Why Digital Detoxes Don't Work (And What Actually Does)

So, here's a mystery that has long interested me. Uh, back when I was working on my 2019 book, Digital Minimalism, I ran an experiment where I recruited around 1,600 people and I had them uh do what I called a digital declutter. So the idea was they would spend 30 days abstaining from the use of what I called optional digital tools and then at the end of the 30 days they would reflect on which tools they really missed and which ones they actually wanted to add back into their lives. I was inspired by Mary Condo when I did this. You know she said if you want to clean out a closet take everything out and then only add back in the stuff that you really need. And I figured we should try to do this with our digital lives as well. So this was a fun experiment. It even ended up being reported on in the New York Times. But here's the mysterious thing. There was a real division in the outcomes of the people who participated. Some people had great success in moderating and controlling their digital behavior going forward, while other people really failed and fell back almost immediately into their old habits. So, what was the difference between these two groups?

Well, here's one of the big things that caught my attention. The group that failed tended to treat the experiment like a digital detox. They hoped that a sufficiently long break from their devices would reduce the allure of devices, but this largely didn't happen. The group that succeeded, by contrast, tended to fill the declutter period with lots of activity and action and experimentation. Right? We're talking about like working on new hobbies or being much more aggressive about socializing or reading or walking or exercising more, doing more self-reflection, journaling, all these type of things. So, they treated this experiment like it was a chance for them to do some analog reprogramming. That's what I called it at the time. So, all right. So, this is the mystery then. Why did this analog reprogramming approach work so much better than simply trying to do a digital detox?

Well, it's Monday, which means it's time for an advice episode of this show, which is the perfect opportunity to look closer at this question. Now, there's a specific reason why I'm tackling this issue right now is because at the moment, as you might see if you're watching, I'm not in the studio, but I'm on vacation. I'm up in the uh the upper valley of Vermont and New Hampshire. And while I've been up here on vacation, I've been reading this book. This is Max Bennett's book, A Brief History of Intelligence, which is a a really fascinating history of the evolution of the brain from the very first multisellular life through modern humans. And when I was reading this book, which gets really in the weeds of neuroscience, I came across an answer to our mystery.

All right, so here then is our plan. I'm going to use the neuroscience that I learned from Bennett's book to answer three relevant sub questions. Number one, what's going on in our brain that makes our phone so appealing? Number two, why did digital detoxes not really work so well and making the phone less appealing? And three, why is analog reprogramming like I saw among those people who succeeded in my experiment something that works much better with our brain wiring? We'll then use this wisdom to isolate some concrete advice that you can use to hack your own brain to spend less time on your phone and more time doing things that matter. All right, we have a lot of sort of brain science geeking out to do here, so let's get started. As always, I'm Cal Newport and this is Deep Questions, the show for people seeking depth in a distracted world.

All right, so let's start with our first sub question here. From a neuroscience perspective, why are you addicted to looking at your phone? All right. So, here's what I learned from Max Bennett's book. I've talked about this before at a sort of higher level, but I'm really honing in now on what is actually going on in the brain. I think it's important to get precise so that our advice can get precise. So, if we really want to know what in your brain is responsible for you picking up that phone more than you want to, it is a truly ancient neural structure called the basil ganglia. When I say ancient, I really do mean ancient. The circuitry of your basil ganglia is basically the same as the basil ganglia that you will find in a lampay fish. Even though our last shared ancestors with the lamp rays are the original vertebraes from 500 million years ago. That's how old this particular part of our brain actually is.

Now what does it do? Well, in his book, Max Bennett calls it the puppeteer of the animal, right? Right. So the basil ganglia uh it takes in input from all sorts of different parts of your brain so it can monitor your actions in the external environment and then its output is connected to the motor circuits in your brain stem. And so most of these motor circuits are inhibited all the time. The basil ganglia can turn off that gate on particular circuits and actually cause you to do specific actual physical actions. Right? So it's like it's the puppeteer that controls your physical actions based on the input that it's getting. Now what's critical is that once you get past the very simplest animals, what what makes basil ganglia so important is that they're connected to dopamine neurons that generate dopamine when exposed to things that generate a reward. And rewards are typically we have these other ancient structures like the hypothalamus that recognize if something is rewarding or not. um uh the dopamine neurons will also withhold dopamine if the activity is non-rewarding or harmful. So the basil ganglia is actually um one of the main things it wants to do is repeat actions that maximize dopamine release. So if it has sort of learned that a particular physical action creates dopamine release, it will be highly uh motivated if you can if I can sort of anthropomorphize the brain which is sort of met up um to repeat that action.

All right, here I'm going to read a quote here. Here's how Bennett actually describes this decision-making process happening within the basil ganglia. So, as Bennett writes, the basil ganglia accumulates votes for competing choices with different populations of neurons representing each competing action, ramping up in excitement until it passes a choice threshold at which point an action is selected. All right? So we have the puppeteer, the basil ganglia that ultimately decides the actions we take and it learns through exposure to past rewards that certain uh actions if it's going to generate a reward, it's going to be much more likely to actually take that action. There's actually even a competition happening within the basil ganglia of potential next actions where whichever relevant dopamine neurons get more excited, then that's where it's going to go. So it wants to go with what it thinks is going to give us the biggest reward.

So why do we pick up our phones? Well, attention economy apps on our phone are very good at consistently producing reward. So the input pattern neurons that are connected to seeing the phone are going to uh activate dopamine neurons often. So the basil ganglia learns when I see that pattern of a phone is nearby or that's one of my possible future actions. I've learned that I'm going to get a reward if I do that because consistently when I pick up the phone, I'm getting a little bit of a reward, a little bit of dopamine. So, I've really strengthened those circuits. So, the vote for picking up the phone gets very strong. You can actually even think about the machine learning algorithms behind something like Tik Tok's uh curation like which decides what video to show you. What it's really doing is building a a model of this system in your brain and trying to figure out the reward it's trying to get is you actually continuing to watch. So it it's it's figuring out what can it show you that most consistently will generate dopamine so that it can get the [clears throat] uh strength and the action it wants. So it's literally sort of hacking the way this system works. All right. So we've we've heard things like this before. I've talked about this before, but it's good to actually have some more neuroscience expertise behind this and recognize this is the basil ganglia with has its own dopamine neuron inputs and and this is what's leading us to pick up our phone.

Okay, with that in mind, sub question number two. Why do digital detoxes fail? So why did the people in my declutter experiment who were just like, I I need to get away from my phone for 30 days so I can lose that addictive appeal. Why did they go back to using their phone just like they did before? Well, when we understand the basil ganglia as our our puppeteer, we realize not being around a stimuli for a little while doesn't change much because think about the role of this, right? Like this is this is uh it implements reinforcement learning within our brain. is how we learn what patterns generate rewards and what patterns we should avoid because they generate harm from an evolutionary perspective. We don't want to forget those patterns quickly, right? So if if I am an early, you know, vertebrae and my basil ganglia has learned that when I see a certain type of plant that there's often like food behind it that's like useful to me. I don't want to forget that. So, okay, if I don't see that plant for a month, but then I come across a part of my sort of Cambrian sea and I see it again, I want to remember like, yeah, that's good. That's where the food is so that I can take advantage of that. So, we don't lose, right? You we we it's not as if these memories of rewards will quickly fade if we don't get exposed to those rewards again and again. So if I uh do a detox, I spend a week without my phone, I might feel better in that week in the sense that I'm not numbing my brain on these apps, but as soon as I see that phone again after this detox is over, my basil ganglia is like boom, reward that wins the vote, puppeteer actions, pick up the phone. Same thing with like a digital Shabbat. Each week I take one day off from the phone. I mean, all this might have immediate benefits, but it's not going to make your phone less appealing. If you really wanted to uh directly hack the reward centers in the basil ganglia, what you would actually have to do is you would need to have a direct harm programmed into picking up the phone. So, you would need to set something up where there was, you know, electrods on your groin or something and every time you touched the phone, it gave you a shock. that would rewire those uh that would rewire those reward neurons very quickly and pretty soon you'd be like, "Oh, I'm definitely not going to pick up the phone anymore." But simply not being around your phone doesn't make that phone any less appealing next time you actually see it. And so this was the issue that was going on with the the participants of my experiment who just treated it as like a white knuckle experience is that the basil ganglia was like, "All right, we're not seeing any phones right now." But when it sees one again, it's like, "Oo, I remember that. There's food behind that plant." And it's just as appealing as it was three weeks earlier.

Hey, let's take a quick break to hear from some of the sponsors that makes this show possible. Look, if you listen to my podcast, then presumably you're interested in ideas for taking control of your mind to produce deep results in a distracted world. But there can be a difference sometimes between the type of scattered advice you might get on a podcast and what would be delivered in a carefully produced class. This is why I recently recorded my very own masterclass course which is called Rebuild Your Focus and Reclaim Your Time. It is a comprehensive look at the type of things we talk about here. Now, I was excited to record this class not just because I wanted to get my ideas out there, but because I'm a masterclass fan myself. just to name a few examples among many. I've listened to the Malcolm Gladwell class, the Aaron Sorcin class, and the Ron Howard's class and enjoyed every one of them. All right, so here's the relevant details you need to know about MasterClass. They offer more than 200 classes across 13 categories taught by the world's best instructors, including yours truly. You can watch the classes, but you can also listen to them in audio mode on your phone, meaning you can learn while you're working out or commuting or doing the dishes. And look, these classes work. Three and four members feel inspired every time they watch. 83% have applied something they've learned to their real lives. Masterclass keeps adding new classes, so there's never been a better time to get in. Right now, as a listener of this show, you get at least 15% off any annual membership at masterclass.com/deep. That's 15% off at masterclass.com/deep. Head to masterclass.comdeep to see the latest offer. I also want to talk about our friends at Gustto. One of the more annoying and timeconsuming parts of running a small business is payroll. You want to focus on producing stuff that matters and instead you're juggling tax forms and calculating withholdings. This is why you need Gusto, which can take all of these headaches off your plate, letting you get back to doing what you do best. Gusto is an online payroll and benefit software built for small businesses. It's an all-in-one remote friendly and incredibly easy to use so you can pay, hire, onboard, and support your team from anywhere. We're talking about automatic payroll tax filing, simple direct deposits, health benefits, commuter benefits, worker comps, 401k, you name it. Gust makes it simple and it has options for nearly every budget. You can get unlimited payroll runs for one monthly price. No hidden fees, no surprises. Look, our ad agency here at this podcast uses Gust to pay us. So, I can tell you from my perspective as someone on the other end, it really makes life very easy. So, try Gust today at gusto.com/deep and get three uh three months free when you run your first payroll. That's three months of free payroll at gusto.comdeep. One more time, gustogustto.com/deep.

All right, let's get back to the show. All right. So, sub question three. Now, we're going to have to get to a new level of geekdom in our neuroscience that I don't think we have yet reached on this show. Our sub question number three is why does analog reprogramming work when digital detoxing does it? So remember, analog reprogramming is my term for um aggressively exposing yourselves to other value producing activities. That the people in my experiment who ended up with long-term success spent the declutter period aggressively reinventing themselves through all sorts of activity and experimentation and self-reflection. All right. So to understand why that actually does work, we're going to have to dive even deeper into the brain functioning and get into the core of the new things I learned from the Max Bennett book.

All right, so let's take this step by step. So in our brain, we have a cortex. This is old. This this evolved relatively early. Um, it's what recognizes things in our world, right? So, it's what allows us to do pretty sophisticated pattern recognition so that we can understand where we are and what's happening and then we can connect that to things we've learned previously about rewards or harms. When you get to the first mammals, however, you get a a new part of the brain becoming much more prominent and it's what we would call the neoortex, which is actually like a layer that's on top of the brain. It covers the old cortex. All right. the neoortex. This is where things start to get more interesting. If we look closer at it, we'll see a big part of the neoortex is what's known as the sensory neoortex. Um, this actually, as far as we know, is running simulations of the real world. It uses these cortical columns to run simulations of all parts of our of the real world. And it's constantly sort of simulating. And one of the things we think happens is it's constantly doing these short-term simulations and making sure that what it thinks is going to happen matches up with what does. So it does a a simulation of what it thinks is going to happen as you step your foot forward and and if that matches what happens, you move on happily. But if it doesn't because like the the there's a loose rock or something, it's immediately that discrepancy is like, okay, problem. What's happening in the real world doesn't match what we thought and we have to put more resources to bear to figure out what's going on. So we have the sensory cortex is like a world simulator.

But then we have the frontal neoortex which has three main sub regions. But the one the the the primary sub region that is going to matter for our discussion here is what's known as the arranular preffrontal cortex or the APFC. Now this is something that evolved with the first mammals. Actually for the very first mammals their frontal neoortex only had an APFC and then these other regions evolved um as mammals got more sophisticated and as we got to primates. Okay. So now we're getting a little bit more complicated. the a granular preffrontal cortex, the APFC. We think what this is used for in part is to figure out what to simulate. So again, we have the sensory neoortex that can run all these pretty detailed simulations. And for the most part, it's just simulating everything it thinks is about to happen just to make sure that the world matches our understanding. But we can use these uh the sensory um the sensory neoortex to run all simulations of stuff that's um not about to happen. And there's other parts of the brain we can also involve in these simulations including other parts of the the prefront the the the frontal neoortex that allow us to do like pretty advanced simulations of the future and what if this happened or what if I went over here how's that what will happen how will that make me feel and we think that the APFC the a granular prefrontal cortex is like the coordinator of these simulations. It's the part of your brain that says, "Okay, I want to explore this possibility and let's see what happens." It's actually pretty cool that, you know, mammals can do any sorts of simulations at all. They can actually measure this in rats. They can do um they they call it visceral trial and error or vicarious rather trial and error where you can actually like see when the rat pauses and they're simulating the different possibilities of the different places they could go in the maze before they then start moving again. So we can kind of pause everything and run these simulations and the APFC is we think in charge of that.

All right. So what is the role of these simulations that the APFC initiates? Well, this is the way uh Bennett explains the simulations connecting to our behavior. So if the APFC initiates a simulation of something that we could do, it basically is feeding those simulations into the basil ganglia which doesn't know if it's seen the results of a simulation or the real world. It's way too simple and primitive. It doesn't know that. It's just being shown stuff through its input. If the simulation leads to a rewarding output then the steps of that simulation are reinforced. So I want to be careful about this because rein this is you know temporal difference reinforcement uh learning is a little bit complicated. Um, but essentially when you you know when you get to a reward state in the type of reinforcement learning that happens in our brain that gets reinforced backwards through the steps that led to that reward state even if they go back relatively far. So even like the initial step that led towards a eventual reward ends up getting reinforced. Right? So this is a model of learning that uh is attributed originally to Richard Sutton figuring it out and then we realize like oh this is really happening probably in the brain um of a lot of different animals including humans and this is actually how the basil ganglia when we say it learns about rewards uh it's doing this type of temp this this this type of reinforcement learning where the the rewards propagate back right so what's happening is actually the the APFC starts a simulation of something um it shows it to the basil ganglia who thinks it's happening. If it leads to a reward, then it reinforces the steps along the way. Then the APFC turns off simulation mode. Now we're back in the real world. Like we're getting in real input about what's actually happening in the real world. And the basil ganglia says, so if we just simulated one possibility, it's like, oh, I just saw that. And if we take this first step here, that's going to lead us, you know, that's just been reinforced because that's leading us down a path to a reward. So it's a little bit complicated. I had to kind of read this a couple times, but essentially by showing the basil ganglia simulation that leads to a reward about something you could do right now, when you go back to the real world mode, it thinks it's back at the beginning again, you just reinforce those steps. So, it's probably going to actually then take those real actions if that reward was really strong. So, the basil ganglia still is the puppeteer that makes all the decisions. So it's like the APFC is like I'm going to show you these movies of like if we went and did this it's going to lead somewhere good so that you'll start reinforcing those type of actions so that when I then say okay now we're back in the real world you're going to follow those actions. So the you have to influence the basil ganglia. You have to convince it that a certain set of actions you're going to start heading down lead you somewhere well. And you do it by just like simulating life and it learns, oh the oh this led somewhere good so I'm going to do that again if I see it again.

So let's put this back now. Let's take this all let's try to connect this back to uh uh our phone behavior. All right. So uh your phone is here and you don't want to pick it up. The basil ganglia knows there's Tik Tok on the phone and picking up the phone is um you know it's reinforced because it's going to give us that little hit and that hit will be have some sort of rewards and so that's what it wants to do. Your APFC at this point can say I'm going to simulate an alternative, right? So, I'm going to simulate going and picking up my running shoes, putting them on, and trying to log training miles, which I'm then going to put in my log and see if I'm making progress towards like getting in better shape, right? In the absence of the simulation, the basil ganglia would just say like picking up my shoes doesn't seem very rewarding, but picking up the phone does. you'll pick up the phone. But the simulation is going to run through this whole simulation that ends in a very um rewarding end state where you finish the run and you have the endorphins and you're you feel a um a sense of accomplishment from having, you know, made progress in your training and that's really rewarding. And the basil ganglia thinks this just happened. It doesn't know the simulation's fake. And so it starts reinforcing in its circuitry all of the steps that led to that reward state all the way back to the initial step of picking up your shoes. Now you you switch back to like I'm I'm in the real world. You see your shoes. You see your phone. Well, that picking up your shoes just got a bunch of reinforcement back from that in state in the simulation. And if it's strong enough, if that reward state was strong enough that you experience at the end of your simulation, picking up the shoes now outvotes the phone and you pick that up and you go and you get the run and you get the reward in the end.

So basically we have to uh expose ourselves to rewards, right? If there's a possible reward that is not only more compelling than picking up the phone, but it's compelling enough that when those rewards back propagate all the way to the very beginning of the whatever steps lead there, it's still really strong. Then we can uh setting down the path to that deep reward can outvote picking up the phone. So it's a little bit complicated what's happening, but it's interesting to think about. So what this tells us and this I think this explains the mystery is that the more you expose yourself to deep rewards from nonphone activities. The more you make it possible to have the simulations of those activities win over the short-term desire to pick up the phone. But the key point from Bennett is you actually have to have experienced these the simulations have to be compelling which means you have to have like experienced these rewards before for that simulation to be compelling enough that you're going to head down that path instead of the the shallower path of picking up the phone. So detoxing doesn't work. What you really need to do is to prepare your basil ganglia so that your APFC simulations will be sufficiently compelling. And this means exposing your basil ganglia as much as possible to real rewards that came from more valuedriven longerterm analog activities. It's a bootstrapping process. The more you do this upfront, the easier it will be to keep doing this going forward and the easier it will be to actually subvert the attraction of the phone. Because really, these reward signals you get from looking at something like Tik Tok are like fine. They're consistent. So, they're very pure, but they're not massive, right? It's like it uh it's the alleviation of boredom and the exposure to novelty. That's what you get, right? I mean like Tik Tok and X is like it's almost you know it's RCOO and its abstraction of just like stuff that's like that's kind of weird it's just kind of like interesting and novel right like so that's that's a a consistent but not super strong reward signal. So real valu-driven analog activities that give you like deeper rewards or help your sense of self etc or big sense of accomplishment. These can way outweigh the phone but you have to have been exposed to them enough time that your basil ganglia knows about them and then therefore it'll rate the simulations of that possibility high enough that you'll take the right first step instead of picking up your phone. Right? So it's not about trying to separate yourself from your phone, but instead about trying to repeatively and repetitively expose yourself to things that are better. That's I think why in my experiment, the people who are very aggressive about activities, the better is that they are basically training their basil ganglia to recognize a bunch of these uh rewards for the valuable activities as being very strong. So that later when the APFC triggers a simulation of going a non-phone route, those simulations are very compelling.

All right. So, this leads us to what's the practical advice if you feel like you're using your phone too often. You need repeated direct exposure to what we can call deep rewards. You need the ability to take steps towards these rewards uh to be both ubiquitous and accessible. So the the you you you really want to sort of surround yourself with opportunities to make steps towards reaping a deep reward. You need those steps to be possible and nearby in order for that to possibly win out. The first step towards whatever you're doing that's not the phone has to win out against pick up your phones. They need to be ubiquitous and nearby. Right? This is why if you get a lot of reward out of art, building a really nice art studio in your backyard is important because it's right there. You can literally just take a few steps and you can be there working on the art. Whereas, if you have to drive, you know, across town to an art studio, that first step is not proximate. So, it's not really very easily able to compete with the phone that is right next to you. It's why having notebooks, we talked about this in a recent episode, having notebooks handy to take notes on some sort of bigger design project or writing project you're working on matters because that's a step you can take right away that's leading you towards a deep reward or reading meaningful books. You have those books with you at all places. Or training, you have the ability to do physical training or exercise, like the stuff you need is at least right there to get started on it. You need to make the paths to these deep rewards accessible and ubiquitous if they are going to compete with your phone. You also need sufficient pathways to deep rewards in your life that you have a sufficient density of options. You probably need three to six different deep reward producing activities that you sort of keep juggling or or in the hopper. They're there as possibilities. I again I saw this with the digital declutter results. It's the people that did a lot during that period without their phone that had the best success after that period ended. Because if there's just one thing you do, there's a lot of situations where that's not something you can really reasonably make progress on and then there's nothing to compete against the phone.

All right. So, if I pull these threads together, I just thought this was really interesting to learn about what's really really going on. And it's all about simulations of possibilities. And if the simulation ends with a really big reward, going down that path can win out. But the only way that the thing at the end of that simulation is going to get received in your brain in the basil ganglia as a real reward is if you've actually experienced it before. And it's that's why I say it's like a bootstapping process. Like you have this initial process of like I'm forcing myself to do a lot of things to generate deep rewards that I wouldn't normally do at this level or density. But the more you do it, the more easy it will be to keep doing it. And then eventually you fall into this rhythm where you're much more interested in pursuing deep rewards than you know seeing someone get hit by a bowl on X is it. So it it bootstraps on each other. So this is why the digital declutter was successful is because it's not because it got people away from their phones. It's because it got people analog reprogramming. It bootstrapped the process of helping the simulations win over the consistent but moderate value signals of picking up the phone. And so stop thinking so much about how do I get away from my phone and think much more about how do I get towards the things that are better. It'll be hard at first. You'll have to force yourself, take some days off, go I don't however you want to do it, but it will get easier if you keep pursuing those rewards.

All right, so there you go. Jesse reporting from my studio in DC. How uh how do you rate my my neuroscience lecture? >> I kind of like it. Have you finished the book yet? >> No, I have it here. I am on page uh No, I'm on page 261. >> Is this part of the thinking research or is it just a random book? >> It is. Yeah. So, so you know, I'm working on this book about thinking and so I'm reading a lot about thinking and its role like the history of thinking. So if I want to know the history of thinking, I realize I need to know the history of the human brain. >> Mhm. >> And that is why Yeah. So that's why I'm reading this book. It's like my fourth or fifth book. My plan is to read like maybe 10 books this summer that are just about thinking, the impact of thinking. I'm just trying to understand thinking as well as possible before I move forward in that new book project I'm thinking about.

Hey, let's take another quick break to hear from some of the sponsors that make the show possible. The problem with these huge tech companies is that that they don't just want your money. They want to know everything about you. Shadowy data brokers make a living compiling detailed profiles of your online activity and then selling them to marketers and other corporate interests that want to control you with targeted ads. They're not selling a product. They're selling you. But you don't have to let them. There is a way to keep your browsing history private. It's an app called ExpressVPN. Now, a VPN works as follows. When you want to access a cider service, instead of directly sending messages to that cider service, you encrypt them and send it to a VPN server that then decrypts your original message and talks to the site and service on your behalf. This means that you know your ISP or someone who's watching your internet activity doesn't know what sites and services you're accessing. And on the other end, the sites and services that the VPN is talking to on your behalf don't necessarily know who it is that originated those messages. Privacy is regained. Now, if you're going to use a VPN, I recommend that you use ExpressVPN. They have plans that start at just $349 a month, which is only 12 cents a day, and it works on all of your devices. We're talking phone, laptop, tablet, you name it. You just tap one button to turn it on, and you're protected. It's that easy. Now, I I personally find it important to use a VPN like ExpressVPN because otherwise you're exposing much more of yourself than you might realize. So, secure your online data today by visiting expressvpn.com/deep. That's expvpn.com/deep to find out how you can get up to four extra months. That's expressvpn.com/deep. I also want to talk about our friends Advant. What's the one thing in business that's spreading as fast as AI? AI risk. Every new tool your team signs up for, every vendor that turns on AI features, every new integration, each one is an opportunity for something to go wrong. And most security programs weren't built for AI's pace of growth. This is where Vanta enters the scene. Vanta is the number one agentic trust platform used by over 16,000 fast-moving companies like Ramp and Cursor and Harvey to ensure that they're always audit ready. And now Vanta is helping companies like yours watch for the risks that show up between audits across your vendors, your AI tools, and your whole environment. How does it do this? The Vant agent works like a 247 GRC engineer in the background finding issues, drafting fixes for you, and uh cutting vendor assessment time by up to 50%. So whether you're a fast growing startup or a global enterprise, Vanta is here to help you automate your security and compliance and earn and prove trust. Get started today at vanta.com/deepquests. That's venta.com/deepquests.

All right, let's get back to the show. Um, but anyways, that's probably enough from me. Let's [clears throat] hear from you like we like to do on Monday episodes as we want to open up our inbox and hear what questions, observations, or interesting things you have to share. Remember, you can always reach us with your own contributions at podcastcalport.com. All right, Jesse, what do we got to to get going today? >> Our first note comes from Paul, who is passing along an article. All right. So, Paul says, "In case you haven't seen it, I was sure you'd find this essay interesting." Well, Paul, I'll be the judge of that. Maybe I won't, but let's see what Paul had to send along. Um, I'm going to bring this up on the screen for people who are watching. All right. So, here's the article. It's from the New York Times. The title is The Loi Way I Broke My Addiction to the Algorithm. There's a sort of claymation picture here. Uh, I did read this. So, Paul, I have seen this before, but let me try to get to um the core of it. All right. So, the author of this article says, "It's why I've been trying to live a summer of Lud, a term I'm cribing from a group that has been papering flyers all over New York City this past few weeks in an effort to get people off their phones." Dot dot dot. Um, for me, less and less has been Okay. I've been staging an experiment for the past few months. I've removed all my social media apps, Instagram, Twitter, Tik Tok, and moved them onto an old iPhone I had lying around my house. I dubbed this phone my scroll phone. I designated a single spot in my home where I could use it, a stripey armchair in my living room, which I have now anointed the scrolling chair. All right, so this is this oped about this idea of all my social media is on one phone that I only use while sitting in a particular chair. And this author said like, "Okay, uh, that's been really successful." Jesse, what's your what's your I have a complicated take on this. What's your guess, though? What's your guess on where I'm going to come down on this suggestion? >> Oh, that's a good question. Pro or con or in between? >> I guess it depends on how long she sits in the chair. >> Uh, yeah. She goes on to say she sits in the chair 17 hours a day. [laughter] >> Con. She just sits there scrolling constantly just ah put in my veins. Um actually it's a hard question to ask you because I have sort of a um mixed reactions. Um the thing I think that is good about this advice if we're going to put on our neuroscience hat is that the uh what you're tuning down is the pattern recognition from the cortex uh the old cortex of the phone and the phone being nearby. Right? So like those apps are not on her normal phone. they're only over by a chair. She's sort of convinced herself that's the only place you use the phone. So, she's not in a a context where she has time or is near the chair. It's just not coming up as an option as much. So, that does probably help, right? Is that that your your brain then is like not really voting for using your phone unless you're like in the room where the chair is and have time to go sit in the chair. Um that also puts a little bit of friction on it. So, that's that is um that is probably somewhat useful. On the other hand, you know, what you really need to do eventually is like what we just talked about, which is analog reprogramming, right? So like what you really need to do is to uh [snorts] instead of just having elaborate ways to try to put gates around this otherwise highly appealing activity is to make more valuable activities more appealing to get to a place where even if your pattern recognizers are sparking about your phone, you're not drawn to pick it up, right? So, I think analog reprogramming um is going to be much more effective in the long term than simply trying to add more friction or gates around phone usage. My bigger issue is that this reminds me a little bit. I mean, I don't know. This has vibes of the alcoholic that has like the super complicated rules around like when and how they drink and I, you know, it's only on these days if like I'm I've seen I'm around this person and only this much and I keep it over here and I keep the alcohol. And in the end, you're like, man, you're going through a lot of effort to to make it seem like you're you're doing something about your drinking issue, but that making sure that like the alcohol stays in your life. Like, it does kind of have that feel a little bit that sort of like the elaborate rules that drunks end up having that that they put around their their drinking. So, if you're having to have a scrolly chair and all these other and an old phone or this or that, I mean, I mean, at some point, you're like, maybe I just shouldn't use social media. So, I'll throw that out there as well. But anyways, I like the idea. I like that it's successful. uh we know from our brain science some reasons why that that we would expect that to be successful. Um, but I don't think by itself it's a full solution. >> Even before like when I was just a fan of the show and you had all your advice about social media, I just basically followed it. So now I never go on it. But if people like text me a link or something, I'll open it because I still have the apps. I just never go on it. So I can see like a link if somebody sends me a link or something like that. So that's all I do. >> Well, that's probably where you want to get, right? Is in some sense you want to be in a place where >> you don't have to have elaborate rules >> and gates that try to keep you away from it. It's just not as appealing to you. You don't I mean, it's all bootstrapping. >> You you you use it less. Other things build more rewards. Um, >> however, I did draw a line a couple days ago. Somebody sent me a Tik Tok link and I was like, I can't open it. So, the next time I saw him, I was like, you have to show me it on your phone because I'm not going to download the Tik Tok app. though. >> Good for you. Good for you. I know it's funny like I have most of these apps somewhere on my phone from various articles I've written in the past where I have to use them or this or that, but like I don't know. I have Tic Tac. I have Tik Tok on here. I think from the article I wrote for the New Yorker last year where I use Tik Tok. Um, and I have zero interest in load like I'm clicking on it now. Does this even work anymore? All right. Clicking on Tik Tok. Update your app. [laughter] I can do that. So there we go. It's like I >> expected me to bury an entire airplane in my backyard with a massive trench. >> At first it looked like nothing. >> All right, there's a video of someone burying an airplane. >> I mean, that's awesome. All right, [laughter] I take it back. Can I tell you what I just saw, Jesse? Is this real? All right. So I just saw a guy and this whole video is only like two minutes long. He buried an airplane in his backyard, like a passenger jet it looked like. Covered it over, put grass on top of it, cut off the front, put like hobbit doors on it, so you can go down this like passage in his backyard and be inside an airplane. >> And it's probably a deep work studio. >> I take it back. We all should be watching Tik Tok. This So, look, this is what I have to contend with. reading my book about the history of brain structures or watching a video of a guy burying a a plane in his backyard. Oh man. Um, okay. What else have we got here? >> Danny, as a follow-up to last week's interview with Brad Stolberg, >> right? Last week we had Brad on, we were talking about optimization and about how overoptimization doesn't necessarily uh lead you, it can make things worse. All right. So, Danny says, "This isn't necessarily a new article or link, but I was listening to your episode with Brad Stolberg about the optimization paradox, and a lot of the discussion revolved around problems that are basically explained by Goodart's law. This is the idea that once an indicator becomes a metric that is targeted, it loses its power as a useful indicator." All right, I feel like I should load this up here. Goodart's law. So, I found the summary of it. I'd heard it, but people mention this a lot in like the context of AI. Um, so I have a summary here. What is Goodart's law? There's three key takeaways. Let's just look at these. Uh, takeaway number one, Goodart's law warns of distorted metrics when tied to goals. It states, "When a measure becomes a target, it ceases to be a good measure, emphasizing how metrics can lose their effectiveness when manipulated to meet specific objectives." The second takeaway, the law highlights the risks of overfocusing on targets. Prioritizing specific metrics can lead to unintended consequences such as gaming the system, neglecting broader goals, or sacrificing quality. And three, the takeaway is balancing metrics with context is key to avoid the pitfalls of good hearts law. Organizations should treat metrics as tools for guidance rather than rigid targets, ensuring they align with long-term objectives and core values. Um, I don't know if that needs to be a law, but I do think that is common sense that if you have a single metric that you're like, this is what I'm pursuing, that doesn't necessarily optimize the actual result that you care about if those two things are different. Okay, I completely believe that because this to me is the core problem

of knowledge work right now. Right? Right.

So, if in my book *Slow Productivity*, I talk about this idea of pseudo-productivity. That the primary way that we try to organize our efforts in the knowledge work context is around the belief that visible effort is a proxy for useful effort. So, the more busy you seem, the more useful we assume you're being. Right? So, there we have a metric: busy, visible activity that a lot of people in knowledge work are trying to maximize. But as we know, if you listen to my podcast or read my books, busyness is often far disconnected from actually producing things of value.

So, we've seen this real clearly, um, with AI and computer programming. There was this period that already has come to an end. But there was this brief period this year after the coding harnesses got successful, in which we had, uh, but before the prices had actually been raised to their actual, real prices, when everything was still heavily discounted. We had a bunch of companies that would say, "The metric we care about for you as a programmer is how many AI tokens you burn using our coding harnesses." So, the more AI tokens you're burning, we will assume the more useful code you are producing. We're going to have leaderboards at many of these companies to see who's burning the most tokens.

This turned out to be a, uh, a terrible metric to optimize for because it's very easy to have to get these these, uh, AI models that burn less tokens and produce code that it checks the code, and this code goes to that code, and here's 30 versions of the code. It doesn't mean that you're producing good code, and it doesn't speed up the rate at which features actually get added or products actually get shipped.

Now, pretty quickly, the token leaderboards went away because, again, as I mentioned, uh, to try to get people to use these harnesses on top of these models, Anthropic and OpenAI were were selling tokens at a real discount. They basically had these unlimited accounts. You pay $200 a month and burn as many tokens as you wanted. It was costing a lot more to actually do this compute because it's expensive. And so when they adjusted to say, "Here's the real price," all those leaderboards went away because you had people who were spending tens of thousands of dollars a month to get to the top of that leaderboard. And again, if it was leading to massive increases in the amount of value being shipped from the companies, then maybe it was worth it. But it doesn't.

And it's kind of the paradox of of AI programming is that AI tools help you produce a lot more code, but they don't necessarily massively speed up the rate at which features get added or products get shipped. The real place you see the major productivity gains is if you're building proof of concepts, if you're trying to, um, hack together prototypes, or if you don't really care about the quality of the code. It seems magical. If you're working on a mature codebase, things are much more complicated.

So, Goodhart's Law, I think, is, I think that is useful. Um, again, I think we got this with, if we rewind the clock with pseudo-productivity, we get like email response times, amount of times you're on Slack, the number of meetings you're jumping in and out of, all metrics you can maximize that aren't directly connected to actually producing value. Um, so I agree, Danny. I think Goodhart's Law, uh, I think that is, that's useful. That's useful, you know, useful terminology for us to throw into the mix here.

All right, Jesse. So, what do we got for question three?

>> Our next note is from Nenah, who is sharing a policy article she thought you might find interesting.

>> Interesting. Yeah. So, Nina said, "I want to share a screen time policy piece someone reached out to me about." Well, hey, you know, nothing gets me more interested than screen time policy pieces. This is like catnip for me. Uh, let's load this up here on the screen for those who are watching. All right. So, this article is showing up in *Health Affairs*. The title is, "Digital Addiction is a Public Health Problem. Is Public Health Law the Solution?" Uh, and we see a group of authors here. Lead author, Sophia Pelieri. All right. The subhead says, "Engagement-maximizing architecture such as infinite scroll, autoplay, and emotionally targeted notifications remain broadly permissible. That gap, however, is now being challenged on multiple fronts."

Um, so I'm going to read a core paragraph here. Uh, let's see, where did I find this? Okay, I read this earlier, so I'm just going to hone in on this paragraph, which I think kind of gets to the, the kind of the big idea here. "Public health law has long addressed harms arising from commercially engineered products by deploying a familiar regulatory toolkit: product design standards, warning and disclosure requirements, advertising restrictions, age-based access controls, uh, surveillance obligations, and funding mechanisms for treatment and prevention. These tools align with different public health objectives. Regulators might prioritize reduced consumption. Um, controlled access restricts availability to ensure that only patients with appropriate clinical indications can obtain prescriptions, or safer engagement through mandatory design standards."

All right. So, what they're saying here, let me find, there's one other paragraph I want to define here. Uh, okay. Okay, so up here it says, "Contemporary digital platforms are explicitly designed to maximize engagement by leveraging well-characterized reinforcement mechanisms. What scholars study gambling have termed 'addiction by design,' including variable reward schedules, emotionally salient feedback, and frictionless continuation in social media algorithmic feeds and AI chat systems. The risk is further amplified by hyper-personalization. Modern recommendation systems continuously infer user preferences, emotional states, and vulnerabilities, tailoring content or conversations in real-time to maximizing engagement, intensifying exposure to emotionally salient or validating stimuli, and reinforcing compulsive use through variable and individualized reward structures."

So, what they're arguing is like, hey, this sounds familiar to other things that we have regulated, and if we look at the way those other regulations have happened, we might see a possibility for how we would regulate screen time. Um, if you read the article more, and I, I read it in some detail earlier, again, they have these models of like tobacco use, um, gambling, and, uh, s- other age opioids, right? And they said we have overlaps between all three of those with screen time, and each of those has particular solutions. So, like with, um, tobacco use, we had age gating, right? It was, uh, kids shouldn't use tobacco because their brains can't handle it, and we have education. We're going to, um, educate about the harms and addictive nature of tobacco. With opioids, it's controlled access. Actually, the state is going to control, uh, who gets access to it and under what circumstances. And then when it comes to gambling, there is design restrictions, right? So, like, there are restrictions on what you can and can't do when you're doing gambling games, right? Like what is allowed, what's not, what you can do with the variable reward schedules or not. There's a lot of restrictions, for example, around slot machines, like how slot machines are allowed to behave or not behave, um, etc. So, they said we have like responses for all three of these things that overlap screen times.

And as I read closer, they said the gambling response is probably the most relevant. So, the in the same way that we have restrictions about how games designed to be addictive in a casino work, we could imagine a world in which we had similar restrictives about digital, um, uh, addiction. So, they say here, "Yeah, so they called it 'design safeguards to mitigate engineered addictiveness.'" I'm interested in this, right? I would say traditionally, I had been skeptical about the idea of regulating reduced addictiveness of technology because my main concern is when I read thinkers and digital ethicists and policymakers talking about engineered addictiveness, they hone in on, they have this mental model where they can hone in on these like particular things you added onto a digital product that made them addictive, and you could just turn them down. This was my issue with it: I don't think that model is correct.

So, they're like, well, you know, in this mental model, they would be like, you know, you know, something like Twitter or TikTok, the issue there is, well, you have infinite scroll, or you have a, a particular way that like the, the new post pop up that's like a slot machine, and that's kind of that's addictive. Or they talk about dark patterns, which is really kind of a nonsense term for like, well, the way it interacts with you, um, emphasizes addictiveness. The whole point here in this type of thinking, in this mental model, is we could turn those features off and then have a version of TikTok or Twitter, um, that wasn't addictive. But that's not actually my understanding of how these things work. I, I think the addictive loop, the thing that makes the basal ganglia always vote to pick up that phone and look at that app, is actually typically just in the core functioning of the app, right? It's, I'm showing you videos, and I'm selecting the video to show you based on things you liked before. It's a very simple feedback loop that pretty quickly locates subsets of the videos in the space of possible videos that, uh, generate a novelty or humor or other type of positive reaction in you, and then you want to keep looking at it. There's not a feature you turn off that makes that non-addictive. That's the, that's the issue with the engineered addictiveness approach is that I think it, this model just isn't correct.

Now, you could say, "We'll turn off the algorithm," but that, the algorithm is just the thing that decides what video to show you next. In the context of TikTok, it's using a pretty basic multi-arm bandit style optimization. If you turn off the algorithm, what's it showing you then? Like, what videos does it show you? So, that's what, that's, that's always been my issue is that like, I don't think the mental model of, um, social media is fine, and then we added addictive features, made it bad, turn those back off again. I just don't think that mental model is right. I think for a lot of people, it's built in. You know, there's this sort of vigilance switch on things like Twitter where for a while, if you're, you know, if you were more of like a left-leaning academic, Twitter was great and exciting, and then it kind of got, it shouldified and got worse. And so you have this paradox, paradise lost idea of like, something must have made this worse. We can go back to the way it was before. And I actually just think it's somewhat, um, fundamental.

On the other hand, this is what's interesting to me is like, okay, let's pull this thread. Like, if we really were serious about no, uh, you know, engineered addictiveness is bad. So, using recommendation algorithms that are using, uh, fine-grained observations of your behavior to decide what to show you to be as engag- like, you can't do that anymore. That's not allowed. TikTok goes away, right? That is what TikTok is. But maybe that's not the worst thing. And if you're Instagram, right, or your Facebook, you go back to follower feeds, uh, these aren't algorithmically curated. So, I'm just seeing a reverse chronological timeline of like stuff that's being posted by people I actually have to follow. That wouldn't be the worst thing. I think these things would be much less addictive. Like the way Twitter used to work is like, you would look at your timeline, um, and if you came back to look at it 10 minutes later, you're like, none of the hundred people I follow who I think are interesting said anything new. There's nothing there. All right, let me move on with my life. New apps will be like, "No, no, no. Hold on, hold on, hold on. Let me show you a, I have a, a someone getting sucked into a drain pipe to show you, or a fight, or someone getting hit in a car." Like, I'll just show you. There's always stuff here to see.

So, actually, the thing I thought was a bug with the engineered addictiveness reduction regulation argument, the fact that like it's just fundamental to how these things work, actually maybe is a feature. Then like, if you can't be engineered addictiveness, it, you can't have a lot of these platforms, or they have to go back to the way they were before when they were interesting but but not, uh, compelling in this way. Our understanding of brain science that we talked about today helps us understand, you know, why this is right. If, like, looking at X, sometimes you get, uh, something new that's interesting from someone you follow, most of the time you don't. Means that it's, uh, reward signal is way more inconsistent, and the power of the votes for picking up the phone to look at X is going to be much less than in a world in which it will always find something on a person using a personalized algorithm to show you.

So, I think that's interesting, but again, there's all devil's advocate. But then someone will say, like, "But what if, like, Netflix wants to recommend shows? Does it mean it can't do that anymore?" Right? And wouldn't we want there to be auto-recommendations? And hey, maybe not. Maybe maybe we want this all to be human-curated. So, I don't know. I, I've had a, Jesse, I've had a complicated history with this type of regulatory path. I don't think people's mental model of engineered addictiveness is right. But the actual model, actually, if we get rid of that, that's a much bigger swing.

>> Mhm.

>> But maybe it's something we have to consider. I don't know. What would you think about like a Netflix? This is not the addictive issue with Netflix because it's not a close, a rapid feedback loop like with TikTok. But if you got rid of recommendations on Netflix, I don't think that would really matter. I think most people are just finding they're happy to be like recommended stuff by people, but I don't know, maybe I'm being naive here.

>> Yeah.

>> All right, what else do we got?

>> Our final question comes from Jessica, and it's about deep work in the age of AI.

>> All right, let's see here. Jessica says, "My law school assigned *Deep Work* as summer reading." Oh, that's cool. We should find out what law school that is, Jesse.

>> Yeah.

>> And I'm reading it. Does Jun know, or is it not?

>> I don't know. No.

>> Okay. Um, and I'm finding it to be incredibly relevant in this current AI-centric climate, despite it being published before the ongoing chokehold that AI has on everyone, especially students. I really wanted to know if you had some new thoughts or a kind of follow-up to the book in the context of gender of AI currently and its effect on deep work. To me, deep work regarding law is more important than ever. Yet, as I read the book, I can't help but wonder how many of my fellow students will only ask ChatGPT to summarize it for them and not bother reading it for themselves. Do you have any thoughts or insight as to the value of deep work now in relation to many falling back on AI for tasks that had previously been worked through manually? The book is as relevant as ever, but I find myself wanting some commentary on deep work's place directly in the context of the heavy use of AI among students.

>> Um, it is a good question because the context surrounding deep work has changed from when I, when I wrote that to now. And here's one of the big changes I've seen in the AI age. In 2016, when that book came out, the number one issue was that people were undervaluing deep work. So, they weren't spending enough time doing true deep work. And because of that, the value they were producing was being artificially capped by prioritizing other things like responsiveness or meetings or pseudo-productivity. You didn't get enough time for actual pure deep work. And because of that, you actually was holding you back how much value you could produce.

Um, the other issue I was trying to correct in that book is that people didn't understand what deep work was. So, even when they thought they were doing deep work, they were doing things like quick checks of email inboxes and Slack channels and not realizing that that was a catastrophe for their cognitive capabilities, that all of that cognitive context shifting was causing lots of issues. And so, if you could be more careful about how you approach deep work and you prioritize it more, you would be happier. You'd produce more value. So, it was it was about people just not doing enough deep work or understanding how to do it.

Well, in the AI moment, we have this other issue which, as Jessica mentions, is people outsourcing things that would normally require deep work, like reading a book. Basically, anything that causes cognitive friction. We're like, "Ooh, I'm, um," which almost always is going to be abstract processing. We're using our brain in ways that we weren't evolved for. So, any time we're we're grappling to understand words or to produce words, or understand mathematics or produce mathematics, um, these are examples of activities that cause cognitive friction. People are turning to AI to try to reduce that friction. So, typically, you would need to use deep work to understand a book in a law school class, but now you could get a summary of it, and you have to expend a lot less energy.

The reason why this is a problem is that it makes you dumber. And if you're dumber, you're worse at deep work for when it actually has to happen. The friction is what you want. That is the feeling of the metaphorical muscle getting stronger. If you want to be stronger, you actually have to do the exercising. And so, if you don't grapple with a book, which is hard and causes cognitive friction, you don't really understand that material very well. And if you don't understand it very well, you can't deploy it very well later when you need to. When whatever you're doing in this context, in your law context, it makes you dumber. You're just not getting the benefits. Deep work gives you benefits in terms of understanding. Same thing with writing. It's uncomfortable to have to yoke together many different parts of your brain to put letters onto a blank sheet of paper. But writing is how your brain categorizes, makes sense, and better refines your understanding of things. This is, I mean, we've been doing it this way for a while at law school. You, you struggle with text. You struggle to write about those texts. That is making your brain stronger so that you can do law. If you have a machine do those things for you, your brain is not getting stronger. You will be a worse lawyer.

It's the equivalent of just having another student do your work for you. Yeah, that's easier, but the whole reason why you're doing that work is because you're trying to create a lawyer brain, which requires a lot of training. Just like if we were going to make you into a special operations operator, we would want you to do all this PT that we're doing at Naval SEALs training because we need your body to get very strong and your endurance to be very high. Otherwise, you're going to struggle when we put you into missions. Same thing if you want to be a lawyer, we need to do cognitive PT, which is going to be reading these terrible books and going through all these footnotes and wrangling with it to try to write clear briefs. And you, and your mind as you write, like, "This doesn't quite make sense. My logic's not quite there." And in all of that, your brain is getting in the Navy SEAL shape because that's what you need for the equivalent of a Navy SEAL mission in our cognitive world, which is like working on a complicated law case.

So, yeah, I used to worry that people just weren't doing enough deep work, and now I'm worrying that they're outsourcing the deep, the obvious deep work that remains and are getting none of the benefits because of that. So, yeah, I do think it is an issue. Uh, you know, it's funny. I have a call right after this about AI policy at the university level. So, I've been thinking a lot about this. Um, and, you know, we need to, I think we're at a point now with AI where we have to think about the human brain, why it's important, and what we value about it, and what we're trying to do about it from a humanistic standpoint. What is the goal? What do we, what do we want to do with our brains and why? And then step back and say, "So, where do I want to use AI or not?" Just like we would do with physical fitness. Like, I could drive around on a Rascal scooter and make sure that like I never moved my my body at all, um, and that would be easier in the moment, but it would, it's not good for my body. It'll make me less healthy, and I'd be more miserable about it. We just got to start thinking that same way about our mind.

I, again, I think the role of AI in education right now should be incredibly limited, right? Incredibly limited. It should be, it should be basically focused only when there's AI-driven tools that's used in a professional, uh, academic context. As you get to the right level of training, you can learn how to use those tools. Otherwise, this is Navy SEAL training for your brain. If you're at law school or you're at a university, it doesn't make sense to bring in polio to lift the weights or to have someone else do the sit-ups for you. So, you know, I think we're going to get better at this. We're, we're all still trying to grapple with this, Jesse, but, um, my new book, which I haven't just an inkling in my eye, my *Indefensible Thinking* book, we'll get into this, but all I'm doing now is reading books that maybe will help me figure out what this book would be about. It's so far from now I can't even think about it. But in theory, it's something I am gonna tackle.

>> But your last book, you signed a two-book contract. Are you gonna do the same thing?

>> No, I want to just sign, I just want to sell this next book.

>> Got it.

>> Um, because I don't know what's going to come next. I, I, I'll do a two-book contract if I have two good ideas. I, I mean, I signed my last two-book deal. It was like basically at the end, it was in the pandemic, and I, it was *Slow Productivity* and *The Deep Life*, and I was like, I knew I wanted to write those two books next. It's many years later that finally we're almost there. U, but this time I just want to write this next book and then see what comes next after it.

>> Yep.

>> Um, all right. Well, let's, uh, before we conclude for today, we like to briefly check in on one of the episodes about what I'm up to. Clearly, I'm not in the Deepwork HQ. Jesse is. Uh, I'm up in Vermont. I'll tell you what's clutch about this place, Jesse. We rent different places each year, uh, for now until I build my my Deepwork HQ North, which will happen. It will happen. Mark my words. This place is walking distance to a trail system.

>> Okay.

>> Which, to me, that's important. I love thinking walks. Thinking walks in the woods are great. And to be able to walk into a trail system, um, and do my thinking walks, that has been, uh, clutch. So, I'm adding that to my list for things that the Deep Work HQ North is going to have to have.

>> Do you have a thinking walk scheduled for today?

>> Um, yeah. I hope so. Let me see. We work on the podcast. I have a call. Then we're going to go into town. Yeah. Tonight, if it doesn't, if the rain holds off, I'll definitely do, I'll definitely do a thinking walk. Uh, definitely do a thinking walk tonight. Yeah, I've been working on an article up here. So, like, I've been, I just submitted a draft. So, that's, that's the only reason why.

>> So, I don't even know what I'm thinking about. Um, I don't know where I am with the book. I think I'm going to have to do like a July book roundup the old-fashioned way next week, Jesse, because I don't know. We put a couple episodes in the can. I don't know what the last book I talked about. I've been reading a lot of books. I'm in the middle of a lot of books. I don't even know. So, I'm going to just wait until next week when I'm back and I have my reading list from home and I'll go back. But I'm, I, I don't know where I am.

>> Just as long as the audience knows that you haven't quit reading is efficient.

>> No, I'm about to finish two books in the next couple of days. I think I finished a couple books right before I left. Um, I think it's, this might be a seven-book month, I think, in the end, but we'll see.

>> But yeah, I don't even, I don't even remember where we are. So, I have not quit. I'm reading more than ever. I'll do an old-fashioned book roundup, um, once I'm back to the HQ next week. Uh, otherwise, hopefully the HQ's doing okay. I ran a lot of experiments in there when you were gone, Jesse. You might notice there's like a growing number of circuits in the in the maker lab area.

>> Yeah.

>> Let me just say this for people who understand when I'm thinking about my Halloween animatronics. I successfully before I left for this trip can now, uh, am able to program, sequence, and X lights that I can then download onto my Falcon Player controller on a Raspberry Pi, which can then connect to a light controller. I'm using Elgato Plus light controller over an Ethernet network, and that controller can then control multiple programmable lights. So, I now have the full work chain in place for doing professional caliber sound, lights, and actuator motor synchronization and control, so that I can achieve my goal for this year of having a fully coherent Disney-style animatronic Halloween scene with movement, lights, and sound all, um, synchronized, and a countdown timer in between executions. So, I have all the technical chain in place. Now I actually have to just, uh, start working on the actual, the actual props. So, that's I've been hard at work of that. I know you follow that closely and you check all my circuitry when you come in. So, hopefully it all looks okay. But I, I am happy about that.

>> As long as you don't turn into a Walt Disney and smoke three packs a day, then you're fine.

>> Uh, I don't know, man. It was creative. That's like a Mason Curry book. What an artist do you have to start smoking through not just three packs a day. Also, this, this, this I'll, I'll have to do as well. He, you know, his he had bad back because he had injuries from playing polo or whatever. And so, by the, the late 40s, early 50s, every day he had a nurse, a full-time nurse named Hazel George. And every day at the end of the workday, uh, she would give him like a pretty extensive back massage in his, uh, office, but she would mix him a whiskey-based drink, which he would drink through a straw because he was on the massage table so he could be drink. And I, I would bet you, I would bet you, you know, Mickey Mouse's shoes that he had a cigarette in the other hand as well that massage. Let's be honest. Like, this guy's got it. He, this guy had it figured out.

>> If you're going to smoke three packs a day, you don't have much downtime.

>> Yeah. You got to get after it. So, think about that. Daily massages while drinking whiskey through a straw and smoking a cigarette. This guy, this guy lived life. Did die at 65. Um, I am reading a Disney book, by the way, that is excellent. It's, uh, I'll talk more about it, you know, next week in the roundup. It's an academic book. It's a crazy book. An academic book, new book, Princeton University Press, by an art historian at UC Irvine named Ronald B. Thencour, uh, that is getting into like the precise use of industrial automation technology deployed for rides in Disneyland and kind of like, there's, there's an academic thesis here about Disneyland as being a place to sort of like expose an econ, expose people to the logics of automation, etc., etc. But man, this thing is research. He is like in the weeds on, and he's an art historian, but he's in the weeds on my type of nerd stuff. Like this type of programmable logic controller was used and wired up in this way for the, the, the Matterhorn Bobsleds, like, just all of the technical details of the engineering behind these rides. I'm like, this is a crazy book. Um, but I am happy it exists. So, I have some good Disney content for sure.

All right, that's enough of that. We should probably call it here. Uh, but thank you everyone who's listening. I believe next week we'll be back in the HQ, right? This is the only one we're recording while I'm on the road.

>> Yes.

>> All right. So, I think next week I will be back in the Deep Work HQ. Um, no. And probably we will have, I don't know. I think it'll probably be an AI reality check. I get so mixed up on what we're recording, what we're not. But, you know what? Just stay tuned. Stay tuned. You never know what you're going to get, but it'll always be good on the Deep Questions feed. So, until next time, as always, stay deep.

Hey, if you made it this far, you must be ready to join my fight for depth in a distracted world. Now, the best way to do this is to join over 125,000 people who receive my email newsletter each Monday. You can sign up at calupport.com/idas. And when you do, I will send you a free guide to my seven best ideas about cultivating a deep life. Sign up today, caler.com/idas.