, robbing garbage. Yeah, it's crazy about the shit you find out there. Joe: [00:55:04] Yeah, that's cool. Harpreet: [00:55:06] Do people tell you that you look like. Joe: [00:55:09] Hmm. That is a good question. Nobody really tells me that. Actually, that's a good question. But yeah, I was thinking about that. Yeah, next question. So, like, everyone thinks that I'm like 20 years younger than I actually am. I will say that. So I just think I'm like a teenager, so. Harpreet: [00:55:25] Yeah, definitely. Joe: [00:55:26] Definitely that I'm like 80 by the way, for the audience. Harpreet: [00:55:30] So let's get moisturized. Joe Looking good. What song do you have on repeat? Joe: [00:55:36] That's a good a good question right now. My kids are listening to a lot of Devo, the old eighties pop band. So I think that's what I have on repeat right now. But I like Devo because. So whatever nerd rock. Harpreet: [00:55:47] So I got to check that out. I don't think I've actually ever listened to Devo. Joe: [00:55:52] Cool. Harpreet: [00:55:53] So what are you currently reading? Joe: [00:55:56] That's a good question. I mean, I'm reading David Deutsch is the beginning of unlike a lot of other books, [00:56:00] I think that's going to take a bit to read through. That is a that's a you've got to think a lot. Yeah. That book which I mean more so than others. And in the background, what I have citizens into is a book about the French Revolution. I'm reading database internals and O'Reilly Liminal Thinking's, another book I'm reading right now. Harpreet: [00:56:16] I interviewed Dave Gray. Did you know that? Joe: [00:56:18] I believe so, actually. I think you mentioned that. Yeah. Harpreet: [00:56:21] Yeah. Talk to him about that. Joe: [00:56:23] Yeah, it's a good book. Yeah, it's really helpful. Harpreet: [00:56:26] Yeah, he's awesome guy, too. Like, the interview got off to an interesting start. It took him a while to warm up, but then he just started warming up. Like the first few minutes felt really awkward. Like the most awkward interview I've had. Joe: [00:56:37] Between Two Ferns. Awkward? Or was it? Harpreet: [00:56:39] It was just it was his his body language was just not he just wasn't opening. And I was like, all right, I got to warm this guy up somehow. But then eventually it something he just he started opening up a lot more of the interview was amazing. Joe: [00:56:52] Trust the boundaries. Harpreet: [00:56:53] Yeah, exactly. Exactly. Yeah. Thing is good that's actually free on audible right now to anybody has audible premium membership it's free but yeah David Deutsch is man the beginning of infinity. I listened through it and I was like, oh fuck, man, I can't understand this shit is difficult to understand. You got to read it. Joe: [00:57:09] Yeah, you don't. You only listen to a book like that. Harpreet: [00:57:11] Yeah, yeah. That definitely going to purchase that book. But there's a couple of resources I picked up to help me kind of digest the concepts, just to try my mind for it. One of them was a interview with Brett Hall and Naval Ravikant, who talk about the beginning of infinity. Joe: [00:57:25] Oh, that's right. They did that. Yeah, it's so good. One, have you seen the. It's like one of the smartest. You get that guy on your podcast and be cool then a hero. He's smart. Harpreet: [00:57:34] Yeah, he's that. He's probably like the voice in the back of my head. Joe: [00:57:38] The other book I read all the time. And I just actually, it's funny, I own a copy of this, but I got another one just like, wrap it in plastic and keep it. Yeah, but it's for Charlie's Almanac, so it's expensive. I'll leave it at that. But this is if I were to pick a desert island book, I would just like this one. Not even hesitating. Harpreet: [00:57:54] I bought the book on your suggestion. It's been sitting on my bookshelf, so I haven't got a chance to actually read through it [00:58:00] in its entirety. I just. I was like, Holy shit, this book is too fancy for me to read. So I just kind of like, have it sitting on the top shelf of my book bookshelf, like, just staring me down. Joe: [00:58:08] I think there's a PDF available of it, so just read that one. Yeah. Yeah, I just got this one because my other one is kind of dog eared. It's pretty old. But I figured, you know, Charlie's old at this point, and so having any collectibles of him is probably worth it. Yeah, but smartest guy. Smartest guy I've ever met by a long shot. The guy is the smart people in this bunker, so. Harpreet: [00:58:30] Yeah, he's got this lecture keynote speech. I don't know if that's some universities. What do you call that thing? Commencement speech? Yes, commencement speech that he did at a university that was talking about mental models and stuff. And people like you should write a book. And then that's where that Charlie's Almanac came from. But real quick, going back to that David Deutsch, beginning of infinity. Yeah. Have you have you watched the Brett Hall talk podcast? Podcast? He does just chapter by chapter, breaks down break downs of the beginning of infinity and fabric of reality. And it just person like me when I read a book, I will not only just read the book, I'll watch lectures of that. I say watch lectures, but like presentations. Ted talks, things like that, podcast interviews, like I would just be all absorbed in that book from any type of way that I can get it. What's what's kind of your process like when you're reading? Joe: [00:59:21] I read fast, really fast, but that's what I retain a lot to. I think it's just reading forever, but so that's that. But I would say for a book like that, I actually take notes and I highlight a lot and I journal about things too. If there's a concept that comes to mind, like I'll write my own thoughts on it. I think it's one of the things where you it's one thing to read, it's one thing to passively read. Active reading really involves thinking about the concepts. I think most books can be passively read because it's just either entertainment or infotainment, but something like like books. You're going to want to put some time into stuff like that if you really want to get out what you're trying to put in. Harpreet: [00:59:57] So yeah, you got to wrestle, wrestle with those ideas. [01:00:00] One book you recommended to me is back here still how to read a book that that book has been extremely helpful for me. When I approach a book now. Joe: [01:00:08] It's really. Harpreet: [01:00:08] Good. Yeah. Just a quick note for anybody listening. Like the basic thing you want to do when you get any book is make sure. Read the table of contents. Go to the index. Look at the index. See what's a reference the most or what's what has the most? Page numbers. Read the the back cover the what's that called? Joe: [01:00:27] The fine or the writing? I have no idea what it's called. I can't remember. Harpreet: [01:00:31] The publisher's blurb. Make sure you read the preface of the introduction and then like the first time you're reading a book, you just you're just skipping through pages, reading a page at a time, maybe chapter at a time, and then go back and do the same thing. But for each chapter and that's been so helpful. And when you take notes like what's your note taking process like. Joe: [01:00:51] I'm old school, I have legal pads. Yeah, I've piles of legal pads everywhere. So I just take notes, just scribble. Yeah. It's not so much that I need to take coherent notes. It's more just like there's an idea, jot it down, because that logs in your memory more efficiently than just saying, Oh yeah, I'll come back to that. Harpreet: [01:01:09] So pointer to to memory or something like that. Joe: [01:01:12] Exactly. Yeah. Harpreet: [01:01:14] I've recently maybe you recommend this book to me to How to take smart notes. You know that I like that a lot. I've been trying to implement that in my life, so I all have just a pile of five by eight note cards by me when I read, and then I'll take like it'll highlight and then whatever I highlight, I will translate it into my own words. And then that gets into put into this box. And then I take what's in this box and then I put it into a obsidian and just these are all the have bidirectional linking in obsidian that just helps create a knowledge graph. It's pretty, pretty interesting. Joe: [01:01:47] That's interesting. One thing I do, though, one thing I want to point out for even for data scientists and people who write code, like just having a piece of paper, you just jot ideas like I always when I'm writing code, I always have a piece of paper or a legal pad, so I just write down [01:02:00] thoughts, flows, that kind of stuff. Just a tip. It works really well. Harpreet: [01:02:04] Yeah, absolutely. I used to I used to spend money on, like, the Moleskine notebook because. Joe: [01:02:09] Everybody oh, yeah. Harpreet: [01:02:10] Moleskine notebook. But then I found this thing. I just got it in the mail. Today there's a pack of ten time works wonders. These journals that are just as nice as the moleskine. But it's even better because the pages are numbered. And not only the pages are numbered, you can create your own index. And I've got a Yeah I bought a pack of ten of these for 80, 80 bucks Canadian. So I'm pretty sure it's a lot cheaper in the US and yeah, I'm excited about it. Joe: [01:02:35] That's really cool. Yeah. Harpreet: [01:02:37] I'll give you a link to it. Let's go to a random question generator and then we'll begin to wrap it up. So first question coming out, the random question generator is what's the best thing you got from one of your. Joe: [01:02:47] Parents childhood free from rules? So I grew up with very few restrictions on what I could do, but I think that taught me self discipline and independence. So. Mountains. Harpreet: [01:03:00] Ocean. Yeah. Yeah. Same here. I grew up in California, and I used to. Joe: [01:03:04] To. Harpreet: [01:03:05] I mean, we used to pick ocean all the time, but nowadays it's mountains with a lake around. That's. That would be my go. Joe: [01:03:11] That's cool. Harpreet: [01:03:12] What talent would you show off in? A talent show? Joe: [01:03:14] Either climbing or deejaying. One of the two. Yeah. Harpreet: [01:03:18] For the DJ party. Ooh, that's a good one. What? Bend your mind every time you think about it. Joe: [01:03:27] I think organizational behavior. Yeah. Every time I think I've got it figured out, I realize that I don't. And nobody ever probably will. So. Harpreet: [01:03:38] Organizational behavior. Break that down. What do you mean by organizational? Joe: [01:03:41] Well, so, yeah, I mean, you see a lot of companies, right? It's a collection of individually very smart people. When you get them together, it sort of takes on a life of its own. Right. And company culture, everything about it. I'm just endlessly fascinated by this because it never works the way it should in theory. [01:04:00] So I'm constantly amazed with the things I see. Harpreet: [01:04:05] One more. What's your favorite piece of clothing that you own? Joe: [01:04:10] I don't know, man. Probably this jacket. I'm just kidding, but. Yeah. That's a good question. I don't really have one, so I try and I try not to get too attached to things. Harpreet: [01:04:20] That's a good way to be, man. Yeah, I spent a couple of weekends just getting rid of shit. Joe: [01:04:26] Exactly. I just. Yeah, same. Just moved. So, like, I don't obviously I don't like certain things that's going to happen, so. Harpreet: [01:04:35] Joe, how can people connect with the work they find you online? Joe: [01:04:38] I mean, so I'm not on any social media except for LinkedIn. So if you look for me on Twitter, Facebook, TikTok or whatever, I'm not there, but you can find me on LinkedIn. Just search my name and then you can find me on YouTube. Ternary data, which is my company, put up a lot of content. They're going to be pumping out a lot more content really soon of courses in tutorials and stuff. And then obviously you can email me at Joe Turner Datacom, as you kind of mentioned, you know, I do like to help mentor. You know, it's tough for me to do it individually. I would say if you want mentorship, this is more of a PSA show, Property Office. I can probably be more effective in a group than I can individually. Spend time is very limited right now, so. Harpreet: [01:05:18] Thank you so much. I'll be sure to thank you all of that all of that information right there in the show notes and a link to the book and everything when it comes out. Joe, thank you so much for taking time. I appreciate you being here. And as usual, my friends, remember, you've got one life on this planet. Why not try to do something big? Cheers, everyone. Joe: [01:05:33] Thanks, Aubrey.