Ep. 353: Summer Schedules

I'm Cal Newport, and this is Deep Questions, the show about cultivating a deep life in a distracted world. I'm here on my Deep Work HQ, joined as always by my producer Jesse. Maybe really our first summer day we've recorded, in the sense that it's 80-something degrees. Is it that hot? It's going up to like 87 or something like this.

I'm getting a new air conditioner installed in our downstairs this weekend. Yeah, because I remember you did it in the upstairs. Yeah. And this new one, I believe, if I'm understanding the marketing copy properly from Carrier, is self-aware. That's how advanced the technology now is in air conditioners.

Like, the one we had is from, like, roughly 1976. The one downstairs. Yeah. I mean, I think there's just, like, a relatively small guy in there with an ice cube and a fan that he would just sort of wave it over. And then this new one... It's Skynet.

As far as I can tell. It's like computers and controls and it's connected to the internet for God knows what reason. I don't know what's going on with this thing. How's the one upstairs doing? It's doing pretty well. So same, you're buying it from the same provider? Yeah, but I'll just say this. If you find me... A debt in my house froze in and my bank account's emptied. Blame the air conditioner. I don't trust it. These things are too smart. We've been getting, speaking of...

We've been hearing from people about the absence of my newsletter, my email newsletter. So brief update on that. I made the decision a while ago to take the newsletter. to the next level make it more serious i was like look this needs to be It really should be a weekly newsletter, well formatted and branded with the right content and edited properly, et cetera, et cetera. So I was like, I'm going to hire someone to work on this with me. And this is going to become a truly properly professional.

The weekly newsletter, which is what it should be, and that's what it's going to be. But basically, as I was going through that process of getting all of my ducks in a row... to move the newsletter to the next level, I just stopped doing it. Because it's like, yeah, we're going to have this nicer version coming, and so let me wait for that. So that's why there's been a delay, but these changes are now upon us. I believe the day this episode comes out, there'll be a newsletter post.

which is God willing, the last newsletter post that I'm doing all on my own. And then there should be a three to four a month rhythm. Continuing from there with the help of my... new newsletter guru so you'll sort of see in the weeks ahead we're gonna we're gonna get a better looking newsletter there's gonna be some more content in the newsletter in addition to my main essay so i'm looking forward to that yeah we're here's the big deal though uh we're naming it

So remember, from the early days, my newsletter slash blog was called Study Hacks back when I gave student advice. And then at some point, we moved away from student advice. We didn't really emphasize Study Hacks. It didn't really make sense. And it was just sort of like... Cal Newport's thing.

It needs a name. And so I've decided I'm going to steal the name I was using for Thursday podcast interview episodes. I'm going to steal that name for the newsletter. The newsletter is going to be called In Depth. With Cal Newport. So what are you gonna call the interviews? I don't know. There's other ideas for that. Yeah, I have other ideas for that.

I think I'm going to call it. They're going to create names. Yeah. Well, what I figured, I did some research on it, and I found both Oprah and Joe Rogan are very popular interviewers, so I'm going to just name it. um, Oprah and Joe Rogan presents and then whatever interview I'm doing. And I figured like it'll be more popular then because people will be like, I like those people.

And then we'll get more listeners. Mad Dog's a good interviewer, too. Mad Dog, Oprah, and Joe Rogan presents a conversation with Michael Easter. I like it. I can do the voices. and then we'll try. That'll go over really well. Anyway, so I'm excited about it. I mean, first and foremost, I am a writer, and I want the newsletter...

I want to be able to put my thoughts down. Yeah, I'd have to respond to people because they were like, am I not on the list? Am I on the list? I checked. I reached and arrived. I know. It's coming. Actually, I am very interested in feedback once it gets going. One side. once a week again. I really do want feedback. You can send it to Jesse at jesse.com. He's more likely to see it than me about like, oh, I like these topics, not these topics because we cover a lot and

There'll be overlap with the podcast. I typically, sometimes not, I go deeper on things in the newsletter. I can be a little bit more nerdy on it. So anyways, if you're not subscribed to that, just go to calnewport.com and subscribe because the weekly email. has begun all right jesse we got a good episode today we got a summer themed Slow productivity, inspired, deep dive. We got some good questions, calls, case studies, and then reaction.

I'm going to react to the most important article about AI that you did not read, but tells us a lot about what's happening now and should also bring your... blood pressure down a little bit if you are worried that we're weeks away from AI conquering the world. All right, so let's get started with our deep dive. For me, today is in some sense the first day of summer. I'm a college professor and my final grades were submitted last week.

I hosted the lunch for the graduating seniors, commencement was on Saturday, the campus has just shut down for the summer season. Unlike most professors at most research universities, however, I don't take summer salary. As soon as the spring semester ends, I am officially off the clock until the fall comes. I've received no paycheck from the university. I receive no paycheck from the grant agencies during the summer. There's no expectations from anyone else. My time is my own.

Here is the schedule that I more or less try to run. during these summers of no external obligation. I try, if possible, to have no professional appointments, meetings, or calls on Mondays or Fridays. I want to begin and end my week with quiet and depth. So the transition into the weekend and from the weekend back into the week is one that is much more gradual and subtle and cognitively rewarding. You're not going from a quiet Sunday to a crazy Monday.

I try when possible to make one of those days, Monday or Friday, an adventure thinking day. where i'll spend at least three or four hours usually somewhere outside in nature being way too hot because i do live in washington dc but maybe going to like the peck toxic wildlife refuge right outside of the Beltway or to Wheaton Regional Park or to the, I like the Rachel Carson Greenway.

to a trail, go for a hike, spend a couple hours outside, go to Rock Creek Park, bring my notebooks and try to really work through ideas. For Tuesday, Wednesday, and Thursday, those days start with deep work. I don't look at an email or a computer until midday. And then in the afternoon, I have an admin block. What I try to do is have 30 minutes on my calendar midday, Tuesday, Wednesday, Thursday, where like a furious task.

Completing machine, I go through small logistical things that have piled up and try to make as much progress as possible. In the afternoons on those three days is when I'll have calls, I'll have interviews. I'll have professional-related appointments, and I try to end those days by far. That is my summer schedule, and I love it, and I can get away with it because I have these really slow summers. But it is a vital part of my success.

It is a step away from the frantic demands of the school year. I get my best thinking done during the summer. I produce on my best ride in the summer. It's when I recharge for the year ahead. So I want to talk to you today about this general strategy of seasonality. I'm going to argue that it's important for basically everyone, especially if you have some sort of knowledge or office work job,

You might not be able to push your summer schedule to the same extreme as mine, but I think it should be different. I'll argue why, and then I'll give you some ideas about how you can do that, even if you have less flexibility than myself. Alright, let's start with the case for the seasonality, taking one season to be different and less intense. I'm going to read something from my book here, The Bible for the Show, Slow Productivity. I wrote about this idea.

Did some good research on this, so I'm going to take a second to take advantage of the research I already did. All right. I'm going to read here. The side-by-side comparison underscores the degree to which our experience of work has transformed during the recent past of our species. Our shift from hunting and gathering to agriculture, the Neolithic Revolution, only really picked up speed somewhere around 12,000 years ago.

By the time of the Roman Empire, foraging had almost completely disappeared from the human story. This reorientation towards agriculture threw most of humanity into a state similar to that of the rice farming Agta. grappling with something new, the continuous monotony of unvarying work all day long.

day after day all right so i'm arguing here i i so i i do a before this i went through a treatment of what the foraging and hunting and gathering life was like in which we spent most of our species existence Drawing on some more recent research from the Ajita people in the Philippines where you had the same people that divided half of the group. went to rice farming and the other half stayed with

There's interesting research from the 2000s where they compared the lifestyle between these two groups that are otherwise very much controlled to be the same people in the same type of geographical region. And they found that, oh, the hunting and gathering foraging group has way more variation in their days. Busy periods, non-busy periods, busy seasons, non-busy periods. So what I'm arguing here is that change with agriculture

Our days when we were working became uniformly hard days. However, I'm going back to reading here, the one saving grace in this scenario is that agriculture didn't demand this homogenized effort the entire year. as the busy sowing and gathering of crops is offset by the quiet of winter. So we went from hunting and gathering and foraging, where even within a given day, there's huge variation in intensity.

This hour is really locked in because we're in the middle of a hunt, but the next two hours we're taking a nap during the midday sun. Then we get agriculture. I said, no, no, when you work, we have this new idea of working sunrise to sunset of equal intensity. our Paleolithic forbearers did not have such uniform intensity, but At a bigger scale, you had busy seasons and less busy seasons. The fall was very busy. The winter was not.

Moving forward in the human story, and I'm reading again here, the Industrial Revolution stripped away those last vestiges of variation in our work efforts. The powered mill followed by the factory made every day a harvest day. Continuous, monotonous labor that never alters. Gone were the seasonal changes and since making rituals. Marx, for all his flaws and overreach, hit on something deep with his theory of entrefundance. I said that perfectly, Jesse, by the way.

estrangement which argued that the industrial order alienated us from our business our basic human nature The workers eventually inevitably fought back against this grim situation. They pushed for reform legislation like the Fair Labor Standards Act passed by the U.S. Congress in 1938, which fixed 40 hours as a standard work week.

limiting the fraction of the data that could be snared in monotonous effort without extra pay. They also form labor unions as a counterbalance to the more dehumanizing aspects of industrialization. All right, so what happens next is the factory in the mill is a brand new thing where we said not only are you going to work all day long, at equal intensity, there are no seasons anymore.

The textile mill doesn't care if it's January versus June. Now, all year round, there's not going to be a change. What I'm arguing here in the book is that this was deranging for people this was like really difficult it's not what our species evolved to do to hard work all day long without change all year round very unnatural and so we had to at least to make it bearable, get things like reform legislation and labor unions to at least try to have some pushback there.

Finally, we get the knowledge work, which brings us to what us as our audience, to our world. So let me read here briefly. Then knowledge work entered the scene as a major economic sector. The managerial class didn't know how to handle the autonomy and variety of jobs in this new sector. Their stopgut response was pseudo-productivity, which used visible activity as a proxy for usefulness. Under this new configuration we took another step backward.

As in the industrial sector, we continue to work all day, every day without seasonal changes, as any such variation would now be received as non-productiveness. But unlike in the industrial sector, in this invisible factory we'd constructed for ourselves, we didn't have reform legislation or unions to identify the most draining aspects of the setup and fight for limits.

Knowledge work was free to totalize our existence, colonizing as much of our time from evenings to weekends to vacations as we could bear, and leaving little recourse beyond burnout or demotion or quitting when it became too much.

Our estrangement from the rhythms of work that dominated the first 280,000 years of our species' existence was now complete. So the argument I'm making there is that we've sort of invented, when it comes to knowledge work, the most unnatural possible way of approaching work. It is as far as we have ever been as a species away from how we spend most of our time as a species. kind of had this steady march away from the way we were wired to exist for the Paleolithic.

Agriculture made it a little farther from that. The Industrial Revolution was even farther from that, and we had to bring in protection so that at least it didn't overwhelm us, and at least we recognized, hey, this is really unnatural, so you better pay us well and extra pay, and we're going to push back.

And then knowledge work got rid of all that and just left with the really unnatural pace. And everyone now pretended like this was good though. And this was somehow like what it meant to be productive. So seasonality...

pushes back against that it goes back towards our agricultural roots even farther back to our forager roots where it says we have busier periods and less busier periods a busy time of year and a less busy time of year when you better match the natural rhythms for which we're wired your life becomes more tolerable and sustainable, and over time you produce better quality work because when you're in this unnatural edifice of continuous invisible factory labor, you just burn out.

And then it just becomes activity for the sake of activity. And yeah, you're working all day long, but what does that work? Well, a lot of it's going to be moving emails back and forth, taking on those unnecessary meetings, because at least it's something that's not too taxing and looks like you have activity.

It's not a good way of producing results, and it's a great way of burning out the human psyche. So I am a big believer in having variation. I will say as an aside, by the way, Jesse, that whole... I won't name names, but that whole sequence in this book, which came out of a New Yorker piece that came out before the book,

starting with the hunting and gatherers, moving through agriculture into the Industrial Revolution, how we got alienated from our rhythms, quoting Marx and Marx's theory of estrangement to talk about what happened. when we finally separated so much from what we were wired for. A book that came out not too recently had that exact same arc, with the same example citing some of the same studies, ending up with Marx and the theory of estrangement or whatever.

I don't think the author was stealing it. I think his research assistant just like came across my new yorker piece at some point or like great here's some ideas they went into their system and this is kind of the problem of the modern way that a lot of these non-fiction writers write where they have a research assistant gather a bunch of stuff on a bunch of topics, and then the writer has a bunch of examples to pull from to make their points.

I think he just digested my New Yorker piece, broke it into its pieces, and then the author was like, oh, these are good research parts, and reconstructed my exact arc, basically, in his book. Did you read the book? Yeah. Have you reviewed it on your monthly box? You're trying to get me to identify who this is. I will not identify who this is.

But yes, it was the baseball book of why. It was a weird tangent that author went on. I don't know what Marxist theory of estrangement had to do with baseball trivia, but I'm on to them. All right. Seasonality. So how do we do this? Well, what I do with my summer schedule is one example, but that's pretty specific to me. I mean, it's like why I became a professor and writer was exactly to be able to do that. flexibility and schedule,

and a job like the writing that could sort of support me to take time off like that and give me like an activity to do. There's a lot of ideas you can do that are smaller that'll give you a taste of seasonality. I have a few to suggest here as you think about the summer ahead. One, consider having one day each week, maybe a Monday or Friday, where for the most part you don't schedule appointments.

Don't make a big deal about this. Don't announce it. Don't tell people when they ask you when you're available that you're not available on Mondays. have plenty of options to suggest. Like, yeah, I have this day, this day. How about here, here, here? Just don't happen to suggest days on this day that you have secretly marked for yourself as a no meeting day each week. It won't always work.

By many weeks, it will. And to have one day a week or two days a week, especially connected to a weekend, where you can get into your work and get caught up and see what's going on at a more of a leisurely entry pace and not have to jump into a thousand meetings really makes a big difference makes the season seem slower

If possible, try to make one of those days into more of an adventure day. Get out of your normal workspace if you can, especially if you are able to work remote and go into nature, go to a museum, go to the mall in D.C. It's what I used to do. make it feel different now still work work on your notebook bring your laptop with you you know still get things done but in a completely different location it just makes the the pace of work seem different um I would also suggest... I lost a page here.

Where did I put? Oh, I see what I did here, Jesse. I put my ads in the wrong place. Okay. I said no meeting days. Yes. All right. Another thing you should think about doing is pushing new project start times. Again, you don't need to make a big deal about this, but you can wind down your projects as the summer starts.

And as you take on new projects, be like, yeah, yeah, yeah, I'm finishing up some things now, but like late August, I'll get that ramped up. That's close enough in the future that people are like, yeah, fine, that's reasonable.

But what you've done is created a bit of a lacuna in your schedule here. There's going to be this period where you're not really starting or in the middle of any big projects. Now, you can't do this all the time because eventually the projects have to get done, so you can't keep pushing them into the future. But once a year, you can start saying,

for a little while when new projects are coming to you. Like, yeah, I'm finishing up a bunch of stuff now, but what if in mid-August I'll take this on? The reality is you're not really finishing up that much stuff, but it does buy you some space. Put your hand down more. And what I mean by that is when it comes to volunteer activities within your work, especially what researchers call non-promotable activities like organizing the internal office birthday parties or this or that.

Maybe you're that reliable go-to person that likes to help out and people like that about you. That's great, but put your hand down for a couple months. Just don't volunteer or say, I don't know, I'm a little busy right now. Again, if you do this all the time, then people might be like, oh man.

He or she looks never useful. It's kind of annoying. You do it for a month or so, for six weeks. No one notices then because like you're mainly volunteering a lot. They won't notice, but it makes these periods feel different. You just have less of this extra work on your plate. This is a strategy I talked about in my slow productivity book. The idea of like a highly autonomous blocker project. You take on a project for the summer that you can keep referring to.

to try to push away other type of work. Like, well, yeah, I can do that, but I'm working on the Maguire report, right? And that's where my head's down on this, but after I finish the Maguire report, I can work on these other types of things.

choose that proverbial Maguire report to be something that's highly autonomous and that you can, in reality, do like an hour every day and stay on top of it. It is a extremely effective strategy people don't know how long work takes most people are like pretty inefficient you take something where you have a lot of flexibility it doesn't generate a lot of meetings it doesn't make you have to be on other people's schedule be efficient and effective

and then use that thing to push everything else off your plate for a little while. You can only get away with that so long, but really we're just looking for a month or two of respite.

Start your day slowly. Apologize only if people notice. Start your day in the coffee shop. Get a little breakfast and coffee. Work there for a while. Just change the rhythm of the day so it just feels... slower it's not as frantic as other times of years like this is psychological but these type of psychological differences

really can add up finally this idea that people keep getting upset about this i'm not sure why just you know every other week go do something midday someday that you don't normally do during the weekday you go see a movie or something that's two and a half hours it's not some big deal people go to doctor's appointments and stuff all the time But it's going to give you a completely different feel for your week when you're in a movie theater at 2 o'clock.

and you know it's a weekday, it makes you feel like you're not just in the normal invisible factory clock and clock out. This is what I do every day. It just gives you a sense of autonomy and slowness and control. And in the summer, do this more often. Go see some dumb movies.

Especially if you're in a situation where you don't have to account for every minute of your time. Again, if you go to movies every day, if you stop work at 2 every day all the time, people will notice. But if once or twice a month you're gone in the afternoon, you're back a little bit late, no one knows, no one notices.

So you can use these smaller tips if you don't have a huge amount of autonomy to introduce some seasonality. Summer is a great time for seasonality. So hopefully a lot of you in the audience are going to join me. and slowing down for the months ahead. Do you ever need more than 30 minutes to do admin work? Probably, yeah. Sometimes it extends more. But I try to just have this mindset of when I get in there, especially for Georgetown stuff, there's not much in the summer.

But queries come up and a dean needs this or that. Just try to be really focused and work through. The opposite problem is what I'm trying to solve more often than not, which is

There's not a lot of stuff in my inbox right now, but try to do 30 full minutes of useful admin work even if there's not just messages to answer. Fill in the rest of the time to get ahead of things or to organize things or to... to preemptively like step out or prod things along so it goes both ways try not to work beyond that 30 minutes too far but also make sure you use the full 30 minutes this is not going to work for most jobs it wouldn't work for my job in september

But in July, because it's an academic institution that's technically shut down and I'm not on salary, it's much more plausible. I also have to deal with writer admin as well. But there, I'm just really good at... All the various teams in my life sort of know in the summer, like, yeah, Cal's not. That's not his. Super accessible time. I guess I thought you were also handling your personal admin.

Yeah, that's different. But that's fine. That's summertime. You have to mow your yard and work on your budgets. I've got to do my taxes next week. I'm doing my budget today. Yeah, but that stuff's fine. That's like you're at home. I don't mind that as much, and that does take a while. All right, well, we've got a bunch of good lister questions coming up. But first, I hear a quick word from some of our sponsors. I want to talk about my friends at Cozy Earth.

I love Cozy Earth. This is one of these products or companies where I use a lot of their products. I absolutely swear by them. In particular, Their sheets are a must-have in our family. We have multiple pairs of these sheets so that when one is being washed, we can switch them into the bed. We have other multiple pairs for the guest bedroom. We travel with these if we're going to be somewhere for more than a week or two. They're very comfortable sheets.

and they're cooling, and my wife and I really have a hard time now sleeping on almost anything else. We have other things as well. We have the duvet cover. We have the towels. My wife has the PJs. I have one of the sweatshirts, which somehow has a cooling effect, just like the sheets, which is really pretty cool. It kind of cools you down when you're hot.

I am a fan of Cozy Earth. I swear by them, and I think you will enjoy them as well. Their mission is to transform your five to nine, the time that matters most, into the most comfortable part of your day. Alright, and you're going to get that with Cozy Earth. Start with the sheet. But check out the other products as well. We gave these away, at least twice we've given these away as gifts. I think I gave some to my mom, and then Julie gave some. We like social... We like these things. We like...

We like cozy earth. All right. They have a hundred day sleep, a hundred night sleep trial rather. enjoy over three months to experience the unmatched comfort of cozy earth if you're not completely in love with your bedding you can return it hassle-free but guess what you're not going to do that they also have a decade-long warranty on their betting products. This means years of luxurious worry-free sleep knowing that your investment is protected.

Luxury shouldn't be out of reach. Visit CozyEarth.com and use my exclusive code DEEP for up to 40% off Cozy Earth's best-selling sheets, towels, pajamas, and more. Trust me, you won't regret it. That's CozyEarth.com, code DEEP. And if you get a post-purchase survey, tell them you heard about Cozy Earth from Deep Questions with Cal Newport. Sanctuary awaits.

At Cozy Earth. Also want to talk about our friends at Shopify. If you run a small business, you know there's nothing small about it. As a business owner, I get it. My business takes up a lot of my time and attention and I am always looking for places to avoid having to waste time dealing with extra products or things that don't work well. this is where shopify enters the scene if you're going to be selling things

You 100% need to be using Shopify. This is just a consistent theme about all of the entrepreneurs that Jesse and I know that sell various things, whether it be online or in the real world. Shopify is a consistent theme, how much they like Shopify. Shopify is the commerce platform behind millions of businesses around the world. And 10% of all e-commerce in the US

We're talking about household names like Mattel and Gymshark as well as to brands that are just getting started. Tackle all your important tasks in one place. Shopify, you could be doing inventory. You can be handling your payments, of course, but also analytics. It can help you with your marketing. It can help you with international sales. They have tools that can work in over 150 countries. You have a real store. They have award-winning point of sale.

And they can connect your online and offline sales all together. Use Shopify if you're selling things. I don't care where you're doing it. or what type of thing you're selling, online, digital, what country, think Shopify. So get all the big stuff for your small business right with Shopify. Sign up for your $1 per month trial and start selling today at Shopify.com slash deep. Go to Shopify.com slash deep. Shopify.com slash deep. All right, Jesse, let's move on to some questions.

First question is from Quez. I'm a teacher in Texas who usually spends summers on long backtracking trips which helps me recharge while walking, listening to good content, praying, and reflecting. Because of a knee injury, I can't do my usual trip this year and don't enjoy too much unstructured time at home. How would you suggest I structure my eight-week summer break and daily schedule to get similar benefits without being able to leave home or walk all day? I'm thinking maybe take up drinking.

Good old-fashioned Texas moonshine. Or whiskey, like an old-fashioned Texas cowboy. That'll get you through the eight weeks pretty well. Okay, so you can't walk all day, but I want to push on this a little more. You say you can't leave home. I don't think he means he can't literally leave his house. He means just like on longer trips, I guess, right? Yeah. Yeah. Because I was going to say a few things I was going to say, but my overarching advice here is have a schedule, have a plan.

Right? Don't go into your day saying like, ah, whatever. The day is kind of off. We'll just see how it unfolds. There's not much I can do because of my knee. It's that type of slow through the mud, like I'm just sort of wasting time, I'm on my phone for three hours this morning, I don't even know where that went, I'm going to watch some TV and see what's going on, like that type of...

stuck in the mud, mired in time mentality can really be relaxing for a day, but over eight weeks I think is really going to be a bummer. So you want to have some notion of a schedule or plan, whatever those details are. This is what I'm doing this summer, and this should probably involve some sort of project or projects you're working on, and these might be more intellectual, cognitive, and physical.

preferably things you can work on outside of your house. I mean, you say you can't leave your house, but you can leave your house. Go different places. Start your morning with the paper at the coffee shop and you're going to the woodshed because you're working on mastering some sort of new...

woodworking tool, you know. Yeah, your knee is hurt. You can't walk all day, but it's not going to be something where you can't walk on it. I mean, even people with knee replacements, they want you up and around. So I want you up and around.

working on some sort of projects that is not just in your house it could be cognitive it could be physical some mix of the two have some initiatives you're going on i'm going to read all these books i'm going to watch every movie that you know, every Ford movie between this 10 year period and like read books about or whatever it is, but have things to do that feels like they're rewarding.

That's what the long hikes were. So you want something that's similarly ambitious, but just not something that makes such a big demand on your knee. And there needs to be a plan here. I think we tell ourselves This was kind of my pushback against the sort of Ginny O'Dell hypothesis about doing nothing. We tell ourselves that's some sort of Rousseauian platonic ideal.

That doing things is an internalization of capitalist narratives and doing nothing is a brave resistance and that's where we're going to rediscover ourselves. But there's only so much of literally doing nothing that we can put up with before we get antsy. And then what happens is we have all these highly polished attention economy tools that are very much happy to solve our problem for us. Like, you're antsy about doing nothing. Look at this screen. Your eyeballs bleed.

We can give you something to do and then time just disappears. And you come out of it feeling depressed. I was just looking at a study. that was featured on NPR Club a couple months ago that was showing these connections between looking at algorithmic content and your mood. And as they, in an experiment, took people off of their phones, essentially, they found this. steady increase in their actual psychological well-being.

so it's you know doing nothing It's not, we think it is very appealing, but really what we don't like is being overloaded, burnt out, overworked. But we're not looking for the opposite of that. Or if we are, the opposite of overworked is not doing nothing, it's a reasonable amount of work. The opposite of

grueling efforts that feel meaningless and pointless is efforts that are meaningful and have a point. The opposite of these things that wear us out is not sitting there on your phone doing nothing all day it's doing less things and doing those things better more meaningful things so just keep that in mind quiz and then keep keep active get that knee healed maybe pt is a big part of the something you're doing this summer and then get back to those hikes as soon as you can

All right, here we have next. Next is from Martin. In episode 352, you answer the question, am I working hard enough to get tenure? I'm a postdoc myself in computer science, but not at an R1 university. A few faculty members here are at the top of their respective fields, but we do not learn the same toolbox and publish consistently as the top venues. How do I discover these secrets that the faculty don't even know my name?

I mean, honestly, Martin, if that's what you want to do, right? You want to publish at the top of your field and gun for top academic positions. You need to work for those people. So if there's a few of those people at your school, you want to find a way to actually collaborate with them. Now, you might not be able to change your postdoc advisor this far into game.

But if you can find a way to collaborate with them, hey, I'm making your life useful. I'm writing a paper with you and doing a lot of the work. That's what you have to do. I mean, it's an apprenticeship system. High in academia is an apprenticeship system. You have to find an apprenticeship. It is very hard otherwise to make your way into this sort of metaphorical guild. So see if there's a way you can write a paper with these top-of-the-line faculties. I mean, good for you for noticing.

Based on last week's answer, that's who you need to be allied with if you have this very narrow goal of publishing the top places and getting access to these type of jobs. So you've got to find a way to actually work with them. It's very difficult to learn these things from afar. All right, who do we got next? Next question is from Will.

Even if the short-form content topics are still relatively deep, do you still caution against short-form content creation and encourage long-form writing as a way to sustainably build an audience? Or do you think today's attention economy is too focused on social media to the point where the barrier to entry is too high for a new successful blog or sub stack?

So I was questioning a little bit his use of the word short and long. Maybe you have a better sense here, Jesse. When he's saying short form versus long form, is long form book? No, long form is a blog or a substack, and short form is like a YouTube short or something like that. Oh, I see. So he's saying, is it too hard?

to the long-form stuff that I talked about before, is it too hard now to start with that? Yeah, like to start your website and stuff like that. Okay, so it's always been hard. Here's what I think has happened. And I don't know, this is maybe a curmudgeonly way of looking at the media landscape. But if you look back 30 years ago, where what's dominating in the printed word media landscape is going to be books and the big glossy Condé Nast magazines that you find on the newsstand.

There wasn't a lot of people. who are like Hey, I kind of want to be in that game. You didn't have to sort of like the average person being like, yeah, I want to be one of the authors on the shelf there at Barnes & Noble or with my articles in Vanity Fair. You had a relatively narrow group of people who wanted to be professional writers, and that was a vocation. It was kind of competitive, but sort of cool if you could get there. But the average person was,

content to be like, yeah, I read that stuff. I think a lot of this changed with Web2 and then particularly after Web2 gave way to social media. where a big part of the pitch of phase one social media, and I'll tell you what phase one means in a second, was you can try on, you can cosplay producing content for an audience.

I'm convinced that this was the original sales pitch of phase one social media, was that Facebook, which was the first big player, they emerged in 2004, and they're generally public in 2006. You have to remember, at this point, Web 2, which had been around since the early 2000s, which was blogs, basically, was pretty brutal.

This technology came along and said, hey, anyone now can publish something. There's not a gatekeeper at a magazine or a newspaper or like a web developer you need to sort of put content. You can set up an account and publish something and it's on the web and anyone can read it. And what happened is a lot of people tried it and no one cared.

Because it's really hard to write things, as it always has been, that people care about. And so a lot of people were disillusioned by blogs. It was embarrassing. It was... whatever it was, demotivating. I wrote and no one came. Then Facebook came along and said, okay, here's our deal. You write on here, people will come. Here's the social contact. You're going to go through the contractor and go through and you're going to indicate all these other Facebook users who know you.

And we're going to tell them about stuff when you publish it. And yeah, the stuff doesn't matter, but there's going to be a social contract that they'll come over and they'll click like and poke and write comments and you'll do the same for them. And we can all sort of feel what it's like.

to be producing ideas and commentary and have an audience that cares about it. So we all can kind of cosplay in a lightweight way what it would feel like to actually be Michael Lewis writing these pieces for Vanity Fair. Or Malcolm Gladwell producing these books that are getting all these readers. And that was a big appeal of phase one social media.

Now that changed when we got the phase two. TikTok brought us into the phase two where we said, forget publishing your own stuff. What really matters is just like put it right into my veins, algorithmic distraction. And we kind of gave up.

caring about posting stuff and what people thought about us and was just like, put it right in my veins until my eyes bleed. That's phase two where we are now. I think that was the original pitch of phase one social media. And so it really... traded on this idea that like Yeah, not only is the ability to publish words accessible to a lot of people now, but like it's...

really like kind of right there you get a couple right breaks and some stuff goes viral and and and it just feels like you're right there you're posting stuff and people you know you have this built-in audience that'll look at it a little bit and anything could go viral at any time and you hear stories about it

And suddenly now this became something that it was like running triathlons or something that like a lot of people could do. And I think it changed our approach to things like professional content production to like, hey, maybe I should do this.

Which again, you don't hear a lot of people in 1985 just casually walk by and see someone reading The New Yorker and be like, hey, I should write for that. That might be fun. I'm going to start doing that. Because there's just less venues and they're really hard. This is being curmudgeonly, but I'm coming back to this and saying, well, what content are you producing? Do you want to be a professional writer? Then you should go down the path of becoming a professional writer.

don't be seduced into this idea that there's some sort of shortcut into that that requires a lot less effort and a lot less hard hoops to make it through where i'm posting short-term content and something takes off and and you know next thing you know i'm a writer on snl or whatever like yeah that happens but really not that often

So is the media attention economy too focused for things like a successful blog or sub stack or a podcast or books? No, I think it's like it's always been. It's very hard. It's no easier, no harder than it's always been. If you want people to read your articles or books or listen to your podcast, which is kind of like an audio version of long articles, it's like it was 30 years ago. It's hard. You have to have an extra expertise and a voice.

and find your audience and what you offer and have a couple swings that finally connect and then build on those. It's the same thing that professional writers have been doing all the time. I don't think it's any harder, any easier than it's been before. We just have this other thing over here, which is the cosplaying that you get in...

attention economy social media, where everyone feels like they're pretty good and they're one lucky break away from being the next Kim Kardashian. That, I think, has warped how we think about content creation. But I'm a big believer in writing professionally. I just put these reality checks on it that it's like very hard. Right? Digital technology did not make this something that is easier for like a huge number more people to succeed at.

It changed the forms in which you can do it. It changed the economic for the people who are good at it. Now, you have more control over how you reach and get to your audience, but it doesn't actually give you a big audience. that has still earned, I think, the ways that it always was. So I don't know. I feel like I'm going on a rant here, Jesse, but it's something I've been thinking about recently is that, again, 30, 40 years ago, it would be more rare that someone just be like, I want to...

be publishing stuff that people read it was like the people who really want to be a writer and now i think it feels more accessible in a way in part because social media like plays on this idea of like people are listening to you and like you're You're like one or two good quips away from like a retweet storm that's going to make you famous. And so I guess my answer is like the entry level is not too high. It's as high as it's always been, which is pretty high.

It's hard. I was having this conversation with someone recently. Like, well, how do you succeed with a podcast? Like, it's hard. Like, typically you have to have something else going that's already big. All these things. It's like Substack. It really helps if you're already well-known. You're already a journalist who's well-known, then you could maybe build a successful Substack. Do you read a lot of people with Substacks?

Um, some... Do you get the email notifications? I just looked at my inbox, yeah. Yeah, I don't use the app. Or sometimes it takes me over to the app. But I just look for my email inbox. I like that form. I mean, email newsletters I think is a great form. Podcasts is a great form. I like these forms because they're independent.

And you're not beholden to super large companies that are using your content. I mean, Substack does a little bit of this. But email, newsletters, and podcasts, I think it's like a golden age of content production. But the people succeeding in these... It's the same people that were succeeding in like the magazine industry in the 1990s. It's like kind of the same number of people and some of the actual very same people, you know, like. Gladwell and Lewis.

It's a podcast. So the forums are different, but I don't think, this is like the misnomer about media revolutions is they don't make it easier to be successful. They just sort of change the terms of success. That's kind of my growing theory. Now that we'll have our new weekly newsletter,

every week. It's going to be good. It's going to have a logo. It's going to have some good links. It's going to have some good links in it. We're going to have links every week. I might eventually bring in from an archive piece at the end or just sort of pull out some classic stuff from the decade or so I've been writing about this stuff. It'll be good.

Got to hone in on my topics, though. Oh, the video for the podcast. There is going to be, like, featuring that week's podcast episode will be right there. So, like, people who are email subscribing will always hear, like, oh, here's what's going on in the podcast this week. Should be good. Yeah. And then on the podcast, what I'm going to do is in a monotone, read the entire newsletter word to word every week.

just all the way through page break. I'll read the punctuation as well, because how else is I going to know? Semicolon space. Just be like that. And that's really what the podcast is going to become. Instead of me, I'm going to have 1980s-era Macintosh. Remember how they could talk with that robot voice? Just reading out my newsletter. I think it'll be just as successful. With an occasional throwing of your French accent with the...

Yeah, and then I'll come in and do a bad French accent. Yeah, I think we've got this all clued in. I think we are about to explode. This world is so hard, by the way. It's so hard to grow podcasts and newsletters speaking of which. I mean, that would grow it, but... What a hard world. All right, what else do we have here? We have Mark. Mark. I'm an actuary with a job that leads to adversarial interactions with the regulators who are often challenging our assumptions and rejecting proposals.

Since filings are infrequent and mistakes are highly visible, it's tough to design a deliberate practice routine to reduce pushback. Mark, you're actually in a situation that's really ripe for deliberate practice because you're getting super clear feedback We are really mad because of acts.

That's great feedback on like, okay, how do I never do X again? So you probably are in a cycle of deliberate practice here that is making you better it's not fast in the sense of you're doing 20 reps a day but like you know each year you're getting probably a dozen or two great pushbacks right so great teaching signals about what's working and what's not working and the rate at which you are improving at writing these types of reports that you're filing is probably

impressive if you're looking at the right time scale. I will throw in, however, if this job is stressing you out, and it sounds like it might be, I mean, it was stressing me out. You're just constantly being yelled at by people who, by default, are mad about what you're going to send them. You might think about getting another job. The key, however, is if doing so, take your career capital with you. Don't quit this job to become a yoga instructor.

quit this job to become an actuary somewhere else, but in a non-adversarial environment, right? So if there's something that's emotionally draining about a position, See if you can pick up your skills and take them somewhere else that removes that thing that is emotionally draining. Because I'm suspecting that might actually be the real problem here. You know, the sparring. I wouldn't like that either. All right, here we go next. Next up is Kayla.

As someone who excelled in humanities and liberal arts, I appreciated your suggestions to read books with opposing views to grapple with different theories and perspectives. However, I'm curious if you have any recommendations for how to improve. quantitative and technical skills as a way to get smarter in a dumb world. I would focus on specific projects that require you to master a new technical or quantitative skill to make progress.

The two types of projects that you'll actually return to is either one that has a professional salience, so you've volunteered to do this thing, and your organization's like, yeah, that's good, do this thing, and now they're waiting for an answer, and to get them an answer, you have to pick up the new skills. That'll focus your mind. Or something that's really fun.

I think it'll be really fun to build this thing that's going to require me to learn circuit design or going to have to learn computer programming. Maybe I'm doing it with my kids. But have a project that you either are interested in for professional financial reasons or for fun, sort of like family connection reasons.

And let that drive you to have to pick up new quantitative skills. Don't just sit with a textbook and say, all right, guys, it's coming to Torx Day, and just sit there and try to memorize theorems. That's not going to be very fun. but having a project is going to make that easier. I like, by the way, that you brought up my suggestion from before because I always like to underscore that. Read two books that are very smart and have the opposite view on something.

Even independent of the content, the dialectical clash of two opposing opinions on something adds just a new layer of nuance to your mind. It just makes you smarter. It makes you see the world in more interesting shades of gray and not black and white.

It's a really interesting experience. Even if you don't care about the topic, seeing that clash just adds like a more sophisticated layer to your brain. And if it's a topic you really care about, then you extra sure want to read the opposing view as well.

Because that's where you're really going to figure out why you care about it. It's going to strengthen your conviction. It's going to make you a better advocate for that. Do not try to avoid stuff that you think is going to be contradicting stuff that you care about because that's not a belief. If it's I'm avoiding contradictory information, that's not a belief acquired in good faith. It's an idol that you're worshiping. You like the idea of being a part of that cult.

So it's great to challenge ideas. It's going to make them stronger. It's also great just to make you a smarter thinker to see what it looks like when you read two very opposing views. All right. Oh, we've got a case study this week where people send in their accounts of applying the type of things we talk about on this show in their own life. If you have a case study, you can send it to jesse at calnewport.com. Do we have a name for this week? I guess not.

We'll just call this Peter. Yeah. Peter said, at the end of 2022, my life reached a turning point. I had just become a father for the first time, released my own music after years supporting other artists, and was deeply involved in serving my local Baha'i community. On paper, I had achieved many of my life goals, but something still felt off.

I realized after talking with my wife that I was constantly distracted thinking about music and work while at home and feeling guilty for not being fully present with my family or in my community service. Social media, I suspected, was a big part of the problem. It'd be funny here if the case study said, so I got rid of my family and now have much more time for my social media and I've never looked back. His name is Sorrel. Sorrel, okay.

Social media, I suspected, was a big part of the problem. Discovering your TED Talk and then diving into your books, especially deep work, was a revelation. It helped me see that my habits needed to change if I wanted to be truly present and effective in all areas of my life. Over the past two years, I've worked hard to put your ideas into practice. I've become much more intentional about how I organize my days, treating my music career like a knowledge workshop with clear boundaries.

Deep work sessions are now a priority, allowing me to focus on improving my core musical skills, which are composing, producing, singing and playing instruments. I'm also more disciplined with my time. I keep social media off my phone.

stick to fixed work hours, and reserve evenings and weekends for my family. This structure has actually made me more flexible, letting me step away from work when needed for family or community projects. I've also picked up new habits like boxing and reading more books, which have brought a steady sense of well.

What I really like about this case study is this sometimes paradoxical sounding idea that having some more structure and intentionality and control of your time makes your life slower and more full and more meaningful.

There is sometimes this idea that's out there that once you begin to think more intentionally about your time, that necessarily means what you're doing is internalizing capitalist narratives and you're building your life into like a stressful scramble of trying to produce, produce, produce. But that's not always what's going on. An unstructured life can be busy and exhausting. An unstructured life can be intentional and more relaxed. And I think that's what we see in Surreal's story here.

I'm going to go back and show you a couple things that I think are important here. His job was flexible, right? As a music production, he's not at an office. working for a boss adding clear boundaries made it much better here's when i work and here's when i'm done without that it just bled over into the rest of his life and was causing real distress and problems.

He makes deep work a priority. There's lots of stuff he could be doing that's probably more fun, and maybe I should be working on the social media pages of the bands that I'm repping and try to make sure that they're getting attention. But he says, no, on a regular basis, I want to deep work on getting better at music.

Long term is probably making him much more successful. Short term is giving him a lot more psychological satisfaction out of his job because it continually connects him back to the core artistic element of what he's doing and why he values it. He keeps social media off his phone.

and weekends and evenings are just for family. That really helps. I'm not working. I'm not on my phone. You're going to connect to your family much more. So these seem like simple ideas, but these little rules and structures made a really big difference in Surreal's life. And so I think it is a fantastic case study. All right, let's do a call. Yep. Hey, Alex Sorto here. I'm a commercial real estate appraiser, and my question has to do with the topic of side hustles.

So, you know, there's a lot of random and conflicting advice on the Internet about side hustles. And maybe some of it is unrealistic and some of it is not. I would love it if I could create and develop a side hustle this year to keep me up float during down times. And I could be something that I actually particularly enjoy, like fashion. or something along that line where it could be a passion as well.

So I just wanted to see what were some of your ideas or best practices for creating a side hustle and implementing it into your regular schedule. I mean like Alex. referred to me as Al. Yeah, I think he knows your name's Cal. You think it's because his name, he was queued up to say Alex. Yeah. And then, yeah. He definitely knows your name. Cal. Yeah. All right. I was going to say Alex. I'll see you in hell, buddy.

I'm going to assume that was a runtime error. I was somewhere recently where I was continually introduced as Kyle. So, would you rather be introduced by Al as Al? The weird thing is, my wife was misintroducing me. I think she just forgot. Side hustles. All right. Three things to keep in mind. One, I typically would say be wary of a side hustle where ultimately what you'll be doing is trading your hours for time. So basically, wage labor side hustles.

They don't scale well. You're typically getting less. money for your time than you could get just putting those hours into your primary job. And it's not what you're really looking for with a side hustle, which is like a flexible and stable and potentially scalable source of side income. So it's tempting because they're easier to get into. It's easier to get people right off the bat to pay you for your time. But if you're effectively just, if I spend five hours

versus 10, I'm going to get half as much money, is not where you want to be in a side hustle. That's not a good side hustle. Two, I would say, care about your existing career capital. The more that you're leveraging rare and valuable skills that you've earned, the more value you can get and the more likely it is you can have a really effective side hustle.

So if you like something but you don't have demonstrably rare and valuable skills in that area, the possibility that you are going to get sort of scalable, time-independent income from that direction is very low. That type of low-hanging fruit just gets... washed out and efficient market. If there's some relatively easy way that you can make a lot of money on some topic you're interested in and don't know much about,

Anyone who doesn't know much about it could make money on that topic, and they would start trying it, and then that would drive down the price, and that market would go away. So the inefficiencies that produce good side hustles usually take advantage of skill. The more rare and valuable skill that you're deploying,

the better chance you have of finding a real inefficiency and getting a really actually attractive side hustle. So don't ignore your career capital. You mentioned fashion. You would really need some sort of career capital in fashion before you think that's going to necessarily be a good side hustle.

Finally, there's an idea from Derek Sivers that I featured in my book, So Good They Can't Ignore You. When evaluating your current side hustle and whether you want, if it's successful or not successful, or whether you want it to become more of a full-time thing, I'm quoting them here, use money as a neutral indicator of value. Are people paying me for this?

People will tell you your idea is good. People will tell you your idea is interesting. People will, for free, like, yeah, I would love for you to give me a tarot card reading or whatever, but they don't like to give away their money. And people only give away their money if they feel like they're actually getting a proportional amount of value for that money. So Derek Sivers had this idea with the side hustles that he...

transformed into full-time roles on multiple occasions throughout his interesting career is that he would use money as a neutral indicator of value. If this is generating money, it's valuable. If it's not, it's not. Should I make this my full-time job? Well, here's the question. Is it generating enough money to be my full-time job?

That seems like it's really simple, but people often ignore that. They want to sort of assign other sources of value to their idea. Like, well, it's really cool. I know someone else who did really well with this. My friends tell me it's awesome. That's not really a reliable indicator of value. Get people to give you money. If they're not, the idea is not that successful. The hustle is not going to be that successful.

All right, Calyx, I hope you enjoyed that advice. Jesse, I sort of reversed it back on him there. I don't know. See you to his name. All right. Do we have another call? Yeah, we do. All right. Hi Cal, I enjoyed your recent podcast looking at how to become a straight A student as I am a college student and that was my initial introduction to you. My question is, you talked for a second that if you were going to rewrite that, that you would...

have a section on focus. And I know that you said that you'd probably never rewrite it, but I was curious what principles you would go over in a section like that. That's something that I really struggle with and I'm trying to figure out how to implement that as a college student.

So my basic framework about focus is in the book and when it changed. There's a core idea in that book, How to Become a Straight-A Student from 2006. It's an equation that says work accomplished equals Total hours spent times intensity of focus. And my point there, like why I made that argument was, One of the easiest ways or most effective ways to get more work done

is to not just increase the hours piece of that equation, but to increase the focused piece of that equation. They're multiplied by each other. Double your focus, you can double your output in the same number of hours. And back then, I was saying what I saw is too many people would study in a low intensity of focus, so sort of at the library with their friends and reading instead of doing more difficult, like passive consumption instead of active recall.

and they would just try to increase the hours component to get the amount of work that needed to be accomplished done. But that's a hard game. You've got to spend long amounts of time doing that work.

And I would say, no, no, increase that intensity of focus, come at this thing with laser-like focus, and you can drastically reduce the hours needed to get the work done. And that's actually more sustainable in the long run because you're not spending all your time studying. That central equation still stands. What's different is the obstacles to intensity of focus today versus 20 years ago.

You have way worse obstacles, in particular, in your hand, in that phone that's next to you while you study. That is a much bigger obstacle than we had to deal with in 2006. I was worried about conversations with people at the same table. I was worried about you just using low intensity techniques. So using study techniques that weren't very mentally taxing. Therefore, they didn't require much focus, but required a lot more time. And walking over to public...

email terminals to check your messages. That's the type of stuff I'm worried about. You have obstacles to boosting that intensity of focus now. that are way more accessible and appealing and disastrous than what we were facing back in 2026. But I would have one rule. one extra rule that I think translates most of the value of that equation to today.

Never have your phone with you when you study. That would be the rule I would add. And don't look at this like this is crazy. You're not an ER surgical attending. You're not the Pope. People will be okay if they can't get in touch with you for 90 minutes. It's okay. You're not that important. Don't bring your phone with you to study. Go to a library to study. Go to a non-popular library to study. Don't have your phone with you.

Now you just don't have access to those distractions. Now you can just apply the advice as written in my 2006 book. Now still you have to use high-intensity study tactics that are going to get you to focus more. Use active recall instead of passive recall. Go to places where you don't have a lot of distractions. All the stuff I talk about now applies equally as well in 2026 as it was in 2006.

as long as you don't have that phone with you. So it would be like my number one advice. You would want to put a giant sign over every library or public study space that says, abandon all ye phones ye who enter here. A little Dante reference right there. And that puts you back on a level playing field, and then you actually have a chance of succeeding with focused, high-intensity studying. Let me tell you the effect of this. It will feel like a superpower.

You will, and I'm not exaggerating, study a third of the time of the other people in your class and you'll get good grades. If you study without your phone and you use smart study techniques like the type I talk about in my book that actually work, they're hard but actually work and not what's easier, and you boost up that intensity of focus part of the equation,

And you do the time management stuff we talked about a couple weeks ago on the podcast that's in that book. You're going to get really good grades. You're not going to work late into the night. You're not going to do all-nighters. You're going to have a lot of free time. So that's it. That would be the number one thing I would add. You just don't have screen time rules. Don't get cute about it. It is just not allowed to come with me when I go to study. That'll make all the difference.

For the newer students in the audience, can you, in a couple sentences, explain to them what Active Recall is? The only way to learn new information that actually works with any sort of effectiveness is trying to produce the information out loud from your own brain as if lecturing a class without looking at notes. That's how you learn things.

The opposite of that is passive recall, which is where you read your highlighted notes. That is an incredibly inefficient way of learning stuff. I'm going to give you an analogy. Let's make the analogy trying to get muscles larger. Passive recall!

is I'm going to hold a shake weight. Have you seen a shake weight, Jesse? They used to sell these in infomercials. It's like a dumbbell where the two ends go, and they shake back and forth, like powered by batteries or whatever. It's shaking your muscles. It's going to kind of make it stronger. I guess you're like kind of holding it up or something. Active recall is like I'm doing curl.

With heavy weights and it kind of sucks because it's burning, but I'm doing them on a really good schedule. Your muscles are actually going to get easier. It's harder, but your muscles are going to get bigger. You can stand around with a shake weight as much as you want. But maybe get a little bit stronger. So there you go. Active recall is the only thing that matters. It sucks in the sense that it's really hard.

But you have to think about hardness as like, it's not literally this, but like the feeling of your neurons reforming. You're forcing your brain into a new configuration where that information is accessible. Lecturing out loud is if lecturing a class. I preach this to my students like five or six times a semester.

I even wrote, like, I'm teaching discrete math. Sorry, I was teaching discrete math to 100 kids this semester. Is that where all those exams are? Yeah, do you see those? I looked at the first question. Yeah, not so bad. Yeah, I don't know. Discrete probability. So I have a, this feels like a FERPA violation, but I was grading this weekend, and so I have all my exams in the HQ, because I was grading in the HQ the other day.

But I always tell them, I wrote an article about this when I was in grad school because I did Well, I got the highest grade I've talked about before, but that's when I realized I might have some math ability when I took large discrete math classes on the ground and got the highest grade in the class. And so I wrote an article about here's how I studied in that discrete math class. And I'm always telling my students, I wrote an article, do this. Do this.

Maybe they do. They do. That's a pretty good grade. That wasn't a hard exam, I don't think. It was a very reasonable exam. So there you go. Active recall. All right. What do we got here? We got a final segment coming up. I'm going to react to something on the internet you should know about. But first, let's briefly hear from another sponsor.

One of the longest sponsors of this show, and for good reason, is our friends at Grammarly. That's because from emails to reports and project proposals, it's more challenging than ever to meet the demands of today's competing priorities without some help. Grammarly is the essential AI communication assistant that boosts productivity so you can get more of what you need to get done faster no matter what it is you're writing.

or what tool it is that you're doing the writing in. I have watched Grammarly over the years embrace AI enabled features that have continued to grow its utility in producing professional and effective communication in your job it'll produce drafts if you want like hey I gotta do a quick draft of this it can take a bunch of notes you paste in there and it can put them into a better summary it can look at your tone

Hey, what is the tone of this email I sent? Why don't you double check that for me? And maybe I want to change it to like a more friendly tone or more professional tone. Maybe this is, you know, you worry about how you're coming across. It's there. It can help you with that.

Maybe you're trying to come up with some ideas. Or, of course, the stuff that Grammarly has always done really well, just check to make sure that your grammar is good and this is coming across as professional. I think of having Grammarly like having a... editor who looks over your shoulder with a bunch of ideas and to help you make your writing even better. You could hire someone to be like, let me help you with this email before you send it. Just think about how those advantages aggregate.

When you have like help with clarity and effectiveness in your communication and all the communication you're doing in your job, that's going to add up to you just being seen as like a much more. capable and effective employees. So it's a fantastic, I think, focused deployment of AI, not some grandiose, you know, this is going to take over the world. Like this is a specific thing that people need to do better in their jobs and we're going to help you do it.

you can write and edit quickly 93 percent of professionals report that grammarly helps them get more work done You deserve to be equipped with the best tools to do your job with the right AI support, and that is Grammarly. So let Grammarly take the busy work off your plate so you can focus on high-impact work Download Grammarly for free at grammarly.com slash podcast. That's grammarly.com slash podcast. You think I would know that word by now?

podcast i also want to talk about speaking of technology our friends at vanta Trust isn't just earned, it's demanded. Whether you're a startup founder navigating your first audit or a seasoned security professional scaling your GRC program, proving your commitment to security has never been more critical or more complex. That's where Vanta comes in. Businesses use Vanta to establish trust by automating compliance needs over 35 frameworks like SOC 2 and ISO 27001.

Centralized security workflows, complete questionnaires up to five times faster, and proactively manage vendor risk. Avanta can help you start or scale your security program by connecting you with auditors and experts to conduct your audit and set up your security program quickly.

Plus, with automation and AI throughout the platform, Vanta gives you time back so you can focus on building your company. Join over 9,000 global companies like Alassian, Quora, and Factory who use Vanta to manage risk and prove security in real time. Here's the good news. My listeners can get $1,000 off Vanta at vantadec.com slash deepquestions. That's V-A-N-T-A dot com slash deepquestions for $1,000 off. Alright, Jesse, let's do our final segment.

I want to talk briefly about an article that's important. You might not realize how important it is I'll bring this on the screen here for people who are watching instead of just listening. It's from the Wall Street Journal. It came out on May 15th. Here's the title. Meta is delaying the rollout of its flagship AI model. The company struggled to improve the capabilities of the latest AI model.

mirrors issues at some top AI companies. So what's going on here is that Llama4, which they're nicknaming Behemoth, had been a long... Long-awaited next big large language model from Meta, they were greatly increasing the amount of compute and data. But when they finish, They realized this thing wasn't that much better. It wasn't better enough versus their last model to justify it yet.

releasing it there's an important quote in here i'm looking through the article now where other companies have been having similar problems all right let me read from the article Meta's recent challenges mere stumbles or delays at other top AI companies that are trying to release their next big state-of-the-art models. Some researchers see the pattern as evidence that future advances in AI models could come at a far slower pace than in the past and at tremendous cost.

Right now, the progress is quite small across all labs, all the models, said Ravi Schwartz-Ziv, an assistant professor and faculty fellow at New York University's Center for Data Science. GPT-5, one of OpenAI's next big technological leaps forwards, was initially expected around mid-2024, the Wall Street Journal previously reported. In December, the journal reported that development of the model was running behind schedule.

All right, and they give some more examples. All right, we can take this off the screen now, Jesse. This is pointing towards a very, very important trend in AI, but the details are a little murky, and I think most people don't know it, and I want to make it as clear as possible. Here's the best way I can explain this. There was in the aftermath of OpenAI starting to make their moves with their GPT family of models, so in particular GPT-2 and 3.

There was this idea, they're called the scaling laws. There is this idea that as you continue to increase the amount of computation and the amount of data that you use to train these language models, their skills would keep massively jumping forward. And in fact, the chart that showed this relationship between resources and ability was exponential. And for a while, it was fitting that.

GPT-3 was like this huge improvement over GPT-2. GPT-4 was a huge improvement over GPT-3. It was better at like everything. It was doing it better. The visions you heard about. a year or more ago about artificial general intelligence, for example, about all work being essentially automated. was people looking forward on this curve.

and saying, this was so much better than that model, and this model was so much better than this model. As we continue to build out the compute and get more and more data, eventually these models will just become so capable that we can just build software agents around them to actually execute the stuff they say we can just say like hey here's a job tell me how you're going to do it and have a software agent actually execute it and basically we can automate everything it will just

be able to understand so much about the world that we can just ask it, describe how you do this or that, and it'll be a general intelligence that can basically do anything that a human can do. This is going to completely change the world. This was the scaling vision was the vision. for what was going to happen. The reality, however, is that the

benefits of scaling are starting to give massively diminished returns. It turns out that this exponential doesn't keep going. Hundreds of billions have been invested in building out compute and in building out the... biggest possible data sets to train these things. But after roughly that GPT-4 point, the improvements begin to become diminishing. That's why GPT-5 hasn't come out yet. It's not good enough. It's not that much better.

That's why Llama 4 Behemoth Project, they haven't released it yet. It's not that much better. And in the cases where they have put this in, you look at like Grok, I guess it's 3 now versus Grok 2, massively more compute. It's a little bit better. So the scaling laws are kind of trailing off.

So we're not going to. There's a dream of like if we keep building bigger and bigger data centers, we don't have to do anything else but get bigger and we're going to have like AGI. That's not happening. I don't think everyone's caught up to that yet. There's still so many of these sort of non-technical articles about like, hey, what are we going to do in the future?

we're going to have no jobs, is they're still looking up this curve. Well, look at how much better this model was than that model. If we just do that five more times, it's the danger of exponentials, the seduction of exponentials.

We do that type of doubling in ability five more times and like all work will be done this way. But that's starting to level out. So where is all the energy shifted? Because they still need to make improvements, right? You've noticed OpenAI has now released all these models that have weird names.

It's GPT-40 and GT.403 and all these different types of weird names. Well, they're having to put their energy into much more narrow development. So a lot of what's happening now is they'll take what they call a foundational model, which will just be an existing model like GPT-4. And then what they're going to do is they get a particular data set about a particular type of task they care about where they can have a lot of examples of questions and right answers.

So computer programming is a fantastic example of this is because you can have a question saying write a computer program that does this and then you can compile the answer and test it and see if it actually does it. Math is another area where they could do this because you can write math problems and have the right answer.

And so they take an already trained foundational model like GPT-4 and they do a tuning step with these type of synthetic data sets with right answers where they ask it these questions and if they... give the wrong answer they use reinforcement learning to kind of move its weights away from that and if it gives the right answer then they use reinforcement learning to make its weights more solidified towards that and they can tune

the base language model to be better at these specific tasks for which they have big synthetic datasets with examples of right answers. That's what they've been doing since the scaling's not working as well. so they're like look this model is good at computer programming why they built a huge synthetic dataset of computer programming examples.

Over here, we have a program that's getting really good at math. Why? Well, because we're hiring a bunch of math PhDs and paying them $100 an hour to write all these problems where here's a problem, here's the answer. We're using that, the TUN foundational language model. The reasoning quote-unquote advances that came recently. was they worked through, built out these big synthetic datasets of

questions that require reasoning, and then step-by-step answers. And then they could use that with reinforcement learning to train certain models to be better at doing something that looked like that type of step-by-step reasoning. So that's what's really going on now.

We're not just getting this like massive general increases in overall capability by just scaling compute and data. So instead, the companies are saying we're looking for these like specific applications where we can have these really good synthetic data sets. and then we'll build bespoke models that are better at that particular thing.

and maybe they get worse at other things. Like the latest reasoning-tuned engine of OpenAI better matched the step-by-step reasoning. They trained it with reinforcement learning, but it hallucinates a lot more. Whoops. Because in trying to actually match that, it helps to make things up more because you can better get a logical seeming chain of answer.

So the wrong thing to do now is that this is not now a general extrapolation. You can't say, like, now it's good at math, and now it's good at reasoning, and now it's good at computer programming, so tomorrow it's going to be good at X. It's like, no, no, no, these aren't additive or aggregate. GPT-4 is kind of like their state-of-the-art base model if you're looking at OpenAI.

and they're looking for specific applications where you can get good data sets for reinforcement learning on particular tasks. Now, this is producing and will produce cool bespoke tools. It's going to be a long product development cycle. You've got to figure out places where you can build these type of tuned models and then build good software agents around them and integrate them into the workflows where people use them.

but it's a far cry from the vision that we people had two years ago like no no just GPT-6 will be Anything you ask, it'll do at a human level. In fact, it can write its own code to make agents around it, and we won't even have to work anymore. That is not the trajectory we're on at the moment anymore. Without the scaling laws, we have to sort of hand-build bespoke models that can do...

specific tasks where we happen to have data for it. There's many things that we do in our lives where there's not going to be good data sets for, but where there are, I think they're going to eventually have really cool tools. This is a cool technology, but it is not. On one of these, I have to write an article about the end of business type of trajectories.

So anyways, this is a really important thing that's going on. This is being discussed a lot among the high executives in these companies, the death of the scaling laws. trying to build out these bespoke skills. I mean, so I'll just say real briefly, so why then did we have like the AI 2027 report come out last month saying now by 2027,

These AIs are going to be in charge of everything, and by 2030, humanity might be dead. Well, I look closer at what was really going on there, because they know... The scaling laws aren't working, right? So what's their tech story for how AI is going to kill humanity in the next like three years, right? If you look closer, it all comes down to a single assumption, which I think is specious.

The tuning we're doing for producing computer programs is getting better, which is true. We have good data sets for that, and computer programs are text, and we can produce of reasonable size good computer code. The foundational assumption of AI 2027 is well, we're better at computer programming with models now than we were before, so maybe if we keep getting better at that in the next year or so,

The models will just be able to build their own. They'll solve the problem of how to build a more technically advanced AI. Yeah, we can't just scale language models. We don't know the tech story for what type of models would be super powerful or super intelligent. But maybe if we just make them better at programming, they'll figure it out in the year 2026.

and then we'll be in trouble. That's a crazy assumption. Anyone who's actually working with the tuning of language models to produce computer code to think like, yeah, that's going to lead to giant breakthroughs and architectures for artificial intelligence. It's a crazy claim.

It's like saying the Wright brothers being like, we just found the new type of propeller and now we can get powered flight now. It used to just be for a little bit on Kitty Hawk Beach. Now we can actually go up and circle the field and come down and do like 10 minute flights. So I don't think we're that far away from interstellar travel. It's different. I mean, they're both flying, but it was completely different levels of complexity and challenges. So anyways, this is an important story.

I'm increasingly believing there's this weird temperature raising on AI that is becoming more disassociated from the tech story. And so not to talk about AI too much, but this, I think, is an important story because understanding that the scaling laws seem to be faltering for language models puts us into a much more traditional business innovation cycle landscape for AI tools right now. Alright, there we go. I'm writing a master. Actually, it might be out by the time this comes out.

On my newsletter, I'm writing a big AI and work article where I'm just dumping a lot of this stuff I've learned in the one article that I can point people towards. So check that out at calnewport.com. It should be live when you hear this podcast episode. Nice. I've just been deep on this in a while. I don't want to keep talking about it all the time. Let me just put down, like, hey, here's what I think is going on. And it's an article, at least in the draft it is now.

here's the places where like ai is really affecting the world of work here's the places where i think it will soon it hasn't been developed yet but will soon here's some places where it's like a little more hazy like how exactly is it going to play out but there could be some cool stuff and here's the stuff that's um crazy town I try to cite a lot of things and make the tech story a little bit clearer.

Look at that article. CalNewport.com. It should be up. All right. Speaking of which, I should probably go work some more on that. So let's call this episode to a close. Thank you all for listening. Back next week with another episode. And until then, as always. Hi, it's Cal here. One more thing before you go. If you like the Deep Questions podcast, you will love my email newsletter, which you can sign up for at calnewport.com.

Each week I send out a new essay about the theory or practice of living. deeply I've been writing this newsletter since 2007 and over 70,000 subscribers get it sent to their inboxes each week So if you are serious about resisting the forces of distraction and shallowness that afflict our world, you've got to sign up for my newsletter at calnewport.com and get some deep wisdom delivered to your inbox. Each week.