Data Engineer Builds AI Sales Tool at Bayer: Real Enterprise AI Implementation

Vance Crowe [00:00:08] Welcome back to the podcast. I'm glad you're here. I'm your host, Vance Crowe. Each, whenever I want, I sit down with interesting people and it just so happened my good buddy, the highly disagreeable data engineer, Rob Long, is in the studio. Welcome to the podcast.

Rob Long [00:00:22] Great. It's good to be here.

Vance Crowe [00:00:24] How does that strike you being called The Highly Disagreeable Rob Long,

Rob Long [00:00:28] It sounds like you're deliberately handicapping me. Oh. So that people will be, have low expectations and then be maybe, maybe be pleasantly surprised. We'll find out. Yeah.

Vance Crowe [00:00:37] I think it's probably better to have low expectations. I know when I'm giving talks, if somebody stands up and is like, this is gonna be the greatest speech you've ever heard. Yeah.

Rob Long [00:00:47] And

Vance Crowe [00:00:47] You're like, no, no. And don't do that at all.

Rob Long [00:00:49] Show don't tell.

Vance Crowe [00:00:51] So one of the things that I've learned from you is this fascinating question that you ask people and then I'm gonna ask you and see where you take it, which is, what is something beautiful that you've observed lately?

Rob Long [00:01:03] Okay. Well there's a whole list of things. So I'll, I'll give you maybe the top two. Just a few weeks ago I was on a flight headed to Germany for, for business. And we were coming in across the bottom coast of England, which you could, we were kind of like looking down across the top of the cliffs into the English Channel. And we turned and went south to cross over into France and Belgium and Germany. And in one moment I could kind of tilt my head and see the both sides and the entire English channel. So like I could see Dover and then the English channel and then Cal a on the other side. And I thought, this is, this is a pretty amazing view that, you know, u until very recently, no one could possibly have. And very few people probably ever get to see that. And I thought that was pretty amazing. And I don't know if this is it, it's maybe more interesting than beautiful. But just today I was packing up my stuff to come into the office and I found a tiny, teeny tiny baby scorpion sitting on the power brick of my mac on, on the desk in my office, which is not outside, you know, but it was like just half a centimeter long. And I think he was there for the, for the heat. And I was like fascinated and interested, also disgusted at the same time. 'cause it was in my house and they are poisonous and are there any more of these?

Rob Long [00:02:34] And how did it get in here? But I don't know, I thought it was a neat little

Vance Crowe [00:02:38] Is that kind are scorpions in Missouri. I didn't even know this was possible. Yeah. I hate that the scorpions in my mind are something I wanna be far, far away from.

Rob Long [00:02:48] Yeah. I don't think they'll kill you, but you don't want to mess with it.

Vance Crowe [00:02:51] Yeah. My wife is definitely allergic to these things. Like, like yeah. Gets stung by one of those and she's never been stung by a, a bee, but we've always had the, like is it the same thing? I think they're in the same genre.

Rob Long [00:03:04] No, they're, they're different chais rates or something like that. I don't know the name of the family, but there's this whole huge, like there are sea scorpion fossils from hundreds of millions of years ago that are like bigger than this table. They're just ridiculous and, and terrifying if you imagine encountering one of those things.

Vance Crowe [00:03:24] Yeah. I was just watching a thing the other day about how the, if you're in a cold environment, if you're in like a polar environment, the cold temperatures, they think slow down the metabolism of some of these giant sea creatures, which is why you get these, like you can have sea spiders in the, in the Atlantic Ocean and they're just kinda little itty bitty things. But if you take that same sea spider and you take it down to the depths near the Arctic ocean, they become these like giant monstrosities of your,

Rob Long [00:03:55] I've heard of this nightmares. The term abyss gigantism comes to mind. I think this is a Wikipedia entry I have read at some point. Yeah, no, that's fascinating though.

Vance Crowe [00:04:06] You have a, a system for categorizing interesting things that you come across or that you read that's somewhat like Wikipedia.

Rob Long [00:04:17] Yeah. This is my personal, you know, collection of notes, which is in markdown, in obsidian is what I use to, to drive it. I also just edit it, edit the files directly to, but yeah, you use the word ontology or you know, a, a taxonomy of things that I keep and it's just, you know, if you work with computers for a long enough period of time, you end up with junk drawers of files everywhere where it's just a bunch of stuff that you've downloaded that you don't want to just throw away. 'cause the moment you delete it, it, you're gonna need it, you know? And it costs you so little to keep four kilobytes in a file somewhere. But the thing that it does cost you is your ability to understand anything that's going on. So if you have just some kind of minimal organization to it, then all of these things can have value. Right. But I, I like to keep, like, you know, we've talked a lot about how I track books that I read and stuff like that. This is the same, the same concept there except I'll keep, you know, articles and just anything of interest and kind of link it into other stuff so that I can come back later and find those things or update them. And it's just nice, a nice rich growing set of resources that I'm kind of gardening, you know?

Vance Crowe [00:05:38] Yeah. I can relate to this in a way that I couldn't before because now using ai I create a lot of like instructions for my AI and projects and websites. And so I have to find a way to store this. And I remember I was losing my mind because I, I was just giving the AI do this command and this one and this one, and I came up with A-J-S-O-N blob, I think is what it's called, where it was just like, you imagine that the way the AI is handling it is you say, don't do this anymore. And it's going back in there and erasing it, but it's not, it's just adding another command. Don't do that. So it has the do this and then somewhere you're putting in, don't do this. And you taught me about GitHubs and repos and checking out information and making updates to it and checking it back in and all the way up until I needed it. This was completely irrelevant to me. Yeah,

Rob Long [00:06:30] It seems like a lot of effort to get started and if it, you're, you don't have any particular reason to do it, then yeah, it is, it is too much. But the moment that you're swamped by markdown files, which now everyone is, you know, we are all markdown now is, is happening right then, then all of a sudden it becomes insanely valuable. The ability to do that and then say, oh my God, what have you done agent? You what in the world you can roll back all the changes to a, some known good state that you've tagged and said, okay, we deployed this to prod last time, so let's go all the way back to that and erase your nonsense and start again.

Vance Crowe [00:07:08] I, you're, you're the one that coined the, the Fantasia concept with AI in my mind, you know, where like you can start using AI being like, ah, I'm making the mops carry all the water for me. And then it just completely goes totally out of control. And I don't know if you have had the AI go out of control on you, but I have had the AI start emailing people and doing shit I did not want it to do.

Rob Long [00:07:32] I can't imagine giving the AI access to my email, my personal email. I would, I would never ever do that. But absolutely the primary, if you, if you, all of a sudden if you're like, boy, this was working just fine and then it just seemed to lose its mind and become crazy, you've almost certainly crossed the line of context window fullness and you just need to save what you're working on, clear it and start a new fresh context window.

Vance Crowe [00:08:03] So maybe a good place to go that would be interesting is you are using AI on a professional level. Not, not just like, Hey, I wanna build a website for my myself. How do you use ai? What is, what is the, what is the professional user of AI different than the casual amateur?

Rob Long [00:08:21] Well, I think there's a couple of different areas that we could be referring to. So like for my personal use as in, you know, getting tasks done for myself and then deploying applications that wrap or use AI themselves for an enterprise. Like, so I work at Bayer, that's no mystery. A project I've been working on for the last, I guess nine, nine or so months now, is something called Sales Companion. So we have a vast sales organization that's out there selling seeds to farmers. And those salespeople, well first of all, they created a new role that they call a sales agronomist, which I think is not new in the industry, but was new for Bayer. And so they wanted to kind of kick this off by turbocharging their ability to, to sell, to help the growers that they're interacting with. And so they commissioned us to build this app. One aspect of it is an agronomy agent, which has all sorts of internal data that we generate from creating all the hybrids. Yeah. So people

Vance Crowe [00:09:30] Not involved in agriculture agronomy is like, Hey, I need to figure out all the problems that could come up with, with growing crops. It could be anything from pests to how do I set up the, the my planter rows and how far I space them apart to harvesting windows. Like everything that is agriculture and economy kind of put together. Yeah. And so this is like you're saying that part of this app is designed to be very specific on helping people resolve the challenges they have to be the most efficient grower that they can be. Yeah,

Rob Long [00:10:05] Absolutely. So, and it's not just a generic thing, but it's primed with all of our internal knowledge and lists of our products and different things like that. So that guy sits on one side and the other side we have an application that's sales companion, that's the actual, you know, it's an iOS application that the, the salespeople use. And so they go in and have a conversation with the grower that's in their territory. They come out and they just kind of do a brain dump into the app. They'll, there's a, you can speak to it, you can type into it, you can take pictures like, so if you've taken written notes, you can just take a picture of that and it'll include that in the whole thing. And the idea is to just, you know, capture as much of the, the knowledge to help these people as possible so that it's not locked up in one person's head who's, who runs that territory and can leave or you know, God forbid die or something. Then you just lose all of this kind of knowledge about how, how things are working. So this captures that and makes it available internally. But then also every time they enter these notes, they get summarized offline and fortified with the agronomy agent. So it will read these things and say, oh, okay, it looks like you're customer here is having a problem with frogeye leaf spot disease on their, you know, their soy. You should recommend de laro to them as a, as a product. 'cause that'll take care of it at this growth stage and blah, blah blah.

Rob Long [00:11:36] Right? Like there's a whole chunk of verbiage around how to explain what it is, how to use it, everything like that. And so that's just there because a lot of people don't have access to that kind of knowledge and aren't gonna dig around in databases and find these things. So this just kind of surfaces a lot more information to them. That's one thing that I'm doing to use ai. So in that, there's various different processes like the agronomy agent, the summarization, you can ask questions so you know of your territory. I, I might say, well, which of my customers are having problems with frogeye leaf spot disease in English? And it will answer you in English like, oh, bill Johnson of Sunny Valley Farms and whatever. And it'll answer you and understands what you mean and will answer your question and simultaneously provide you links to the relevant notes that you took that referenced the thing that you asked it about. So it's a lot like the Google AI enabled search that they have. So it's like trying to just answer your question right away. But if your goal was to go find some other resource, you still have those there. And the way that we use those is good old fashioned software engineering. So we're deploying applications that connect to the APIs that provide AI models and providing them context that come from our databases and various other systems and kind of marshaling all that stuff the way you do personally, I use a whole lot of different AI products.

Rob Long [00:13:17] So, you know, Claude code is my default, you know, go-to for building stuff on my own time at work. We have a different harness, but I'm also using the Claude model through that harness. That's something that we built internally that's specifically designed for the way that we work. And you know, just kind of the, for like random asking a question kind of stuff, I still have chat GPT on my phone that I'll, you know, like, what's the meaning of this? Go look this up for me. Kind of stuff like that. I went, I used briefly for a while, an open claw instance, which I really, really enjoyed. I had a lot of fun with that. But the moment that Anthropic turned off the API access for the subscriptions and I I, I wasn't using it for something valuable enough to pay thousands of dollars a month for the API access costs that are outside of the subscription. And so I just pulled a few workflows that I was interested in there. Like there's one particular blog that I follow for AI and kind of, you know, cloud computing sort of stuff, and I just have it crank every day, pull posts from there and generate a podcast out of that and read it, publish it. And then I, I read that so, or listen to it every day for myself. So like I automated that one little piece without the whole open claw apparatus.

Vance Crowe [00:14:47] I mean, I, I am using open claw way less than I was before. There was a period of time time where I was spending anywhere from 50 to $200 a day because I was just having it build stuff and it was great, but I'm now way more stable and, and use a different system. But on that open claw, when you or any system, if you wanted to make a, I have a blog, it comes in every day and I want to turn it into an audio podcast that I can listen to. What's the steps to take to get to that?

Rob Long [00:15:17] Boy, that's a good question. The steps that I have taken are probably not the steps that most people would take. So my goal was to do this entirely on my own machine. So, you know, I wrote a pipeline in Go, which is a programming language or I didn't write it, I had Claude write it, right? And the whole purpose of this, or the whole like leaning of is, is to get as much of that work done as possible. Not using the AI, because the AI is expensive and error prone, right? Like there are certain pieces that only can be done by it, but everything else I automate and just make it a retriable thing. 'cause everything that's over a network will eventually fail. So you have to be able to retry things. So I pull down, I have a little tool that I, I just found somewhere that will take a website that you pointed at and turn it into a markdown file, right? So I just have it go have it, pull down the main site of the blog and then have it read through there for a new article, pluck that article out, pull that URL out, and then it has links to like five or 10 other sites. Like that's the nature of this blog. The guy's just kind of saying, here are some good articles I read today in a brief summary. And then I have this thing go and pull all of the original sources of those articles. And then that chunk of markdown I hand to a model which then reads it and prepares a script based on, you know, a prompt that I have.

Rob Long [00:16:55] And then that script I have read by a text to speech model that I'm running on my local machine. So I've got a very moderate GPU in there, which is like 12 gigabytes of VAM, which is not a lot, but it's enough to do this kind of a thing. So I have it then read that. And then it also will come up with a prompt for an image to generate. And I just kind of have a bunch of random science fictiony kind of tropes that I have it, you know, basically use. And it'll generate a new picture for everyone. And so like each episode of the podcast that that's produced has this kind of randomly generated sci-fi looking picture to it. 'cause it's all about AI and computers and stuff, so why not? And then it publishes it.

Vance Crowe [00:17:47] Oh, so it publishes it in an, in an accessible place.

Rob Long [00:17:51] Yeah, it's a podcast, you know, so dss.io is where I, I, I made this site, which is, you know, it's a whole other angle that we could talk about if we wanted to. But I publish it on Nostr, which I say Nostr now because the computer understands when I say Nostr and not when I say Nostr, which is how I used to say it. So I've actually, in order to be understand stood I do that. But, so I'm publishing a noser article and then the DRSS website will take an account on Nasta and turn any articles that are published there into an RSS feed. And this is just a way that you can get messages from a source. And in fact is what podcasts are. They are an RSS feed that just has pointers to audio files. And then your RSS, or pardon me, your podcast reader or listener just goes and pulls those MP three files and plays them for you. So I just built all these interlocking pieces. I've been working on parts of this for 10 years at this point. It's just kind of a, a hobby that I do. But

Vance Crowe [00:19:06] Yeah. You said that the computer hearing you Nostr instead of Nostr, which we've talked about this in past episodes, but how much are you using verbal your voice to communicate with ai?

Rob Long [00:19:19] It depends on the context. If I am, I, I say this if I'm doing real work wi, which means like, you know, programming, you know, manipulating data, doing things, not at all, just there's zero. But if I'm prompting an ai, I do find that the audio input works best for me because my, the, the rambling nature of speaking you, it doesn't, it it's not, it's not a drawback. Like if I, if I start giving you details of something, like I'm trying to explain something I want you to do. If I just keep laying more and more things on the pile, you eventually will run out of Ram. Listen. Yeah. You stop, understand, stop listening.

Vance Crowe [00:20:03] Yeah, listen,

Rob Long [00:20:04] And you'll, you'll forget the first things I told you and mix things up and whatever. But that's not the case with the computer, right? It's like, I have this, you have a curve where it's like the more detail you're giving it, the better an answer you're getting, but then it kind of has diminishing returns, but there's still returns. So it doesn't, it's fine. You can just keep talking and keep talking and adding more detail and adding more detail. Not indefinitely, but I find that it is better. And then like if I'm driving, I find that the time, like I can be interacting back and forth researching something without actually having to sit there and read stuff. I can say, what's a library that I can use to turn a website into a markdown file? You know, like, is there anything written in go, no, I don't like rust, let's go here, let's, you know, I can kind of interactively research things and get things done without having to look at the telephone.

Vance Crowe [00:20:59] I, I am the same way on the, you know, research or I don't, I don't know what I would call it. It feels too easy to be research, you know, I often think of research as like sitting in a library, what I was doing in grad school, like cycling through things. But my understanding, my like comprehension goes way up when something is spoken to me, like my auditory, my ability to remember things that I've heard is way higher than almost anything else that I can do. And like I, I was like, all right, I'm reading the Bible with my girls. They're, the Macedonians are brought up and I'm like, who are these people? And then you can just start asking AI and it's just like having the greatest professor you've ever had. And then anytime you're like, wait, wait, I don't understand. Tell me, tell me about that. He never gets frustrated. He's never like, let me finish this point first. You know? And like it's enabled me to have a much better understanding of like, areas of history and, and like, I've done this on the path of religions and hard forks in Christianity and Alexander the great and very, you know, the things I'm interested in right now. And I do it for hours at a time.

Rob Long [00:22:11] So that's an interesting turn to take there into history and maybe research and comprehension. I, I am not like that. If I hear things I can pluck interesting concepts and kind of save one little direction or pointer to something. But the content is not, I don't memorize or, or retain things from hearing very well. It's much more reading is how I do that, like visually looking at stuff. But beyond that, I feel like How would you characterize your, your knowledge of history? Like, I feel like, you know, your college educated individual, you know, you've probably had some raft of history courses. Do you feel like you have a good grasp of it?

Vance Crowe [00:22:59] No, I mean, I would say that ai, my grasp of it in the last year has gone up exponentially more than whatever I learned in all of high school and all of college, you know, graduate school because I'm able to like zoom the camera lens in and zoom it back out and get a better scope of it all, but very, very low. And

Rob Long [00:23:20] It's probably adjacent to or linking to ideas that are relevant to you. Yeah,

Vance Crowe [00:23:25] Stories that I care about.

Rob Long [00:23:25] That's why you're there, right? But that's, that's kind of where I'm going with this, is the more stuff that you have in your, in the hopper for history, like Alexander the Great, right? Like I couldn't tell you what year that he was the whatever, somewhere

Vance Crowe [00:23:44] Around 300 bc

Rob Long [00:23:46] 300, yeah. It would've been after 400, right? 'cause Aris, he was a pupil of Aristotle. And, but that's the thing, you have all these kind of like markers that you can think of. Like obviously zero ad is a, is an interesting starting point where you can be like, alright, there's the birth of Jesus is here, and then you have like 400 BC the like high point of Athenian culture, you know, Plato, Aristotle, these guys. And you know, you can go further back and further forward and there's kind of like points and items in history in different locations in the world. And the more of those things, you know, the more that you can like find something adjacent that you already know about and kind of link it to that.

Vance Crowe [00:24:33] And when you have the like, connection between things, the sparkle of electricity in my brain just goes nuts. Like this, this will make me sound dumb, but I really didn't know that Greek culture preceded Jesus. It was just like, I just didn't know, you know? I know you talk about the Romans in, in, you know, all throughout the New Testament, you know, whatever. But it did not dawn on me Romans after the Greeks and the Greeks before Jesus. So the idea of Alexander the Great being Greek and going into the Persian Empire and tearing it apart, I had no concept until I was like, wait a second, who are, who's this? And why is the, why is the book of Maccabees describing a, you know, the post Alexander the Great War? So then I was starting to talk with AI about it. But exactly to your point, it's the markers that give you the, the context to make it so things are sticky.

Rob Long [00:25:28] Yeah, yeah, that's right. It's, it's so much fun to think about those things, right? And like that there was a Greek empire crossing this whole region pretty much the, almost the entire footprint of the Roman Empire minus the like western pieces of it, and then a whole lot more to the east too. But like, you know, everyone in that area spoke Greek. And then the Romans came, you know, like, I don't know, a hundred BC something like this timeframe and took over pretty much all of the gr like they took over Greece and, and encompassed all of that stuff. And then Latin becomes the lingua franca. But people in the Eastern Mediterranean still retained that Greek origins, those cultures. And so later on, a thousand years later when the Roman Empire breaks up, you have this eastern and western split Byzantium and Rome, right? And the Byzantine half is the former Greek speaking parts that were all Greek Orthodox and you know, eastern religion versus the western part. And so you just say you have these two units and it, like, all these things echo through history, you

Vance Crowe [00:26:39] Know? That's so interesting. 'cause if somebody had said, you know, who were the Byzantine at say, I don't know, maybe the afterthought of the Persians right after they were, you know, smashed by Alexander, maybe they came back, I don't know. But it's the Greek influence. Yeah, yeah. So the, okay, very

Rob Long [00:26:56] Interesting. You know, Istanbul, Constantinople,

Vance Crowe [00:26:58] I've heard the song, but no context for it.

Rob Long [00:27:01] Constantinople was what it was called. And it was the capital of the Byzantine Empire. This was the seat of Greek imperial government up until 1453 when the, the Turks came and took it over.

Vance Crowe [00:27:16] And that's when it became an Islamic, that's

Rob Long [00:27:19] When it became Istanbul.

Vance Crowe [00:27:21] Okay.

Rob Long [00:27:22] Yeah.

Vance Crowe [00:27:24] Yeah. I don't have any context for this. I, my focus has been post Roman and following the, you know, we've been, you and I have been talking about how, you know, you have the

Rob Long [00:27:37] Forks

Vance Crowe [00:27:38] In, in, in any kind of ideology, right? We talk about Bitcoin and then there's forks of Bitcoin and, and this is where people take what you, you had the original code and they start making adjustments to it, to the degree that the people on the original forks say, we're not into that. And then people that split over here, and then at over time the fork becomes its own branch or whatever that is. And I've been, I had spent quite a bit of time focusing on when did Protestantism come about and like, when this fork came out, how, how did that fork off of the Catholic church? So I've spent way more time northern area post Roman and know nothing of the Byzantine Empire at all.

Rob Long [00:28:21] That's interesting. Yeah. That's a And then those forks fork, and they fork each other and it just keeps proliferating, right? And then you end up with this kind of bimodal setup where there's one that wins as like the original recipe, the, the, the, the, the, the, the real true thing. And then the others are trying to compete with that to be different and be, have some other aspect that has more of a draw. So, and then we've mapped that onto Catholicism and Protestantism and then all the various things that flow from that to Bitcoin and Litecoin and then Ethereum and all the things, you know, Bitcoin, everything that goes downstream from all that wild stuff. And you end up with a, a similar pattern where there's like the network effect on the original one, and then the other ones have to try to offer some kind of crazy new feature to draw in people.

Vance Crowe [00:29:21] Well, and then to your point before about having hooks that allow you to understand something deeper, I mean, I had actually a very good religion class in high school where they were talking about where did Protestantism come and who were the puritans who were, but because it didn't get connected to anything, it was like islands of knowledge that like you could, I could only understand as like, oh, the Puritans, they believed in predestination and that made them work really hard and like, but couldn't connect it to any concept. But when you actually understand, oh, Luther broke off from the Catholic church and as soon as he did that, now people are like, oh, we can do that. I didn't know we could do that. And then you get Calvinism and then you know, how that broke out and then where did Calvinism and Catholicism collide and have all of their conflicts? Like to me, I think, and maybe this is just me, like the, this concept of hard forks having an actual model in my brain for that, if somebody would've taught me history based on the concept of forking, I think I would've gotten it a lot better. But I also maybe just didn't get it. 'cause I wasn't paying attention.

Rob Long [00:30:30] I mean, I was that way with literature in high school, right? Like I, I was supposed to read Moby Dick in, in my high school literature class, and I remember reading a few pages and just being like, I don't, I don't care about this at all. This is not, and it's like a combination of, it's not relevant to me yet in my life, I'm not prepared to try to draw meaning out of this. You know, it's arcane and uninteresting to me to something that's pushed on you that you're just like, eh, I I just wanna get through this. And then I came back later in life and read it and it's one of my all time favorite books, right? But I don't know. And also just having different modes to have these things cross crossed paths in like, this is a funny diversion. But, you know, I've played a lot of video games when I was a younger person, not, not in the last, I don't know, 10 or 15 years, but, you know, most of them were like simulations or like, you know, I'll say Europa Universalis IV. And I'll shout out Jeremy Osh, who I think knows what I'm talking about, where you're, you start in the year 1419 and you're like one of several hundred possible little principalities all around the world that you can pick and play as, and you go for like 400 years and like you have little, it's kind of like risk. Have you ever seen that how the world's divided up into territories? Only They're a lot smaller than that. And you know, you have soldiers and you know, domestic policy and foreign policy, and then you get like historic leaders for that.

Rob Long [00:32:04] So if you were, you know, the French, you might have King Louis the four as your leader in that time in history and his like attributes that help you match roughly what history believes about these people. So like, if you sit there and play this game for hours and hours and hours over, you know, days or months of your life, you end up like, you know, internalizing some of these things. Not to say that they were perfect or absolutely correct, but like, I don't know, I'm pretty good at maps, you know?

Vance Crowe [00:32:36] Yeah. Like some, somehow, like a more intuitive first person perspective of

Rob Long [00:32:41] Yeah. It makes it relevant to you of

Vance Crowe [00:32:42] These experiences. Yeah. Yeah. The, i I am so deeply grateful for a game. We've probably even talked about it on this podcast, SIM City 2000. Like that game singularly helped me understand so much about like, how humans work. Even just like cities or just like, oh, you need power. Oh you, you're gonna need some water here. And like, it's so funny 'cause if you encounter somebody else that played that game, you have instant connection, right? Like there's, there's like, there's probably a lot of games that are like that, but Sim City 2000 in particular Yeah.

Rob Long [00:33:16] Was

Vance Crowe [00:33:16] That way.

Rob Long [00:33:17] No, I, I remember particularly the, the book, the instruction book for that was like, not just a, here's how you can add power lines, here's how you save your game. But like, it had quotes and like interesting bits about how people think about cities in general. And I remember just being fascinated by this. Like, I've read that book many times as a kid just 'cause I was so into that. But yeah, that's, that whole Will Wright, you know, software toys, right? It's meant, it's meant to be this thing where you can kind of tweak various aspects and see what happens. It's not so much that you are building explicitly a in a perfectly specified thing and you know exactly the outcome. It's more like a garden that you're tending and you're delighted to see interesting things that happen. Like where is there high, you know, property values and where, you know, educated citizens and other interesting aspects that just can just kind of emerge from these things.

Vance Crowe [00:34:17] What's the name of that talk where he talks about the gradients? Do you remember that?

Rob Long [00:34:21] No.

Vance Crowe [00:34:22] Do you know what I'm talking about? The

Rob Long [00:34:23] Talk? Yeah. It was like an hour and a half long. It

Vance Crowe [00:34:25] Was a big long talk. So though it's like a mind blowing, like helping you understand that there's a, like basically, you know, you could, what is it? One axis is like fun and one axis is work or something. I don't

Rob Long [00:34:38] Remember. Like yeah, it was, no, it was like social life and work life.

Vance Crowe [00:34:40] Yeah.

Rob Long [00:34:40] And it was like, you know, you could maximize one and then you just get zero on the other. And so that's not great, but there's like, splitting it perfectly up the middle is very hard to do. That's a narrow thing. Like, it's pretty easy to just max out one or the other ones. But there's these deep alleys between those two angles. So if you try to go up the middle and, and do both, it's a lot harder. You, you're gonna fall off and go into one of the, one of the gutters,

Vance Crowe [00:35:12] The term fitness landscape comes to mind. Is that this?

Rob Long [00:35:16] Yeah, sort of.

Vance Crowe [00:35:17] Yeah. I not that long ago was using a concept that you and I have talked about with local optimas.

Rob Long [00:35:24] This

Vance Crowe [00:35:24] Is one of those things that when you draw it for people, they're like, oh man, that explains so much of my whole life, right? Where well, you do a better job explaining it than me.

Rob Long [00:35:33] Yeah, no, this is the part where, I mean, obviously we're not gonna do this, but if I, I had a whiteboard, I would, you know, I'll just kind of do this up here, right? So we'll just imagine a profile of a landscape and we'll say there's a little hill here, let's call it a foothill and then a huge mountain here. And the height, the alt, your goal is to get the highest altitude. So one simple algorithm is to say, I move for forwards and if I go up, I keep going. And the moment I start going down again, I stop and I'm, I've now reached the highest point, right? So, but in effect, but that strategy does is bring you right to the top of that little foothill and then you stay up there in that little spot. Whereas if you could see further on the landscape, you'd see, well, I can de optimize, I can go downhill for a little bit and then I can go way further uphill. But if you just have a very simple algorithm that's hill climbing, then you can't do that, right? So that's, that's kind of that,

Vance Crowe [00:36:35] That concept and then applied to life. It's to say like, if all you do is just like, hey, as long as I'm getting better, I'm doing this thing. But then when you hit that like, oh no, every time I go forward I go downward. You want to like adjust and say, is it worthwhile? There may be things I need to unlearn. There may be habits that got me to this local optima that are gonna keep me from being able to go up to the, to the rest of that mountain. And being willing to take a look at whatever you're, you're doing and saying, all right, is it worth it to go down some in order that I can go up way higher in the future?

Rob Long [00:37:11] Yeah. And you can freight in a whole bunch of other concepts that we've talked a lot about, like, you know, time preference, whether you prefer to have goods immediately in the present, or you're willing to wait for a longer period to have better things in the long run. And so if you're like just straight hill climbing is like hedonism, you're just pursuing straight to the goal. But if you have a little bit of fortitude, a little bit of time because you've planned things out, then you can take random walks, right? So if you're going downhill, or maybe this is a three dimensional map, it has, you know, an X and a Y and then a Z height, right? So you're trying to maximize your Z height and you can wander around in the X and y dimension because you can't see the entire map. Like that's maybe the, that's another part of this, right? Is that yeah, if you know the fitness landscape perfectly, then you can just go straight to the highest point. That's, that's easy to do, but that's rarely the case with complicated things you have to do. You have to sample it at some location and measure it. And that implies a cost of time or effort. And so where should you do that? What, what is your strategy? And there's a whole, you know, organiz a whole section of mathematics and data science that is about, you know, solving these complex differential equations or whatever they might be to find the right spot on that map or explore it iteratively.

Rob Long [00:38:43] And basically you have to have the strength to be able to withstand going downhill detim not being at the highest thing you can find right now to, to be able to find something even better, right? So like, if you're strong, you can afford grace to people that you encounter, right? It's like just bringing it back to like human interactions. I could be very distrustful and suspect everyone I encounter, or if I know that I can handle myself and that I've got, I'm, I'm gonna be fine. You can just kind of grant people some grace and let them find, find what it is that they know that you need to know or how, how you can get along together, you know?

Vance Crowe [00:39:25] Yeah. I think that's one of the things that when you're young, it's hard to imagine being able to, the stronger I am, the easier it is to be kind, the easier it is to be open or generous, right? And, and like, because you maybe want to be seen kind, generous and open, but if you're very vulnerable, this is bad, this is bad strategies, you know, but the stronger you are then the more flexibility. And that's true across all domains. So as you're describing this, a thought is coming to my mind, which is today on social media I was talking to, I was observed a guy that I, I've never met him in, in real life, but we've interacted a bunch of times on X and he was basically saying like, the AI data centers, this is catastrophe, this is all this is leading towards is data surveillance and it's not helping us. When I see that, like, I think, okay, tell me more about this, right? Like, is there something, you know, that I don't know, it like doesn't cost me very much to, to ask and the, the feedback that I get is like, this is obvious they are trying to surveil us using this, first of all. Like, how, how do you, how do you think about that? I mean, they're probably not wrong in the sense that more data capture, you know, flock cameras, things like this, but it's, it seems like there is a trade off there.

Rob Long [00:40:51] I mean, we're already thoroughly, thoroughly observed from every angle and have been for a long, long time, right? Like there's no, all of the existing data centers are holding all everything you're doing right? All, every website you go to, every move of your mouse is being recorded in some cases and sent elsewhere. I understand and myself do not like to be subjected to surveillance. So I take steps to avoid that, but at the same time, I feel like there's productive ways to engage with these things. So if I could get a model that was as smart as Claude is and run it locally and not involve a third party, I would do that because that would afford me to be free from surveillance. And I, I do have those, they're just ridiculous. They're just not good enough. And I guess I could say that like two years ago if I were using a model that I can run locally on my machine, I would be blown away and think what in the world happened to computers? This is amazing. But after using the best models, it feels like you're just going to a kindergarten to hire somebody. It's not, it's, it's difficult to know that there's something better and, and like just deal with something going in circles chasing its tail. So I'm willing to make the trade off for some aspects of what I'm doing. Like my podcast generating workflow, right, that I told you about.

Rob Long [00:42:24] I don't care if philanthropic knows how I do that, that that doesn't bother me at all. I'm not worried about that. You know, that's fine. There are things I guess that I, I don't want them to know about, you know, like for example, my, my obsidian vault, you know, I would like to keep some aspects out out of their hands from that. So that, that's like my private domain. I, I don't write things down there that I don't want other people to know because it's a computer and it, it can happen. But yeah, generally I keep the, you know, other people's AI out of there.

Vance Crowe [00:43:03] Yeah. I, any fear that I had about AI I think gets put aside for its massive productivity and not, not even just productivity, the skill enhancement that it gives me, like, the ability to do things that I am actually not capable of doing, doing without AI is, is amazing. And the jump, I think I'm gonna do a podcast here again. I did a podcast a while back where I had the AI interview me on a podcast. Yeah,

Rob Long [00:43:33] Yeah.

Vance Crowe [00:43:34] I'm, I think I'm gonna do it again because it is going to be staggering. It was like, that was one of the most popular podcasts I've ever done. People wrote me like, I cannot believe this is unbelievable. But the distance it's gone from whenever I did that a year and a half ago till now is so mind boggling. Like, AI is so much better. It's even so much better since I was doing stuff in January to now. Yeah. It's like staggering how much better it is.

Rob Long [00:44:01] What would I, you mentioned before that the people's kind of fear about the data center stuff. Is it really just surveillance that they're concerned about? Like I don't, I I used to be on XII just closed it up. I'm, I don't do any of that type of social media anymore, but like, like what is the fear out there? Like, I, I see it's hard,

Vance Crowe [00:44:25] It's hard for me to name I, I mean, among the people that I think are respectable Yeah. Hardworking, productive people. Jared McDaniel being one of them, this other guy goes by Mizzou Tiger. These guys are producing goods in the real world, right? Like the Jared McDaniel raises cattle, this real thing. And so like, that's why I don't just blow it off, but they're like, they, they describe it as worthless. Like if you go and say, well, what do you mean like, it, it can do these things? And it's like, well, people should be doing these things. We, we need more hard goods, we need more. But outside of that, I mean, I even kinda said this, like, to me the alarmism sounds like anti GMO or, or you know, climate change alarmists where you're like, once you get an inch beyond, I'm uncomfortable with this. The reasoning doesn't appear to be there. I've, I've not had anybody. But you know, you see all this people like booing the, you know, I saw that the commencement speech Yeah. The commencement speeches and they're booing the, the people that talk about ai, which to me it show, it represents, they are actually afraid, you know, they're afraid of the future. I'm not discounting their fear, but it also says, if you're doing this from a university, it means the university horribly failed them. You know, like put them in a position where they were not cognitively ready to deal with a world that's rapidly changing and they paid, they probably went into debt to not be prepared.

Rob Long [00:45:57] Yeah, I don't know. I,

Vance Crowe [00:45:58] I, I feel like I've heard

Rob Long [00:46:00] I all kinds of things myself from people just by where I live. I'm, I'm out in Franklin County, Missouri here, and like there was a, there was a data center that was slated to go in there and it got shut down, you know, and I, I don't know if there was a rezoning request or something, that was how they stopped it. But I, I don't know, I feel like also the people who produce things in the real world tend to be fairly, you know, libertarian or at least respectful of private property. And I don't know how you square that, right? If we're saying, I wanna build a building on property that I own and put computers in there, where's the problem with this? You know, like, I'm not saying there isn't, can't be one, I'm just not aware of it. And like, we would like that. I imagine people want there to be more businesses, more economic activity happening generally. So what's wrong with this economic activity?

Vance Crowe [00:46:59] The

Rob Long [00:47:00] Only

Vance Crowe [00:47:01] Complaint that like registers with me where I'm like, ah, I'm, I am a little nervous about that, is the price of electricity, right? Because, because we are hyper regulated on who can produce electricity, right? And we've not invested in it. And so we have shitty infrastructure and we have not been like gaining steam, pun intended, right? To, to build more out. Like yes. If you all of a sudden start plugging way more computers in, the demand for electricity is gonna go way up, whereas the supply stays the same, it's going to, it's gonna cause rates to go up. That's something I'm concerned about. But that answer to is build more power plants, right?

Rob Long [00:47:42] Or deregulate some of these things or allow people who want to build data centers to also build power plants, right? Like I, I thought that was a concept that was floated a few months ago at least, that there was gonna be this. And the same with water, right? Like there's this complaint about how much water they consume. And I like, I remember hearing all of these things in the past about Bitcoin, you know, it's like, oh, these servers are consuming more power than, you know, Switzerland or something like this. And it's like, okay, but you don't want it. But clearly a lot of people find some economic utility and it, otherwise they wouldn't just be pissing all this electricity down the drain for no good reason. You know,

Vance Crowe [00:48:24] I think there's something to do with like fear of the future and being left behind. I, we had a couple over to our house the other weekend, amazing couple, very intelligent, when we started talking about ai, they were like, oh man, this is gonna make such an unfair future that we don't want it. Right? Unfair, like, like that people, and I was like, let's be more specific. Like who unfair to, and these are well traveled people, people that have lived abroad. And they were like, well what about the people living in Africa or Ecuador or, you know, and I was saying, this is amazing.

Rob Long [00:49:01] Now they have the AI available to them.

Vance Crowe [00:49:03] Yeah. I mean I remember you and I talked about this before, like the, the getting lights into someone's house so their children can read at night is far superior to them sitting out in a parking lot reading under a street lamp. And even way better than sitting by candlelight not being able to read this is like that, you know, not just on steroids, but on rocket ship, right? Imagine the African child that doesn't even need a library card anymore, doesn't even need local resources. If he can have electricity and access to a terminal that gets him the internet, he could build things that are beyond our wildest dreams.

Rob Long [00:49:41] Yeah. I don't know. I don't know. I don't know. And maybe we shouldn't go too far down the road of imagining the arguments, you know, and arguing against them. I don't know. I asked you, right? Because I, I encounter this so frequently and I just, I can't get a mental model of what's going on there from people. And I don't, I don't, I've not pre like cornered someone and been like, tell me what you're really upset about. You know, like, I can't, I can't get there.

Vance Crowe [00:50:07] Well, I, I was in the, in the process of like exploring, hey, I want to bring on another young person. I've got some things that I want them to do. But one of the critical skills they would have to have is they have to be motivated to use AI to amplify what they're doing. And the feedback I heard back was, hey, a lot of people in my network, they're, they're reluctant to use AI because they feel like this is going to take away jobs. It's going to make it so they can't make a living. And I, I want to appreciate that, but when I use ai, I am like, it makes

Rob Long [00:50:39] Me

Legacy Interviews client testimonial (ad segment) [00:50:40] So much better. Yeah. Like I am able to produce at so much higher of a level that I don't understand it. And in fact, you could take a person who has just the median IQ and give them abilities that they had never had before.

Rob Long [00:50:56] Yeah.

Vance Crowe [00:50:57] So it doesn't make it, you know, we're two people are, that's what I feel very similarly like. Yeah. I don't know. Well, you know what, let's, let's take a break for just a second and, and let talk a little bit about Legacy Interviews. I bought a legacy interview because when I was listening to the podcast and heard about it, I thought it was a great idea. And then my mom listened to your podcast and she said she wishes she had that for her grandmother. And I said, well, why don't we do that for you and dad? And she never considered it for them, but it just quickly became like a really good idea that they were excited about. My parents never in their life had such a long period of time where they spoke about themselves. At first they were uncomfortable with that, but then quickly became comfortable and just realized how much of themselves they could get out there in that period of time where they are the focus. My parents enjoyed the experience, particularly my dad. He kind of struggled through his life of thinking he wasn't successful, just didn't have the material wealth that he felt that he should have. And then as he was telling his story, talking about his family and his life as a whole, he realized that he had this extraordinary success, but he never really thought of it as success. It was this moment where his whole life, like it all made sense now. And here's a man that just turned 80 and it's all because they just sat down and spent a few hours talking about their lives with you.

Vance Crowe [00:52:28] My mom was so grateful to have this on video, to really see their thoughts of their own life from start to finish. And we've got that forever. Yeah. So I was just telling you that I, I was so happy the other day because a person was like, I was like, Hey, how did you hear about Legacy Interviews? We just had this great conversation. They're booking one for their parents. And she was like, oh, well I've always wanted to do this. And when I would go on Google, I could never find the thing that I wanted. And then the other day I was sitting in front of my AI with Gemini and I just typed in, I want somebody to record my family, you know, telling their family stories, how, how should I go about doing this? And AI told me about Legacy Interviews and I was like, yeah, that's

Rob Long [00:53:13] Pretty good. No, I actually, I'm a happy customer of Legacy Interviews myself very as you well know. But yeah,

Vance Crowe [00:53:21] I,

Rob Long [00:53:22] You know, my in-laws, my wife's parents live in Minnesota, so, and we're here in St. Louis and we don't get to see them as frequently as we do my parents. And you know, we were kind of talking about how, how do we convey the value that they have to, to our kids to get to know them and hear their stories. And like, I know the answer to this. This is Legacy Interviews. And so we just put the money down and did one for her parents and one for my parents. And we were super happy with it.

Vance Crowe [00:53:54] So you haven't seen the interviews yet, right? No,

Rob Long [00:53:56] Not yet.

Vance Crowe [00:53:56] How, how was, how was the experience for them? What they

Rob Long [00:54:00] Well, for my wife's parents, they, they loved it. They had a great time. I think they enjoyed the setting and being able to talk about themselves all day. Not to mean that they're, you know, anything bad about them there, but that's just makes people generally feel good. Yeah. Course generally happy as you know, I think my parents are more reserved about that. My mom, at least she's where I get my disagreeable streak from

Vance Crowe [00:54:29] And

Rob Long [00:54:29] My dad is the, is the social talkative one. But they had a good time too, and I'm looking forward to it, seeing it. Yeah.

Vance Crowe [00:54:37] I, I often describe like even if you don't get the video, I mean that's why people do it, is to get the video, but the experience in and of itself, like when do you get a chance to be like, yeah, we had this thing. And I often prompt people like, you know, you think that the stories that are gonna make you the happiest to talk about are the ones where you already, like you knew how to do it and you succeeded. But actually the stories that make people like laugh and look at their spouse and smile are the ones that are, are the exact ones that you and I'll be laughing about with our spouse where you're like, remember how terrible that was? Remember how, how bad everything looked. And then, because the best stories are the ones where things were going down and then you made some change and you figured out your way out of it. But the human experience is like, okay, we got through that terrible thing, now we've got the next thing we've gotta deal with and the next thing. So you could go your whole life and never actually look back on what you've done.

Rob Long [00:55:30] Yeah, no, that's right. That's like what is comedy is the falling and then rising action, you know, it's like that stuff that's not truly tragic right. Is ends up being comedic 'cause you overcame it through the rest of your life. And that's the, that's the looking back part.

Vance Crowe [00:55:48] So I am, I, you know, as you know, I teach these classes on in space communicating, and one of the sections is on storytelling. And I'm like, you know, people are the most embarrassed when they're telling their resume story about like, Hey, how did you get this? You know, how'd you get to this job? And the true story is like, I was going along and Oh, I got fired. Or, you know, oh, I had this problem. Right? And they think I've gotta cover that up in my story. And I'm not suggesting people should go out and tell the, the terrible stories. But truly, when you tell your introduction in a way where you're like, I was going along and I thought I knew how everything worked, and like, oh no, I, I crashed into something that didn't work. And now all of a sudden I was in this unexpected chaos. My, I thought I was gonna have to move my whole family to California because we were gonna do this other thing. And it was this chance encounter that I had where I really listened to what this guy said, and I made a phone call and, oh, my whole life changed as a result of it. And that's why I'm here. That's not only a more interesting story for people to remember. It also, if you show some level of vulnerability in, in an introduction to yourself, people are like, that shit happened to me too.

Rob Long [00:56:56] No, exactly. Right. Like, I, I hire a lot of people, software engineers, and one of the things that I'll ask is, what is, what is the worst mistake you've ever made? Or give me an example of a horrible mistake that you've made. And it's a really good, because like the, the person who's like, oh my God, let me tell you, I, I thought I was real clever. I set this thing up and then I destroyed the production server and we were down in production and I had to recover it from scratch. And it was this and that. And the person who's willing to tell you that story, you know, that they, they're at least in the right headspace, right? Like, they might not be the best person in the world, but they've actually had their hands on meaningful problems, had a real problem occur. And they're not afraid to tell you that they made a mistake. If somebody is like, nah, I don't think I've ever made anything. I think it's like, well then what are you doing? You know, really? Like, what are you really doing? It's a pretty good sorting mechanism.

Vance Crowe [00:57:58] That is, that is very good for, for hiring. I mean, it's, as you're saying this, I'm thinking like, I've, my, one of my work ones I can remember, I was like so proud of myself 'cause I was beating down a vendor on pricing, right? Just beating him down. And then we went to like, go to production and send this thing off. And I didn't actually look at the proof. And the way he made it cheaper was he just degraded the quality to the degree that I could have done it better with a Xerox, right? So I had on behalf of my company, you know, I, I mean, I remember going into my boss's office and being like, I got him down by 40%. And then we started getting receipts back from this mailing that I did, and it was like, so horrifying. And then

Rob Long [00:58:40] You're

Vance Crowe [00:58:40] Like, what lesson did you learn? Well one, you know, if you beat somebody up on price too much, they're gonna give you crappy work. And two, like, why didn't you check the proofs? Like, why did you listen to this guy? Yeah.

Rob Long [00:58:50] And

Vance Crowe [00:58:50] Like, but then, you know, this guy has done work and, and demonstrating humility is maybe the most important thing. 'cause humility is the ability to know that there is more to learn, right? Yeah. Like that there's, there like, oh, I don't have everything figured out. No,

Rob Long [00:59:07] I, yeah. My, my brief story along those lines is I used to work at the Genome Institute at Washington University, and we have this, you know, big compute farm and there's like one place where the reference genome, so this is like the human genome reference that they have that all of these jobs that are running in parallel all the time, cranking through all these samples doing cancer research, by the way, are referencing this. And they're taking reads off the machines and mapping them to this file and saying, okay, I found a little piece that came from chromosome two, 200,000 bases in whatever, and it's just doing this whole time. And my task was to prepare the next version of this reference genome to deploy out there. And I was started by, you know, I'll, I'll just, I'll just go to the, to the fun detail here because maybe one or two people out there will be listening in the Unix command line. So I wanted to like see what were the names of the chromosomes in this file. So I catted the file and g gripped the output for the lines that contain these chromosome names. And I made one critical error. I left off a backslash character, which escapes things. Now every line, every line with a chromosome name in it is led by a greater than sign. But on the Unix command line, this will redirect the output of a command into a file.

Rob Long [01:00:41] And so what I ended up actually doing was grepping the names of these chromosomes and, and overriding this, you know, multi gigabyte file with like, you know, 23 lines of the names of the chromosomes. And so everything in production starts failing. All these jobs that are doing this research are just start just falling off of line. I'm like, oh God, oh God, what did I do? And it, it, you know, it took me just a second to figure out what was going on. 'cause I could see it. And, you know, it was, it was something that could be fixed, but it was not a pleasant experience as you might imagine.

Vance Crowe [01:01:20] What was the lesson outta that?

Rob Long [01:01:23] The lesson was to, you know, measure twice, cut once. Yeah.

Vance Crowe [01:01:26] You know, you

Rob Long [01:01:27] Just gotta look at it and stop before you hit return when you're, when you're playing with a loaded gun, which is the Unix command line. And remember kids, Claude has that, that's the bash tool that it has. That's what that is. And it can do anything and make it look like it's doing anything else. It's amazing, the indirection, it can be super sneaky.

Vance Crowe [01:01:53] So that question I just asked you, this is like a question that I, I hued out of the earth from having experiences where you're talking to somebody and they're telling you about, you know, some experience they had and you can't touch it, right? Like, I, I couldn't talk to you about Unix. If I start telling you my code problems, it'd, it'd be like, you know, either, you know, some kind of measuring contest or, or me like, you know, one-upping you or so saying, you know, I don't like, but in the, in the case of if you talk to somebody and you're like, they're telling you about an experience you can't relate to, if you just ask them, what, what was the lesson in that? It does not matter the context. It could be a very serious dark situation that they were describing. And they will stop. And oftentimes for the first time in their life, think about what the lesson was out of that and, and, and be delighted to have gone down the dark path with you.

Rob Long [01:02:49] Yeah. Right on.

Vance Crowe [01:02:52] You have a lot of ways that you have, like, you have people that report to you. That's where you kind of, the way I understood it was that you asked people that were kind of nervous. What's something beautiful you've seen, you've also now talked about like what are some mistakes that you found? What are other questions you, you use or ways that you interact with people as a leader?

Rob Long [01:03:15] Oh, that's an interesting question. I don't, well, I'll say this. I run a standup meeting three days a week that as this, it's a ridiculously large group for a standup meeting typically. 'cause it's kind of a cross team collaboration. But I really try to keep a cadence going so that we don't let people kind of just blah, blah, blah about stuff. Otherwise you end up with like, you know, if I say, Hey, how's, you know, issue 1, 2, 3 going, friend, tell me. And then you'll get like, yeah, I was kind of thinking about this thing over here. I thought I would do something over there. And it's like, it's noticeable if you keep everybody in the pattern of just tell me what you did yesterday, what you're going to do today. And if you have any blockers, this is the classic how you do a standup meeting. Then you can hear when somebody is bullshitting you, right? It's like, and it, and, and it'll happen. And it's like, I think my disagree ability helps me in this case to kind of push people and say like, Hey, I'm not hearing what you did yesterday. Can you go back and cover that? As opposed to just like, oh, okay, and then move on. You know? 'cause then you just, then they learn that they can snow you and just bullshit and not do work. That's not exactly what you're,

Vance Crowe [01:04:47] I mean that, I think that's a really good one. Like, I think that a lot of times when people are reporting back to you, if they aren't giving you what you want, it's often because there's not been a clear articulation of like, how to give me what I want. And then once you have that established, if they're not giving you what you want now, you know, like, oh, it's 'cause you're trying to go around the difficult question. Yeah. And like, this is, this is valuable.

Rob Long [01:05:13] I don't know if it's because people are used to just doing what they want and presenting it as what the boss wants or the leader or whoever is accountable. But I don't think that the average person wants to just give you, just doesn't know what you want. I think it's, well at least in my experience, I make it clear what I want. And there's still a lot of people that don't give me that. You know,

Vance Crowe [01:05:38] I think that it has to do with the way people reference information when you ask for it. Like the way that they are experientially going into their proverbial file cabinets and like, how is it that they're pulling back out this information? Is it, you know, you and I have talked about this a lot of times. Is it top down, Hey, I'm, that's you, right? Like, I'm looking at it from what do I need? And then, or is it experiential first person? Like, okay, let me walk through yesterday I did this and then I did that and then I did this. And I think that when you have two different styles of the way that you reference information or what you want, that, that can often come into conflict. 'cause you don't want the story. Yeah. And I only want the story.

Rob Long [01:06:23] Yeah, that's right. I don't want to hear about your adventure. Right? Like, I just want to know what, what are we doing here? I don't know. I I'm less optimistic. I feel like most of the time it's people trying to avoid doing work. Like if I mumble through my, my update to you, then I can screw around and not do anything and just kind of like, oh, I'm working on this thing. Yeah.

Vance Crowe [01:06:47] But I don't think I, I guess I would disagree in like, I don't think it's that people don't want to do work. I think what I have observed is procrastination comes from not knowing what to do next. And so people will sit and spin and go look up sports scores and, you know, do anything they can because they don't know what to do next. And I have found that the way that I short circuit that is I'm always like, all right, what is the singular, smallest, tiny step I can take towards resolving this? And I can now short circuit my way outta procrastination almost any time. But I don't think most people can,

Rob Long [01:07:23] Yeah. I don't know what causes the work to not be done. Exactly. But I don't know, I, I'm less confident about people. I think there's just a, a much narrower distribution of people who are interested in being productive. And I think most people can figure things out if they want to, but they just don't want to.

Vance Crowe [01:07:46] Yeah. But I think that there's just, I think people don't have as much command over their like, what to do next system. I mean, don't get me wrong man. I've been frustrated by people that fuck around and tell you they're gonna do work and tell you they dish work.

Rob Long [01:08:02] I should specify, I'm putting this in the, the corporate environment. This isn't like a group of entrepreneurs trying to work out some kind of deal

Vance Crowe [01:08:11] Together business. No, I mean, I'm, I'm saying whether it's me running my businesses here or working in corporate America, like, I certainly get frustrated with, you know, dip shits that are just not getting stuff done. Right. No, no doubt. But one of the things that I have discovered is, I think it is not willfully not wanting to do work. I think it is like a program that just starts spinning them around in circles. 'cause they don't know what to do. And if they knew what to do, they would be a different person.

Rob Long [01:08:40] Yeah. I mean, I feel like if I were not in corporate America though, my solution to this would be, I don't care why you're not doing it. I want you away from me because you did not yourself figure out how to ask the right question. Like, there's a level of autonomy that I think is required for you to be productive, at least in my experience is almost all in software development. Right. And I know that it, it can be a weird different way of working than other jobs, other, other professions that people do. But that's kind of the, the area that I'm talking about. And I feel like if you are not productive, you're a dead weight loss. Like, we'll move faster just by letting you go

Vance Crowe [01:09:25] See that. But that's, that's where like my experience has shown me, like, you're probably right that you should cut people faster than you want to. Right. But that there is just such a high cost as an employer. It's, it's like you have to find these people that you're going to hire. Yeah. Then you've gotta bring them on and then you've got to like help them, you know, whatever. If they were leaving their last job, they're getting into this job now you've gotta give them context. You've gotta, so there is absolutely like a how much of this should I tolerate? Because the amount of work that I'm gonna have to do to get another person up to speed is not inconsequential. Is actually, particularly if it's just a

Rob Long [01:10:05] Man show show. This is not relevant though, right. If if you have somebody that's a negative, adding a negative progress to your work, just firing them and not replacing them means you are making more progress.

Vance Crowe [01:10:18] I mean, I guess it's just not that sim you know, like I, I just don't think a along a one single person's work output is not like, are they negative or are they positive? There is an overall score maybe. But if you have, they're supposed to be working on multiple things, right? They're delivering multiple things to you. And if they stop doing this thing that is time consuming, but very simple, but they're messing up all these other ones, then the time consuming. But very simple one becomes your task now. And it it, you could be in a spot where you're like, if I do that now, all these other goals get,

Rob Long [01:10:52] I think you're missing the point. I think what I'm saying here, and, and I'm not saying all areas are like this necessarily, but these tasks, so, you know, you know what a Kanban board is, right? So this is like Trello.

Vance Crowe [01:11:07] Oh yeah. Okay. Right?

Rob Long [01:11:08] Yeah.

Vance Crowe [01:11:09] Yeah.

Rob Long [01:11:09] And you have a column that's your backlog. These are issues that we've written up, we've added detail to, and we all agree this is the right approach for this. And they're ready to go. And generally engineers pick from the top of that list and work the next issue that's at the top of the list. 'cause they're sorted by priority. Okay, great. So if there's somebody who takes that issue and will no matter what it is, easy, hard, whatever the task is, will never ever complete it on their own. Like at some point in the weeks of them spinning in circles looking at that issue, they're gonna take your best engineer off to the side. Randy, I'm talking to you. And they're gonna, they're gonna take his time and say, Randy, how should I do this? And then he'll say, do it like this. And then they'll say, how do I do it like that though? And it's like, well do this, do that, that this is the thing that, that's just total waste. Like you can just take that person out of there and then when Randy has time, he'll do the issue.

Vance Crowe [01:12:14] Okay. I I can see that, that point entirely. And as I am hearing my own point of view and like looking back on my own experience, I would, I would've been better off to let go of the dead weight. Yeah. But it is fucking hard. I 'cause you're like, I'm going to have to endure serious pain Yeah. In order to make this happen.

Rob Long [01:12:37] Yeah. And especially if you're talking about a corporate environment and full-time employees, not contractors, like contractors are, the one thing you have going for you there is that you can just not renew the contract. Right. You don't have to go through a process of performance improvement plan and tracking this and doing that to, you know, it's ridiculous,

Vance Crowe [01:12:59] Honestly.

Rob Long [01:13:01] Anyway.

Vance Crowe [01:13:01] Yeah. I mean this is good. We were definitely, definitely on a spot where we get to, to demonstrate our shared disagreeableness. You know, it's worth saying for anybody that doesn't know Rob and I is that Rob is, on a scale of zero to a hundred of disagreeable is a zero. But what surprises people is I am a one on the disagreeable. Yeah. So it's not like, it's not like Rob bad disagreeable guy because I'm actually very disagreeable, but have covered over it with, I don't know, my, my weird love of language and I don't know.

Rob Long [01:13:34] And that's a percentile, right? So it's like there's, for vance's score of one, there's like one person in the, in the room of whatever, a thousand people, a hundred people not No, it's more, right. That would be more disagreeable than Vance. And that's, that's me. And then there's,

Vance Crowe [01:13:51] You

Rob Long [01:13:52] Can't go any further down the scale.

Vance Crowe [01:13:54] You are the Asim tote. Right? The the like where the, the distribution just goes on and on.

Rob Long [01:13:59] That's the limit goes to infinity.

Vance Crowe [01:14:01] I'm out there.

Rob Long [01:14:02] Yeah.

Vance Crowe [01:14:05] Well, what are you paying attention to these days?

Rob Long [01:14:09] Well, springtime has been just beautiful. I pay a lot of attention to it. And I've really like, like I said, I live in a, a rural area of Franklin County and just a random thing at at dusk, there's been this bird that I heard and I'm like, what in the heck is this thing? And I could hear it in my house, like the, the TV is on. My wife and I are sitting there and we're like, what is this? We thought it was a coyote or something at first and went outside and it's like, oh, that's a, that's a whipper wheel. And it was just sitting, you know, somewhere around the house. But it was so incredibly loud. It was wonderful. There's so many different birds that make calls and show up in the springtime that are just not here at any other time.

Vance Crowe [01:14:58] What time of day was this?

Rob Long [01:15:00] It's like a half an hour after sunset. Okay.

Vance Crowe [01:15:02] It's like,

Rob Long [01:15:02] Not entirely dark, but getting dark. Okay.

Vance Crowe [01:15:05] So like,

Rob Long [01:15:05] And it's just, you know, making this call.

Vance Crowe [01:15:08] I mean the, the birds in my area are, they're, they're caco, right? Like, they're so and beautiful, but like if I'm recording like a lot of my friends, I'll record a voice memo to send to them. And like, it's hard to sometimes hear me over them. And it was not like this in central Illinois, like where I grew up, it was not Missouri is, or at least where I live in Missouri, is like 10 x bird diversity in different calls.

Rob Long [01:15:36] That's interesting. I wonder why that is. If it's that we just have a lot more forests around, but that's allowing this or what? Or if it's the north south,

Vance Crowe [01:15:46] I've always felt that the, yeah. The north south of where we live, where, you know, where we are in St. Louis or this kind of where you're in Franklin, like it's not the north, but it's also not the south. And so you get this bigger diversity where I have all the trees that I had in central Illinois growing up, but then I also have a whole bunch of more diversity of things that were more southern that we didn't have at all. And that's gotta have an impact on the bird,

Rob Long [01:16:10] Bird

Vance Crowe [01:16:11] Population. Yeah.

Rob Long [01:16:12] I should add the, for the, the people interested in the birds. I actually was listening to the calls and I, a friend of mine helped me figure out that it wasn't actually a whipper will, but it was something called a Chuck Wills widow. And it's like both of these names are an Ada pia of their call. So like a whipper will is like whipper rip, you know, it makes this weird sound that somebody somewhere heard as whip poor will. And then the Chuck Wills widow, it has this like, so it's like almost the same, but it starts with a different word, a different sound that comes outta the bird. And apparently their range is like a hundred miles south of where I'm at. So this is a new, or I never heard 'em before, but they're not, they're not known to be in this area. Climate

Vance Crowe [01:17:04] Change, man.

Rob Long [01:17:04] Yeah.

Vance Crowe [01:17:07] I, my first instinct would be to pull out my phone and do one of those bird ID apps. Do you use those?

Rob Long [01:17:15] No, I just looked up a whipper will call on YouTube. So I mean, we're still ending up in the same black hole there, but, you know, and then, yeah, I shared it with a friend of mine who's, who's been out to my place and appreciates birds. And he was like, oh no, I used to hear that on our farm down in Van Buren, Missouri. And that's a, that's a Chuck Wills widow.

Vance Crowe [01:17:38] Wow. So

Rob Long [01:17:39] Like, and I went and listened to that one. Like absolutely it is

Vance Crowe [01:17:42] That ai, I have one like the Merlin app is what it's called. And you can just set it out and it will, you can see the waveform and then whenever it hears a bird, it will pull up a photo of that bird.

Rob Long [01:17:55] Oh, neat.

Vance Crowe [01:17:56] And then it will like, so if it's like, there's like three different birds, it's just like popping up whichever one is going. And it is so incredibly helpful. You know, we were talking before about the markers, right? Like to have an image and hear the call at the same time. Because, you know, sometimes when you're trying to, you're hearing birds, you're like, I don't, where is it? What is it? You know? But when you have the image and that bird that maps to my brain so much tighter, and then amount of calls that I can now identify has gone way up because of that.

Rob Long [01:18:26] That's pretty cool. I've never seen that. I'll have to give that a try.

Vance Crowe [01:18:29] Yeah. It's, it is as good as the picture. This app, the one that does the plants, which is another one that has like, it, that exploded my knowledge of trees.

Rob Long [01:18:40] Absolutely. No, I use, I still use that.

Vance Crowe [01:18:43] So the big talk in all of social media right now is ticks are everywhere. Yeah. They're, and they're, they're getting worse. What's been your experience with the ticks?

Rob Long [01:18:52] Yeah, I don't like 'em. They're, I feel like I live at ground zero for these things. I'm sure that's not true, but there's just a lot of them in the woods. Last year there were not a lot. There was, they were very light. And I don't know if there's like a, you know, oscillating pattern to this or what exactly, you know, if it's like wetter than the next year, there's gonna be a bunch of them or how it works. But a, a friend of ours, David Aki was telling me about a thing that exists where it's a, it's permethrin that's in like cotton and they put it in like a cardboard tube, like a toilet paper roll and just set it outside and mice will come and take the cotton and bring it to their nests. And that's where ticks are most frequently breeding and increasing their numbers. And so over the course of time, like a season, you know, it doesn't work immediately, but this will apparently significantly reduce the, the tick population,

Vance Crowe [01:19:59] Man I'm super interested in, in anything that can reduce them. Because we were out at a lake house and my daughter and, and my wife and a friend took the girls all on a walk and they came back. And I have never seen that many ticks in my life. It was like, unbelievable. I think on my one daughter there was 20 ticks. Oh

Rob Long [01:20:21] My god. Like this. It

Vance Crowe [01:20:22] Was everywhere.

Rob Long [01:20:22] Seen ticks

Vance Crowe [01:20:23] Inside their arms. It was multiple kinds of ticks. It was like,

Rob Long [01:20:27] I mean like the little bitty the nims.

Vance Crowe [01:20:28] Yeah, yeah, yeah, yeah. Where you're doing like the, you know, you're searching through hair and you're like, every single freckle could, could possibly be a, it was wild, man.

Rob Long [01:20:36] There's a, and like, I don't know, probably part of my disagreeableness comes along with a strong disgust reaction. And like the, the, once you find one, I feel like six, 12 hours after that, any little, you know, thing that you feel, you're like, yeah, what is that? That's another tick.

Vance Crowe [01:20:56] Get

Rob Long [01:20:56] Get outta here.

Vance Crowe [01:20:57] That's gotta be, I wonder, I wonder what that neurological thing is. 'cause I know like, I have the same thing if I've like observed poison ivy Yeah. And I like walked through something and then I'm like, I don't know if I touched it or not. And then you go and watch it and I'm like crazy absurdly allergic to poison ivy.

Rob Long [01:21:13] Yeah. Yeah.

Vance Crowe [01:21:14] And then if I start to feel, you know, scratch on my elbow and I'm like, oh no, what is that? Oh. Oh, what is that?

Rob Long [01:21:21] Yeah.

Vance Crowe [01:21:24] Started reading Arabian Nights the other day. Oh,

Rob Long [01:21:28] That's interesting.

Vance Crowe [01:21:28] As a recommendation from a friend and have you read it? No. As an adult. Okay. So no, I'm

Rob Long [01:21:35] Aware of

Vance Crowe [01:21:35] It, but remember I would highly recommend it was not appropriate for my fifth 5-year-old.

Rob Long [01:21:42] Yeah. That's, I I can imagine. I bet there's concubines and beheadings and Oh

Vance Crowe [01:21:46] Yeah. All kinds of good stuff. So like the, I'm, I'm, I'm, I'm five pages in maybe not even five, three pages in. It's already getting good. You know, they're one sultan talking to another, or one king talking to another. And then the one king is coming to the, the palace of his brother. And he forgets something and he goes back and his wife is laying with another man. And then you're like, oh, okay, let me, we go see how I can shape and censor this. And then I'm like, all right, well that, we're just gonna move past that. Right. She, my daughter doesn't really understand, I can skip past that. You go a few pages later and he observes that his brother, who is also the king, his wife is now in the garden, you know, cheating on him with another man. And like, you're just like, what is going on here? So I have decided that I will be first reading the book and probably not for the girls. It's too bad. 'cause I'd really hyped it. I'd been like,

Rob Long [01:22:43] Yeah,

Vance Crowe [01:22:44] Arabian Nights, you know, just like Aladdin. And it was, I I, I'm sure there are stories in there that do not have quite that much the violence Violet can handle that. But the, the, is

Rob Long [01:22:56] That where

Vance Crowe [01:22:56] The cut holding doesn't

Rob Long [01:22:57] Work? Oh God. Is that where Alibaba and the 40 Thieves comes from? Is that one of the stories?

Vance Crowe [01:23:03] I think so, yeah. Yeah. So like the, the, the thrust of the Arabian Nights is there was a king who, he was unhappy with a wife. Somebody did something bad to him, so he killed her and then he started having a new wife and every morning he would cut off her head. So he was just like going through wife, after wife after wife. And there was one that realized, oh, I've gotta do something different. So she started to tell him a story and the story was so good that he was like, tell me another one. And she's like, oh, I can't, I'm too tired. But I'll tell you another one tomorrow night. Okay. And so then he lets her live and as long as she keeps telling him stories, then this

Rob Long [01:23:39] Is how television works. Right. They put a hook at the end of the show. Yeah. And it's a cliffhanger. And it's like, well, you gotta find out next week.

Vance Crowe [01:23:47] So I, I went in knowing that that was the, the like, overall. But I think that I had asked him, is it a children's story? And I think I walked past his advice. So I think this is all my fault

Rob Long [01:23:59] To show him the, the Popeye episode of Alibaba and the 40 Thieves. That'll, that'll, that's good enough. That's high enough culture.

Vance Crowe [01:24:06] What books are you reading these days?

Rob Long [01:24:10] A lot fewer lately, actually. Not, not by conscious decision, but just kind of AI monetizes every moment of my time. Like I am able to push further and do more work on computer related projects, which is the bulk of what I work on. So it just, and it feels like an insanely fertile and productive and chaotic time in the field of, you know, software engineering and just computers generally. And I feel like I've just invested a lot of my attention into how these things work. And yeah, I mean, I probably should swing back to more frequent reading. I have Treasure Island sitting on my, on my stack among other things. But before that, a Michael Levin paper

Vance Crowe [01:25:03] Okay.

Rob Long [01:25:04] That I wanna read. So

Vance Crowe [01:25:05] Our, our shared friend Kate Crosby, has a theory that AI is going to make future generations illiterate. That they will not be using written language. They won't read nearly as much. It strikes me as like, it seems like I use the computer in reading a lot. Yeah.

Rob Long [01:25:24] But

Vance Crowe [01:25:25] I, I could imagine a world in which our children are the last generation where most of the kids no longer have the same reason to read. It's not as quick, it's not as robust. I'm not saying it's a good thing. I'm just saying I, I could imagine a world in which that occurs.

Rob Long [01:25:41] Yeah. I don't think it's impossible. I don't know how likely it is. I mean, I think a lot of predictions like that end up are like useful ways of framing things and thinking about things. If not a literal, like no, like no one will, will lose the ability to read or something. Or like, maybe the importance of it in culture will go down. But I, it's hard to imagine that will in, you know, become mass illiterate. You know, I feel like there's, like, to interact with the ai, yes, you can have it talk to you, right? It is possible to do, but it's also annoying and terrible. Like, I, I, I much prefer to read if I have the, like, if I'm not driving, you know what I mean? Like, if I'm, I would like to sit at a desk with a very large monitor and just read like, this is how I learn things. People talking to me are just giving me the answer is not helpful to me.

Vance Crowe [01:26:39] Yeah. I mean, I, I mean, I believe literacy has an important part in the role of human beings, but it's, it's more conceivable to me after having used AI and having the same experience you are where like, the only reason I'm reading books right now is 'cause of my children, right? Like, if, if not for my children, I would be, you know, 'cause as soon as I put 'em to bed, I'm like, all right, let's get back to work. Let's get to do these things. But like, mostly with ai, but like, I think, at least in my own experience, I remember there was a time when, you know, I was completely allergic to owning books. I was like, why would I want these analog things? Look how much space they have. I can't search 'em. I have to carry 'em around. And now I'm like, you fucking idiot. Like those, how many books did you get rid of that were, you know, your favorites? You're gonna have to go rebuy.

Rob Long [01:27:28] Yeah. I love the idea of them being inert, that there's no sneaky editing happening. Like if you have the, you know, Amazon or, or you

Vance Crowe [01:27:38] Know, the Kindle reader or whatever.

Rob Long [01:27:39] Yeah. Or Audible or something. It's like, oh, if a certain turn of phrase or way of thinking about things gets out of style, they can just edit it away from you and you'll not know what it was like, you know? But

Vance Crowe [01:27:54] Yeah.

Rob Long [01:27:56] Well,

Vance Crowe [01:27:57] I've really enjoyed this conversation. Thank you for coming by.

Rob Long [01:28:03] Yeah.

Vance Crowe [01:28:03] Any predictions about the future? Come back in a year? What do you think the world will look like in a year?

Rob Long [01:28:10] Boy, that's a really good question. Probably there'll be, you know, the dollar will buy less stuff per unit.

Vance Crowe [01:28:18] Oh man. The purchasing power goes down. You talk about any of this. Yeah,

Rob Long [01:28:22] I mean I don't have a whole lot to say that we haven't said before, but it's kind of the iron rule of fiat currency, right? Like it's generally trending downwards in purchasing power. 'cause you can just make more of it. So that's true. And we're spending a lot, a lot, a lot of money. So as a government, you know, the United States, so that's un you know, there's not, we're not collecting more money, we're just spending a lot more. So that's happening technology wise. I don't know. That's the really interesting thing. I feel like the, the way that we work has so changed and is in the process of changing right now. That it's almost impossible to say what it will look like six months from now of like what will an a software engineer, how will they do their job? I guess Just that there's so much going on there. Yeah.

Vance Crowe [01:29:18] Joscha Bach just on Monday said, when I asked him, you know, what, what will AI be like in a year? And he's like, who, who knows? How could you possibly know?

Rob Long [01:29:25] Right? Right. And then you get to, you know, it sounds like a science fiction concept. And it used to be recursive self-improvement, right? Where the AI is helping to improve the ai. I I don't know why you would say we're not there, right? Literally right now where you have, you know, anthropic, at least the people that write Claude Code are using Claude Code to write Claude Code, right? And they're not really writing code anymore. The model is writing the code and they're a layer above that writing a specification for what the code is. So the, the software engineer in that, you know, if everything were like that, which is not yet, but that's kind of the trend. The software engineer becomes a software engineering lead where the agents are like the engineers and the engineer is now saying like, here are the features I want to see. Here's what, here's what the definition of done is. Here's the kind of, you know, in variance we need to maintain throughout the whole process. All of these sort of things that a lead would typically do. That's what the engineer will do in order to cause the agents to produce code. So that's a wild, you know, that, you know, if you're not writing software, this is probably just sounds like, you know,

Vance Crowe [01:30:44] I I,

Rob Long [01:30:44] Blah blah, blah, but

Vance Crowe [01:30:45] The, the quality of the ai, so I built a bunch of stuff from January till, I don't know, late mid-April, right? And then I was doing it through Open Claw and then I started using More Claw just within its own terminal.

Rob Long [01:31:02] Yep.

Vance Crowe [01:31:02] And I started having it evaluate the things that it was doing. And so it already had a description of what I wanted and like it could go in and test these things. The improvement that has happened in two weeks from me being like, let's go look at this code that I wrote. And it's like, oh, this is causing this problem and this is causing, and how much better the whole thing is. It's like, it's so mind boggling that the computer is, the AI is fixing code that AI wrote and now it's more resilient and it's better and it picks up like, oh, we should also fix this problem that you don't see here, but it will become a problem later. And you're like, how in the world does it spot those things?

Rob Long [01:31:41] Yeah. I mean that's the, the power of large data sets, right? Like if you mentioned GitHub before, this is a huge treasure trove for these model companies to say like, and especially internally GitHub and why things like Cursor, which is like a, you know, know a, a software development tool specifically that wires in a bunch of AI directly to it. They have such a good feedback loop because they're seeing people call to the model and say, here's what I'm trying to do. And then the model does something and then the people say, yes, I like it. No I don't like it. Or Here's what's wrong with it. So they're getting this closed loop of training data that they can continuously improve on. And that's kind of the name of the game here is like, how do you get these sources of data? So the software itself, the value of that is trending to zero. 'cause you can just will it into existence. You just say what you want and you can get it. That's a huge over exaggeration. I realize that. But so the, the only things of value are the, the hardware and the model to run it on, to run on that hardware and sources of interesting data. And that's about it.

Vance Crowe [01:32:55] So this kind of makes me reflect back to the conversation we were having about literacy, right? Like if you are an, an AI agent lead, right, software lead, and you're not writing software and over and over and over again the best hires are to, to have a person that's the lead as opposed to the software writer. Will people in two generations still be writing code? Will will that be an art that will get lost?

Rob Long [01:33:23] I don't know. It's possible. And the example that everyone will bring up, myself included, is what's called assembly code. So when we first invented computers in the late forties, early fifties, you interfaced with the computer in binary. So ones and zeros flipping literal switches and setting them and then saying go. And then we got to the point where you could actually use a teletype machine to enter characters into it and things like this. But the programming was very, very low level. Like take something from this register and add it to something in that register and then that's it. And that's all you get to do. And you have to figure out how to, and like it's true that you can make, you know, Microsoft Windows out of all of that. And, but the amount of complexity and, and level of detail, one thing I would compare it to is the, like what's the right level of of understanding of something that helps you grasp it better? So physicalist or materialists will say, okay, it's all just corks at the bottom of it, right? There's just some kind of elemental particles and physics describes everything useful about the universe. And it's like maybe technically true, maybe, I don't know if it is or not, but let's assume it is to go from that to me predicting where will an atom that's part of Vance Crow be 12 hours from now, right?

Rob Long [01:34:56] Like, okay, well one answer is like we do hundreds of trillions of calculations and simulations and calculate trajectories and do all of this stuff. Or I can recognize that Vance Crow is human being and 12 hours from now he's gonna be in his bed sleeping. That's where that atom's gonna be. So that's like a much more useful level of abstraction than it's all quirks. So the same is true with assembly code. So it's like the super low level stuff. And right now when you write computer code, you are not writing assembly, you're writing a higher level language like C or Java or go or Rust or something like this. And that is then compiled down into assembly code. So for a period of time everybody was writing assembly code and now virtually no one is, there are a few things here or there where somebody's writing a video driver or something that interfaces with the, the hardware in a, in a way that has never been done before or needs to be absolutely perfectly as fast as it can possibly be. But they're very rare and I think that's likely to be the case. We'll just, we just added another layer on top and instead of like another more refined computer programming language, we have the English language, which just describes the thing that we want. And now we have a thing that compiles that English into software or whatever we compile it into in the future.

Rob Long [01:36:31] You know,

Vance Crowe [01:36:33] That is a fascinating explanation of, of how that could play out. That seems very conceivable to me. And I imagine that's probably true of a lot of things, you know, like it could be in the physical world too, right? This is, we used to make cars doing this thing and then, then we started explaining to the machine what we wanted the the thing to do and it builds the car for you.

Rob Long [01:36:57] Yeah, yeah. There's an interesting concept that I, I find myself using it a lot when dealing with like, support issues at, at work and that's the XY problem. And you end up having somebody showing up saying, I can't, you know, I can't do x, I'm not able to do X. And usually it sounds very strange to me as the person who's trying to help them use the system that I had built. And the the problem is they don't know what the right thing to do is. Ultimately they know what they're trying to do and they have a wrong idea about how to get there. So if I say like, you know, I'm not able to, I'm, I'm failing to come up with a good example here. I'm sorry. But essentially it's like I, I don't even know how to ask you how to help me. I'm just asking you this strange other question. Like I can't, I can't hammer this nail with my telephone. My telephone is I need a new telephone. I need a new telephone. That's the ex, right? And you're like, okay, but why do you need a new telephone? It's like, well it doesn't hammer these nails, right? And you're like, well that's obviously stupid. You don't know what you're doing. But with these complex things, like nobody knows all of the details and they shouldn't need to know all of those details. So it's like the skill of hearing somebody say, I'm, I have this weird problem. And they're like, what, what even are you trying to do? Why are you, why do you have that problem? What is your goal? And like push back the like weird idea about, you know, hammering the nail with their telephone.

Rob Long [01:38:33] They're

Vance Crowe [01:38:33] Just dead example. They're dead fire 'em.

Rob Long [01:38:35] Yeah, there you go.

Vance Crowe [01:38:37] No, but I take your point. I mean like that's something I, I actually believe I'm like kind of primed to do is like, when we run into a conflict, it's always like, what is the, what are we trying to do here? Right? Like if you start asking me

Rob Long [01:38:50] Questions, this is a good expression of interest based negotiations. Yeah,

Vance Crowe [01:38:52] That's good. Yeah, I mean, and you know, like with my wife and I, when we start getting into conflict, it's like, alright, okay, hold on. What is it that we are trying to achieve here? And like the number of conflicts that that reduces goes way up because like there's a lot of times when you're like mad about this, you know, or getting emotional about whatever the specific friction is that's going on right here. And then once we get aligned on like, what are we trying to do here? Then it's like, oh, oh, okay, I see Now I'm, I

Rob Long [01:39:22] Could, let

Vance Crowe [01:39:22] Me go get the hammer. Yeah,

Rob Long [01:39:23] Yeah, there

Vance Crowe [01:39:24] You go. I think that driverless cars are gonna really, really change culture, whether that's in a year or not. I, I believe that this is going to be a radical change to society. I think like all of our society, modern suburban city living is based on having enough room for everybody to have these, these vehicles be personally owned and driven around and parked. And, and I think that when you're able to hop into a pod in an urban environment and be able to ride an exercise bike or do homework with your children or have your children delivered to you from some place to another or have your groceries delivered, I think the, I don't wanna call 'em productivity gains 'cause it's not directly that, but it's like the, the, the gain on human attention is going to just completely be altered from what it is for what, how we spend a significant percentage of our time now.

Rob Long [01:40:32] Yeah, it's interesting. Like I remember having these conversations, I don't know, 15 years ago at least, something like that and thinking that it was, you know, a year or two away that we would, we would get there but we just never did. Right. The horizon kept going further and further away. But yeah, I remember all of that stuff of like, well we can eliminate massive parking garages and like all the space taken up by parking lots.

Vance Crowe [01:41:00] Parking lots, yeah.

Rob Long [01:41:01] Just, you know, ugly and hard, not hard to maintain but like takes labor,

Vance Crowe [01:41:05] They're, yeah, it's expensive,

Rob Long [01:41:06] You know? Yeah.

Vance Crowe [01:41:08] Have you been in a, in a self-driving car?

Rob Long [01:41:11] No, but I did see the ones that Amazon rolled out. I was in Las Vegas like a month ago for the Google Cloud platform, you know, show that they put on out there and that weekend they had a bunch of these and they looked zr like they look like not a car, like the, there's just people, sorry, there's just people inside of there and it looks like a pod and both the front, the front wheels and the back wheels can turn and they were just shuttling people around to the conference from various hotels and like, they were, apparently they were free. We didn't try to get in any of 'em 'cause we had stuff to do. But it would've been an interesting thing to do.

Vance Crowe [01:41:57] I've been in Teslas, Wes Kearney and Zach Smith and done both of the self-driving and like, it was awesome. It was like amazing. I, and I liked the experience of driving a car, but I sense that that will be like riding horses or something. Like it's gonna be so something you do for fun. Yeah, something that like if you can afford the insurance on a human driving it versus the, the self-driving

Rob Long [01:42:21] And there'll be like a narrow sliver of people who actually still need to do it. Like people who need to write assembly code, they'll be like cowboys and, you know, I don't know, forklift drivers or something, you know.

Vance Crowe [01:42:34] But just so wild, like to me, I, my, my brain crackles with electricity when I think about the things I could do if not driving, like exercising in your vehicle, the spending time with your children, I'm sure it would actually mostly be spent just on your phone, but

Rob Long [01:42:51] Yes. Screwing around just more downtime.

Vance Crowe [01:42:55] No,

Rob Long [01:42:56] I, it's, it's promising. It's interesting.

Vance Crowe [01:42:58] Any thoughts on Bitcoin since the last time we talked?

Rob Long [01:43:02] I mean, you know, keep, keep buying Bitcoin, that's it. That's, that's the whole thing. Just accrue what you can and figure out a way to get paid in Bitcoin. That's, I don't know if, I think we've probably hit that point, but, you know, I've got some eggs in my car that I'm gonna get, you know, our good friend Benjamin will give me Bitcoin in exchange for and I'm happy to, to do that on a small level, you know,

Vance Crowe [01:43:30] A few dozen. Yeah, I mean I've I've done it with Legacy Interviews and feels so good. Yeah, they just, they just get parked in a, in a little account and well

Rob Long [01:43:37] That's one thing I've like, I remember coming to it and looking at it and seeing like, okay, well Bitcoin's, you know, purchasing power relative to other things is growing at this rate. How can I beat that other than just putting money into Bitcoin? And the answer is, whatever you're doing, you need to be standing on top of Bitcoin. Like if you're your, your value add, you get paid in Bitcoin, you're inherently beating 'cause you're already getting Bitcoin for it. And then whatever your value add is for it, your margin now you're making on top of that. So to me that's the, that's kind of the last word for as far as I'm concerned is figuring out how to get paid in Bitcoin.

Vance Crowe [01:44:16] What do you think about this comment that people have made about, hey, if we don't get people to start spending Bitcoin, like then it's not gonna succeed. Jack Dorsey's famous saying this, I

Rob Long [01:44:27] Don't know. I I think it's absurd to, to do the, like splitting hairs and say that is this a, you know, a mortal or venial sin or whatever, you know, like this kind

Vance Crowe [01:44:37] Of

Rob Long [01:44:38] Ontological, you know, categorization of things. Is it really a currency or whatever? I think that's absurd and I think I spend Bitcoin all the time and I think, I think if there's a a use to spending Bitcoin, people will do it. Otherwise, there's nothing wrong with just stacking and not spending, that's fine too. You don't have to be a martyr to go out there and spend the Bitcoin. But it's also possible to do things like using the strike app to spend from cash that you keep there into Bitcoin so that you will just say, you'll just have cash in an account at Strike and say, alright, I'm gonna buy something now, but I want to buy it in Bitcoin. Then they will take on your behalf that cash that you gave to them. They will buy the Bitcoin and send the Bitcoin to somebody. So you are able to not have to think about capital gains taxes 'cause you're just paying cash, but you're pumping energy into the Bitcoin economy by bidding up Bitcoin essentially.

Vance Crowe [01:45:43] Yeah, yeah. The, it's, it's all great. Like two big pieces of news that I did not have a chance to digest, like happened right before this podcast started. One Apple has now said that apps can allow Bitcoin to be used within their app ecosystem.

Rob Long [01:46:01] I saw that and I was deeply confused by that too. Like I initially was like, oh this is them wanting to get their 30% cut. But then I think Kate actually raised the point that, yeah, what strike is already on there? So how's that work? Like that's Bitcoin.

Vance Crowe [01:46:19] Yeah, I, I don't know. I mean I didn't look into it very deeply. I just had seen it and thought it was interesting and then the other one was the announcement of of, you know, they've been talking all about this US Reserve. I have no idea what any of that means. I didn't look into it at all, but

Rob Long [01:46:33] Yeah, I'm not worried about it. Yeah,

Vance Crowe [01:46:34] The at the end

Rob Long [01:46:35] They either will or they won't and fine. You know,

Vance Crowe [01:46:38] I kind of feel that way too. Alright man, this has been fantastic. Thank you so much for, for stopping by. Yeah,

Rob Long [01:46:45] Thanks for having me.

Ad segment / testimonials (multiple voices) [01:46:46] Alright, that's gonna do it for the podcast. Thank you for tuning in right now. I am just opened up the class for Interest-Based Communication. This is where I teach people how to talk about their ideas so that other people wanna listen so you can share, understand, get through conflict, and have better interactions with people. If you're interested in learning more, you can watch this video or go to Articulate Ventures slash IBC for interest-based communicating. We'll be back whenever I find another interesting person to sit down with. My name is Heath Hunter. I'm a managing director at Broadview Group in our business, your success is dependent on how you're able to effectively tell your story to relate to your customer or your potential partners, and we needed a course that could train our staff on how to do that. I was looking for something much less than what Vance delivered. I literally just wanted him to come in and help us a little bit with storytelling and he came back with this curriculum, which was awesome. How to craft a memorable introduction, build a story around our firm and how we personally weave into that story. How do we continue the conversation through asking good questions, how to negotiate in a collaborative way versus a competitive way so that we maintain that relationship that we've worked so hard to build. Our relationships are everything. To me, interspace communication is really understanding the perspectives and the point of view of others being able to meet them where they are.

Ad segment / testimonials (multiple voices) [01:48:23] Interest-Based Communication to me is creating a dialogue with someone else sitting across the table from you. It allows each party to really describe what they care about and work together to accomplish that for each party. A lot of people, myself included before I took Vince's class, really went in with higher weight on what was important to me. Everybody has a tendency to just focus in on the price, but there are hundreds of other components that come into something as complex as a transaction for a company. What are they actually interested in? What do they hope to achieve and what does success look like for them walking out of a negotiation? Those could differ widely from how they're showing up. If we can somehow get on the same page, you can create these collaborative solutions that are a win for everyone, we've definitely woven that into the process. We're trying to ask those deep probing questions on what are their motivations, what do they care about? Really understanding who they are, introducing yourself to better communicate who you are and your background. What makes a good introduction? This is something that I, I really appreciated and I've been in the workforce for, you know, 27 plus years and done lots of introductions. I've heard lots of introductions, making a good introduction. I think everyone just assumes that you should just kind of start talking about yourself and go into your title and what you do and frankly, that's typically not very memorable or interesting. Going through this course and learning about storytelling and the arc of a story and how to embed that in your introduction was honestly like nothing I had ever heard before.

Ad segment / testimonials (multiple voices) [01:50:02] And it makes a lot of sense, right? Because if you're a good storyteller, you tend to sort of capture an audience and you want to be able to do that with your introduction, sort of that hook that you need in in a story. For me, the big thing was having faith that sitting back and really paying attention to the conversation will help draw out a connection before I would spend my time trying to, okay, what do I have in common with this person? Oh yeah, this is a chance to interject and and establish that link. Ironically, just kind of stepping back and saying, I'm just gonna really pay attention to what's in front of me. It really has impacted the conversation because you can get deeper faster. I used to think it was all about how interesting I could come across, but what I've learned and honestly what Vance taught us, the most interesting folks that you run across, ask really good questions. That's been a game changer for me. Instead of trying to focus on myself, focusing fully on the other person and really just listening for like, what are the details that they're giving me? Because there's a reason that they mention that. We assume that folks can kind of see what we see in our heads. And what this course taught us about diagramming your ideas is to put it on paper, put it out there so the other parties can see it. We were meeting with a group that weak, and we were trying to communicate different ideas and outline a vision for this potential business. Instead of doing it in a PowerPoint format, we did it on a whiteboard. It was transformational. The CEO, he, he just lit up. We actually handed the pen over to him and we said, Hey, show us where we've got it wrong, or show us where you have additional ideas, the level of engagement, you know, went through the roof and he got up and you could see he was almost giddy.

Ad segment / testimonials (multiple voices) [01:51:42] Having that limiter of the speed of your hand, you're bringing them along in, you know, your idea you're trying to convey. So the practicality of the course, the mirroring, the diagramming, the exercises. I think that's one of the things that I appreciated most about the course. Super candidly, I've been through a lot of these in my career, and you walk out of there and you're like, well, that was a waste of time. What I can tell you about what Vance brought to the table. I, at no point ever felt that way. I was able to walk out of every single session and put to work immediately what we had practiced that day. He has a great way of connecting concepts so that you have this, you know, holistic framework, so to speak, to connect with people, which is something that we do all the time, but deliberate practice of it really helped up my game. I would recommend that everyone take this course, and I would venture to say that this has been more impactful for me at home than at work. Pretty much everybody is having conversations and negotiations at home. I feel super lucky and blessed that we were able to have this course at work and that it sort of bleeds into home life. I think that's kind of the best of both worlds. Any corporation or any firm, no matter what the size I think would benefit from something like this, who I'd recommend this to, is a small group of professionals who are really dedicated to self-improvement. Interest-Based Communication, to me, is a holistic way to build relationships. Can benefit anyone.